Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Social Sciences 77

Doing Things in Relation to Machines Studies on Online Interactivity ANDERS OLOF LARSSON

ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2012

ISSN 1652-9030 ISBN 978-91-554-8328-9 urn:nbn:se:uu:diva-171757

Dissertation presented at Uppsala University to be publicly examined in Ekonomikum, Hörsal 2, Kyrkogårdsgatan 10, Uppsala, Wednesday, May 16, 2012 at 09:15 for the degree of Doctor of Philosophy. The examination will be conducted in English. Abstract Larsson, A. O. 2012. Doing Things in Relation to Machines: Studies on Online Interactivity. Acta Universitatis Upsaliensis. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Social Sciences 77. 97 pp. Uppsala. ISBN 978-91-554-8328-9. The Internet is often discussed in conjunction with various notions of interactivity. Recently, conceptualizations of a “Web 2.0″, mainly focusing on harnessing user-generated content, have grown increasingly common in both public discourse and among researchers interested in the continued growth and transformation of the Internet. This thesis approaches the use and non-use of online interactive features by societal institutions. Specifically, the thesis focuses on online newspapers and online political actors, studying the practitioners working within those institutions and on their respective audiences. Consisting of four empirical studies, the thesis is informed theoretically by the application of conceptual tools pertaining to structuration theory. In Anthony Giddens’ original conception, structuration theory posits that social structure is recursively shaped (and possibly altered) as human agents choose to re-enact certain modalities of specific structures. By changing their uses of the rules and resources made available to them by structure, humans are given agency in relation to overarching, macro-level structures. Giddens’ writings have also been contextualized to the study of information technology use by Wanda Orlikowski, who has mostly focused on organizational research. Combining insights from Giddens and Orlikowski, the thesis suggests that most Internet users are enacting a “structure of audiencehood”, entailing somewhat traditional consumer behavior, rather than a “structure of prosumerism”, which would entail extensive uses of the interactive features made available online. Similar traditional use patterns are discerned for practitioners. The thesis suggests that we should not be surprised at relatively low levels of use of interactive features by practitioners and audiences in these contexts. While the chosen areas of study are often surrounded by expectations and “hype” regarding the consequences of online interactivity, institutionalized news and politics can be said to represent stable structures – structures that have functioned in similar ways for extended periods of time, and, thus, are not so easily amended. Anders Olof Larsson, Uppsala University, Department of Informatics and Media, Kyrkogårdsg. 10, SE-751 20 Uppsala, Sweden. © Anders Olof Larsson 2012 ISSN 1652-9030 ISBN 978-91-554-8328-9 urn:nbn:se:uu:diva-171757 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-171757)

’Without music to decorate it, time is just a bunch of boring production deadlines by which bills must be paid’ - Frank Zappa

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals. I

II

III IV

Larsson, A.O. (2012). Interactivity on Swedish newspaper web sites – What kind, how much, and why? Accepted for publication in Convergence: The International Journal of Research into New Media Technologies. Published online before print. Larsson, A.O. (2011). Interactive to me – Interactive to you? A study of use and appreciation of interactivity on Swedish newspaper web sites. New Media & Society, 13(7). Larsson, A.O. (2011). “Extended infomercials” or “Politics 2.0”? A study of Swedish political party web sites before, during and after the 2010 election. First Monday, 16(4). Larsson, A.O. and Moe, H. (2011). Studying political microblogging: Twitter users in the 2010 Swedish election campaign. Accepted for publication in New Media & Society. Published online before print.

Reprints were made with permission from the respective publishers. Additionally, parts of the cover paper have been refashioned into the following journal articles: Larsson, A.O. (submitted). ‘Rejected bits of program code’ – Why notions of “politics 2.0” remain (mostly) unfulfilled. Accepted with minor revision for publication in Journal of Information Technology & Politics. Larsson, A.O. (submitted). Understanding non-use of interactivity in online newspapers – Insights from Structuration theory. Resubmitted with minor revisions to The Information Society.

Acknowledgements

It has been said before, many times - writing a PhD thesis is indeed solitary work. For the most part, anyway. Luckily, the life of a PhD candidate also includes interactions with plenty of other people – colleagues, supervisors, co-authors, students, friends and family. It all starts at home. Mom and dad – you’ve provided me with the best possible start in life, as well as with continuous support throughout my studies. The genetic and social combination of a civil servant mother and a journalist father has proven successful with regard to my chosen line of work, and your loving care at “Hotel Hemma” has helped me rejuvenate and relax. As for life at Ekonomikum, Göran Svensson deserves special mentioning and great thanks for initially believing in my potential as a teacher and as a researcher, for helping me out at the very beginning and for countless music experiences. Later, my trio of supervisors has been helpful, albeit in different ways. Else Nygren has provided important suggestions and hints with regard to everything from data analysis to general writing skills. Stefan Hrastinski proved especially helpful at the very beginning of my PhD work, being very generous with his time and a supportive co-author (although our paper did not make it into the thesis after all). My main supervisor, Pär J. Ågerfalk, deserves a special mentioning. Pär: thank you for believing in my potential as a researcher, and for letting me do my thing. You gave me my chance, and for that, I am grateful. Also, thanks for Figure one. Thanks also to head of department Mats Edenius for your encouragement and support. Similarly, special thanks go out to Lars Engwall and Birger Rapp, the founders of the Swedish Research School of Management and Information Technology, that so generously funded my entire PhD period. Thank you both for believing in me and for providing insightful comments during MIT seminars and PhD courses. Similarly, Claes Thorén has provided many useful comments on earlier versions of this thesis – thank you for this. A big “thank you” is extended to my colleagues at the department of Informatics and Media. In particular, Eva Enefjord, thank you for your patience in helping me to (almost) grasp the mysterious workings of the travel reimbursement management system. LG Svensk has provided continuous mac support, and Pierre Hjälm has been continuously helpful in installing all kinds of weird server applications on my behalf. Ann Skogh-Gunnarsson and

Anna-Lena Torstensson Kåberg gave me some important advice when I was put in charge of designing my own course, for which I am thankful. Speaking of which, thanks again to Stefan Hrastinski and to Anneli Edman, who encouraged me to arrange my own undergraduate course. A great experience indeed. I am grateful towards Ilona Heldal for agreeing to serve as discussant during my final seminar. Josef Pallas has been helpful in informing me regarding future possibilities. I enjoyed teaching together with Cecilia Strand, and we seem to have rather similar tastes in movies – always a good sign of a solid person. To my fellow MIT PhD students, Fredrik Bengtsson and Tobias Nyström, I wish you the best of luck in finishing your respective theses. The same goes for the other PhD students at the department. Christian Fuchs – thanks for your insightful comments on earlier drafts of two of the papers included here. Christian Christensen – thanks for putting me in touch with the right people. Beyond the department, David Sörhammar deserves special mentioning for his love of pancakes. Moving beyond Ekonomikum and Uppsala, Mark Aakhus – thank you for agreeing to serve as opponent for my dissertation defense. Hallvard Moe has served as a continuous source of inspiration on visits to Bergen and via Skype. Special thanks go to the San Diego Air and Space Museum for generously providing me with the cover picture. Finally, I’d like to extend a big thank you to all my family and all my friends, from all spheres of my ife. Sølvi Elise – your everyday support and love is more important to me than I am able to express in writing. I wouldn’t have been able to go through with this without you in my life. This was fun! Can I keep doing this?

Uppsala, March 2012 Anders Olof Larsson

Contents

Introduction ................................................................................................... 11   From Web 1.0 to Web 2.0 ........................................................................ 12   Aim of thesis ............................................................................................. 14   Institutions online ..................................................................................... 16   Media organizations ............................................................................. 16   Political organizations .......................................................................... 20   Structure of thesis ..................................................................................... 27   Research Strategy .......................................................................................... 29   Methodological considerations ................................................................. 34   Theoretical considerations ........................................................................ 35   Generalizability ........................................................................................ 37   Summary of included studies ........................................................................ 40   Study I....................................................................................................... 40   Study II ..................................................................................................... 44   Study III .................................................................................................... 49   Study IV .................................................................................................... 52   Structuration theory ....................................................................................... 57   The duality of structure............................................................................. 58   Routines and institutions .......................................................................... 61   Critique ..................................................................................................... 63   Structuration Theory and ICTs ................................................................. 65   Use of and contribution to structuration theory ........................................ 70   ‘Most people don’t do things very differently’ ............................................. 72   Understanding non-use – three modalities of structure ............................ 76   Three types of enactment .......................................................................... 80   Conclusions .............................................................................................. 83   Limitations and suggestions for future research ....................................... 84   References ..................................................................................................... 87  

Introduction

The purpose of this introductory chapter is twofold. First, the reader is provided with an overview of the broader thematic area to which this thesis makes a contribution. This is done by reviewing the recent developments of the World Wide Web – developments that are popularly, and, to some extent also academically, understood as a transition from a “Web 1.0” to a “Web 2.0”. Second, the presented broader theme is narrowed down to introduce the aim and the research questions that serve as guides for the thesis at hand. Following this, a literature review on research undertaken regarding the specific cases to be scrutinized in the thesis – online news media and online politics – is provided. The chapter concludes with a section outlining the structure for the rest of the thesis. This thesis deals with a technological phenomenon that many of us make use of, often on an everyday basis – the Internet (Buick, et al., 1995). And yet, while most of us have become quite entangled in the World Wide Web, utilizing various online services for business and leisure time activities alike, certain aspects of these services remain largely unused. Specifically, the thesis places its focus on the use of online interactivity in the context of what is often categorized as societal institutions. Arguably a very broad term, the concept of an institutions can be popularly understood as “established ways of doing things” or, in the words of Anthony Giddens, ‘institutions by definition are the more enduring features of social life’ (1984: 24). Beyond the abstraction of theoretical definitions and discussions, we can distinguish what could perhaps be labeled a more “common sense” definition of the term under discussion. In other words, when we think of the concept of institutions, what comes to our minds? Assuming that the reader of this thesis has a background in the social sciences or the humanities, longstanding, established organizations with some variety of power vested in them would perhaps be a suitable forecast of the thought processes activated by this request. Consider, for example, the Catholic Church – an organization that for centuries on end virtually controlled mainland Europe through a consistent liturgy and by means of efficient communication channels (for that particular time period, at least). Furthermore, consider organized parliamentary politics as we know it, societal practices that grew out of feudal society, embraced democracy and that today enjoys persistence in most 11

western nation states. And finally, consider the mass media, purveyors of information (and, in some questionable cases, “truth”) to the public, gatekeepers and agenda-setters, who today hold somewhat weakened positions, considering the challenges to their privileged positions often posed by the Internet. Church, politics and media – all three can be considered institutions that have played important parts in shaping the development of western societies during the course of the last few centuries. Being certain that there are plenty of adequately skilled theologians who would be interested in an analysis of how the church has made use of the interactive features made available by the Internet to further their communicative efforts with their different constituencies, the decision was made to leave this specific case behind. While it is an interesting topic, the church will not be scrutinized in the thesis at hand. Instead, focus will be placed on the other two institutions discussed above – politics and mass media. Specifically, the thesis studies how practitioners working within political organizations and media organizations approach the Internet with regard to its potential for interactivity with their respective audiences. Conversely, the thesis will also take the reverse perspective in order to assess how audience members in the political and media contexts approach the presented possibilities for interaction. By studying how institutional practitioners and their respective audiences make use of online interactivity, the thesis contributes to the understanding of how historically important societal actors took the step to go online – and what consequences this appears to have had. The next section discusses the online developments that have been taking place during the last couple of years – a developmental process that, at least according to some, has supposedly led from a “Web 1.0” to a “Web 2.0”.

From Web 1.0 to Web 2.0 In the approximately twenty years that have passed since the spread of the Internet through western societies during the mid-1990s, the use of it has developed in different ways – a manner of progress that is similarly valid also for changes regarding the Internet itself. While the more technological side of these developments is beyond the scope of the thesis at hand, some of the more non-technical aspects of these elaborations are all the more essential. It might be practical to approach these developments in reverse chronological order, thus starting with the latter of the two suggested phases of online developments. The notion of a “Web 2.0”, a second, conceivably upgraded, version of the World Wide Web has been prevalent in popular as well as academic discussions since at least 2005, when a white paper entitled 12

What is Web 2.0? Design Patterns and Business Models for the Next Generation of Software was published by Internet advocate and consultant Tim O’Reilly (2005). While O’Reilly is not a scholar himself, this particular publication has enjoyed popularity within scholarly circles, as it is frequently quoted and used as a starting point of sorts for researchers interested in online developments. In the paper, O’Reilly outlines a set of principles and practices that he suggests together make up the collective rationale behind the 2.0 dictum. Specifically, he proposes eight core patterns of Web 2.0. These are as follows: 1. 2. 3. 4. 5. 6. 7. 8.

Harnessing collective intelligence Data as the next ‘Intel Inside’ Innovation in assembly Rich user experiences Software above the level of a single device Perpetual beta Leveraging the long tail Lightweight software and business models and cost effective scalability

As the individual items on the list implies, the concept of Web 2.0 can largely be understood as ‘an umbrella term for a number of new Internet services that are not necessarily closely related’ (Warr, 2008: 591). In addition, while many of the ideas and concepts listed by O’Reilly are worthy of further explanation and scrutiny, the purposes of the thesis at hand merits delimitation. With the focus of this thesis in mind, concepts pertaining to such more technical concepts shall be set aside. Instead, what is deemed to be most important here is the concept addressed at the very beginning in the list presented above – harnessing collective intelligence. Sometimes referred as ‘the core pattern of Web 2.0’ (Warr, 2008: 592), the practice of harnessing collective intelligence, of allowing users to play key roles in the co-creation of web sites and services, is indeed often pointed to as one of the most important aspects behind the 2.0 idea. Certainly, the media-savvy among todays organizations have been quick to capitalize on what could be described as the “2.0 brand” – For example, libraries have adopted the term, resulting in the paraphrase “Library 2.0” (i.e. Casey and Savastinuk, 2007; Curran, Murray and Christian, 2007; Dobrecky, 2007; Gross and Leslie, 2008) and similar configurations of the apparently popular labeling can be readily found in several other areas of professional activity. While the “Library 2.0” example might seem a bit off track given the topic at hand, it is used here primarily as an example of the popularity of the 2.0 label. With this wide-spread use in mind, one might be somewhat surprised to find how little actually appears to be new about this supposedly novel way of organizing and operating presences on the World Wide Web. 13

While the technical platforms that allow the Internet to function are constantly subject to new and sometimes revolutionary developments, the goals behind what could be described as the more operational side of the web, the view of the web that most of us interact with in our everyday lives, appear to have changed little. While the technical as well as the exterior appearance of the Internet in general has definitely been augmented since the mid-1990s (and, quite possibly, also since last month), it has been proposed that the underlying principles guiding the 2.0 rhetoric are perhaps not that innovative. Indeed, the suggestion has been made regarding the 2.0 label that ‘nobody even knows what it means’ (Tim Berners Lee quoted in Anderson, 2006), signaling doubts regarding the supposed novel aspect of this alleged second version of the World Wide Web. Similarly, the point has been made by Tim Berners Lee that a 2.0 paradigm offers nothing radically new in terms of outlook for web design, since participation and mass collaboration has been inherent in the Internet from its outset (e.g. Harrison and Barthel, 2009: 158159). The basic view of web 2.0 as ‘about participating rather than about passively receiving information’ (Tapscott and Williams, 2006: 37) could thus be said to rather similar to the ideas and thoughts that were allegedly inherent also in the initial, “1.0” version of the world wide web. The differences, then, are not pertaining so much to underlying ideology as to online technological developments helping to make what is sometimes called the read-write web (e.g. Gillmor, 2004) a possibility.

Aim of thesis This thesis is based on four empirical studies, reported in four journal articles. Two of the included studies deal with the topic of online news, and two are focused on online politics. As discussed in the previous section, these areas of activity could be considered as baring institutional qualities – and as such, the ongoing activity within these respective contexts are dependent not only on the work performed by the practitioners within each profession, but also on the modes of use employed by their respective audiences. Through both sides respective uses of the interactive features made available by the Internet, further emphasized by means of the previously described “Web 2.0” principle for web design, the activities within both thematic areas are often pointed to as undergoing changes or even challenges due to the supposed external influences from the assumed increase of user activity. The aim of this thesis, then, is expressed as follows:

14

How do practitioners working within societal institutions and their respective audiences make use of the interactive features made available by the Internet? This overarching aim has been broken down into two research questions that have guided the work performed and reported here: RQ1: How do practitioners within societal institutions make use of online interactivity? RQ2: How do their respective audiences make use of the interactive features offered to them? By focusing on not one, but two societal institutions, the thesis will be able to make stronger claims about the use of online interactivity by such actors and their respective audiences. Moreover, the choice of these particular institutions can be further merited by the fact that these are areas where the Internet has often been pointed to as having a particularly large influence (e.g. Bowman and Willis, 2003; Foot and Schneider, 2006; Gillmor, 2004). As will be made evident through the individual studies contained herein, the thesis makes its case by means of large-scale empirical data collection and analysis. This analysis is complemented by a literature review that outlines previous work performed on the use of interactivity by media actors, political actors and their respective audiences. Moreover, the thesis is informed theoretically by employing structuration theory, as introduced by Anthony Giddens (1984) and further developed by Wanda J. Orlikowski (Orlikowski, 2000). The specific use of structuration theory and the contributions to theory development as actualized in the thesis is discussed at length in the fourth chapter of the thesis. The contribution of this thesis is twofold. First, an empirical contribution is made by unearthing and addressing a series of research gaps pointed out by previous studies performed in the contexts under scrutiny here. In this way, the literature review presented later in this chapter provides a background from which to assess the individual contributions of the four studies included in the thesis, as well as the thesis itself. Second, a theoretical contribution is made by employing conceptual tools pertaining to structuration theory in order to provide an overarching theoretical assessment of the results obtained from the four empirical studies. As will be discussed further in the fourth chapter, suggestions for future employment of structuration theory have been made by previous researchers. This thesis follows up on one such suggestion, made by Jones and Karsten (2008), by using structuration theory according to their suggestions.

15

Similarly, a suggestion to use the theory in institutional settings was proposed by Orlikowski herself (2000: 423). While the theoretical contribution of the work presented here is discussed in detail in the fourth chapter, the empirical contributions of each study included in the thesis is specified each of the individual studies, as well as in the third chapter. The following section presents a review of research performed looking into the uses of online interactivity in the contexts under scrutiny here.

Institutions online ‘Institutions by definition are the more enduring features of social life. In speaking of the structural properties of social systems I mean their institutionalized features, giving 'solidity' across time and space.’ (Giddens, 1984: 24)

While the basic principles of structuration theory are discussed at length in the fourth chapter, the quote provided above offers a further indication of what is meant by institutions according to this perspective – stable, societal bodies that are based on continuous human activity and that have enjoyed such longevity for a considerable period of time. This thesis focuses exclusively on two such established societal actors – the media sector (exemplified here with online newspapers) and online political actors (exemplified as parliamentary political actors). As will be shown in the literature review pertaining to each of these cases, these are institutions that have often been thought of as being especially affected by the advent of the Internet – be it Web 1.0 or Web 2.0. Furthermore, the literature review provides the reader with short introductions to the rationales behind the included studies, providing specific contexts for the empirical contributions of the thesis. Additionally, the review is organized with the previously presented research questions in mind, using subsections that clearly deal with practitioners and audiences for each of the specified cases.

Media organizations Since the spread of the Internet in western countries during the mid-1990s, hopes and fears were plentiful regarding its supposed influence on journalistic traditions and practices. While the claim has been made that journalism has always been influenced by technological innovations (Pavlik, 2000: 229), views regarding the assumed effects of the Internet on the journalistic 16

profession have arguably been rather dramatic, with references to the impending ‘death of the printed newspaper’ (Thurman and Myllylahti, 2009: 1) as a not entirely uncommon part of the rhetoric (e.g. Fortunati and Sarrica, 2010). Whereas casualties of printed newspapers have (unfortunately, one might add) been part of the digital reality faced by todays news organizations, other newspapers have nonetheless made the move to the online environment in an attempt to meet the challenges and opportunities posed by the Internet. Such online ventures have often been part of a larger transformation; moving print-focused organizations towards more overarching roles of “media houses”, attempting a variety of diversifying actions in order to more efficiently face the new digital environment. While a variety of digital initiatives have been part of the rationales for numerous newspapers since the spread of the Internet, research on online newspapers have mostly shown that the majority of such initiatives have largely resulted in what could be described as status quo situations. Apparently, professionals within the news business have ‘tended not to realize the potential of new technologies, thus affecting limited change across the industry’ (Mitchelstein and Boczkowski, 2009: 567). Similarly, audience members have adopted a rather reluctant stance towards becoming engaged ’as content co-producers’ (Boczkowski, 2002: 278), expressing skepticism towards using the opportunities for interaction on newspaper web sites (Bergström, 2008). As interactivity has been described as the defining character of the Internet (Downes and McMillan, 2000; Kiousis, 2002), these reported behavioral patterns merit further investigation. Is it all hype? While some online newspapers have embellished their online presences with a variety of interactive features (e.g. Chung, 2004; Mitchelstein and Boczkowski, 2009), the argument is sometimes made that most online news actors still appear as ‘cautious traditionalists’ (Chung, 2007: 53), thus adopting a somewhat conservative stance towards the Internet (e.g. Boczkowski, 2004; Cohen, 2002; Quandt, 2008; Robinson, 2010). Similarly, audience members are not necessarily making use of the interactive features made available to them on the newspaper web sites. The following two sections outline research done on journalists and audiences online, respectively. Journalists online Early research often suggested that newspaper web sites tended to be characterized by what is sometimes referred to as “shovelware” journalism, a term that suggests a transferring of content from the offline to the online platform with little or no adaptation to the characteristics of the Internet (e.g. Schultz, 1999; Tankard and Ban, 1998). By the mid-2000s, Boczkowski (2004) noted that change was perhaps on its way regarding this rather traditional principle. Specifically, he suggested that the cumulative transformation effect that 17

online newspapers seemed to be undergoing, perhaps as a result of increasing online competition between newspapers or increasingly aggressive other online actors, had resulted in more imaginative newspaper web sites. Likewise, Greer and Mensing (2006) found an increase in the employment of interactive features, albeit minimal, in their longitudinal study of online US newspapers. Similar results, indicating incremental changes, have been reported by other researchers (Chung, 2004; Mitchelstein and Boczkowski, 2009). While these developments might have challenged the notion that interactivity is ‘counterintuitive with the principles of traditional journalistic culture’ (Domingo, 2008: 698), journalists have, as pointed out earlier, for the most part conveyed a rather cautious stance towards the Internet. While some have expressed a more positive view of the novel possibilities for reader interaction offered (Bowman and Willis, 2003; Gillmor, 2004; Pavlik, 2001), most accounts detailing journalist’s views and actions have suggested an approach somewhat akin to the aforementioned conservative stance. Employing an ethnographic approach, Domingo (2008) studied the work practices in four online newsrooms, finding that firmly established journalistic routines tended to counter the notions of increased interaction with the audience. Similarly, Singer (2010) used an online survey in order to gauge the attitudes towards various forms of user-generated content among British journalists. Her results suggested that while most journalists appear at least somewhat positive towards the inclusion of user-generated content on their respective web sites, there is an apparent need for the provided content to be ‘carefully monitored’ (Singer, 2010: 127) – a costly gatekeeping task, both in terms of personnel and time (Thurman, 2008). Thus, interactivity in the online newsroom becomes a difficulty that most journalists and editors need to deal with on an everyday basis. As for more recent developments, pertaisning to the advent of the so-called Web 2.0 (O’Reilly, 2005), we are starting to see studies on the impact of blogs as well as micro blogs (i.e. Twitter) on journalistic practices. For example, in her influential study on how political journalists appropriated the blog format of online publication, Singer (2005) found that journalists tended to normalize the blog channel to make it conform to established work routines. Likewise, in his study of how British newspaper The Guardian made use of their blog for audience interaction, Matheson (2004) concluded that while the blog at least potentially challenged the traditional, established one-way relationship between journalists and audience members, the specific case of the Guardian was ‘not in any way revolutionary’ (Matheson, 2004: 460), suggesting an appropriation of the blog format to a conventional, timehonored variety of journalism. Much like the blog format, its derivative micro blog (in particular Twitter) has been pointed to as having potential influence over the journalistic 18

profession. Taking as their starting point the aforementioned study by Singer (2005), Lasorsa et al (2011) found similar patterns of use among the “jtweeters” (journalist Twitter users) they studied that Singer found among journalist bloggers. In other words, while journalists might be quick to start using new means of mediated expression, their specific approaches to such new channels are often adapted to established forms of journalistic practice. These results suggest that while some journalists are indeed “early adopters” of social media and other features that allow for audience interaction, most journalists tend to adhere to what has been previously described as “conservative” or “traditional” journalistic thinking (e.g. Chung, 2007; Domingo, 2008; Domingo, et al., 2008; Gillmor, 2004; Vobic, 2011). In sum, research performed during the past decade indicates that most media practitioners are perhaps closer to the notion of online journalism as ‘TV news on a computer screen’ (Dahlgren, 1996: 64) than some critics and pundits might suggest. While scholars have shown that journalist attitudes towards employing interactive features in their everyday work routines might be subject to age, where younger journalists tended to be more positive (Chung, 2007), there still appears to be an inherent tension in the reporteraudience relationship (Robinson, 2010: 141). As made clear by the outline presented above, most research on online newspapers seem to conclude that their approach to interactivity is careful at best. While some researchers have attempted to chart out the types of interactive functionalities available on newspaper web sites, rather few scholars have assessed what features of newspaper organizations that seems to have influence over the degree to which they use interactive features on their web sites. This research gap is addressed in this thesis by means of study I, entitled Interactivity on Swedish newspaper web sites – What kind, how much, and why? As the title implies, the study makes a contribution to research on online newspapers and their utilization of interactivity by providing a largescale investigation of Swedish newspapers. The study rationale and results are further discussed in chapter three. News audiences online Initial studies regarding audience opinion on online interactivity tended to report on audience members as expressing mostly positive and curious attitudes towards the new phenomenon. For example, an early study conducted in the USA indicated that users tended to feel a sense of commitment towards sites that offered more interactive features (Shyam Sundar, 2000). Similarly, other scholarly work concluded that more interactive features on web sites usually resulted in more positive audience attitudes towards the site (Aikat, 1998; Kiernan and Levy, 1999; McMillan, 2000; McMillan, Hwang and Lee, 2003). For newspaper web sites, it has been suggested that 19

interactive features would lead more visitors to spend time on the site (Gerpott and Wanke, 2004). As such, while attitudes towards interactivity appear to have been positive, levels of use have consistently been found to be low in a majority of studies. While notable exceptions of increased audience interaction in the online newspaper context is available (e.g. Bruns, 2010; Deuze, 2006), other empirical accounts have suggested different opinions. For example, Hujanen and Pietikainen (2004) used a multi-method approach, employing surveys and in-depth interviews in their study of how young Finns perceived and used opportunities for interaction in the online news context. Finding rather low levels of interaction, the results suggest that this non-usage of interactivity might stem from the fact that respondents saw the news web sites as “untouchable”, a finished product arranged by professionals, in little or no need of amendments from the audience. In Sweden, similar results indicating mostly non-use from the audience side were reported by Bergström (2008), suggesting that these traditional views on the journalist-reader relationship might stem from the ‘strong position of the printed newspaper in the Swedish context’ (Bergström, 2008: 77; see also Hallin and Mancini, 2004). In sum, research performed has showed that most journalists and audience members do not partake in the interactive capabilities of the Internet to any larger degree. These combined results indicate that the transition to a more ’dialogical’ (Deuze, 2003: 207) form of journalism might be farther away than sometimes suggested. Furthermore, the suggestion has been made that even if audience members do not use the interactive features made available to them, they might appreciate the fact that they are offered the possibility to interact (e.g. Deuze, 2003). The premise of such “indirect” effects of interactivity is evaluated in study II included in this thesis. Entitled Interactive to me – Interactive to you? A study of use and appreciation of interactivity on Swedish newspaper web sites, the study employs large-scale survey methods to query online newspaper audience members regarding their use and appreciation of interactive features in the online newspaper context. The full rationale behind this second study is provided in chapter three.

Political organizations Much like with the previously discussed media sector, there has been plenty of discussion on the Internet and its supposed potential to ‘reshape political communication and campaigning’ (Lilleker and Malagón, 2010: 26). Indeed, early studies suggested that the Internet would be ushering in a third age of political campaigning (Blumler and Kavanagh, 1999), bringing with it increased and ongoing contact between politicians and electorate. Similarly, 20

almost a decade later, analogous rhetoric is prevalent, discussing ‘postmodern’ or ‘professionalized’ campaigning through the Internet (Zittel, 2009). Coupled with the establishing of terms like e-democracy or e-representation (Coleman, 2005), much of the early rhetoric around the consequences for political activity in the wake of the Internet was characterized by a rather optimistic tone (e.g. Castells, 2001; Coleman, 2005). Indeed, in a very early study from 1997, Barnett noted that most of this rhetoric was ‘relentlessly upbeat’ (1997: 194). Nevertheless, this view of the Internet as the “the magic elixir […] to reinvigorate the masses to participate in the process of government” (StromerGalley, 2000: 113) does not appear to be unique for the digital era. Much like for the profession of journalism, technological development has often been perceived as a ‘key driver of change in the electoral area’ (Gibson, Lusoli, and Ward, 2008: 15), suggesting the development of new, more interactive roles for politicians as well as the electorate. However, as pointed out by Kalnes (2009: 251), one should be careful not to overemphasize these supposed changes at the expense of continuity. Indeed, most research into online political communication has given weight to a view of stability and continuity with regard to usage patterns of political actors and citizens alike. While an online presence of some sort could be considered almost mandatory for politicians seeking office (as suggested by Druckman, Kifer, and Parkin, 2007: 426), and while the abundance of political information and opportunities for political mobilization of citizens are surely acted upon by certain citizens, most researchers seem to conclude that we are predominantly seeing a transferring of offline patterns into the online context, giving way to ‘politics-as-usual’ (Margolis and Resnick, 2000). With recent tendencies towards a “Web 2.0” rationale of online activity, placing more focus on the user and user participation (e.g. Cormode and Krishnamurthy, 2008; O’Reilly, 2005), propositions regarding the Internet and its potential for political rejuvenation have been raised yet again (e.g. Lilleker, et al., 2011; Lilleker and Malagón, 2010). The apparent online success of the 2008 Obama campaign for the US presidency has further invigorated such claims (Kalnes, 2009: 264; Wattal, Schuff, Mandviwalla and Williams, 2010). Adopting the view of the politician and the citizen respectively, the following sections presents a discussion on the popular hypotheses of innovation and normalization with regard to online politics. Specifically, the first of these hypotheses suggests the introduction of novel practices for both politicians and citizens through employment of the Internet, while the latter of the two implies a status quo of more traditional roles for both identified groups.

21

Online politics – providing innovation or normalization? Although there are exceptions, many scholarly efforts on online political communication have taken some dichotomization of the aforementioned hypotheses as their starting point. While the pairing of innovation and normalization hypotheses are arguably among the more common of these dichotomies, there are of course other varieties of this nomenclature – perhaps most notably, the couplings of optimist/pessimist (Bentivegna, 2006) and shift/enhancement (Jackson and Lilleker, 2009; Stromer-Galley, 2000). Regardless of the label used, these oppositions appear rather similar in their twofold outlook – the former suggesting an overarching pattern of change in the relationship between politicians and their respective constituencies, the latter a more conservative outline, with little or no change following the introduction and the continued development of the Internet. As mentioned briefly above, the innovation hypothesis suggests that the impact of the Internet on political communication will be a powerful one, providing ‘radical systemic changes’ (Kalnes, 2009: 251) for both citizens and politicians. While definitions and formulations of these hypotheses are abundant, Schweitzer (2008) outlines a series of specific assumptions that proponents of “both sides” in the proverbial confrontation tend to adhere to respectively. First, according to the innovation hypothesis, the Internet is believed to make information dissemination more efficient, give rise to more dialogue between voters and politicians through the employment of various online interactive features, overall ushering in more sophisticated approaches to web design and similarly matching online publication rationales. Beyond these more structural aspects of web development, the innovation hypothesis also suggests developments pertaining more to content and political discussion, such as providing more room for policy discussions, decentralized debate and less political “attacks” and foul play through the supposed quality of online argumentation and instant fact-checking (Schweitzer, 2008: 451). Second, following the normalization hypothesis, proponents here suggest that offline structures and forces will fundamentally shape political activity on the Internet (e.g. Margolis and Resnick, 2000; Resnick, 1998; Vaccari, 2008b). Again, Schweitzer (2008) outlines some of the basic claims often included in definitions of the hypothesis at hand. As for structural aspects, the hypothesis suggests an underutilization of interactive functions at the hand of the politicians, as well as a neglect of interactive options, adding up to lower degrees of web site sophistication than as suggested by the proponents of the formerly discussed hypothesis. Focusing on aspects concerning online content, the normalization hypothesis provides even more contrast to its innovation counterpart, foreseeing an increase in self-referential campaign coverage, a focus on the main candidates up for election as well as a 22

prevalence of slander and other forms of what is often referred to as negative campaigning (Schweitzer, 2008: 452). Politicians online As pointed out by Kleis Nielsen, ‘the Internet’s potential for political mobilization has been highlighted for more than a decade’ (2010: 755). However, for this potential to come to fruition, it must be acted upon by the politicians up for election, as well as by their respective electorates. Discussing US presidential politics, Gibson pointed out that by the year 2000, the Internet ‘had become virtually ubiquitous’ (Gibson, 2004: 98) – a development that, at least according to popular debate, seems valid for other contexts as well. Indeed, political actors in the US context would appear to be more apt towards employing Internet features than politicians in other contexts (e.g. Gibson and McAllister, 2011; Wattal, et al, 2010). Utilizing in-depth interviews with senior e-campaigning managers for the candidates during the 2004 US presidential election, Vaccari (2008a) found that the various online activities undertaken by democrats and republicans alike tended to target supporters and activists, rather than attempting to persuade undecided voters. Furthermore, Vaccari’s respondents suggested the increasingly important role of the Internet for political communication, predicting adoption throughout entire campaign organizations instead of merely among those staffers responsible for online affairs. The US context, arguably one of the most commonly studied when it comes to political communication, has also been subject to a number of comparative research efforts. For example, Hara and Jo (2007) studied the use of online functionalities (defined by the authors as fundraising, civic participation and e-mobilizing) by candidates up for election in the 2004 US and 2002 South Korean presidential campaigns. Results indicated that traditional representative mechanisms tended to have large influences over the scope of online representation, suggesting that the ‘equalizing effect’ (e.g. Vaccari, 2008c) of the Internet was perhaps not as persistent as some had proposed. Comparing the 2008 parallel campaigns for President of the USA and Prime Minister of Canada, Jaeger et al (2010) compared the uses of ICTs by the campaigns to organize and disseminate their messages, finding differences between the two countries, as well as internal differences in the employment of the Internet. Perhaps not very surprisingly, the results further strengthened the view of the 2008 Obama campaign as ‘defined by its creative use of new technologies’ (Jaeger et al, 2010: 79). Reporting somewhat different results, Kleis Nielsen (2010) found the use of mundane internet tools (like email and search functionalities) to be more important than the use of emerging tools (such as web 2.0 services) or specialized tools (like custom designed campaign web sites) in the practices of staffers working for

23

two congressional campaigns, suggesting that as such, ‘the mobilizing potential of the Internet will remain potential’ (2010: 758). Broadening the scope to the European context (as suggested by Gibson, 2004: 97; Lilleker and Malagón, 2010: 26; Schweitzer, 2008), a number of research projects have indeed been performed. In a study comparing the 2002 and 2005 German national elections, Schweitzer (2008) found minor parties to be ‘reluctant to jump on the Internet bandwagon’ (2008: 464), given their apparent underutilization of structural features allowing for voter interaction and mobilization when compared to the major parties. As for the United Kingdom, Jackson and Lilleker (2009) concluded that political actors had been somewhat slow on the uptake of a ‘Web 2.0 style of political communication’ (2009: 232), and that a paradigm of “Web 1.5”, offering a comfortable segue between the two extremes, was in the making (2009: 247). Similar to the results reported from the US, Gibson et al (2008) compared online candidate campaigning in the UK and Australia and found patterns of major party dominance in both countries. In southern Europe, reports from recent election campaigns seem to correspond to the practices outlined above. In France, Lilleker and Malagon (2010) studied the levels of interactivity on the web sites of presidential candidates, concluding that the presupposed changes to political communication following the employment of the Internet were ‘minor and at best incremental’ (2010: 39). Focusing on the same election, Vaccari (2008c) analyzed the characteristics of the web sites of parties and presidential candidates, finding significant gaps between major and minor political actors, giving further evidence against the claim that the Internet would function as a equalizer of sorts between larger and smaller parties. With this in mind, Vaccari suggested that ‘despite the media hype, online electioneering in France is still at an intermediary stage’ (2008c: 1), discerning an approach somewhat akin to the “Web 1.5” mode discussed previously. As for the 2006 Italian parliamentary elections, the same author found partial support for the normalization hypothesis, reporting on specific variance in the utilization of the Internet based on party ideology, voter demographics and media strategy (Vaccari, 2008b). Overall, though, Vaccari described the uptake of Internet functionalities among political actors as ‘slow and half-hearted’ (2008b: 75). Finally, the Nordic region of Europe is often pointed to as characterized by high levels of election turnout as well as a high percentage of Internet penetration and use. It follows from this that the use of the Internet on behalf of politicians in these countries might be more advanced than as reported from other areas, so as to match the alleged digital competencies of the constituency. For the 2003 Finnish parliamentary elections, Strandberg (2009) found that while other variables were certainly at play, belonging to a major party emerged as a strong predictor for candidate web site functionalities and sophistication. Four years later, during the 2007 Finnish elections, Carlson 24

and Strandberg (2008) focused on the use of the video sharing site YouTube, concluding that while the site gave voice to certain minor candidates up for election, only 6 per cent of the total number of candidates shared videos through YouTube. In Norway, Kalnes (2009) studied the adaption of Web 2.0 features (such as blogs, Facebook and YouTube) before and during the local elections of 2007. Results indicated ‘a slight e-ruption’ (2009: 261) in relation to the innovation hypothesis, suggesting incremental changes in electioneering practices. In sum, then, results from a decade or so of research seem to indicate that ‘an architecture of participation is in the process of being built, but perhaps in a limited form’ (Kalnes, 2009: 261). However, Jackson and Lilleker (2009) point out that as more voters become more competent with online practices and use of Web 2.0 features, politicians might find themselves forced to more clearly integrate such features into their campaign efforts. As outlined above, many of the studies performed are focused on one election, or comparisons between two elections in the same country, or in two different countries. The empirical contribution made here with regard to this is evident in study III, entitled “Extended infomercials” or “Politics 2.0”? A study of Swedish political party web sites before, during and after the 2010 election. The study features a longitudinal analysis of Swedish political party web sites before, during and after the 2010 parliamentary elections, focusing on party use of different types of interactive features (as informed by Foot and Schneider, 2006). The study provides a novel approach to research on how political parties make use of their web presences, as well as new insights into how these uses are played out during the course of an election year. The complete rationale behind study III is provided in chapter three. Political audiences online Much like in the previous section, it seems feasible to start with the US context, as the majority of studies have been performed there, and as the Internet has been pointed out as influential for citizen’s political information gathering and involvement during both the 2004 (Foot and Schneider, 2006) and the 2008 (Smith, 2009) presidential campaigns. Providing more in-depth data regarding the 2008 campaign, results reported by the Pew Research Center indicated that 55 per cent of US adults used the Internet to get news or information about that particular election. The researchers further concluded that younger citizens tended to be among the most involved in online political processes, but that this type of involvement was found to some degree across virtually all generations (Smith, 2009).

25

Furthermore, while claims about the potential of Web 2.0 features like various social media services are often aired in conjunction with political events (e.g. Baumgartner and Morris, 2010), criticisms regarding their supposed rejuvenation effect on the political engagement of young citizens in particular have been brought about as well. Among the more stern critics are Evgeny Morozov, who has focused mostly on the many “social media revolutions” that have taken place in various totalitarian countries during the “arab spring” of 2011 (Morozov, 2011). While his focus was beyond the scope for the established democracies placed under scrutiny here, the labeling of online political activity as unengaged ‘slacktivism’ without any tangible results (Morozov, 2009: 13) has made its way into western political discourse as well. However, the argument has also been made that this alleged “blackboxing” of emerging patterns of online political engagement is a simplification, and that online political activity is ‘at worst harmless fun and can at best help invigorate citizens’ (Christensen, 2011). Moving to the European context, focus here has indeed been mainly placed on younger voters. Presenting data from a transnational survey on university students in three countries (Italy, Spain and the Netherlands), Calenda and Meijer (2009) concluded that while the use of various Internet tools tended to reinvigorate political participation in the respondents, this participation tended to take on what the authors labeled as traditional forms. As such, ‘the online political world is indeed a natural extension of the offline world’ (Calenda and Meijer, 2009: 893) – a statement that gives at least partial confirmation to the normalization hypothesis. Also in the Netherlands, Hirzalla et al (2011) studied the use of Vote Advice Applications during the 2006 Dutch parliamentary elections, finding support for the innovation hypothesis among younger respondents, while older voters tended to behave more conservatively (see also Bakker and de Vreese, 2011). As mentioned earlier, the Nordic countries are often pointed to as characterized by high levels of ICT use and similarly high levels of voter turnout. Again, this might suggest likewise high levels of Internet use for various political purposes. In Finland, the previously mentioned study by Carlson and Strandberg (2008) on the use of YouTube during the 2007 presidential elections also detailed the use of the platform at the hands of the voting public, concluding that videos featuring presidential candidates did for the most part not ‘create any significant viral buzz’ (2008: 171). Utilizing data from the 2005 Norwegian Election Study, Karlsen (2009) found that younger voters were more likely to go online for political information than their older counterparts (see also Karlsen, 2011). Again, we see the focus on young voters in many research projects – a focus that appears also in studies performed in the Swedish context. In a study of Internet use during the 2006 Swedish parliamentary elections, Bergström (2007) found that younger cohorts tended to use the Internet to gather political information more often 26

than their older counterparts. Specifically, those who make use of the Internet for political purposes tend to be the same people who are politically engaged in the offline realm. A rather obvious result, perhaps – as well as a result that could be said to give some merit to the normalization hypothesis. As made clear from the presentation above, notable examples of innovative use of the Internet for political purposes have been reported on behalf of both citizens and politicians. However, the bulk of the research performed seem to largely support the normalization hypothesis (see also Schweitzer, 2011), thereby giving merit to Kalnes’ suggestion that the apparent changes and developments in digital political campaigning and engagement, significant as they may seem, should not be ‘overemphasized […] at the expense of continuity’ (2009: 251). In the context of this thesis, matters pertaining to the citizen’s role in the political context are dealt with in study IV, entitled Studying political microblogging: Twitter users in the 2010 Swedish election campaign. As the title implies, the study places its focus on the Twitter platform and provides a structural analysis of its use during the 2010 Swedish parliamentary elections. The study thus provides the research community with insights regarding political use of social media such as Twitter in an established democracy, as well as with a new means for data collection and analysis of Twitter messages. As with the previously mentioned studies, the rationale and findings of study IV are described at length in the third chapter of this thesis.

Structure of thesis Following this first introductory chapter, a second chapter labeled Research strategy provides a conceptual overview of the work performed. As already mentioned, this thesis consists of four empirical studies, published in four journal articles. The second chapter starts out by discussing the relationship between each of the included studies to the overarching aim and research questions that have guided the work performed. The succeeding subchapter introduces the methodological considerations dealt with while undertaking the work presented here. Similarly, the following subchapter introduces the theoretical considerations of relevance here. The final subchapter provides a discussion regarding how generalizability might be assessed in the context at hand. The third chapter provides extensive summaries of the four empirical studies that make up the bulk of the thesis at hand. The reader is introduced to the rationales, methodology, literature reviews and results of each paper, altogether providing an overview of the empirical work performed.

27

The fourth chapter contains an introduction to structuration theory as conceptualized by Anthony Giddens and Wanda J. Orlikowski. Specifically, the chapter introduces the terminology and conceptual tools used in the thesis. Furthermore, the chapter provides the rationale for employing this particular theoretical perspective, pinpoints the theoretical contributions made by the thesis, and considers some of the criticisms sometimes brought to the fore when discussing structuration theory. Finally, the fifth chapter provides an analytical discussion based on the results from the empirical papers. Specifically, terminology pertaining to structuration theory, as introduced in the fourth chapter, is employed in order to provide a theoretical understanding of the collective empirical findings presented in the individual studies. By providing such a theoretical reading of the empirical data, the thesis makes a contribution not only to structuration theory, but also to scholars and practitioners interested in the continuing development of the World Wide Web. Following a subchapter with conclusions, the chapter finishes up with a discussion of the limitations of the work performed and provides some suggestions for future research.

28

Research Strategy

The purpose of this chapter is to present the conceptual design behind the research performed. First, each of the four included studies is discussed with particular regard to their individual contributions, as well as their contribution to the aim of the thesis. Each study is also gauged regarding their relevance for researchers and practitioners. Second, I discuss the methodological considerations dealt with during the empirical work. Similarly, the third section of the chapter provides a discussion on the theoretical considerations undertaken. The final section discusses how the concept of generalizability has been assessed. As mentioned previously, the empirical material that make up the bulk of this thesis consists of four empirical research studies, reported in four journal articles. While these studies are summarized extensively in the next chapter, in what follows I provide a conceptual outline of how the four studies relate to the overarching research questions as presented in the former chapter. As a starting point, Table 1 provides an overview of the four studies, detailing their respective titles, aims, material, methods, themes and foci.

29

Table 1. Studies included in the thesis. Study I– Interactivity on Swedish newspaper web sites – What kind, how much and why?

Aim ‘to examine what kind of interactive features are available on the websites of Swedish newspapers, and what factors seem to influence the utilization of those features’

Material Data on presence or non-presence of interactive features according to Chung’s typology

Method Content analysis, regression analysis

Theme Online news

Focus Practitioners (RQ 1)

II – Interactive to me – Interactive to you? A study of use and appreciation of interactivity on Swedish newspaper web sites

’to study variations in newspaper website visitor use and appreciation of interactive features.’

Survey data on use and appreciation of interactive features according to Chung’s typology

Online survey, Exploratory factor analysis, regression analysis

Online news

Audiences (RQ 2)

III – “Extended infomercials” or “politics 2.0″? A study of Swedish political party Web sites before, during and after the 2010 election

How do the political parties use interactive features on their web sites during an election year?

Data on presence or nonpresence of interactive features according to Foot and Schneider’s typology

Longitudinal content analysis, descriptive statistics

Online politics

Practitioners (RQ 1)

IV – Studying political microblogging. Twitter users in the 2010 Swedish election campaign

‘to study participation in political debate on Twitter’

99,832 tweets

Descriptive statistics, social network analysis

Online politics

Audiences and Practitioners (RQ 1, 2)

30

First, the classification into specific themes is recognizable from the discussion and literature review presented in the first chapter. Two studies each pertain to the two identified themes of online news (studies I and II) and online politics (studies III and IV). As discussed in the introductory chapter, these thematic areas are of specific interest for the thesis at hand. As such, the four studies all individually relate to one of the two specified institutions. Second, focus here refers to the units of analysis under scrutiny in each of the studies. The terminology of “Practitioners” and “Audiences” is familiar from the research questions introduced in the first chapter – thus, Table 1 presents each study with labels as relating particularly to one of the two. By making an analytical distinction between the professionals working with online newspapers or within political organizations on the one hand (labeled here as “Practitioners”), and the Internet users visiting these online presences on the other (labeled here as “Audiences”), the dynamics of structuration theory is touched upon more clearly. Similarly, much of the literature review presented in chapter one revolved around such a delimitation for both themes of interest. While Gidden’s original outline regarding institutions can be understood as an essentially ongoing, collective construct between those that work within them on a day-to-day-basis and those that somehow take part in in the activities of each institution, this does not make an identification and analytical delineation between the different actors involved in the reproducing of structures an impossibility. Rather, the argument is made here that such a separation has helped in taking an overarching view of the individual studies performed, provided clear links between the research questions and the studies performed, and allowed for a means of comparing results from both thematic areas. As mentioned previously, the detailed outlines of the four included studies will be provided in the third chapter. The emphasis here, however, is placed on how each of these individual studies fit into the conceptual design of the thesis at hand. In the following, the included studies are briefly introduced with particular regard to their relevance to the aim of the thesis at hand, their scholarly relevance to their specific field of study, and their relevance for practitioners in applicable professional fields. Study I is entitled Interactivity on Swedish newspaper web sites – What kind, how much and why? As the title of the study implies, the aim is ‘to examine what kind of interactive features are available on the websites of Swedish newspapers, and what factors seem to influence the utilization of those features’. The focus of study I on the product of online journalists clearly places the study in line with research question number one, with its focus on the roles of practitioners.

31

Moving beyond the contribution to the thesis, the paper also makes a specific contribution to research on online journalism. By complementing what could be labeled as a descriptive analysis of what types of interactive features tend to be employed in the online news context with an attempt to explain the influences for such uses, the study answers the call for research to look beyond what is seen on the web page (Boczkowski, 2002: 277; Chung, 2007: 43-44). Besides its scholarly relevance, practitioners might find the study useful as it provides a critical exposé over Swedish online news – offering overarching insights into digital journalism that could be beneficial for the everyday working conditions of the digital newsroom. Study II is entitled Interactive to me – Interactive to you? A study of use and appreciation of interactivity on Swedish newspaper web sites. Focusing again on the theme of online news, the aim ‘is to study variations in newspaper website visitor use and appreciation of interactive features’. Thus, the focus is placed on the online newspaper audience, an emphasis that directs the study in line with research question number two. Besides the contribution to the thesis, study II also responds to questions and calls posed by previous, similar research efforts. For example, the study presents an identification of factors that appear to predict audience use of interactive features in the online news context (suggested by Chung, 2008: 673), and furthermore provides insights regarding user attitudes towards interactive features in the specified context (suggested by Chung, 2007; Hwang and McMillan, 2002; Sohn and Lee, 2005). Moreover, the results on use and appreciation of interactive features should be of relevance for practitioners within the online news industry who seek to amend their practices and provide their audiences with meaningful online experiences. Study III is entitled “Extended infomercials” or “politics 2.0″? A study of Swedish political party Web sites before, during and after the 2010 election. As outlined in Table 1 above, the theme shifts from online news to online politics. Study III is somewhat akin to study I in that it goes beyond a descriptive analysis of the web pages of the respective institutions, trying to unveil some of the supposed “inner workings” that appear to have influence over what we as audience members see when we visit these particular web pages. Specifically, the study features a longitudinal analysis of the web pages of Swedish political party web sites before, during and after the 2010 Swedish parliamentary elections. Focus here is once again placed on the practitioners – looking into the different uses of the web pages of political actors during the 2010 election year. With regard to the two research questions that have guided the work performed, study III relates specifically to the former of these, focusing on practitioners. Besides its contribution to the thesis, the study makes an individual scholarly contribution by providing unique longitudinal data on these mat32

ters, collected during the course of an entire election year. Moreover, its relevance to practitioners is similar to those specified for paper I – by providing data on year-long performance of political web sites, the study should prove useful for political actors who seek to engage their respective audiences on- and off the electoral season. Finally, study IV is entitled Studying political microblogging. Twitter users in the 2010 Swedish election campaign. Much like study III, the final study to be includd in the thesis deals with the theme of online politics, but differs in two important aspects from all three previous studies. First, while studies I through III all deal with web pages in different ways, study IV places its focus on a particular social media application, namely the microblog service Twitter. Second, as can be seen in Table 1, the paper also takes a different approach when it comes to its focus – while studies I through III have placed their respective emphasis on either practitioners or audiences, the fourth study applies what can perhaps be described as a “bottoms-up” means of data collection, focusing on both groups. As discussed in the literature review presented in chapter one, social media services such as Twitter have been heralded as baring with them a potential for rejuvenation of political debate – allowing for more users to be heard in online discussions of supposed societal relevance. Study IV tests these propositions by providing a large-scale structural analysis of Twitter users during the 2010 Swedish parliamentary election campaign. Based on the results of a social network analysis, the most frequent users are identified and classified according to a typology that is derived iteratively. Besides its contribution to the thesis in addressing both research questions, the study is also of relevance to the broader field of political communication and e-government research. In particular, it introduces a methodological means by which structural analysis of Twitter user networks can be performed. Additionally, study IV approaches the seemingly endless scholarly and societal discussions on what roles different ICTs can have for political discussion, dissemination and campaigning (e.g. Wattal, et al., 2010). By providing pertinent results on such uses, the study should be of high relevance for practitioners working with political campaigns – as well as practitioners working with online communication more broadly. In sum, the previous paragraphs have detailed the conceptual design of the thesis and the four studies included herein. Specifically, the relational patterns between each of the four included papers to the overarching research questions presented earlier have been described. While the following chapter provides extensive summaries of all papers, the previous sections have discussed the relevance of each of the studies to the thesis itself, the individual contributions of the studies, as well as the significance of the work performed for practitioners within relevant applied fields. Moreover, while de33

tailed descriptions of the methods employed for data collection and analysis are offered in the individual papers as well as in the summaries provided in the next chapter, the next section discusses the overarching methodological considerations dealt with during the empirical work.

Methodological considerations Besides detailing the aims of the included studies, and how these studies relate to the research questions guiding the thesis, Table 1 provides an overview of the materials and methods employed for data collection and analysis in each of the studies. Although the included studies make use of a number of different means of data collection and analysis, they can all be described as belonging on the quantitative end of what can perhaps be labeled as a continuum of quantitative and qualitative methodological approaches. Indeed, delimitations between the end points of such a dichotomy between quantitative and qualitative methods can be frustratingly difficult to pin down. Drawing on the sometimes fiery debates regarding this matter, Dey suggested that ‘it is difficult to draw as sharp a division between qualitative and quantitative methods as that which sometimes seem to exist between qualitative and quantitative researchers’ (1993: 4). While difficulties pertaining to delimitation certainly exist, most attempts to describe the differences between the two extremes tend to include some definitions of quantitative research methodology as focused on ‘numerical analysis to illustrate the relationship among factors in the phenomenon studied’, while qualitative methods tend to ‘emphasize the description and understanding of the situation behind the factors’ (Chen and Hirschheim, 2004: 204). With this distinction or, rather, continuum in mind, I have consciously chosen to work with (more) quantitative methods rather than with their (more) qualitative equivalences. This choice was made partly due to personal preference. Be that as it may: I do not consider myself a ‘quantitative researcher’ (Dey, 1993: 4), finding it hard or even refusing to see the merits of other perspectives on the processes of data collection and analysis (what Gibbons, 1987: 1 has referred to as 'methodological monism'). The studies included in this thesis appeared to lend themselves particularly well to such methods. These approaches have allowed me to take a wide-ranging grasp of the empirical situations under scrutiny. However, such a grasp has necessarily carried with it a loss of insights on a closer level of inquiry. Nevertheless, I argue that by employing different types of methods, I have been able to clearly address the research questions that have guided the work presented here. 34

While these types of methods are indeed powerful and have provided interesting and relevant insights into the phenomena under scrutiny, there are limits to the conclusions that can be drawn based on quantitative data. Of course, this claim certainly goes the other way: had a more qualitative approach been selected, other possibilities as well as delimitations would have come into play. For example, for studies I and II, research designs focusing on interviews with journalists and readers respectively, dealing with how they approach the interactive features on the sites with which they work or get their news, is an interesting opportunity – which is also pointed out in the individual studies themselves. Similarly, the topic under scrutiny in study III could also have lent itself to a more qualitative form of inquiry – an ethnographic design, perhaps inspired by Kleis Nielsen (2010), could certainly have served as a suitable starting point for a more qualitative variety along these lines. Study IV, then, represents a somewhat different approach than its three preceding papers. While segments of the analysis certainly make use of what can easiest be described as quantitative methods and approaches, the classification of user types and other delimitations made are not as clear-cut, method-wise. These definitory difficulties point again to the suggestion by Fry et al that ‘the matter of method is not an either/or proposition, it is a matter of degree of emphasis’ (1981: 145). By making use of both traditional quantitative analysis as well as social network analysis in combination with ‘big data’ (e.g. boyd and Crawford, 2011; Bruns, 2011; Manyika, et al., 2011), study IV could be said to take a step beyond the often confounded border between quantitative and qualitative approaches in order to study new forms of media such as Twitter (as suggested by Chin, Junglas, and Roldán, 2011). By using a variety of quantitative methods, I can provide the “big picture” of an empirical phenomenon. Through the use of these methods alone, however, I cannot make more detailed claims regarding individual user reactions or on specifics pertaining to individual cases in my studies (Kaplan, 1964). Such insights, while definitely relevant and interesting, are beyond the scope of the work presented here.

Theoretical considerations As shown in Table 1 and as further detailed in the next chapter, each of the studies included in this thesis are informed by typologies, classification schemas or by suggestions adhered from previous research. In order to more clearly articulate the contributions to the thesis, and to provide a means for 35

meta-analysis of the empirical results reported in the individual studies, this thesis employs key terminology related to structuration theory in order to discuss the combined results from the included studies. Indeed, scholars like Kaplan (1964) and Merton (1967) suggest that theory can provide us with answers to questions of why a specific empirical phenomenon occurs. The assertion made here is that structuration theory can provide such enlightening and possibly also explanatory interpretations of the empirical results reported in the studies. The use of structuration theory as situated in this thesis is depicted in Figure 1.

Aim of thesis

Interpretation

Structuration theory

Research questions

Study

Study III

I Study II

Study IV

Figure 1. Conceptual design of thesis.

Figure 1 provides a conceptual overview of the work performed, detailing the flow between the different parts of the thesis. Starting in the top left corner of Figure 1, the aim of the thesis is broken down into two research questions. The aim, and the specific formulations of the research questions (focusing on the use of online interactivity by practitioners and audiences respectively) were introduced in the first chapter of this thesis. The research questions are then dealt with in the four empirical studies, summaries of which are provided in the following chapter. Finally, structuration theory is employed in order to inform the interpretation of the empirical results from the four studies in accordance with the aim of the thesis – How do practitioners working within societal institutions and their respective audiences make use of the interactive features made available by the Internet? As such, structuration theory is not used explicitly in the four included studies – rather, it is employed as a means to offer theoretically informed interpretations of the summarized empirical results from these studies, providing 36

an overarching view of the work performed. Thus, in the context of the thesis at hand, structuration theory is understood in accordance with what Gregor (2006) has labeled a theory for explaining, ‘showing others how the world may be viewed in a certain way, with the aim of bringing about an altered understanding of how things are or why they are as they are.’ (2006: 624). In fashioning the thesis in accordance with the conceptual design outlined above, I agree with Sutton and Staw who ‘argue for greater theoretical emphasis in quantitative research’ (1995: 383). While the four included studies all make use of specific individual theoretical or typological appropriations, informing the data collection processes and analyses in each study, the employment of structuration theory has provided a means of interpreting the quantitative results presented in the studies, moving beyond what could otherwise perhaps be considered an accumulation of empirical findings. The conceptual design has been constructed in order to help avoid ‘underweighting the theoretical contributions of quantitative research’ (Sutton and Staw, 1995: 383). Indeed, empirical data by themselves cannot be considered theory (Bacharach, 1989: 497; Weick, 1995: 387), and the need for more theoretically informed research, performed quantitatively or not, is touched upon by several other authors. Perhaps somewhat scalding, arguments have been made that research publications need to be more than ‘storage devices for obtained correlations’ (Sutton and Staw, 1995: 379), more than entries into an ‘r-square sweepstakes’ (DiMaggio, 1995: 391), more than ‘business journalism with regression’ (Davis and Marquis, 2005: 335). In sum, by employing terminology and conceptual tools associated with structuration theory as described above, I have made the effort to appropriate my empirical results beyond their specific contexts and make a clear contribution to theory. Drawing on these efforts, the following and final section of this chapter offers an outline of how generalizability was approached in the thesis.

Generalizability As pointed out by Gregor (2006), it is often seen as rather difficult for social scientists to make extensive claims – or even “laws” – regarding generalizability in the vein of our colleagues in the natural sciences. This is supposedly due to the ‘very large number of conditions that might impact on any outcome’ (Gregor, 2006: 616). With specific regard to the generalizability pertaining to the type of theory employed in this thesis, DiMaggio (1995) has 37

described such theories as ‘device[s] of sudden enlightenment’ where the point is ‘not to generalize, because many generalizations are widely known and dull’ (DiMaggio, 1995: 391). Nevertheless, the argument is made here that social scientists should still strive for some notion of generalizability in their scholarly efforts. For this particular thesis, the work undertaken can be conceptually understood or framed as two case studies – one dealing with online newspapers (studies I and II), one with online political actors (studies III and IV). Thus, by using structuration theory as outlined above and as further specified in the fourth chapter of this thesis, I have been able to trace tendencies and mechanisms pertaining to this theoretical perspective in two different, albeit similar cases. As previously mentioned, structuration theory is not easily classified as a ‘theory for explaining and predicting’ but rather as a ‘theory for explaining’ (as suggested by Gregor, 2006). As such, generalizability as a means of predicting precise outcomes in other case settings, i.e. other societal institutions, than those covered here can perhaps be seen as beyond the scope of the work performed. However, by showing how the theory can help explain empirical phenomena in not one, but two case study settings, the argument is made that this thesis is indeed able to make some claim of generalizability from case study findings to theory (Yin, 2003), thereby corroborating the tenets of structuration theory beyond its popular appropriations in the Information Systems field (e.g. Jones and Karsten, 2008). Indeed, as many researchers appear to be in agreement that rigorous generalizations can be based on knowledge claims from a single case study (e.g. Lee and Baskerville, 2003; Walsham, 1995; Yin, 2003), the present thesis, understood as two case studies, should be able to at least partly generalize to other, similar contexts. As suggested by Seddon and Scheepers (2011), ‘if the forces within an organization that drove observed behavior are likely to exist in other organizations, it is likely that those other organizations, too, will exhibit similar behavior’ (2011: 17). While the topic under examination here was not placed on the organizational (or meso) level, this line of argument should be relevant also for macro contexts, such as the ones under scrutiny here. As the individual studies included show, similar tendencies can be seen in the two different case settings under scrutiny here – thus, perhaps those similar tendencies are exhibited also in other contexts. In conclusion, Lee and Baskerville state that ’whether research is conducted quantitatively or qualitatively, there is only one scientifically acceptable way to establish a theory’s generalizability to a new setting: It is for the theory to survive an empirical test in that setting’ (2003: 241), effectively arguing that generalizability is almost an impossibility if it is not carried out through extensive empirical research. With this in mind, this thesis cannot make any extensive claims regarding broader generalizability, predicting specific out38

comes in other contexts. However, while structuration theory can be described as a ’sensitizing device to view the world in a certain way’ (Klein and Myers, 1999: 75) and thus as hard to operationalize and test empirically, this thesis has employed conceptual tools pertaining to structuration theory in two separate empirical situations with similar analytical results. In closing, Seddon and Scheepers remind us that ‘generalizing to other settings based on findings from a single case always needs to be done with great caution: it is so easy to jump to the wrong conclusions’ (2011: 17). I agree with the above sentiment and resolve that as I base my conclusions on theoretical interpretations of not one, but two case studies, I am able to at least partially generalize from my findings to theoretical propositions (e.g. Walsham, 1995; Yin, 2003). However, given this design, the possibility to generalize to other empirical contexts has to be viewed as limited, given the caveats discussed above. Future research can perhaps show if the observed tendencies in the institutions under analysis here do indeed appear also in other, similar contexts.

39

Summary of included studies

This chapter provides extensive summaries of the four empirical research studies that make up the bulk of the thesis. Specifically, each study is assessed by focusing on their individual rationales, data collection procedures, methods employed and main results. Additionally, the previous versions of each study (conference proceedings, workshop presentations) are disclosed.

Study I Larsson, A.O. (in press). Interactivity on Swedish newspaper web sites – What kind, how much, and why? Accepted for publication in Convergence: The International Journal of Research into New Media Technologies. A previous version of study I was presented at presented at the New Media and Information: Convergences and Divergences conference, Athens, Greece, May 6-9 2009. As the title of study I implies, the emphasis here is on the web presences of Swedish newspapers. Specifically, the study tries to fill a research gap identified by previous scholars, mentioned briefly in the first chapter of this thesis: while plenty of research efforts have attempted to map out what kind of interactive features are available on the web sites of newspapers, rather few have made attempts as to assess factors that appear to have influence over the utilization of such features. Specifically, the aim of the study was to examine what types of interactive features are available on the web sites of Swedish newspapers, and what factors pertaining to these newspapers that seem to influence the employment of such features. In the paper, of interactivity was defined according to a four-part typology of online interactivity specific to the context of online newspapers developed by Deborah Soun Chung (Chung, 2008; Chung and Nah, 2009). The typology is based on empirical studies of US newspaper web site visitors, and posits that four different types of interactive features are distinguishable in the online newspaper context. 40

First, human interactivity refers to those features that facilitate interpersonal communication in different ways, making it possible for web site visitors to chat, e-mail or in other ways contact other readers or the journalists at hand. Second, medium interactivity details the more technical features available, features that allow users different choice options when browsing web sites. Medium interactivity can thus denote rather basic features, like on-site navigational possibilities, or more advanced functionalities, like interactive news graphics or video streams, providing the user with more choices for how to fashion their online news consumption experience. The third and fourth types of interactivity are essentially conceptualized as combined forms of the first two. Accordingly, the third type is labeled human-medium, indicating a set of features aimed towards allowing user contributions to the site, thus “engaging users as content co-producers” (P. Boczkowski, 2002: 278). Common features under this heading include functionalities that allow users to submit their own news tips, stories or photos to the newspaper web site, or features that can be likened to offline “Letters to the editor”. Finally, the fourth identified type of interactivity is labeled as mediumhuman, and places its focus on features that allow users to customize various aspects of site content to their respective liking. Common functionalities found under this heading include customized topics, headlines as well as different options for search and news updates or alerts. The four-part typology is summarized in Table 2 below.

41

Table 2. Summary of Chung’s four-part typology of interactive features on newspaper web sites. Type of interactivity

Chung (2008), p. 666

Human

‘human interactive features facilitate interpersonal communication online’ ‘medium interactive features generally allow readers more control or choice options in experiencing news stories’ ‘human/medium interactive features allow users to express their personal opinions’

Medium

Humanmedium

Mediumhuman

“medium/human interactive features allow users to customize news to their liking”

Chung and Nah (2009), p. 858 ‘interactive options that promote human interactivity facilitate userto-user mutual communication’ ‘Features representing medium interactivity rely on technology to allow users to select and elicit choice options’

Function on site

Example features

’Users can communicate with peers and staff’

‘Chat, discussion forums, “e-mail-afriend” feature’

Users can experience content in a variety of ways

Video streams, News graphics, mobile version of site

‘human/medium interactive features that allow users to submit customized perspectives and opinions further provide the audience with a sense of ownership’ “Medium/human interactive features, or features that provide interactive tailoring, allow users to personalize information to their liking”

Users can contribute own content

E-mail-links to journalists, “Reader news tip” Reader blogs, Reader news, pictures etc.

User customization of site looks, content and use

Customizable content, RSS feeds, e-mailalerts, SMS alerts

The typology outlined in Table 2 provided a set of guidelines for the content analysis that was employed to facilitate data collection for the dependent variables – i.e., the degree to which newspapers use different types of interactive features on their web sites. Inspired by previous, similar research (e.g. Gerpott and Wanke, 2004; Massey, 2000), as well as by classic titles detailing the method at hand (Krippendorff, 2004), a representative selection of Swedish newspaper web sites were content analyzed in order to grasp the extent to which features classified according to the typology were used on these sites. Features were identified based on an extensive literature review,

42

sought after on the selected sites, and classified according to the typology introduced above. To gauge reliability, a second coder was trained in order to recode a random 20 per cent sample of the included sites. Intercoder agreement ranged from .88, a score well above the common threshold of .80 (suggested by Krippendorff, 2004) to perfect agreement, all using Holsti’s formula (Holsti, 1969). The overall figure of intercoder agreement was .96. In order to fulfill the aim of discerning what factors might influence the use of interactivity, a literature review was first performed. The assumption was made that since the web sites under scrutiny were all part of newspaper organizations, previous research focusing on factors affecting newspaper performance in general should be of interest also here. The literature review suggested that six different factors should be employed as independent variables for the analysis to be performed – thus, these six factors where identified as possibly having influence on newspaper company operation. These factors, and their respective operationalizations, were: newspaper size (operationalized as circulation of parent newspaper), media ownership (operationalized as net result of owner), region (operationalized as local or national region of operation), competition within region (operationalized as total N of newspapers within each respective region), length of web presence (operationalized as N of days the newspaper had had a web site), number and age of staff (operationalized as N and mean age of web staff). Data for the first three variables were gathered through document analysis, while data for the last three variables were collected by means of a small questionnaire distributed via e-mail to the editors-in-chief of the newspapers included in the study. To avoid what Boczkowski commented on as “the lumping together of apples and oranges” (2002: 276-277) in analysis, i.e. the apparently common practice of treating different types of interactivity as one and the same, multiple regression analyses were performed in order to assess the influences of the independent variables on the dependent variables. The latter set of variables were formalized as a series of indices, one for each of the four identified types of interactivity, and one for the total number of interactive features identified on the newspaper web sites. In short, the results indicated that the only variable having a significant influence on the utilization of human interactive features was the one measuring circulation, a variable that emerged as significant for all types except for the medium variety. Perhaps not very surprisingly, the size of the parent newspaper appears important for web site performance. For human-medium features, the age variable proved to be significant, indicating that younger web staff leads to features that allow the readers to 43

involve themselves more actively in the web site. As suggested in the paper, this particular result could stem from different journalistic work ideals in different age groups, with younger journalists challenging the work routines established by their ‘cautious traditionalist’ (Chung, 2007: 53) older colleagues. While the remaining categories of medium and medium-human interactive features were both significantly predicted by the variable measuring the number of site staff, this was the only independent variable that emerged as significant for medium interactivity. This indicates the importance for a newspaper to sustain a sizeable staff in order to augment their web site with the more technically challenging features included under this label. Finally, results indicated that on average, Swedish newspaper web sites offer a little more than a third (35 per cent) of the interactive features identified in the study. This particular finding merits the question of whether or not this should be considered a rather low or high percentage – a question that could be the focus of comparative or longitudinal research efforts, also taking other contexts than the Swedish into consideration. The study finishes off by discussing some of the limitations dealt with during the work performed – perhaps most importantly, that the readers and their usage and opinions are not assessed. How do they feel about the features (not) offered to them?

Study II Larsson, A.O. (2011). Interactive to me – Interactive to you? A study of use and appreciation of interactivity on Swedish newspaper web sites. New Media & Society, 13(7).

Previous versions of study II were presented at the International Conference on Information Systems (ICIS 2010), St. Louis, Missouri, USA, December 12-15, 2010 and at the 2011 Conference of the International Communication Association (ICA 2011), Boston, Massachusetts, USA, May 26-30, 2011. The limitations described in conjunction with paper I served as inspiration to undertake the work presented in study II. As stated above, the analytical focus here was placed on newspaper web site visitors; specifically, how these visitors make use of and appreciate the different types of interactive features available on these types of sites. While previous research have reported audience use of interactive features on online newspapers at rather low levels (i.e. Bergström, 2008; Hujanen and Pietikainen, 2004), and found similar, somewhat reluctant 44

views towards the employment of such features among journalists (e.g. Chung, 2007; Domingo, 2008; Singer and Ashman, 2009; Thurman, 2008), scholars have posited what could be called an “indirect” effect of interactivity, pointing to the possibility that while visitors on newspaper web sites might not actively engage in interactivity, writing their own news pieces or uploading their own pictures to the site, they might have positive views about the very presence of those features, about the mere possibility to interact. Consider the quotation offered in Deuze (2003: 214), where a user states that ‘I do not want to – or feel the need to – participate in an online discussion or email the editors, but other people can, and its really cool that the site offers users that option’, indicating that such indirect effects might indeed be at work. With this in mind, the aim of study II was to study variations in newspaper website visitor use and appreciation of interactive features. The study was performed by means of a large-scale online survey, focusing on questions regarding use and appreciation of interactive features. The same four-part typology used in paper I was employed here, thus distinguishing between use and interactivity of human, medium, human-medium and medium-human features. In the survey, example features relating to each of the types were provided. Respondents were asked to specify how often they used such different types of features when visiting a specific newspaper web site, as well as to what degree they appreciated those features. These questions were measured by using seven-point Likert-type scales, with higher scores indicating higher levels of use or appreciation. Aside from questions regarding interactive features, the survey also included a series of questions regarding socio-demographics, social engagement, media use and digital literacy, all based on established measurements and motivated by the supposed effect on user activation attributed to these variables by previous research. The survey was administered with the help of LimeSurvey, an opensource survey tool. It was hosted on a server at the department of Informatics and Media, Uppsala University. Respondents were encouraged to partake in the survey via links leading to it placed on two newspaper web sites: aftonbladet.se, one of the most popular web sites in Sweden, under the administration of Sweden’s most popular tabloid newspaper, and sydsvenskan.se, the web site connected to Sydsvenska Dagbladet, one of the most popular regional newspapers in Sweden. Aftonbladet hosted the survey for eleven days, while Sydsvenskan did so for eighteen days, altogether resulting in 1343 respondents turning in complete answers. The collected data was subjected to three waves of analysis. First, descriptive statistics provided insights regarding the distributions of the variables measuring use and appreciation of interactive features. Second, in an attempt to uncover any underlying patterns of use and appreciation, those specific 45

variables were subject to an exploratory factor analysis. Third, in order to provide explanatory evidence, the extracted factors were used as dependent variables in a series of multiple regression analyses, where the previously mentioned variables regarding socio-demographics, social engagement, media use and digital literacy were employed as independent variables. The three waves of analysis will be dealt with in the following. In order to provide an overview of the respondents’ self-reported use and appreciation of interactive features, mean comparisons for both the use and appreciation measures were reported along with standard deviations. A series of Wilcoxon signed-rank tests were also reported in order to gauge the significance of the measured differences between values of use and appreciation. The overall result was one of low values – while respondents tended to appreciate features more than use them, all the means for the included measurements were reported at rather low levels. For example, while the commenting functionality was not used to any greater extent (M=1.52, SD=1.13), respondents tended to appreciate this feature to a higher degree (M=3.66, SD=1.84). This pattern of slightly higher values for appreciation than for use tended to repeat itself for each of the other features. The differences were small, albeit statistically significant, for all the use-appreciate pairings save for one (use/appreciation of video/audio/slideshow features). In sum, the average mean for the “Use” measurements was reported at 1.85, while the same statistic for the “Appreciate” scales was found to be 3.43. Given that these measurements were assessed on seven-point Likerttype scale (with a “7” indicating high levels of use or appreciation), these results do indeed indicate that the availability of interactive features do not appear highly salient to the average newspaper web site visitor. As mentioned earlier, exploratory factor analysis was employed in order to uncover any underlying patterns or modes of usage in the data. Five factors were extracted, together accounting for 50.2 per cent of the variance. The five factors were labeled according to their loadings on underlying variables and are presented in Fel! Hittar inte referenskälla. below.

46

Table 3. Five extracted factors and their underlying variables.

Appreciates features:

Uses features:

Factors The Bystander Email article to friend, Chat, Reader blogs/news, Contact staff, Links to similar content, Share content to SNS

The Prosumer

The Lurker Comment

The Filter

Comment, Chat, Reader blogs/news, Contribute image/video, Contribute news tips, Contact staff

Reader questions/polls, Video/audio/ slideshow, Links to similar content

Email article to friend, Share content to SNS, News updates

The Critic Reader blogs/news, News updates, Customization

As mentioned before, the influences of the independent variables on the extracted factors were assessed by means of a series of multiple regression analyses. By gauging these influences we can provide a picture of newspaper web site visitors and substantiate visitor types based on the factor analysis. In what follows, the five identified visitor types are described with the results presented in Fel! Hittar inte referenskälla. and the results from the regression analyses in mind. First, The Bystander is easiest described as a rather passive visitor, who does not tend to use, but rather appreciates, a number of different features. As per the results of the regression analysis, The Bystander is likely to be a young, internet-savvy female characterized in her on-site behavior by irregular visits to several different parts of the site. Second, the factors associated with The Prosumer indicate that this particular visitor type is arguably the most active of the five, making use of several of the interactive features identified in the study. According the regression analysis performed, The Prosumer, tends to be a male reporting high levels of digital literacy. As for his visit patterns, he tends not to visit the site on an everyday basis, although when he does, he tends to spend a considerable amount of time there. Third, rather few of the independent variables helped predict The Lurker, a visitor type characterized mainly by rather passive behavior. As can be seen in Fel! Hittar inte referenskälla. above, The Lurker makes use of several types of features, but it could be argued that these types of features are 47

perhaps not very demanding to use. Consider, for example, features like reader polls or links to similar content: While they are indeed interactive according to the applied typology, they are perhaps not as demanding or inviting to actual on-site participation as some of the other features included (for comparative purposes the usage pattern of the The Prosumer can be considered). Fourth, The Filter is so labeled because of the apparent tendency to use features that allow the visitor to function as an information hub of sorts, sharing content on social network sites (abbreviated “SNS” in Fel! Hittar inte referenskälla.), and e-mailing articles to friends. Given the results of the regression analysis, The Filter tends to be a female, quite focused visitor, mostly visiting specific parts of the newspaper web site. This could relate to the result that The Filter tends to use news update services, and thus does not need to visit as many parts of the site in order to gather the information necessary. Fifth, and finally, the factors underlying The Critic can be said to demonstrate some level of skepticism or criticism towards online newspapers. Following the regression analysis, The Critic is often male, further characterized by higher age and lower level of education, who appears as a rather intense visitor of newspaper web sites, spending quite an amount of time on the site when he visits it. Taken together, the results indicate that while there are visitors who make use of what could be described as the more demanding features offered by newspaper web sites (i.e. The Prosumer), most site visitors appear rather uninterested in the opportunities to interact. While the majority of the features scrutinized in study II were not used by the respondents to any larger extent, the features that actually were used could perhaps be classified as somewhat “low” in their degree of interactivity, demanding considerably less energy than higher forms - features that require ‘more effort in that individuals must do more than clicking or selecting in order to actively use them’ (Chung, 2008: 661; see also Stromer-Galley, 2004). Taking this distinction between higher and lower forms of interactivity into account, most visitors seem to be satisfied with keeping their use of higher forms of interactivity to a minimum, instead focusing mainly on the less demanding ‘bells and whistles’ (Deuze, 2003: 214) often available on newspaper web sites. Before discussing the limitations of the performed study, the suggestion is made that newspaper web site visitors thus could perhaps be understood as “slow learners”, taking their time to adapt to the possibilities for interaction made available by the Internet. With the results from study I fresh in mind, the same argument could perhaps be made when discussing some of the conservative tendencies relating to journalists and their employment of these types of features.

48

Study III Larsson, A.O. (2011). “Extended infomercials” or “Politics 2.0”? A study of Swedish political party web sites before, during and after the 2010 election. First Monday, 16(4). A previous version of study III was presented at the Evaluating Social Media workshop held in conjunction with the Internet Research 11.0 Conference (IR11), Gothenburg, Sweden, October 21-23, 2010. With study III, the theme of the thesis shifts from online news to that of online politics. As the title implies, the study draws on the hopes and expectations that are often aired regarding the Internet and its supposed potential for political purposes. As made clear in the literature review provided in the first chapter of this thesis, many of these initial promises were eventually largely disproven by empirical research (i.e. Chadwick, 2006). Thus, research on the topic at hand tends to give less weight to a paradigm of a “shift” in political engagement through the use of Internet technology, instead suggesting that the Internet helps in ”enhancing” already existing patterns of civic engagement, generally regarding information provision rather than increased opportunities for voter participation (Vaccari, 2008a, 2008c; Ward and Vedel, 2006). This twofold heuristic relates to the previously mentioned dichotomy of “innovation” on the one hand and “normalization” on the other, as discussed in the first chapter of this thesis. As the title further implies, the paper presents a longitudinal study of Swedish political party web sites during the election year of 2010. While the election itself was held in September of said year, political campaigning tends to be prevalent throughout the entire election year, or at least during the months preceding election day. The choice of Sweden was motivated by the findings and suggestions of previous research, where many studies have focused on US or UK contexts (Chadwick, 2006; Coleman, 2004, 2005). As Sweden boasts high Internet penetration rates among its citizens (Facht, 2008), and as political actors in the country apparently took to the Internet rather early (Gibson, 2004), the country makes for an interesting case study. Similarly, the longitudinal focus was also motivated by the findings as well as recommendations of earlier research efforts – while many studies have focused their analyses on the web presences of political actors during times of electoral campaigning, most such studies have primarily analyzed up to three different iterations of the respective web sites. As such, no yearlong longitudinal study taking these issues into account had been published at the outset of study III. Thus, the work presented in study III wanted to capture the ‘initial or tentative approaches’ (Lilleker and Malagón, 2010: 31) of the political actors, and therefore employed longitudinal analysis of said web 49

sites in order to trace their developments before, during and after the election. In order to accommodate this, data collection was performed throughout the 2010 election year. Specifically, a web site archiving tool (SiteSucker for Mac OS X) was used in combination with PDF printouts (utilizing WebSnapper for Mac OS X) of the individual political party pages, which were archived three levels deep. In essence, this means that the first or “start” page, as well as the main second and third pages were subject to archiving. The archived sites were then content analyzed based on a four-part typology of features employed on political party web sites, the idea being to be able to assess the “ebb and flow” of these different types of features during the election year. The typology used was suggested by Foot and Schneider (2006) and further complemented by other types of features identified by similar studies. Moreover, the web sites under scrutiny – i.e. the web sites of all major Swedish political parties – were visited a priori in order to search for any features not identified by previous research efforts. As for the typology guiding the data collection, it suggests four types of features often found on political web sites. The four types of features are described below. First, informing features provides information on party issue positions as well as plans and priorities for governing. In the context of online campaigning, techniques for informing often involve providing documentation regarding party history, ideology and talking points (Foot and Schneider, 2006: 50 ff; Kann, Berry, Grant and Zager, 2007; Vaccari, 2008c). Second, involving features are defined by Foot and Schneider as ‘facilitating interaction between site visitors and the campaign organization’ (2006: 70). While the basic techniques for involving, such as opportunities to contact the campaign through e-mail, social media (like Twitter or Facebook) or on-site “contact functionalities” are quite common, features often described as more advanced (such as discussion boards or live chats with politicians) are repeatedly found to be rather uncommon (Vaccari, 2008c: 1). Third, connecting features facilitates interaction with other online actors through the use of hyperlinks (Foot and Schneider, 2006: 103). In linking to other online actors, the campaign can be described as deeming those actors important in some way (Rogers and Marres, 2000: 16-17). The practice of connecting also provides the visitors of campaign web sites with a way to place the party in a specific cognitive context or frame (Foot and Schneider, 2006: 105), making the party ”understandable” in relation to the other, linked to societal actors. Fourth, Foot and Schneider define mobilizing features as ‘using the web to persuade and equip campaign supporters to promote the candidate to others, both online and offline’ (2006: 131). By motivating supporters to take concrete action and promote the candidate or party, site visitors can be made 50

to function as campaign organizations on to themselves (Bimber and Davis, 2003: 59). Parties can engage in mobilizing practices by providing site visitors with a number of tools, ranging from ”tell-a-friend” on-site functionalities to supplying campaign materials for on- and offline distribution. Advanced forms of mobilizing might involve parties encouraging their supporters to blog, tweet or act online in some other way in order to spread the party message (Foot and Schneider, 2006: 143 f). The typology suggested by Foot and Schneider (2006) provided a set of guidelines for the content analysis performed. Based on the typology, as well as the review of previous research as described above, a coding scheme was developed identifying 39 features for the informing category, 18 involving features, 16 for connecting features and 9 features for the mobilizing category. In total, then, 82 features were coded as being present or non-present in the archived web pages. Reliability for the coding was gauged by means of the employment of a second coder, who analyzed a random 20 % sample of the material. Following the recommendation of Lombard et al. (2002), Krippendorf’s α was used to determine the levels of reliability for the variables. Specifically, the KALPHA macro for SPSS (constructed by Hayes and Krippendorff, 2007) was employed in order to facilitate testing. The values of α ranged between .65 and 1.0, with a mean agreement figure of .95. Consequently, the results of the coding process were considered reliable. In an attempt to move beyond a descriptive approach, focusing only on the degrees to which the examined sites utilized the identified features or not, data regarding supposed influences for use of these features were also collected. Based on the results of previous research efforts, this involved utilizing variables assessing party size (operationalized as the share of votes received during the 2006 parliamentary election), parliamentary status (in or out of parliament), temporal influence (sites are believed to be utilizing more features as election day approaches) and ideological influence (suggesting that different ideological persuasions might result in different levels of utilization of features). Results indicated that while the indices for informing and connecting functionalities were not subject to much change during the election year, their involving and mobilizing counterparts appeared more dynamic throughout the examined time period. These last two groups of features tended to reach their respective high points during the election month of September, after which they both gradually decreased, indicating less motivation on part of the political actors to involve and mobilize the electorate after the election.

51

In sum, the results from study III suggests that while some political actors try to maintain more long term relationships with citizens, most parties choose to concentrate their efforts to engage the voting public to the more intensive campaign period building up to election day, after which the presence of these features on the studied sites decreases. As for the supposed influences on party utilization of these features, the analysis performed suggested that differences were mostly pertaining to party size, parliamentary status and party ideology. Overall, while the ‘stop-start’ nature of political activity, arranged around election cycles, has made its mark also in the online realm (Gibson, 2004: 102), the results further suggests a need to go beyond the dichotomy of normalization and innovation as discussed in the introductory chapter of this thesis, and to instead see the case presented here as an example of a leeway between these two perspectives – albeit with a slight lean towards the more conservative normalization hypothesis.

Study IV Larsson, A.O. and Moe, H. (in press). Studying political microblogging: Twitter users in the 2010 Swedish election campaign. Accepted for publication in New Media & Society. A previous version of study IV was presented at the 19th European Conference on Information Systems (ECIS 2011), Helsinki, Finland, 9-11 June, 2011. The second study dealing with online political activity, and the final one to be included in the thesis, places its focus on the role of social media within the context of parliamentary elections. As with study III, the context under scrutiny here is Sweden and the 2010 parliamentary election. While various social media applications have been used more passively as indicators and variables for studies I through III, study IV deals more exclusively with one such Internet service. Specifically, the paper introduces a novel approach to the gathering and analysis of Twitter messages sent during the final campaign period leading up to the aforementioned election, focusing on a structural analysis of the use patterns of high-end users. Social media in general and Twitter in particular is one of the latest in a long line of information- and communication technologies often thought to possess the potential for more or less radical systemic changes when it comes to enabling political participation (e.g. Smith, 2009). As pointed out by previous research, studies are needed that go beyond the conceptualization of 52

Twitter and similar technologies as interesting novelties (Honeycutt and Herring, 2009: 10). Study IV thus attempts to answer the apparent need for empirical studies of the Twitter phenomenon - especially in the context of political uses. Indeed, the literature review presented in the introductory chapter of the thesis at hand sketched a rather bleak history of the impact of the Internet on political activity and engagement. Can a novel service such as Twitter be said to contribute to a broadening of public debate, or does it mainly serve as yet another arena for those who could be classified as established societal actors? Twitter is sometimes referred to as a microblog, and as such, can be likened to a diminutive version of a regular blog. Blogs have indeed drawn a lot of attention from scholars interested in the political potential of online communication services (e.g. Kerbel and Bloom, 2005; Sweetser Trammell, 2007), and much of this attention is now focused on services like Twitter. As for Twitter’s structural properties, the service allows users to send short messages (“tweets”) of up to 140 characters to a network of followers. The act of following a user on Twitter is not necessarily reciprocal - a user can follow any number of other users, although the user being followed does not necessarily have to return the favor. Besides sending undirected messages (sometimes identified as “singletons”), users have utilized the open-ended character of Twitter to go beyond the uses intended by the service providers. Besides the singleton type message, which is undirected and visible to all, Twitter users utilize the @ sign in order to facilitate conversations. Posting a message including @USERNAME indicates that the message is intended for or somehow relevant to a specific user. Retweets (RT) refers to the practice of redistributing a tweet originally posted by another user. For analytical purposes, study IV employed the typology proposed by Kwak et al (2010) to distinguish between Singletons (an undirected message); a Mention or a Reply (@ sign followed by a user ID) or a Retweet, as mentioned earlier (marked with “RT”). Tweets also frequently include so-called hashtags, where the # character is in conjunction with a word or a phrase in order to signal a theme. The use of hashtags allows for searching the “Twittersphere” for tweets of specific interest. In order to study the use of Twitter during the 2010 Swedish election, the hashtag #val2010 (Swedish for #election2010) was identified as the most commonly used for purposes of discussing the upcoming election. This hashtag, then, served as a guide for data collection. With election date set to September 19th, the process of archiving tweets was initialized on August 17th and aborted on September 22nd. This particular time period was selected in order to capture the final weeks of election campaigning, as well as some of the post-election activity. By utilizing the YourTwapperKeeper application (TwapperKeeper, 2010), data could be easily archived and downloaded for analytical purposes. In total, 99 832 53

tweets were archived during the specified time period and thereby made subject to analysis. Analysis was performed looking into the top uses of each category of tweets – Singletons, @ messages and retweets. By focusing specifically on such advanced users and visiting their Twitter profile pages, we were be able to say something about who these users were – where they to be considered as established in civil society, or could they be said to constitute new voices, taking part in online discussions? First, though, an overarching analysis showed that over half of the tweets sent during the specified time period were sent on election day – specifically, most of these tweets were sent on election night, supposedly when users where taking part of the post election media coverage. Moreover, the majority of tweets sent during the entire time period were singletons (N=60 088, 60,2%), followed by retweets (N=32 780, 32,8%) and @ messages (@; N=6 964, 7%). With these results in mind, the claim that Twitter has become ‘more conversational’ (Honeycutt and Herring, 2009: 10) would seem invalid in the context at hand. Moving beyond these more overarching results and focusing instead on the specific Twitter practices as identified above, we turn first to the practice of sending undirected messages, i.e. singletons. While the top user identified in this regard can be classified as providing political comedy (its name, “all_insane” is a play of words hinting towards “alliansen”, the collective label for the right-wing alliance of four political parties that have collectively ruled Sweden for two consecutive terms of office), the remainder of users identified here tended identify themselves on their Twitter profile pages as politicians, journalists or political bloggers – arguably users who could be defined as at least somewhat established in Swedish public debate. Second, for the practice of sending directed messages by means of the @ character, the open source software package Gephi (Bastian, Heymann and Jacomy, 2009) was employed. By using social network maps created with Gephi, the top users of the @ character could be easily discerned and classified according to their specific use patterns. Drawing on established terminology (i.e. Shannon and Weaver, 1949), these top users were classified as either senders (meaning that their frequency of messages sent were relatively higher than messages received), receivers (frequency of messages received relatively higher than messages sent) or sender-receivers (user appeared as more reciprocal in sending and receiving @ messages). Results indicated that while the sender group consisted almost exclusively of political actors on the national level, the two remaining groups also included users identifying themselves as journalists, political bloggers and politicians at local and regional levels.

54

Third, Gephi was further utilized in order to analyze the retweeting networks inherent in the collected data. Applying a similar approach to the one described above, a social network map of the top RT users was produced. Based on the results from the social network analysis, and applying a similar scheme to the one discussed above, the top users of the retweet functionality were then classified as belonging to one of three overarching groups. Retweeters tended to retweet the messages sent by others more often relative to their own messages being redistributed. Elites were so labeled based on their tendency to be retweeted relatively more often than they themselves redistributed messages sent by other users. Finally, networkers appeared more mutual in their retweeting practices. In the categories of Retweeters and Networkers we found mostly users affiliated with niche political parties (i.e. the Pirate party), anonymous pundits or journalists, whereas the accounts classified as Elites tended to belong to well-established political parties, journalists or IT professionals. In sum, the overall results presented in study IV indicate that the core users of the #val2010 hashtag employed quite diverse uses and engaged themselves in different network connections with each other. However, it became apparent from the data presented here that many of the identified top users can be said to enjoy rather privileged positions in their respective professional capacities of journalists, politicians and so on. Taken together, then, the findings from study IV indicate that Twitter falls somewhat short of the expectations held by those most enthusiastic on behalf of the democratic and disruptive potential of new web tools. Be that as it may – the findings also indicate the potential of the Twitter platform as a means of outreach for minor, partly marginalized actors. While major political parties and actors appear to have a hard time adapting to the reciprocal nature of @ message and rewteeting practices using Twitter, these identified means of conversation and networking appear to play some part in the use patterns of minor political actors as identified in this study. Study IV finishes up with a discussion regarding possible directions for future research into political (as well as other forms of) microblog research. Arguably, the work presented in the study provides valuable insights into usage patterns when it comes to political Twitter use. However, given its methodological approach, it fails to take into account what is actually being said in the messages sent via the Twitter platform. As such, it could very well be feasible to supplement the structural approach employed here with more qualitative advances, focusing on the semantics rather than on the structural. Finally, we argue that it is important to acknowledge the relative scarcity of Twitter users. What we are studying are advanced Internet users and their patterns of dissemination and interaction. Nonetheless, by studying early 55

adopters (e.g. Rogers, 2003) like these, we might be able to get a glimpse of how use of microblogs or other, similar services will be shaped in the future – and of how relatively marginalized political voices can, at least partially, make themselves heard on a somewhat wider scale.

56

Structuration theory

In this chapter, I present and discuss the theoretical perspective to be employed for the overarching analysis of the empirical studies that make up the bulk of the thesis at hand. As previously mentioned, I have employed structuration theory to guide the analysis. Developed by British sociologist Anthony Giddens, structuration theory has travelled well beyond the disciplinary boundaries of sociology and has provided insights for researchers in a multitude of other fields, such as information systems. The apparent popularity of structuration theory does not mean, however, that its use is entirely unproblematic. In the following, I focus on both the suggested strengths and weaknesses of structuration theory. I also provide a discussion regarding my own use of the theoretical perspective, and how it relates to my chosen research cases. ’We should see social life, not just as society out there or just the product of the individual here, but as a series of ongoing activities and practices that people carry on, which at the same time reproduce larger institutions.’ (Giddens and Pierson, 1998: 76)

Efforts in the social sciences are growing ever more interdisciplinary and the field of information systems (IS) is no exception. In order to grasp the fastpaced developments of the online world, insights from more than just one scholarly tradition are duly needed (Orlikowski and Barley, 2001: 145). While this thesis has its starting point within the IS discipline, my aim here is to provide interesting results for scholars working in other fields and disciplines as well. With this in mind, I have adopted a broader scope in my search for theoretical guidance. Indeed, Pozzebon et al (2009: 2) point out that IS researchers have often sought theoretical inspiration from related scientific fields, engaging with or adopting specific thoughts and arguments in order to assist in their endeavors (see also Baskerville and Myers, 2002). Similarly, Evans and Brooks (2005: 216) mention the usefulness of adopting social theory into the IS discipline. Of particular interest, it would seem, has been the work of British sociologist Anthony Giddens, professor emeritus at the London School of Economics. While many facets of Giddens’ diverse body of work has been adopted by IS researchers (Pozzebon, 2005), it is his theory of structuration 57

that appears to have enjoyed the most popularity in the IS field. The claim by Evans and Brooks that ‘social theory in general and structuration theory specifically have been used extensively in IS’ (2005: 215) is further strengthened by a number of literature reviews on the use of structuration theory within the field (e.g. Jones, 1997; Jones and Karsten, 2008; Rose, 2000), altogether giving further merit to the view expressed by Jones and Karsten: ‘the suitability of structuration as a vocabulary for understanding IS phenomena should […] by now be a matter of record.’ (2008: 149). Although the advantages of approaching empirical data with structuration theory appear obvious, Giddens’ writings on the topic have been subject to a fair share of criticism. Before discussing the more specific points of criticism, the next section will provide the reader with an overview of structuration theory as conceptualized by Anthony Giddens.

The duality of structure First, a point of clarification needs to be made. The intention here is not to provide an extensive, in-depth account of structuration theory and its inner workings. Aside from Giddens’ original accounts (1979, 1984; Giddens and Pierson, 1998) a number of authoritative texts on the topic are readily available (e.g. Jones, 1999; Loyal, 2003; Stones, 2005). Rather, what follows will focus on the specifics of the theory that have proven most relevant to IS researchers in general, and to the research topic dealt with here in particular. As such, I follow the example set by Andersson (2010: 67) who used structuration theory as a ‘tool-box […] pick[ing] concepts and levels of granularity’ at will – a view seemingly also proposed by Giddens himself (1989: 294; see also Rose and Scheepers, 2001). In a series of texts dating back to the late 1970s and early 80s, culminating in the publication of The Constitution of Society in 1984, Anthony Giddens outlined the theory of structuration. The theory was developed in order to address what is perhaps one of the most permeating and challenging issues within the social sciences (Cohen, 1989; Giddens and Pierson, 1998): the roles and respective hegemony of the structuralist and hermeneutic schools of thought. Specifically, Giddens set out to ‘put an end to each of these empirebuilding endeavors’ (1984: 2) by recasting the mutually exclusive dualism of said perspectives as a recursively constituted duality of structure. While functionalist or structural traditions of scholarly thought generally favors structure or the macro level as having more influence over social change (Giddens, 1993: 4), the opposite approach of a hermeneutic or interpretative 58

school of thought places more power in the hands of the individual human agent. Both directions were deemed as equally inept to explain the complex processes of societal development. Instead, Giddens suggested that insights from both views are duly needed in order to provide a fuller picture. Enter Giddens’ concept of the duality of structure, basically stating that the two schools of thought briefly introduced above essentially feed into each other rather than being mutually exclusive. Specifically, Giddens acknowledges that while human agents always act within structures that simultaneously constrain and enable human activity (Giddens, 1984: 177; Pozzebon, 2005: 1358), the structures themselves are dependent on the human activity they govern. Structure can thus be understood as made up of ‘rules and resources […] recursively implicated in the reproduction of social systems’ (Giddens, 1984: 185). As for rules and resources, these are described as the ’techniques or generalizable procedures’ that agents apply in their ongoing enactment of social structure (Giddens, 1984: 21). Rules, then, can be explained as explicit or implicit formulas for action, whereas resources are defined as what agents bring into this action (knowledge, abilities etc.). Hence, structure provides agents with a way of doing things in the more general sense – or with a way to ‘go on’ in life as Giddens puts it himself (1984: 21). As mentioned above, the longevity of structure depends on how it is enacted by agents – how we as humans choose to engage with more or less established notions of how to do things in social life. Since structure is understood as a product of ongoing, repeated human action, it does not exist per se. Rather, structure is understood as a collective memory of established ways of doing things, as ‘instantiations in [reproduced social] practices and as memory traces orienting the conduct of knowledgeable human agents’ (Giddens, 1984: 17). Thus, structure is not visible in itself as it is essentially ’all in our heads’ (Andersson, 2010: 56) – individually and collectively. As such, while the structures that govern are essentially invisible to the human agents involved in these ongoing processes, we can, however, discern the traces and results of structure, its instantiations in everyday life. As structures depend on agents to reproduce them in order to prevail, it follows that agents, according to Giddens, are granted the potential to transform structures by employing the provided rules and resources differently by re-enacting the structure in different ways. This capacity to make a difference stems from the view proposed by Giddens of human beings as knowledgeable and capable agents. Rather than being at the mercy of hegemonic structural properties, agents are seen as very conscious of ‘a great deal about the conditions and consequences of what they do in their day-to-day lives’ (Giddens, 1981: 281), and as capable to at least potentially act in a different way than the structure at hand would immediately suggest. This ‘transformative capacity’ (Giddens, 1984: 15) of the agent is perhaps best understood on 59

a more philosophical level, meaning that in essence, ‘even the most insignificant person on earth has some capacity to make a difference’ (Andersson, 2010: 56). The available literature provides numerous examples on how structuration plays out in everyday life – each more imaginative than the other. For example, Rose and Scheepers (2001) interpret a practice very well known to many academics – that of the conference presentation – using structuration theory. The authors argue that as a conference delegate, you draw on certain rules (in this particular case, rules could be interpreted as the time limit posed upon you by the conference organizers, or the dress code presumed to be adhered to whilst presenting) and resources (for example, access to and skills in using certain software [e.g. PowerPoint or Prezi]) when presenting. The assumption is that the other delegates, as well as the presenter, have a shared understanding of how a conference presentation ought to be performed with regard to such aspects as dress codes, ways of organizing and interpreting the ideas being presented, explicit or implicit codes of seminar behavior, and so on. By performing the presentation in accordance to the statutes of the structure, it is strengthened through re-enactment. However, should the conference delegate choose to diverge from established ways of presenting, ever so slightly, the potential for change at the hands of agents has been acted upon. The structure has been reproduced differently, potentially inspiring other presenters to diverge from the norm and, over time, contributing to changing the structure itself. While the example above will hopefully clarify the internal mechanisms of structuration-at-work, processes of structuration can also be assessed by means of the commonly used illustration available in Figure 2.

Structure

Signification

Domination

Legitimation

Modality

Interpretative scheme

Facility

Norm

Interaction

Communication

Power

Sanction

Figure 2. The dimensions of the duality of structure (after Giddens, 1984: 29)

60

Figure 2 provides an outline Gidden’s original account, which proposes that structure on an analytical level can be understood as consisting of three major dimensions (as shown in the first row of Figure 2, labeled Structure): Signification, Domination and Legitimation. Each structural dimension has a corresponding dimension of agency (as shown in the third row, labeled Interaction): Communication, Power and Sanction. The middle row, labeled Modality, is where agency and structure meet and where, essentially, the duality of structure can be made visible through the inter-linkage and reproduction of agency and structure. As such, agents ‘draw upon the modalities of structuration in the reproduction of systems of interaction, by the same token reconstituting their structural properties’ (Giddens, 1984: 29). The three dimensions pertain to different aspects of structure, where Signification deals with meaning; Domination with power as exercised through control of resources; and Legitimation with moral orders like norms, standards or rules. As the arrows in the figure suggest, the dimension are ‘inherently intertwined’ (Andersson, 2010: 57) and thus separated only for analytical reasons. While the discussion regarding the relationships outlined in Figure 2 have hitherto been rather general and non-specific regarding the use of these concepts in the thesis at hand, a more precise treatment of these terms will follow at the end of this chapter, utilizing also the views put forth by Orlikowski (2000). As pointed out by Rose and Scheepers, the illustration in Figure 2 is often employed as a starting point for analysis using structuration theory, regularly using the concepts discussed in the figure as a checklist of sorts (2001: 223). Even if such an exact use will not become an actuality in this thesis, the figure has provided conceptual tools for other researchers interested in the theory – conceptual tools that will be used also in this thesis. Summing up, then, Evans and Brooks (2005: 215) stated that ’in and through their actions, actors produce the conditions that make their activities possible’, pointing to the duality of structure and action that is at the very core of structuration theory. Next, the concepts of “routines” and “institutions”, central to this thesis, will be dealt with.

Routines and institutions ‘In structuration theory “structure” is regarded as rules and resources recursively implicated in social reproduction; institutionalized features of social systems have structural properties in the sense that relationships are stabilized across time and space.’ (Giddens, 1984: xxxi)

61

As discussed above, structuration theory posits that agents have the possibility to reproduce structure differently, thereby potentially contributing to alter a specific structure. However, as social structures grow more stable and human agents become comfortable within those structures, most of the actions undertaken by agents take on the form of routine - ‘the habitual, taken-forgranted character of the vast bulk of the activities of day-to-day social life’ (Giddens, 1984: 376). Most of our time, we repeat the same or similar actions over and over again – go to work, eat lunch, work some more, have dinner, go to sleep – these are the routinized facets of the durée or the temporality of day-to-day life. Indeed, routinization can in this regard be understood as ‘integral to the continuity of the personality of the agent’ (Giddens, 1984: 60). Such predictability can be said to provide agents with ‘ontological security’ (Pozzebon, 2005: 1357). Thus, routines help us make sense of the world – not only in our immediate environment, but also on a more overarching, macro level. The concept of “institution” can be understood as a specific set of enduring routines – as ways of doing things that have been prevalent for an extended period of time. In principle, this means that the further that social practices extend through time and space, the more common they become over geographical and temporal distances, the more established they become – and the more likely to be regarded as institutionalized features of social life. This postulated relationship and its combined effect on the stabilization of social practices can be demonstrated with a simple, albeit clarifying illustration, as seen in Figure 3.

62

Space Widespread

Time Brief

Enduring

Local

Figure 3. Illustration of social practices stabilizing through time and space (Rose and Scheepers, 2011: 225)

The processes depicted in Figure 3, constituting the longue durée or temporality of social institutions, can be exemplified with some of the established practices of western society like market capitalism, political elections or the media sector – all of them understood as structures, in turn dependent on the actions of agents, having enjoyed longevity throughout the time-space continuum illustrated in Figure 3. As the “embeddedness” or bite of such structures increase, they gradually reach institutional status.

Critique While structuration theory has proven to be a powerful tool to illuminate empirical situations and to analyze complex data (Walsham and Han, 1991), it has also received its fair amount of criticism. Aside from what could be classified as more general comments, suggesting difficulties in grasping the basic tenets of the theory (e.g. Evans and Brooks, 2005: 219; Jones and Karsten, 2008: 129), one especially pertinent area of critique is focused on the theory’s perceived “black-boxing” of social mechanics as discussed by, 63

among others, Archer (1996). By proposing the duality of structure, Archer argues, Giddens does little to advance our knowledge about the relations between the macro and micro levels of structural conduct. On the contrary, structuration theory merely ‘throws a blanket’ over the concepts structure and agency, effectively preventing ‘investigation of what is going on beneath’ said blanket (Archer, 1996: 688-689). The problem described above is sometimes referred to as conflation, pointing explicitly to the difficulty within structuration theory of documenting an institution separate from agency (e.g. Barley and Tolbert, 1997). As pointed out earlier, institutions can be popularly described as “established ways of doing things” – a description that, although perhaps rather simplistic, appears to work well with Gidden’s own conceptualization as outlined in the present chapter. In relation to this, Giddens’ views on structures and institutions as, essentially, collective mental constructs, has been described by Thompson as loose and abstract (1989). Other critics chime in, arguing that these views proposes institutions as being something ‘never concrete’ and ‘essentially processual’ (Archer, 1996: 689), making it difficult to “pin them down” for the purposes of analysis. Admittedly, Giddens somewhat fluid account of social interaction and the structure it pertains to – ‘understood as the fitful yet routinized occurrence of encounters, fading away in time and space, yet constantly reconstituted within different areas of time-space’ (1984: 86) – can be seen as both a ‘theory of hope’ (Andersson, 2010: 57), suggesting that structural change is possible, or as ontologically obtuse (e.g. Archer, 1996). All in all, these and other criticisms collectively build up to a perceived difficulty in applying the theory in empirical situations (Archer, 1995), causing Pozzebon to pose the question of whether Giddens’ thoughts make out the ‘boon or bane’ of empirical research (Pozzebon, 2005: 1355). Indeed, Giddens himself acknowledges that the theory touches upon what might be classified as more philosophical matters (Giddens, 1989), and other writers have suggested that structuration should be understood as a ‘meta-theory’ of sorts (e.g. Jones and Karsten, 2008: 134) – as a way of seeing social practice (Orlikowski and Robey, 1991), as a rationale for approaching empirical data or as a way of thinking about the world, rather than as a fixed set of empirically testable propositions, readily available for operationalization (Weaver and Gioia, 1994; see also Gregor, 2006). These views apparently resonate well with what seems to be the original intent of the author himself. Indeed, Giddens has suggested that theories like the one under scrutiny here should be viewed more as ‘sensitizing devices than as providing detailed guidelines for research procedure’ (1989: 294). Such sensitizing devices could help guide the researcher in identifying particularly interesting empirical phenomena (Myreteg, 2007), and could moreover function collectively as a ‘source of concepts’ (Pozzebon, 2005: 1368) 64

for understanding the various processes at play in specific empirical contexts. As pointed out earlier, these uses of structuration theory have evidently been fruitful for researchers in IS as well as for scholars working in other disciplines. Certainly, Stones (2005) argue that Giddens’ own preoccupation with ontology-in-general (as opposed to ontology-in-situ) does not necessarily disqualify structuration theory from having an important role to play in more situated analytical contexts. In sum, then: while the tenets of structuration theory provides a set of powerful conceptual tools with which to zone in on and carefully analyze specific empirical material and contexts, it cannot be considered a flawless social theory. Then again, is there such a thing as a flawless, perfect theory? While different theories help us make sense of the world and our objects of study in different ways, help us tell different stories, it is important to clearly acknowledge the anomalies that one finds within each theoretical perspective, and the critique that each perspective is thus deserving of. Given the focus of this thesis on societal institutions, clearly a key aspect of the theory under scrutiny here, the decision was made to employ structuration theory to help inform my empirical results on a theoretical level. The meeting places between structure and action, the modalities, have been of specific interest in the analysis. Also of relevance for the decision to make use of the theory was the fact it has been proven and appropriated for the social scientific study of Information- and communication technologies (ICTs), as discussed in the next section, which is focused on the situated use of structuration within IS and related fields.

Structuration Theory and ICTs ’[…] as they do these things in relation to machines and so forth these are the stuff out of which structural properties are constructed’ (Giddens and Pierson, 1998: 83)

As a general social (meta-)theory rather than a theory specific to the study of Information- and Communication technologies, structuration theory is usually amended or adapted in some way before applied in specific empirical or disciplinary contexts. For social scientists interested in the societal role of ICTs, Giddens’ apparent ‘neglect of the technological artifact’ (Jones and Karsten, 2008: 128) has constituted some controversy in adapting the theory. Notwithstanding some sparse commentary on the digital economy and knowledge society (e.g. Giddens and Pierson, 1998; Hutton and Giddens, 2001), Giddens’ writings in general and about structuration in particular 65

have been described as unspecific about technology (Monteiro and Hanseth, 1996). While there appears to be some alleged thematic difficulty in relating structuration theory to the study of ICTs, the quote presented at the beginning of this section has not only provide this thesis with its title, but it is also pointed out by Jones and Karsten (2008: 150) as indicative of the roles of technology as implicated in the reproduction of structure. It follows from this that “Machines”, computers, are pointed to by Giddens as conceivable resources for human agents undertaking the ongoing processes of structural conduct. As agents “do things” with computers and other ICTs, they are provided with a new means for structural reproduction – and possibly also structural change, through their repeated actions. The quote indeed implies the usefulness of structuration theory for the study of technology use, a potential that has not been overlooked. In particular, the potential of structuration theory to further understanding in situations characterized by technological change has been singled out (Gynnild, 2002; Wanda J. Orlikowski, 2000). Given the topic of the thesis at hand, then, the theory seemed appropriate for further scrutiny and application. Within the IS field, as well as in other, related fields, two specific appropriations of structuration theory can be distinguished. The first of these asserts a view of structure as embedded in technology, suggesting that structural properties are inscribed in IT artifacts, effectively setting up “rules” for how said artifacts can and should be used. Examples of scholarly work employing such an embedded view include a series of studies following the approach of Adaptive Structuration Theory (AST), as penned by DeSanctis and Poole (e.g. DeSanctis and Poole, 1994; Poole and DeSanctis, 2004). Theorists inspired by AST work according to the notion of structures as existing independent of agents. Thus, when an IT artifact is fashioned, it is designed with a specific intended use. Given these inscribed starting points, analysis of technology use can be understood as “faithful” or “unfaithful”, and focus is placed on the degree to which a technology is used accordingly, and the degree to which this use matches the specified expected outcomes (DeSanctis and Poole, 1994: 130). Of particular interest here is the work of Wanda J. Orlikowski, who over the course of ‘two seminal papers’ (Pozzebon, et al., 2009: 4) published in Organization Science (1992, 2000) developed her view of structuration theory in relation to ICTs and organizations. Arguably starting out with an embedded view as discussed above, Orlikowski’s original 1992 account takes as one of its focal points the view that designers ‘build into technology certain interpretive schemes […], certain facilities […] and certain norms’ (1992: 410), effectively subscribing to the embedded view of technology as isolated from human agency and viewing the IT artifact as a finished prod-

66

uct, that by way of its design stipulates certain forms human agency (e.g. Barley, 1986; Walsham, 2002). Arguably, the embedded view described above differs from Giddens’ original account of structuration theory, most importantly in relation to the role of human agency. While an embedded approach usually recognizes that agents are not entirely at the mercy of the structures rooted in technology, the basic tenet of structures as embedded appears to diverge quite significantly from Giddens’ position that structure is virtual and dependent on its recursive relationship with human agents – the duality of structure. Indeed, Jones point out that ‘to suggest that structure may be somehow fixed into the technology is to separate it from agency’ (1999: 127), with such a view effectively turning Giddens’ suggested duality back into oppositional dualism. Moreover, in one of his rare direct statements relating to technology, Giddens points out that it ‘does nothing, except as implicated in the actions of human beings’ (Giddens and Pierson, 1998: 82). Thus, while an embedded view of structure has provided scholars with useful insights into empirical phenomena (as suggested by Jones and Karsten, 2008: 152; Orlikowski, 2000: 404) such a view does not appear to remain faithful to the foundations of the original theory (Evans and Brooks, 2005: 219; see also the recent debate in MIS Quarterly: Jones and Karsten, 2008; Jones and Karsten, 2009; Poole, 2009). Remembering that while users can and do use technological artifacts in accordance with how they were designed, users also frequently exhibit tendencies towards circumvention of technological properties, non-use or even inventiveness, coming up with new and perhaps unforeseen ways of employing ICTs for their professional or leisure-time purposes. Accordingly, the second appropriation of the theory at hand suggests an emergent view of structure. It takes as its starting point the human agent and how s/he ‘enacts emergent structures through recurrent interaction with the technology at hand’ (Orlikowski, 2000: 407). Here, the later work of Orlikowski takes a leading role. In her influential 2000 paper Using technology and constituting structures: A practice lens for studying technology in organizations, she describes her revised approach, arguably very different (Halperin, 2005; Jones and Karsten, 2008) from the 1992 account briefly described above:

67

‘When humans interact regularly with a technology, they engage with (some or all of) the material and symbol properties of the technology. Through such repeated interaction, certain of the technology's properties become implicated in an ongoing process of structuration. The resulting recurrent social practice produces and reproduces a particular structure of technology use. Thus, structures of technology use are constituted recursively as humans regularly interact with certain properties of a technology and thus shape the set of rules and resources that serve to shape their interaction.’ (Orlikowski, 2000: 406-407)

Thus, as agents “do things in relation to machines”, these machines become an integral part of structural reproduction, where agents act in relation to ‘the institutional contexts in which they live and work, and the social and cultural conventions associated with participating in such contexts’ (Orlikowski, 2000: 410). Certain practices, performed with or without computers at hand, might be seen as faux pas or inappropriate within certain contexts. The knowledgeable, reflexive agent, as suggested by Giddens, will act accordingly. While this view of technology as gaining structural properties only after being adopted in practice is more in line with Giddens’ original account than the embedded view, it certainly needs to be put in relation to what we can could the physical world and the properties thereof. As Giddens puts it: ‘you can’t just walk straight through a wall’ (Giddens and Pierson, 1998: 82). It follows from this that each IT artifact provides specific properties, suggesting the physical (or digital, for that matter) limits for what can and cannot be done with it. However, contrary to the embedded view, the use of these properties is not predetermined (meaning that there are no “right” or “wrong” modes of usage) and as such, it does not structure human behavior until it comes into play at the hands of agents. As an example, Orlikowski notes that while software tools such as word processors and spreadsheet applications are mainstays in almost every personal computer, our regular day-to-day use of such tools typically involves us utilizing ‘at best, 25 percent’ of the functionalities made available to us by them. She further argues that the development of the World Wide Web at the hands of its users must be considered to have reached well beyond the expectations of several early Internet pioneers (Orlikowski, 2000: 407). With the popularity of the web 2.0 dictum in mind, the traditional roles of designers as well as users of technological artifacts, such as web pages, could be expected to be subject to change (e.g. Sjöström, 2010). Given the emerging approach of Orlikowski’s appropriation of structuration theory, she proposes a “practice lens” through which to view technology use (or non-use) as implicated in the continuous, recurrent recreation of structure. This framework of “technologies-in-practice” is based on Gidden’s 68

Agency

Structure

original account of the dimensions of the duality of structure as outlined previously, and can perhaps easiest be described as an adaptation of said dimensions to the context of ICT use. Orlikowski’s model over the enactment of technologies-in-practice is depicted in Figure 4.

Technologies-in-Practice (rules and resources instantiated in use of technology)

Facilities e.g. hardware, software

Norms e.g. protocols, etiquette

Interpretive Schemes e.g. assumptions, knowledge

Other structures enacted in the use of technology

Ongoing, situated use of technology

Figure 4. Enactment of technologies-in-practice (Orlikowski, 2000: 410)

Much like Gidden’s 1984 outline, Orlikowski places the structural level at the very top of her model. For Orlikowski, then, technologies-in-practice is here understood as the structure, regulating and at the same time dependent upon the ongoing employment of rules and resources instantiated in the use of technology. As discussed earlier, structure is dependent on modalities, where agents interact with the rules and resources made available to them by structure. Building on Giddens original theorizing, Orlikowski further specifies and contextualizes the three types of modalities, which together make up the links between agency and structure. First, Facilities refer to the properties of the hardware and software under scrutiny – ‘those provided by its constituent materiality, those inscribed by the designers, and those added on by users through previous interactions’ (Orlikowski, 2000: 410). Second, Norms refer to protocols and etiquette pertaining to technology use, with users drawing on their respective contextual obligations as well as related ‘meanings, habits, power relations [and] norms’ (Orlikowski, 2000: 410) when appropriating technology. Finally, 69

Interpretative Schemes refer to the ‘skills, power, knowledge, assumptions, and expectations about the technology and its use’ (Orlikowski, 2000: 410) that the agent could be expected to have. Following this outline of structuration theory in general and as employed by Orlikowski, the use of and contribution to structuration theory within this thesis is discussed in detail in the next and final section of the current chapter.

Use of and contribution to structuration theory In one of the more recent literature reviews on the use of structuration theory within the IS field, Jones and Karsten outline several ‘opportunities for future structuration IS research’ (2008: 147-149). In the following, I touch upon some of the opportunities identified by these and other authors, discussing how the thesis at hand relates to some of the suggestions made. Specifically, I will focus on three areas of recommendations: The first dealing with how structuration theory is understood and used, the second relating to research scope, and the third pertaining to opportunities for interdisciplinary efforts. The first suggestion for future research that this thesis acts upon deals with how structuration is used – specifically, studies are suggested to approach analysis through ’sympathetic but critical engagement with Giddens’s work […] without abandoning its central claims’ (Jones and Karsten, 2008: 148). With this claim and the discussion provided above in mind, I employ the view of structure as emergent from technology. My intention with this is based not only on the fact this approach appears to remain more faithful to Giddens’ original account, and that it will allow me to draw from the insights of Orlikowski’s 2000 interpretation, but also that an emergent view provides a more suitable starting point to theoretically gauge both use and non-use of technology (as pointed out by Andersson 2010: 59-60). The second suggestion concerns research scope in relation to employment of the theory. Jones and Karsten point out an interesting dilemma in relation to this: while structuration theory was constructed as a general theory of society with its analytical focus arguably at the macro level, most structurational IS research has placed its focus elsewhere, paying ’little attention to the broader social context of the phenomena that it addresses’ (2008: 139). Indeed, it should come as no big surprise that most structurational IS research as well as most research in IS in general appears to have been performed at the meso or organizational level (Evans and Brooks, 2005: 215; Rose and Scheepers, 2001: 225). While the need to understand the influence of ICTs in social life and not only within organizations has long been recog70

nized (e.g. Walsham, 1993; Walsham and Han, 1993), it is only recently that such studies have become more pertinent (Pozzebon, et al., 2009: 2). By placing the analytical focus of this thesis on the societal or macro level, I contribute to a better understanding of the supposed role that ICTs are sometimes thought of as having in reproducing structure. While I cannot claim to present an exhaustive, all-encompassing account of such practices, the empirical material presented in this thesis should suffice to provide the research community with at least some insights into how these mechanisms come into action. The third suggestion relates to disciplinary alignment – specifically, the already suggested increasing need for interdisciplinary insights in order to provide richer understandings of societal phenomena like those under scrutiny here. This relates to the debate surrounding the status of the IS discipline as a reference discipline or not (e.g. Baskerville and Myers, 2002). Without recapitulating this debate in toto, it would seem that an engagement with the original tenets of structuration theory would help interested researchers connect with debates in other research fields, possibly encouraging ‘reciprocal exchange with other fields that have similarly sought to apply structuration in their work’ (Jones and Karsten, 2008: 152). As pointed out above, while Orlikowski’s work emanated from organization studies, her insights and findings have been useful for many researchers outside this particular field. Since the issues discussed in the present text evolve around areas often dealt with by related disciplines (e.g. media studies, communications and political science), my hope is that the work presented here can provide further linkage between the IS discipline and what I see as my three reference or aligned disciplines. Thus, I would like to suggest an expansion of the claim by Orlikowski and Barley that ‘much can be gained from greater interaction between IT [Information Technology] and OS [Organization Science]’ (2001: 147) to also include other related disciplines into the fold. Research on ICTs is not performed exclusively within the IS discipline. By opening up the discipline to outside influences, much like has been done before, researchers will be provided with complimentary sets of terminologies and conceptual tools with which to further their endeavors. Of course, such openness is not without difficulty. However, given the proper ontological and epistemological alignments across disciplinary borders, I feel that the benefits of interdisciplinary work are too promising not to be acted upon in some way.

71

‘Most people don’t do things very differently’

In this, the final chapter of the thesis, I first provide a brief summary of the overarching results from the empirical studies. With this as a starting point, I then move on to discuss the presented results utilizing terminology pertaining to structuration theory as conceptual tools. Specifically, I draw extensively on the three modalities of structure in order to frame my empirical results. Following this discussion, the subsequent section discusses the combined empirical results with regard to what Orlikowski has labeled as three types of structural enactment. The aim of the thesis is then reiterated and related to the preceding discussion before the chapter and the thesis itself finishes with a discussion regarding the limitations of the work performed. Finally, I suggest some opportunities for future research efforts on themes similar to those dealt with here. The title of this chapter begs an explanation. As probably expected, it is a quote from an interview, where the interviewee is Wanda Orlikowski, talking about her experiences of doing research on IT and its role during organizational changes. Below you will find the quote set in its original context: ’[…] most people don’t do things very differently. Most of my studies, if you look at them, are more about people doing more of the same. In fact, finding very little degrees of freedom for people, either by choice or – coercion’s a strong word – but by the force of their situation, because of career paths, or because of political pressures, or finding it very difficult to try and do things differently. Very often, even though the espoused reason is to try and change the way we work with this new technology, in reality the practice has often been more of the same with technology.’ (Scharmer, 1999)

In the interview segment, Orlikowski talks about how the bulk of empirical research into ICTs and organizational change has provided little evidence of technology as having a potential for inducing change in work patterns. Of importance here is to remember that bulk of Orlikowski’s research has been performed in organizational contexts. As pointed to in the quote above, and as made evident throughout her body of work, we should not be surprised at the extent to which career paths or internal organizational pressures can take hold over an individual co-worker’s incentive to make novel use of the tech72

nological tools available to him or her. Again remembering that scholars have pointed out the perceived usefulness of structuration theory on the societal level of analysis, I argue that in essence, the same views expressed by Orlikowski in the quote above will be useful when assessing the combined results of the empirical studies presented here. As made evident in the third chapter of this thesis, the included studies have indeed shown that ‘Most people don’t do things very differently’. I should like to build this last chapter around a discussion of these combined results as interpreted using concepts derived from structuration theory. Specifically, the argument made is that we as readers and journalists of online newspapers (relating to studies I and II) or as citizens and politicians (relating to studies III and IV) are not, to any larger extent, using the Internet in order to challenge the existing structures of what could perhaps easiest be described as different types of “sender-receiver” relationships. Instead, we are mostly using new technology to do more of the same, thus, using Gidden’s terminology, largely reproducing the structures pertaining to these activities in a relatively traditional manner, thereby further institutionalizing the media industry and organized party politics. Journalists are still mostly acting as “traditional” journalists; politicians too. The same goes for citizens, understood here as readers (in the context of online newspapers) and citizens (for online politics). As pointed to in the quote from Orlikowski above, structures are not that easily amended – especially not when they’ve enjoyed prevalence for longer periods of time. While these forms of organized human activity have certainly undergone internal developments throughout the years, their basic functionality has been and is still fashioned in pretty much the same way. Thus, structuration theory, and the results presented in this thesis, would suggest that such stable, institutionalized practices do not simply change overnight, or as a result of the introduction of some new technological innovation. Focusing first on the theme of online newspapers, study I dealt with trying to discern what factors seemed to have influence over the utilization of interactive features on newspaper web sites. As shown in the study, the average Swedish newspaper web site tended to use a little more than one third of the features included for analysis in the study. Whether or not this is to be considered a high or low ratio could be subject for debate – the argument could perhaps be made that given the amount of media attention regarding audience participation in journalistic contexts, this figure, along with the more detailed results offered in the study itself, could have been expected to be somewhat higher. The interpretation could be made that while most newspapers still follow what could be described as somewhat traditional journalistic norms, tendencies that challenge these norms are indeed in the works. Consider, for example, the supposed “age effect” reported in study I, suggesting 73

that younger journalists tend to be somewhat more open to utilize more challenging forms of interactive features on the web sites that they manage. While study I suggested that journalists were not utilizing interactivity to degrees that might have been expected, it did not take into account the consumers of the news product – the readers, or more specifically, the newspaper web site visitors. Enter study II, where focus was duly placed on the activities of these visitors. In some ways, this second study yielded results that were reminiscent of the results presented in the first one. While the aims and methodology were arguably different for the two studies, we can clearly see how most of the identified visitor personas appear as mostly passive and uninterested in taking part in what is argued to be the more challenging aspects of online interaction in the specific context, rarely leaving the confounds of what could be described as “audiencehood”. However, much like the tendency for younger journalists to employ more demanding interactive features (as reported in study I), similar variations could be discerned with regards to the readers. Consider The Prosumer, arguably one of the more active identified visitor types. While this particular type only make up for one out of five specific visitor types, it still shows that some newspaper web site visitors are making rather extensive use of the interactive features offered to them. Second, regarding the theme of online politics, study III featured a longitudinal analysis of the web pages of Sweden’s major political parties. While the overall findings indicated that most political parties tended to act somewhat unadventurous with regard to letting their respective constituencies take on more active roles through the web pages during the election campaign, results also indicated some novel approaches to web campaigning. Specifically, while the statistic for involving and mobilizing features clearly fell following election day, these values were on average still kept near the .5 coefficient - indicating that nearly half of these types of features were employed even when no election was at stake. Of course, these reported values are to some extent due to an inflated mean, given the comparably larger scores shown by a few of the studied parties. Nevertheless, it says something about electioneering processes – maybe they have not undergone the radical progressions often pointed to by some pundits, but the results presented here could be seen as examples of ongoing, incremental change. Finally, study IV focused specifically on one of the currently popular social media applications – Twitter – and its use during the 2010 Swedish parliamentary elections. Specifically, tweets containing the relevant hashtag #val2010 (Swedish for #election2010) were archived and analyzed, focusing on mapping out the structural relationships between top users of the service at hand. While the findings indicated that the majority of these top users identified themselves as already having established roles in society, results also revealed that some of these users were affiliated with minor political 74

parties (mainly the Pirate Party), or were otherwise relatively unknown with regard to status in civil society. As such, while much of the “happy-golucky” rhetoric regarding the political and participatory potential of the Internet was at least in part disproven by the study, the potential for nonestablished actors to make their respective voices heard was also clearly discerned. As shown in the studies presented here, as well as in the literature review presented in the introductory chapter of this thesis, journalists, politicians and their respective audiences appear to be having some difficulty in adapting to the suggested new forms of interaction that have come about as a result of the introduction of the Internet. To be fair, there have been important changes in the relationship between senders and receivers, with increased audience interaction in the journalistic process being reported by researchers (e.g. Carpenter, 2010; Steensen, 2009). Similar results regarding political participation have also been reported, as discussed in the introduction of the thesis at hand. In this regard, journalistic as well as parliamentary political practices have apparently undergone changes through the development of online practices. However, as this thesis has shown, the bulk of empirical research on the topics at hand still suggests that the attitudes of practitioners, understood here as politicians and journalists, towards increased user interaction are mostly somewhat conservative. This suggests that both professions, their respective audiences, and thereby societal institutions at large, still have some way to go before their web presences can truly become ‘shared spaces’ (as suggested by Singer and Ashman, 2009: 3), collective products, created in participation with an involved audience. Something is indeed happening – but the structures that be have not, apparently, undergone sufficient amendments yet. Journalists are not the only ones involved in the profession of journalism. There has to be someone on the receiving side of things – someone to consume the journalistic end product. Likewise, political actors need an audience – someone to listen to their message and, possibly, vote accordingly. In the context of this thesis, we cannot forego the web site visitors: the readers, who through online developments such as the increasingly popular Web 2.0 rationale supposedly have more input and power when it comes to amending and contributing to the different online experiences. The research presented here has shown that while opportunities for such visitor inputs could be described as at least somewhat increasing, rather few audience members appear to take the time to engage with the more demanding of such opportunities. Most visitors on newspaper web sites appear content to remain just that – visitors, who appear rather jaded and uninterested in the supposedly ever-increasing opportunities to provide the newspaper 75

with various types of input. While this thesis has not made such in-depth inquiry into the uses of interactive features by visitors of political web sites, the results presented by previous research and in the studies presented here provide quite a similar view when it comes to this particular category of web sites.

Understanding non-use – three modalities of structure Research suggests limited use of interactive features on behalf of both senders and receivers in most of the cases accounted for above. With Orlikowski’s model for technologies-in-practice (reproduced in Figure 4) in mind, the studied relationships between practitioners and audiences can be further undestood through the three modalities she draws from Giddens – Facilities, Norms and Interpretive schemes. As previously mentioned, modalities can be described as the pathway between the structure and the agent, where the interaction between these two levels takes place and is made discernible for analytical purposes. The three modalities, and how they can be comprehended in the context of this thesis, are discussed in the following. As mentioned earlier, Facilities refer to the characteristics of the technical artifacts, the hardware and software, at hand. For media organizations, this would refer to their respective web presences and how they choose to use them. For their audiences, facilities would denote the technical equipment available to them. As for the context of online political activity, a similar sketch could be drawn. Indeed, as computers and Internet access have both become more available and accessible across several socio-economic strata, the assertion can be made that more and more people are getting access to at least some form or level of online experience. The same goes for media companies and established democracies in the western hemisphere, where most research on matters pertaining to this thesis has been performed. Thus, the technical facilities – the hardware and the software necessary for interaction between sender and receiver – are in place, ready to potentially spring into action at the click of a computer mouse button. However, as the results of the empirical studies included in this thesis suggests, these types of interaction do not regularly take place. Be that as it may, we are nonetheless seeing an admittedly less clear-cut sender-receiver relationship, since some ‘power users’ (Orlikowski, Yates, Okamura and Fujimoto, 1995) on both sides have started to amend the structures that govern these everyday processes. Consider, for example, the more active visitor types identified in study II, or the engaged politicians pointed to in study IV. Still, most agents, be they readers or journalists, citizens or politicians, do not appear to use these features to any larger degree. This 76

pattern of mostly non-use appears somewhat akin to Orlikowski’s example regarding the use and non-use of word processors and other, similar applications discussed briefly above, where she suggests that ‘in our regular use of these tools, most of typically utilize, at best, 25 percent of these tools functionality’ (Orlikowski, 2000: 407). It would appear, then, that this approximation is a valid description also for the empirical results presented in this thesis. Why is this? Why are we not seeing more changes taking place in the contexts under scrutiny here? As this thesis has shown, the claims that the Internet would assist in increasing political participation among previously unengaged citizens (e.g. Castells, 2007: 255), giving way to a new era of ‘postmodern campaigning’ (Vaccari, 2008a: 648), were indeed somewhat overstated. Similar findings have been reported regarding the journalistic context, as shown earlier. Indeed, simply because the technology is in place does not mean that change regarding human agency will take place – such an assumption would border on something similar to technological determinism. Perhaps we can move towards an increased understanding regarding these questions by focusing on the modality of Norms – to be understood as established ‘codes of conduct’ as suggested by Orlikowski (2000). As discussed earlier, the norm for journalistic practice still appears to be that of the longstanding, established one-to-many form of news reporting and dissemination. While recent developments have indicated that changes to this work routine might be on their way, due perhaps to changes in journalistic schooling (Deuze and Dimoudi, 2002) or to more general aspects of generational differences (as shown in paper I, this thesis), research has mostly found interactivity to be an ‘uncomfortable myth’ (Domingo, 2008: 688) for many journalists and other media professionals. Similarly, the widespread norm for political practice for both the elected and the electorate still appears to be that of established mass communication from the politician to the electorate. While research has indicated that changes to this regular routine might be on their way, due to ideological differences among politicians (e.g. Gibson, et al., 2008: 26; Vaccari, 2008c: 15), incumbency status or party size (e.g. Druckman, et al., 2007: 429; Schweitzer, 2008: 457; Strandberg, 2009: 837-838) or voter demographics (e.g. Vaccari, 2008b: 75), the research performed, including the results presented in this thesis, has mostly reported on various forms of online interaction between citizens and politicians at rather low levels. Most opportunities for interaction at the hands of politicians could be likened to what Foot and Schneider describe as ‘kissing virtual babies online’ (2006: 108), indicating limited and perhaps somewhat superficial use of the possibilities made available. In sum, the ‘uncomfortable myth’ of interactivity suggested by Domingo (2008: 688) found in newsrooms apparently extends its grasp also to agents 77

within the political context. As the results presented in this thesis has shown, most agents, regardless of context, are not engaging with interactive features to any greater extent. Instead, in our everyday lives, most of us appear to be reproducing what could be labeled as different structures of “audiencehood” – remaining readers and voters, consumers of news and politics, largely uninterested in the possibilities for online interaction with institutions that are, at least in some cases, readily available for us. However, we can also discern changes in this regard. Study I suggested a different, perhaps more challenging, approach to the utilization of interactive features at the hands of younger media professionals. Similarly, age differences where in place also for the different types of newspaper website visitor types discussed in study II. Study IV found that non-established political actors (mostly members of the Pirate Party) tended to employ Twitter in a somewhat different way than their more established counterparts. These findings could be interpreted as ongoing norm changes – while more established actors are largely adhering to habitual norms of structural conduct, we are also seeing gradual changes to these norms, perhaps creating openings for further structural amendments. As for the third modality identified by Giddens and implemented by Orlikowski, Interpretative Schemes denotes the ‘skills, power, knowledge, assumptions, and expectations about the technology and its use’ (Orlikowski, 2000: 410) or the ‘stocks of knowledge’ (Giddens, 1984: 29) drawn upon by the user during technology-supported practice. For the purposes of the topic under scrutiny here, interpretative schemes can be understood as the knowledge and assumptions held by politicians and citizens respectively regarding the concepts and functions of information technology use for political purposes. This particular modality is sometimes understood as the level of technological knowledge held by an individual user regarding a specific system (i.e. Orlikowski, 2000). However, this perhaps slightly limited view misses out on the beliefs, assumptions and perceptions of technology use as held by the user. As shown in this thesis, such underlying beliefs about online political practices are sometimes held high – not only by practitioners and audiences, but also by the many academics interested in these topics. Indeed, while some politicians and citizens might perceive the online realm as carrying with it increased opportunities for interactions and deliberation, these perceptions appear to clash with the norms of technology use for various political purposes as discussed earlier. In other words, while our assumptions regarding the role of technology for such purposes might suggest otherwise, the norms (as discussed earlier) often associated with these uses appear as somewhat restraining. With some notable exceptions, it would seem from the research overview previously offered in combination with the results presented in this thesis that most agents, be they practitioners or audience members, tend to share similar interpretative schemes regarding 78

interactivity in the journalistic and political milieus respectively – signifying assumptions and knowledge of a rather traditional, established variety. Thus, the identified situation for online newspapers could be described as a ‘lack of vision’ (Vobic, 2011) on both sides of the proverbial gate-to-bekept. Indeed, if journalists adhere to traditional norms and schemes, and audience members do the same, referring to user contributions in the online journalistic context as ‘John Doe fussing about’ (Hujanen and Pietikainen, 2004: 396, see also Bergström, 2008), we should not expect the relationship between journalists and readers to be altered dramatically anytime soon. Similarly, in the political context, both involved parties appear as slightly reluctant and as involved in a ‘trial-and-error process of adopting the new media’ (Vaccari, 2008c: 2), indicating a careful approach to potential structural amendment. Given the insights to be gathered from previous studies employing structuration theory, the levels of non-use as opposed to use found in the empirical studies reported in here should not come as much of a revelation or surprise. In the aforementioned interview with Scharmer (1999), Orlikowski indeed suggests that ’[…] most people don’t do things very differently. Most of my studies, if you look at them, are more about people doing more of the same’. This perhaps somewhat glum observation appears to hold true also for the topics placed under study here, as most agents involved in the journalistic or political contexts tend to reproduce established structures rather faithfully, thereby further strengthening the ‘bite into time and space’ (Giddens, 1984: 171) enjoyed by what can be referred to as the institutions of journalism and parliamentary politics respectively – arguably social practices that have been fashioned in similar ways for extended periods of time. The combined results of more then a decade of research into online journalism-reader interaction and online political communication with constituents paint a mostly rather austere picture to those who champion the idea that readers and voters alike necessarily want to be further involved in the processes that shape their everyday news provision and political realities, and that journalists and politicians alike would be able to easily adapt to the new possibilities for interaction offered to them through the Internet. Taking the audience side into consideration, Roscoe argues that ‘we are so used to conceiving of ourselves as consumers and audiences that we tend to impose that identity ourselves onto a new medium without considering the other possibilities it offers.’ (1999: 681). Indeed, similar views are discernable when assessing the institutional side of things, as made evident by the results presented in this thesis. The claims made by Roscoe regarding the supposed potential for audience activation made available by the Internet in 1999 seem equally valid today, more than a decade later.

79

Nevertheless - the research performed here as well as elsewhere has also indicated some potential for structural change at the hands of agents in both contexts. As mentioned previously, the interactive features cannot be understood solely as being ‘rejected bits of program code’ (Orlikowski, 2000: 410), as tendencies for certain agents, be they journalists or readers, politicians or citizens to employ these features at higher degrees were also discerned.

Three types of enactment By now it should be clear that the results presented in this thesis, as well as the results from previous, similar research efforts, does not fully support the repeatedly heard claims that various web design rationales would bring forth further audience interaction in the contexts under scrutiny here – allowing for more audience control over online content, as well as the supposed amended reproduction of structure. However, while approaching these results theoretically using the three modalities of facilities, norms and interpretative schemes as analytical tools allows us to identify specific areas of interest, such an analysis does not provide us with a more overarching theoretical abstraction of how societal institutions make use of their online presences with regard to interactivity. For these purposes, Orlikowski (2000: 421-425) identifies what she labels as three types of enactment as evident from how the modalities specified above play out in specific empirical contexts. These enactments, labeled Inertia, Application and Change, represent different modes or levels of use of specific technologies in different settings. Again, while Orlikowski originally derived the terminology employed here in the setting of organizational research, the argument made in this thesis is that these concepts can be useful also in other settings, such as those dealt with here. In what follows, the three types of enactment are presented in combination with a discussion regarding how they can be understood in the context of the thesis at hand. While the particular names of the enactment types provide some indication as to what they entail, the terms at hand need to explicitly specified. First, Intertia is described by Orlikowski as a type of enactment ‘where users choose to use technology to retain their existing ways of doing things’ (Orlikowski, 2000: 421), utilizing the technologies with which they work in order to do more of the same, thereby reinforcing and preserving a structural status quo. Such conservative use of ICTs can indeed be distinguished in the studies presented here – we can point to the fairly traditional use of the web by political parties as identified in study III, or similar patterns of use as exhibited by newspaper organizations in study I. However, as made clear 80

earlier, we can also discern other, more innovative, modes of usage in the presented empirical material. The second type of enactment identified by Orlikowski is labeled Application – ‘where people choose to use the new technology to augment or refine their existing ways of doing things’ (2000: 422). By exhibiting tendencies of what could be labeled as more creative or innovative uses of the technologies at hand, users reinforce and enhance the structural status quo, challenging certain norms and taken-for-granted routines, while at the same time retaining other aspects of structure. Examples of application are undeniably prevailing in the results presented here – consider, for example, the more active user types that were identified in study II, or the denizens who took to Twitter in order to engage in political discussions as identified in study IV. Be that as it may; we are still, this thesis argues, a long way away from more exhaustive structural transformation. Such patterns of more overwhelming structural change is identified by Orlikowski in the final and third type of enactment, which she appropriately labels as Change. Compared to the two previously presented types, the Change variety entails rather dramatic modes of ‘drawing on and transforming existing institutional, interpretive, and technological conditions over time, and, in this way, significantly changing the organizational status quo’ (Orlikowski, 2000: 423). As such, a more active and thoroughly innovative agent appears to be a prerequisite for this particular mode of enactment to come into play. While the results presented here have indeed indicated some rather apparent challenges to the institutional structural status quo, with examples of practitioners and public taking on new roles in both cases covered being present, these examples are arguably not abundant, consistent or systematic enough to be easily classified as examples of change as defined by Orlikowski. Employing a trichotomy such as the one used here will entail some rather tricky delimitory definitions. Orlikowski herself acknowledges this problem, clarifying that ‘the three types of enactment characterized here are not comprehensive or exhaustive, but suggestive of the kinds of comparisons that may be made across the conditions and consequences associated with people's use of technologies’ (2000: 423). Where, for example, should the line be drawn between enactment as application and enactment as change? Arguably, the latter of these should entail some almost revolutionizing aspect, giving way to structural change. While the inertia type of enactment is perhaps more easily defined and identified, delimitations between the other two types are perhaps not as clear-cut as one might want them to be. While we can find examples of both inertia and application in the empirical studies included here, examples of outright change are arguably few and 81

far between. This is perhaps not very surprising. Considering the previously discussed need for ontological security, the suggestion that we as human agents need somewhat stable routines in order to make sense of our respective environments, this could entail that the structures under scrutiny here would undergo change in a rather gradual manner. Again, consider further the fact that the newspaper industry and organized politics have both functioned in pretty much the same way for considerable amounts of time. These modes of human activity should be considered as stable, strong structures, and to expect an enactment of change simply because a new technology enters into the fray of everyday work and leisure life would be simplistic. The empirical findings identified agents that took the step beyond the recognized structures, agents that through their actions in relation to online newspapers and political actors started to challenge and amend the established ways of doing things. Consider, for example, audience members like “The Prosumer” in the context of online newspapers (study II), or a comparably active Twitter user (a “Sender-Receiver” or a “Networker” in the context of political Twitter use, see study IV). I am also referring to the actors on what could be called the “supply side” – agents that, from their perspective, are amending and challenging the established ways of managing, for example, an online newspaper by letting the readers have more say in a variety of ways (as shown in study I), or those professionals responsible for the web sites of political parties who choose to employ mobilizing and involving features even though no election campaign is under way (see study III). These users, found among both practitioners and their respective audiences, can thus be considered “power users”, as discussed earlier, and merit further research in this area. The main results of this thesis indicate that audience interaction in the professional contexts at hand is perhaps not all that contemporary hype makes it out to be. Indeed, as the title of this final chapter reads, most people don’t do things very differently. In closing, however, one could very well argue that we are at a very early stage when attempting to fathom the overarching societal consequences of interacting with these technologies. To illustrate, the urge to paraphrase Nietzsche is indulged upon - ‘The press, the machine, the railway, the telegraph are premises whose thousand-year conclusion no one has yet dared to draw’ (1996: 378). By applying such a historical perspective on more recent developments, we can, over time, gain insights regarding the longue durée of social institutions like those of journalism and parliamentary politics – institutions that, as this thesis has shown, are not that easily amended.

82

Conclusions This thesis has studied the use and the non-use of what has been described as one of the key features of the Internet - interactivity. As discussed in the introductory literature review, and as dealt with in the individual studies included, the supposed relevance of this interactive potential has been revitalized by notions of "Web 2.0" modes of web design – harnessing user participation in various ways, shapes and forms. This thesis has shown that such activities – at the hands of both audience members and their practitioner counterparts – are perhaps not as easily integrated into established notions of conduct, into the structures guiding everyday activity, as one might be led to believe. To reiterate, the aim of this thesis was formulated as follows: How do practitioners working within societal institutions and their respective audiences make use of the interactive features made available by the Internet? For practical reasons, this aim was broken down into two related research questions: one relating to practitioners working within societal institutions, and one pertaining to their audiences: RQ1: How do practitioners within societal institutions make use of online interactivity? RQ2: How do their respective audiences make use of the interactive features offered to them? By keeping an analytical distinction between audiences and practitioners, the thesis has been able to trace patterns of structural conduct in both groups of agents. As discussed in the previous section, these patterns mostly resemble what Orlikowski (2000) has labeled an application form of structural enactment. This form of enactment ‘results in the reinforcement and enhancement of the structural status quo’ (2000: 422), with traces of novel modes of use woven into the mesh of mostly traditional modes of use (or non-use, for that matter). As such, while both practitioners and audiences appear mostly latent with regard to online opportunities for interaction in the contexts under scrutiny, we are also seeing ‘power users’ (Orlikowski, 2000: 406) at play, who draw on ’existing institutional, interpretive, and technological conditions over time and reproduc[e] them in an enhanced or improved form’ (Orlikowski, 2000: 423). Such power users could be interesting for future research efforts 83

to study closely. This suggestion for future research, along with five others, are further elaborated on in the following, final section of the thesis.

Limitations and suggestions for future research While the individual studies included in this thesis all feature their respective sections dealing with limitations and recommendations for future research endeavors, this final section of the final chapter of the thesis will present some of the research ideas that have grown out of the empirical work performed or dawned on the author during the process of writing. Indeed, more insights are required in order to increase our understanding of the identified processes of ongoing, incremental structural change. Gauging these matters should be of interest to scholars working in a variety of disciplines, as well as to practitioners in the identified contexts. For interested researchers, six suggestions for future research efforts are provided below. First, the need for longitudinal studies in the research contexts touched upon in this thesis should be stressed. While two of the papers included here make such longitudinal efforts (papers III and IV), future research looking into online behavior and trends should attempt to more clearly adopt such a perspective for data collection and analysis. This suggestion could be acted upon in a number of different ways. For example, the empirical case in paper IV was the 2010 Swedish election. Future research could be well advised to choose the approach presented in paper IV (or something akin to it) and attempt to analyze political Twitter use in other, similar or dissimilar countries. By employing similar means of data collection and analysis, researchers will be able to move beyond the “snapshot in time” characteristic that could be said to distinguish many current studies. Second, such longitudinal studies should be performed with careful recognition of previous research. By following in the footsteps of previous scholars with similar interests, we can hone our research questions and data collection rationales so as to build a cumulative, collective body of work. In times and contexts characterized by rapid developments, such as the online milieu, adhering to previous work can appear rather difficult. While Twitter and Facebook are currently arguably two of the more popular online phenomena in western Europe and the US (e.g. Bruns, 2011), this condition is obviously subject to change. A lot of attention is geared towards these two platforms – from the general public as well as from academia. However, five years from now, it is unclear whether or not these services will still be enjoying the popularity we are seeing today – given the recent history of the Internet, it is 84

likely that they will have been replaced with some other, novel services. History, as it unfolds, might prove us wrong regarding this particular prediction. Still, such a premonition should be enough for researchers to place their analytical focus not on the platforms per se, but rather on the behavioral patterns that are expressed on and through them. For example, paper IV in this thesis detailed the political uses of Twitter by focusing on the more active users during the 2010 Swedish parliamentary elections. It did so in a manner that focused more on user’s basic communicative behavior rather than on the specifics of the Twitter service. The terminology and means of analysis used here should be feasible also for future studies, looking into the uses of other Internet services. Just to clarify: while there is certainly a place for studies that delve more deeply into the inner workings of specific online services, researchers should also make sure that their research questions are “stress-tested” for future iterations. Third, this thesis has focused on use of online interactivity by societal institutions, exemplified by online newspapers and online political actors. While these are arguably areas of human activity that are often believed to have been especially affected by the Internet, there are of course many other areas that should be studied in this regard as well. While there certainly is a need to “stay on target” for future research efforts, focusing on the coming developments in the specified areas, the need to branch out and focus on other areas of similar interest should be also acknowledged. Such broadening of research scope might entail looking into what could be described as sub-parts of the identified areas – for example, local and regional government use of interactivity should be of interest. This broadening of scope could also entail looking into other areas of business and civil society, focusing on their ongoing efforts to utilize of the online possibilities for interaction. Furthermore, international comparisons would also be feasible. By comparing results from different geographical contexts, researchers will hopefully be able to speak with more confidence about the matters at hand. Fourth, ‘power users’ (Orlikowski, 2000: 406) or ‘early adopters’ (Rogers, 2003) could be expected to function as opinion leaders or ‘intermediaries’ (Orlikowski, et al., 1995; Woolgar, 1996), providing different “readings” of the technologies at hand, taking the lead in augmenting their respective uses of it. In so doing, they can be said to provide examples of enhanced use, paving the way for challenging established relationships between audiences and practitioners, thus reproducing differently the structures pertaining to these relationships. While we are far away from seeing a clear “structure of prosumerism” in the contexts at hand here, studying such power users more closely could provide the research community with additional insights regarding such practices and the agents who undertake them. 85

Fifth, while the thesis at hand has provided theoretical insights as to why limited amounts of online interaction in institutional contexts take place, it has done so in a somewhat unconventional manner. The theory at hand has not been painstakingly appropriated to a specific empirical context and assessed thereafter. While this was not the goal with the thesis, the suggestion is made that future research look into opportunities to test structuration theory in a more direct, empirical research setting. As already mentioned, Giddens’ apparent ‘neglect of the technological artifact’ (Jones and Karsten, 2008: 128) has resulted in some disagreement as to how the theory could be put to use. Indeed, structuration theory has been described as a ’sensitizing device to view the world in a certain way’ (Klein and Myers, 1999: 75) and thus as hard to operationalize and test empirically. The insights presented here, particularly with regard to the appropriation of the modalities (facilities, norms and interpretative schemes), could prove helpful for future attempts at operationalization. Sixth, and finally, while the theoretical perspective employed in this thesis proved useful in assessing the empirical results achieved, the topic at hand could certainly be approached with a number of different theoretical perspectives at hand. As such, it is not unreasonable to imagine that Habermasian public sphere theory (1989), perhaps one of the more popular theoretical perspectives employed while studying online political action, could serve as a suitable starting point for future research efforts. Similarly, insights from the original texts as well as the appropriations of texts by theorists like Pierre Bourdieu (e.g. Johnson, 2006; Russell, 2007) and Michel Foucault (e.g. Shankar, Cherrier and Canniford, 2006; Willcocks, 2006) have proven useful for analytical purposes, as has the use of actor-network theory (e.g. Latour, 1991; Rose, Jones and Truex, 2005; Schmitz Weiss and Domingo, 2010; Zhang, 2010). Different conceptual tools, derived from different theories, can assist researchers in interpreting empirical results in different ways, thereby providing different views on the same empirical material. Discussing the study of online news, Mitchelstein and Boczkowski state that researchers should ‘[broaden] the aspects of the phenomena that deserve examination, the tools utilized to learn about them, and the analytical perspectives that explain the resulting findings’ (2010: 1094). The suggestion is made here that such broadening should be of relevance also for other fields of inquiry.

86

References

Aikat, D. D. (1998). News on the Web: Usage Trends of an On-line Newspaper. Convergence, 4(4), 94-110. Anderson, N. (2006). Tim Berners-Lee on Web 2.0: “Nobody even knows what it means." Retrieved September 1, 2010, from Ars Technica http://arstechnica.com/news.ars/post/20060901-7650.html. Andersson, A. (2010). Learning to learn in e-Learning : constructive practices for development. PhD thesis, Örebro universitet, Örebro. Retrieved from

http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-10354

Archer, M. (1995). Realist Social Theory: the Morphogenetic Approach. Cambidge: Cambridge University Press. Archer, M. (1996). Social Integration and System Integration: Developing the Distinction. Sociology, 30(4), 679-699. Bacharach, S. B. (1989). Organizational Theories - Some Criteria for Evaluation. Academy of management review, 14(4), 496-515. Bakker, T. P., & de Vreese, C. H. (2011). Good News for the Future? Young People, Internet Use, and Political Participation. Communication Research, 38(4), 451470. Balsvik, G. (2004). Information technology users : studies of self-efficacy and creativity among Swedish newspaper journalists. Stockholm: Economic Research Institute, Stockholm School of Economics. Barley, S. R. (1986). Technology as an Occasion for Structuring: Evidence from Observations of CT Scanners and the Social Order of Radiology Departments. Administrative Science Quarterly, 31(1), 78-108. Barley, S. R., & Tolbert, P. S. (1997). Institutionalization and structuration: Studying the links between action and institution. Organization Studies, 18(1), 93-117. Barnett, S. (1997). New media, old problems - New technology and the political process. European Journal of Communication, 12(2), 193-218. Baskerville, R. L., & Myers, M. D. (2002). Information systems as a reference discipline. MIS Quarterly, 26(1), 1-14. Bastian, M., Heymann, S., & Jacomy, M. (2009, May 17 - 20). Gephi: An open source software for exploring and manipulating networks. Paper presented at the Third International ICWSM Conference, San Jose, California. Baumgartner, J. C., & Morris, J. S. (2010). MyFaceTube Politics: Social Networking Web Sites and Political Engagement of Young Adults. Social Science Computer Review, 28(1), 24-44. Bentivegna, S. (2006). Rethinking politics in the world of ICTs. European Journal of Communication, 21(3), 331-343. Bergström, A. (2007). Internetval 2006 - För de redan frälsta. In S. Holmberg & L. Weibull (Eds.), Det nya Sverige : trettiosju kapitel om politik, medier och

87

samhälle : SOM-undersökningen 2006 (pp. 395-404). Göteborg: SOM-institutet, Göteborgs universitet. Bergström, A. (2008). The Reluctant Audience: Online Participation in the Swedish Journalistic Context. Westminster Papers in Communication and Culture, 5(2), 20. Bimber, B. A., & Davis, R. (2003). Campaigning online : the Internet in U.S. elections. New York, N.Y.: Oxford University Press. Blumler, J. G., & Kavanagh, D. (1999). The Third Age of Political Communication: Influences and Features. Political Communication, 16(3), 209-230. Boczkowski, P. (2002). The Development and Use of Online Newspapers: What Research Tells Us and What We Might Want to Know. In L. Lievrouw & S. Livingstone (Eds.), The handbook of new media : social shaping and consequences of ICTs (pp. 270-286 ). London: SAGE. Boczkowski, P. J. (2004). Digitizing the news : innovation in online newspapers. Cambridge, Mass.: MIT Press. Bowman, S., & Willis, C. (2003). We Media: How audiences are shaping the future of news and information. Reston, Va: Media Center, American Press Institute. boyd, d., & Crawford, K. (2011). Six Provocations for Big Data. Paper presented at the A Decade in Internet Time: Symposium on the Dynamics of the Internet and Society, Oxford, United Kingdom. http://ssrn.com/abstract=1926431 Bruns, A. (2010). From Reader to Writer: Citizen Journalism as News Produsage. In J. Hunsinger, L. Klastrup & M. Allen (Eds.), Internet Research Handbook (pp. 119-134). Dordrecht, NL: Springer. Bruns, A. (2011). How Long Is a Tweet? Mapping Dynamic Conversation Networks Ontwitterusing Gawk and Gephi. Information, Communication & Society, 1-29. Buick, J., Appignanesi, R., & Jevtic, Z. (1995). Cyberspace for beginners. Trumpington: Icon. Calenda, D., & Meijer, A. (2009). YOUNG PEOPLE, THE INTERNET AND POLITICAL PARTICIPATION -- Findings of a web survey in Italy, Spain and The Netherlands. Information, Communication & Society, 12(6), 879 - 898. Carlson, T., & Strandberg, K. (2008). Riding the Web 2.0 Wave: Candidates on YouTube in the 2007 Finnish National Elections. Journal of Information Technology & Politics, 5(2), 159 - 174. Carpenter, S. (2010). A study of content diversity in online citizen journalism and online newspaper articles. New Media & Society, 12(7), 1064-1084. Casey, M. E., & Savastinuk, L. C. (2007). Library 2.0: a Guide to Participatory Library Service. Medford, OR. Castells, M. (2001). The internet galaxy : reflections on the Internet, business, and society. Oxford: Oxford University Press. Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1(1), 238-266. Chadwick, A. (2006). Internet politics : states, citizens, and new communication technologies. New York: Oxford University Press. Chen, W. S., & Hirschheim, R. (2004). A paradigmatic and methodological examination of information systems research from 1991 to 2001. Information Systems Journal, 14(3), 197-235. Chin, W. W., Junglas, I., & Roldán, J. L. (2011). Some considerations for articles introducing new and/or novel quantitative methods to IS researchers. European Journal of Information Systems, 21(1), 1-5. Christensen, H. S. (2011). Political activities on the Internet: Slacktivism or political participation by other means? First Monday, 16(2).

88

Chung, D. S. (2004). Into Interactivity? How News Websites Use Interactive Features. Paper presented at the International Communication Association annual convention, New Orleans, LA. Chung, D. S. (2007). Profits and Perils: Online News Producers' Perceptions of Interactivity and Uses of Interactive Features. Convergence, 13(1), 43-61. Chung, D. S. (2008). Interactive Features of Online Newspapers: Identifying Patterns and Predicting Use of Engaged Readers. Journal of ComputerMediated Communication, 13(3), 658-679. Chung, D. S., & Nah, S. (2009). The Effects of Interactive News Presentation on Perceived User Satisfaction of Online Community Newspapers. Journal of Computer-Mediated Communication, 14(4), 855-874. Cohen, E. L. (2002). Online journalism as market-driven journalism. Journal of Broadcasting & Electronic Media, 46(4), 532-548. Cohen, I. J. (1989). Structuration theory : Anthony Giddens and the constitution of social life. Basingstoke: Macmillan. Coleman, S. (2004). Connecting Parliament to the Public via the Internet. Information, Communication & Society, 7(1), 1 - 22. Coleman, S. (2005). Direct representation: Towards a conversational democracy. Institute for Public Policy Research (IPPR): London. Coleman, S. (2005). New mediation and direct representation: reconceptualizing representation in the digital age. New Media & Society, 7(2), 177-198. Cormode, G., & Krishnamurthy, B. (2008). Key differences between Web 1.0 and Web 2.0. First Monday, 13(6). Curran, K., Murray, M., & Christian, M. (2007). Taking the information to the public through Library 2.0. Library Hi Tech, 25(2), 288-297. Dahlgren, P. (1996). Media logic in cyberspace: repositioning journalism and its publics. Javnost/The Public, 3(3), 59-72. Davis, G. F., & Marquis, C. (2005). Prospects for organization theory in the early twenty-first century: Institutional fields and mechanisms. Organization Science, 16(4), 332-343. DeSanctis, G., & Poole, M. S. (1994). Capturing the Complexity in Advanced Technology Use: Adaptive Structuration Theory. Organization Science, 5(2), 121-147. Deuze, M. (2003). The Web and its Journalisms: Considering the Consequences of Different Types of Newsmedia Online. New Media & Society, 5(2), 203-230. Deuze, M. (2006). Participation, Remediation, Bricolage: Considering Principal Components of a Digital Culture. The Information Society, 22(2), 63-75. Deuze, M., & Dimoudi, C. (2002). Online journalists in the Netherlands: Towards a profile of a new profession. Journalism, 3(1), 85-100. Dey, I. (1993). Qualitative data analysis : a user-friendly guide for social scientists. London: Routledge. DiMaggio, P. (1995). Comments on" What Theory Is Not". Administrative Science Quarterly, 40, 391-397. Dobrecky, L. P. (2007). Towards library 2.0: blogs, rss and wikis. Profesional De La Informacion, 16(2), 138-142. Domingo, D. (2008). Interactivity in the daily routines of online newsrooms: dealing with an uncomfortable myth. Journal of Computer-Mediated Communication, 13(3), 680-704. Domingo, D., Quandt, T., Heinonen, A., Paulussen, S., Singer, J. B., & Vujnovic, M. (2008). PARTICIPATORY JOURNALISM PRACTICES IN THE MEDIA AND BEYOND. Journalism Practice, 2(3), 326-342.

89

Downes, E. J., & McMillan, S. J. (2000). Defining Interactivity: A Qualitative Identification of Key Dimensions. New Media & Society, 2(2), 157-179. Druckman, J. N., Kifer, M. J., & Parkin, M. (2007). The Technological Development of Congressional Candidate Web Sites. Social Science Computer Review, 25(4), 425-442. Evans, J., & Brooks, L. (2005). Understanding Collaboration Using New Technologies: A Structurational Perspective. The Information Society, 21(3), 215-220. Facht, U. (2008). Nordicom-Sveriges Internetbarometer 2007. MedieNotiser, (2). Retrieved from

http://www.nordicom.gu.se/?portal=mr&main=info_publ2.php&ex =266

Foot, K. A., & Schneider, S. M. (2006). Web campaigning. Cambridge, Mass.: MIT Press. Fortunati, L., & Sarrica, M. (2010). The Future of the Press: Insights from the Sociotechnical Approach. The Information Society: An International Journal, 26(4), 247 - 255. Fry, G., Chantavanich, S., & Chantavanich, A. (1981). Merging Quantitative and Qualitative Research Techniques - toward a New Research Paradigm. Anthropology & Education Quarterly, 12(2), 145-158. Gerpott, T. J., & Wanke, H. (2004). Interactivity potentials and usage of German press-title web sites: An empirical investigation. Journal of Media Economics, 17(4), 241-260. Gibbons, M. T. (1987). Interpreting politics. Oxford: Basil Blackwell. Gibson, R. (2004). Web campaigning from a global perspective. Asia-Pacific Review, 11(1), 95 - 126. Gibson, R. K., Lusoli, W., & Ward, S. (2008). Nationalizing and Normalizing the Local? A Comparative Analysis of Online Candidate Campaigning in Australia and Britain. Journal of Information Technology & Politics, 4(4), 15 - 30. Gibson, R. K., & McAllister, I. (2011). Do Online Election Campaigns Win Votes? The 2007 Australian “YouTube” Election. Political Communication, 28(2), 227244. Giddens, A. (1979). Central problems in social theory action, structure and contradiction in social analysis. London: Macmillan. Giddens, A. (1981). A contemporary critique of historical materialism. London: MacMillan. Giddens, A. (1984). The constitution of society : outline of the theory of structuration. Cambridge: Polity Press. Giddens, A. (1989). A Reply to My Critics. In D. Held & J. B. Thompson (Eds.), Social Theory of Modern Societies: Anthony Giddens and His Critics, (pp. 249301). Cambridge: Cambridge University Press. Giddens, A. (1993). New rules of sociological method : a positive critique of interpretative sociologies (2nd ed.). Cambridge: Polity Press. Giddens, A., & Pierson, C. (1998). Conversations with Anthony Giddens : making sense of modernity. Cambridge: Polity Press. Gillmor, D. (2004). We the media: grassroots journalism by the people, for the people. Sebastopol, Calif. ;: O'Reilly. Greer, J., & Mensing, D. (2006). Evolution of Online Newspapers: A Longitudinal Content Analysis, 1997–2003. In X. Li (Ed.), Internet Newspapers: The Making of a Mainstream Medium (pp. 13–32). Mahwah, New Jersey: Lawrence Erlbaum Associates.

90

Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3). Gross, J., & Leslie, L. (2008). Twenty-three steps to learning Web 2.0 technologies in an academic library. Electronic Library, 26(6), 790-802. Gynnild, V. (2002). Agency and structure in engineering education: perspectives on educational change in the light of Anthony Giddens' structuration theory. European Journal of Engineering Education, 27(3), 297-303. Habermas, J. (1989). The structural transformation of the public sphere : an inquiry into a category of bourgeois society. Cambridge: Polity Press. Hallin, D. C., & Mancini, P. (2004). Comparing media systems: three models of media and politics. New York, NY:: Cambridge University Press. Halperin, R. (2005). Learning technology in higher education: A structurational perspective on technology-mediated learning practices. London: London School of Economics and Political Science. Hara, N., & Jo, Y. (2007). Internet politics: A comparative analysis of U.S. and South Korea presidential campaigns. First Monday, 12(9). Harrison, T. M., & Barthel, B. (2009). Wielding new media in Web 2.0: exploring the history of engagement with the collaborative construction of media products. New Media & Society, 11(1-2), 155-178. Hayes, A., & Krippendorff, K. (2007). Answering the call for a standard reliability measure for coding data. Communication Methods and Measures, 1(1), 77-89. Hirzalla, F., van Zoonen, L., & de Ridder, J. (2011). Internet Use and Political Participation: Reflections on the Mobilization/Normalization Controversy. The Information Society, 27(1), 1-15. Holsti, O. R. (1969). Content analysis for the social sciences and humanities. Reading, Mass.: Addison-Wesley. Honeycutt, C., & Herring, S. C. (2009). Beyond Microblogging: Conversation and Collaboration via Twitter. Paper presented at the 42nd Hawaii International Conference on System Sciences, Waikoloa, Big Island, Hawaii. Hujanen, J., & Pietikainen, S. (2004). Interactive Uses of Journalism: Crossing Between Technological Potential and Young People's News-Using Practices. New Media & Society, 6(3), 383-401. Hutton, W., & Giddens, A. (2001). On the edge : living with global capitalism. London: Random House . Hwang, J., & McMillan, S. (2002). The Role of Interactivity and Involvement in Attitude Toward the Web Site. Paper presented at the Conference of the American Academy of Advertising, Auburn University, Auburn, AL. Jackson, N. A., & Lilleker, D. G. (2009). Building an Architecture of Participation? Political Parties and Web 2.0 in Britain. Journal of Information Technology & Politics, 6(3), 232 - 250. Jaeger, P., Paquette, S., & Simmons, S. (2010). Information Policy in National Political Campaigns: A Comparison of the 2008 Campaigns for President of the United States and Prime Minister of Canada. Journal of Information Technology & Politics, 7(1), 67-82. Johnson, R. (2006). Bourdieu and the Journalistic Field. Journal of Communication Inquiry, 30(3), 276-279. Jones, M. R. (1997). Structuration theory and IT. In W. L. Currie & B. Galliers (Eds.), Rethinking management information systems (pp. 103–135). Oxford: Oxford University Press.

91

Jones, M. R. (1999). Structuration Theory. In W. Currie & R. Galliers (Eds.), Rethinking management information systems : an interdisciplinary perspective (pp. 103-135). Oxford: Oxford Univ. Press. Jones, M. R., & Karsten, H. (2008). Giddens’ Structuration Theory and Information Systems Research. MIS Quarterly, 32(1), 127-157. Jones, M. R., & Karsten, H. (2009). DIVIDED BY A COMMON LANGUAGE? A RESPONSE TO MARSHALL SCOTT POOLE. Mis Quarterly, 33(3), 589-595. Kalnes, Ø. (2009). Norwegian Parties and Web 2.0. Journal of Information Technology & Politics, 6(3), 251 - 266. Kann, M. E., Berry, J., Grant, C., & Zager, P. (2007). The Internet and youth political participation. First Monday, 12(8). Kaplan, A. (1964). The conduct of inquiry. New York: Harper Row. Karlsen, R. (2009). Online and Undecided: Voters and the Internet in the Contemporary Norwegian Election Campaign. Scandinavian Political Studies, 9999(9999). Karlsen, R. (2011). Still Broadcasting the Campaign: On the Internet and the Fragmentation of Political Communication with Evidence from Norwegian Electoral Politics. Journal of Information Technology & Politics, 8(2), 146-162. Kerbel, M. R., & Bloom, J. D. (2005). Blog for America and civic involvement. Harvard International Journal of Press-Politics, 10(4), 3-27. Kiernan, V., & Levy, M. R. (1999). Competition Among Broadcast-Related Web Sites. Journal of Broadcasting & Electronic Media, 43(2), 271-279. Kiousis, S. (2002). Interactivity: a concept explication. New Media & Society, 4(3), 355-383. Klein, H. K., & Myers, M. D. (1999). A set of principles for conducting and evaluating interpretive field studies in information systems. MIS Quarterly, 23(1), 67-93. Krippendorff, K. (2004). Content Analysis: An Introduction to Its Methodology (2nd ed.). Thousand Oaks, CA: SAGE. Kwak, H., Lee, C., Park, H., & Moon, S. (2010). What is Twitter, a social network or a news media? Paper presented at the Proceedings of the 19th international conference on World wide web, Raleigh, North Carolina, USA. Larsson, A. O. (2011). Interactive to me - interactive to you? A study of use and appreciation of interactivity on Swedish newspaper websites. New Media & Society, 13(7), 1180-1197. Larsson, A. O. (2012). Interactivity on Swedish newspaper websites: What kind, how much and why? Convergence: The International Journal of Research into New Media Technologies. Lasorsa, D., Lewis, S., & Holton, A. (2011). Normalizing Twitter. Journalism Studies, 1-18. Latour, B. (1991). Technology is Society Made Durable. In J. Law (Ed.), A Sociology of Monsters: Essays on Power, Technology and Domination (pp. 103131). London: Routledge. Lee, A. S., & Baskerville, R. L. (2003). Generalizing Generalizability in Information Systems Research. Information Systems Research, 14(3), 221-243. Lilleker, D. G., Koc-Michalska, K., Schweitzer, E. J., Jacunski, M., Jackson, N., & Vedel, T. (2011). Informing, engaging, mobilizing or interacting: Searching for a European model of web campaigning. European Journal of Communication, 26(3), 195-213. Lilleker, D. G., & Malagón, C. (2010). Levels of Interactivity in the 2007 French Presidential Candidates' Websites. European Journal of Communication, 25(1), 25-42.

92

Lombard, M., Snyder-Duch, J., & Bracken, C. C. (2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28(4), 587-604. Loyal, S. (2003). The sociology of Anthony Giddens. London: Pluto Press. Manyika, J., Chui, M., Brown, B., Bughin, J., Dobbs, R., & Byers, A. H. (2011). Big data: The next frontier for innovation, competition, and productivity. San Francisco: McKinsey Global Institute. Margolis, M., & Resnick, D. (2000). Politics as usual : the cyberspace "revolution". London: SAGE. Massey, B. L. (2000). Market-based predictors of interactivity at Southeast Asian online newspapers. Internet Research-Electronic Networking Applications and Policy, 10(3), 227-237. Matheson, D. (2004). Weblogs and the epistemology of the news: some trends in online journalism. New Media & Society, 6(4), 443-468. McMillan, S. J. (2000). Interactivity is in the Eye of the Beholder: Function, Perception, Involvement, and Attitude Toward the Website. In M. A. Shaver (Ed.), Proceedings of the 2000 Conference of the American Academy of Advertising (pp. 71-78). East Lansing, MI: Michigan State University. McMillan, S. J., Hwang, J. S., & Lee, G. (2003). Effects of structural and perceptual factors on attitudes toward the website. Journal of Advertising Research, 43(4), 400-409. Merton, R. K. (1967). On theoretical sociology : Five essays, old and new. Including Part 1 of Social theory and social structure. New York: Free Press. Mitchelstein, E., & Boczkowski, P. J. (2009). Between tradition and change: A review of recent research on online news production. Journalism, 10(5), 562586. Mitchelstein, E., & Boczkowski, P. J. (2010). Online news consumption research: An assessment of past work and an agenda for the future. New Media & Society, 12(7), 1085-1102. Monteiro, E., & Hanseth, O. (1996). Social Shaping of Information Infrastructure. In W. J. Orlikowski, G. Walsham, M. Jones & J. I. Degross (Eds.), Information Technology and Changes in Organizational Work. London: Chapman and Hall. Morozov, E. (2009). Iran: Downside to the "Twitter Revolution". Dissent, 56(4), 1014. Morozov, E. (2011). The net delusion : the dark side of internet freedom. New York: Public Affairs. Myreteg, G. (2007). Förändringens vindar : en studie om aktörsgrupper och konsten att välja och införa ett affärssystem. Uppsala: Företagsekonomiska institutionen, Uppsala universitet. Nielsen, R. K. (2010). Mundane internet tools, mobilizing practices, and the coproduction of citizenship in political campaigns. New Media & Society, 13(5), 755-771. Nietzsche, F. (1996). Human, all too human. Cambridge: Cambridge University Press. O’Reilly, T. (2005). What is Web 2.0? Design Patterns and Business Models for the Next Generation of Software, from http://www.oreillynet.com/lpt/a/6228 Orlikowski, W., & Barley, S. (2001). Technology and institutions: what can research on information technology and research on organizations learn from each other? MIS Quarterly, 25(2), 145-165. Orlikowski, W. J. (1992). THE DUALITY OF TECHNOLOGY: RETHINKING THE CONCEPT OF TECHNOLOGY IN ORGANIZATIONS. Organization Science, 3(3), 398-427.

93

Orlikowski, W. J. (2000). Using technology and constituting structures: A practice lens for studying technology in organizations. Organization Science, 11(4), 404428. Orlikowski, W. J., & Robey, D. (1991). Information Technology and the Structuring of Organizations. Information Systems Research, 2(2), 143-169. Orlikowski, W. J., Yates, J., Okamura, K., & Fujimoto, M. (1995). Shaping Electronic Communication - the Metastructuring of Technology in the Context of Use. Organization Science, 6(4), 423-444. Pavlik, J. (2000). The Impact of Technology on Journalism. Journalism Studies, 1(2), 229–237. Pavlik, J. V. (2001). Journalism and new media. New York:: Columbia University Press. Poole, M. S. (2009). RESPONSE TO JONES AND KARSTEN, "GIDDENS'S STRUCTURATION THEORY AND INFORMATION SYSTEMS RESEARCHED". Mis Quarterly, 33(3), 583-587. Poole, M. S., & DeSanctis, G. (2004). Structuration Theory in Information Systems Research: Methods and Controversies. In M. E. Whitman & A. Woszcynski (Eds.), Handbook of Information Systems Research (pp. 206-249). Hershey, PA: Idea Group. Pozzebon, M. (2005). Challenges in Conducting Empirical Work Using Structuration Theory: Learning from IT Research. Organization Studies, 26(9), 1353-1376. Pozzebon, M., Diniz, E., & Jayo, M. (2009). Adapting the Structurationist View of Technology for Studies at the Community/Societal Levels. In Y. K. Dwivedi, B. Lal, M. D. Williams, S. L. Schneberger & M. Wade (Eds.), Handbook of Research on Contemporary Theoretical Models in Information Systems (pp. 1833). Hershey, PA, USA: IGI Global. Quandt, T. (2008). (No) News on the World Wide Web? Journalism Studies, 9(5), 717-738. Resnick, D. (1998). Politics on the internet: The normalization of cyberspace. In T. C & L. TW (Eds.), The Politics of Cyberspace (pp. 48–68). New York: Routledge. Robinson, S. (2010). Traditionalists vs. Convergers: Textual Privilege, Boundary Work, and the Journalist--Audience Relationship in the Commenting Policies of Online News Sites. Convergence: The International Journal of Research into New Media Technologies, 16(1), 125-143. Rogers, E. M. (2003). Diffusion of innovations (5. ed.). New York: Free press. Rogers, R., & Marres, N. (2000). Landscaping climate change: a mapping technique for understanding science and technology debates on the World Wide Web. Public Understanding of Science, 9(2), 141-163. Roscoe, T. (1999). The construction of the World Wide Web audience. Media Culture & Society, 21(5), 673-+. Rose, J. (2000). Evaluating the contribution of structuration theory to the information systems discipline. Paper presented at the ECIT 2000 Conference. Rose, J., Jones, M., & Truex, D. (2005). Socio-Theoretic Accounts of IS: The Problem of Agency. Scandinavian Journal of Information Systems, 17(1), 133152. Rose, J., & Scheepers, R. (2001). Structuration Theory and Information System Development - Frameworks for Practice. Paper presented at the 9th European Conference on Information Systems, Bled, Slovenia. Russell, A. (2007). Digital communication networks and the journalistic field: The 2005 French riots. Critical Studies in Media Communication, 24(4), 285-302.

94

Scharmer, C. O. (1999, September 7, 1999 ). ["Awareness is the First and Critical Thing" - Conversation with Professor Wanda Orlikowski]. Schmitz Weiss, A., & Domingo, D. (2010). Innovation processes in online newsrooms as actor-networks and communities of practice. New Media & Society, 12(7), 1156-1171. Schultz, T. (1999). Interactive Options in Online Journalism: A Content Analysis of 100 U.S. Newspapers. Journal of Computer-Mediated Communication, 5(1). Schweitzer, E. J. (2008). Innovation or Normalization in E-Campaigning? A Longitudinal Content and Structural Analysis of German Party Websites in the 2002 and 2005 National Elections. European Journal of Communication, 23(4), 449-470. Schweitzer, E. J. (2011). Normalization 2.0: A longitudinal analysis of German online campaigns in the national elections 2002-9. European Journal of Communication, 26(4), 310-327. Seddon, P. B., & Scheepers, R. (2011). Towards the improved treatment of generalization of knowledge claims in IS research: drawing general conclusions from samples. European Journal of Information Systems, 21(1), 6-21. Shankar, A., Cherrier, H., & Canniford, R. (2006). Consumer empowerment: a Foucauldian interpretation. European Journal of Marketing, 40(9-10), 10131030. Shannon, C. E., & Weaver, W. (1949). The mathematical theory of communication. Urbana, Illinois: University of Illinois Press. Shyam Sundar, S. (2000). The Internet - Multimedia Effects on Processing and Perception of Online News: a Study of Picture, Audio, and Video Downloads JOURNALISM QUARTERLY, 77(3), 480-499. Singer, J. B. (2005). The political j-blogger: 'Normalizing' a new media form to fit old norms and practices. Journalism, 6(2), 173-198. Singer, J. B. (2010). QUALITY CONTROL. Journalism Practice, 4(2), 127-142. Singer, J. B., & Ashman, I. (2009). "Comment Is Free, but Facts Are Sacred": Usergenerated Content and Ethical Constructs at the Guardian. Journal of Mass Media Ethics, 24(1), 3-21. Sjöström, J. (2010). Designing Information Systems : A pragmatic account. Uppsala University, Uppsala. Retrieved from

http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-130661

Smith, A. (2009). The Internet's role in campaign 2008. Pew Internet & American Life Project: Pew Research Center. Sohn, D., & Lee, B. K. (2005). Dimensions of interactivity: Differential effects of social and psychological factors. Journal of Computer-Mediated Communication, 10(3), -. Steensen, S. (2009). The shaping of an online feature journalist. Journalism, 10(5), 702-718. Stones, R. (2005). Structuration theory. Basingstoke: Palgrave Macmillan. Strandberg, K. (2009). Online campaigning: an opening for the outsiders? An analysis of Finnish parliamentary candidates' websites in the 2003 election campaign. New Media & Society, 11(5), 835-854. Stromer-Galley, J. (2000). On-line interaction and why candidates avoid it. Journal of Communication, 50(4), 111-132. Stromer-Galley, J. (2004). Interactivity-as-Product and Interactivity-as-Process. The Information Society: An International Journal, 20(5), 391 - 394. Sutton, R., & Staw, B. (1995). What Theory Is Not. Administrative Science Quarterly, 40(3), 371-384.

95

Sweetser Trammell, K. D. (2007). Candidate campaign blogs - Directly reaching out to the youth vote. American Behavioral Scientist, 50(9), 1255-1263. Tankard, J., & Ban, H. (1998). Online newspapers: living up to their potential? Paper presented at the Annual Convention of the Association for Journalism and Mass Communication, Baltimore, MD. http://list.msu.edu/cgi-

bin/wa?A2=ind9810B&L=aejmc&P=R8021

Tapscott, D., & Williams, A. D. (2006). Wikinomics : how mass collaboration changes everything. New York: Portfolio. Thompson, J. B. (1989). The Theory of Structuration. In D. Held & J. B. Thompson (Eds.), Social theories of Modern Societies: Anthony Giddens and his Critics. Cambridge: Cambridge University Press. Thurman, N. (2008). Forums for citizen journalists? Adoption of user generated content initiatives by online news media. New Media & Society, 10(1), 139-157. Thurman, N., & Myllylahti, M. (2009). TAKING THE PAPER OUT OF NEWS. Journalism Studies, 10(5), 691-708. TwapperKeeper. (2010). Your TwapperKeeper - Archive your own Tweets, from

http://your.twapperkeeper.com/

Vaccari, C. (2008a). From the air to the ground: the internet in the 2004 US presidential campaign. New Media & Society, 10(4), 647-665. Vaccari, C. (2008b). Research note: Italian parties' websites in the 2006 elections. European Journal of Communication, 23(1), 69-77. Vaccari, C. (2008c). Surfing to the Elysee: The Internet in the 2007 French Elections. French Politics, 6(1), 1-22. Vobic, I. (2011). Online multimedia news in print media: A lack of vision in Slovenia. Journalism, 12(8), 946-962. Walsham, G. (1993). Interpreting information systems in organizations. Chichester: John Wiley. Walsham, G. (1995). Interpretive Case-Studies in Is Research - Nature and Method. European Journal of Information Systems, 4(2), 74-81. Walsham, G. (2002). Cross-cultural software production and use: A structurational analysis. Mis Quarterly, 26(4), 359-380. Walsham, G., & Han, C. (1991). Structuration theory and information systems research. Journal of Applied Systems Analysis, 17(2), 77–84. Walsham, G., & Han, C. K. (1993). Information systems strategy formation and implementation: The case of a central government agency. Accounting, Management and Information Technologies, 3(3), 191-209. Ward, S., & Vedel, T. (2006). Introduction: The potential of the Internet revisited. Parliamentary Affairs, 59(2), 210-225. Warr, W. A. (2008). Social software: fun and games, or business tools? Journal of Information Science, 34(4), 591-604. Wattal, S., Schuff, D., Mandviwalla, M., & Williams, C. B. (2010). WEB 2.0 AND POLITICS: THE 2008 U.S. PRESIDENTIAL ELECTION AND AN EPOLITICS RESEARCH AGENDA. Mis Quarterly, 34(4), 669-688. Weaver, G. R., & Gioia, D. A. (1994). Paradigms Lost - Incommensurability Vs Structurationist Inquiry. Organization Studies, 15(4), 565-590. Weick, K. (1995). What Theory is Not, Theorizing Is. Administrative Science Quarterly, 40, 385-390. Willcocks, L. P. (2006). Michel Foucault in the social study of ICTs - Critique and reappraisal. Social Science Computer Review, 24(3), 274-295. Woolgar, S. (1996). Technologies as cultural artefacts. In W. Dutton (Ed.), Information and Communication Technologies: Visions and Realities (pp. 87102.). Oxford: Oxford University Press.

96

Yin, R. K. (2003). Case study research : design and methods. Thousand Oaks: Sage Publications. Zhang, W. (2010). TECHNICAL CAPITAL AND PARTICIPATORY INEQUALITY IN EDELIBERATION -- An actor-network analysis. Information, Communication & Society, 13(7), 1019 - 1039. Zittel, T. (2009). Lost in Technology? Political Parties and the Online Campaigns of Constituency Candidates in Germany's Mixed Member Electoral System. Journal of Information Technology & Politics, 6(3), 298 - 311.

97

Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Social Sciences 77 Editor: The Dean of the Faculty of Social Sciences A doctoral dissertation from the Faculty of Social Sciences, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Social Sciences.

Distribution: publications.uu.se urn:nbn:se:uu:diva-171757

ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2012

Paper I

Convergence: The International Journal of Research into New Media Technologies http://con.sagepub.com/ Interactivity on Swedish newspaper websites: What kind, how much and why? Anders Olof Larsson Convergence published online 9 January 2012 DOI: 10.1177/1354856511430184

The online version of this article can be found at: http://con.sagepub.com/content/early/2012/01/04/1354856511430184 Published by: http://www.sagepublications.com

Additional services and information for Convergence: The International Journal of Research into New Media Technologies can be found at: Email Alerts: http://con.sagepub.com/cgi/alerts Subscriptions: http://con.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> OnlineFirst Version of Record - Jan 9, 2012 What is This?

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Article

Interactivity on Swedish newspaper websites: What kind, how much and why?

Convergence: The International Journal of Research into New Media Technologies 1-19 ª The Author(s) 2011 Reprints and permission: sagepub.co.uk/journalsPermissions.nav DOI: 1354856511430184 con.sagepub.com

Anders Olof Larsson Uppsala University, Sweden

Abstract This article examines what kind of interactive features are available on the websites of Swedish newspapers, and what factors seem to influence the utilization of those features. Using Chung’s typology of interactive features, we can discern four types: human (features that facilitate interpersonal communication), human–medium (allowing users to express their personal opinions), medium (allowing users choice options in experiencing news stories) and medium–human (allowing users to customize news to their liking) interactive features. Factors believed to have influence over interactive features are tested using statistical analysis. Even though different factors tend to influence different types of interactivity, results indicate that the most interactive newspaper websites belong to large, national newspapers with younger web staff. Keywords citizen journalism, interactivity, journalism, online newspapers, Sweden

Introduction Following the rapid spread of internet usage during the mid–1990s, established media worldwide began expressing fears regarding the consequences of the expansion of the new medium. Among other things, it was suggested that print media especially would not fare well in the new digital era (Thurman and Myllylahti, 2009). Beliefs were also aired that the interactive potential of the internet would come to change, or at least influence, the practice of journalism, moving it from a one- to a two-way communication process (Chung, 2007, 2008; Hujanen and Pietikainen, 2004), leveling the playing field between journalist and reader, sender and receiver. Today, many media outlets are indeed experiencing economic difficulties, while at the same time, many are investing resources into providing informative, interactive and varied websites for their audiences. This

Corresponding author: Anders Olof Larsson, Department of Informatics and Media, Uppsala University, Box 513, 751 20 Uppsala, Sweden Email: [email protected]

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

2

Convergence: The International Journal of Research into New Media Technologies

article will study the websites of Swedish newspapers; more specifically how these websites make use of what is often described as the defining character of the internet medium: interactivity (Downes and McMillan, 2000; Gillmor, 2004; Kiousis, 2002). The need to investigate how newspaper websites make use of interactive features in countries other than the USA has been acknowledged by previous research (Chung, 2007: 577; Mitchelstein and Boczkowski, 2009). In addition, the Swedish media market might be especially interesting to study in this regard, mainly because of relatively high overall newspaper readership, as well as widespread internet penetration and usage (Carlsson and Facht, 2007; Facht, 2008). Although much work has been done looking into what types of interactive features are offered on newspaper websites, this work has mainly been descriptive, and the question of why such features are employed has not been clearly addressed (Boczkowski, 2002: 277; Chung, 2007: 43–44). With this in mind, the aim of this study is to examine what kind of interactive features are available on the websites of Swedish newspapers, and what factors seem to influence the utilization of those features. This article is organized as follows. First, earlier research on internet newspapers and the study of interactivity on newspaper websites are discussed. Following this, factors that are often thought to influence website features such as interactivity are identified. After the subsequent methodology section, the results of the study are presented and discussed. The article finishes with a description of the limitations of the study and provides a few suggestions for future research projects.

Background The body of research on online newspapers and online journalism offers a multitude of research areas and approaches (see Mitchelstein and Boczkowski, 2009: for a review). Likewise, many studies on online interactivity have been published, identifying a number of different views on the topic. Although journalism has been described as ‘always influenced by technology’ (Pavlik, 2000: 229), the influences provided by recent technological developments seem to have taken some time to gain ground in the journalistic work environment. The next section offers a research review on interactivity in the journalistic context. Since interactivity has been conceptualized in a variety of ways as well as used in many different scientific disciplines, the section after that will provide an overview of some of the ways that this concept has been used within the social sciences.

Newspapers on the internet – cautious traditionalists From early on, it seems that the established media have indeed taken on the roles of ‘cautious traditionalists’ (Chung, 2007: 53) when dealing with the internet medium. In one of the first studies on online newspaper appropriation of interactive features, Newhagen and Cordes (1995) showed how media professionals had difficulty adapting to the underlying potential for two-way communication that characterizes the new medium. Although some journalists expressed an interest in the more direct way of communicating with readers made possible by the digital development (Chung, 2007), these ambitions seemed to be thwarted by what might be described as ‘conservative’ or ‘traditional’ journalistic schooling and thinking (Domingo, 2008; Gillmor, 2004), as well as the professed merits and non-merits of online publishing. For example, in his survey of journalists at The New York Times, Schultz (2000) found that of those questioned, the majority did not take an interest in the Times’ own forums. The use of interactive features, such as user discussion forums, is often seen as increasing the workload for journalists because of the perceived need for

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

3

fact- and language-checking reader contributions (Chung, 2007; Thurman, 2008). The hesitancy described earlier could perhaps also be understood as a result of the business-like, increasingly commercial, non-egalitarian character of newspapers (Martin and Hansen, 1998). Similar studies seem to produce comparable results to the ones discussed here, indicating that journalists and editors-in-chief alike are somewhat bewildered by the new medium, seeing threats as well as possibilities (Boczkowski, 2004; Quinn and Trench, 2002). The lack of interaction has led some researchers to conclude that most newspapers use their online versions as mirrors of their printed counterparts, where content from the latter is simply ‘lifted’ into the former without substantial augmentation. This transferring of content from print to web is often referred to as ‘shovelware’ journalism, indicating that little or no adaptation to the new medium has been undertaken (Chyi and Sylvie, 1998; Riley et al., 1998; Schultz, 1999). However, Boczkowski (2004) points out that one should not underestimate the cumulative transformation effect that online newspapers seem to be undergoing – perhaps as a result of increasing online competition between newspapers, online durability or increasingly aggressive other online actors, recent studies have found increased levels of interactivity on various newspaper websites. Greer and Mensing (2006) reported an increase of interactive features, albeit minimal, in their longitudinal study of online US newspapers, and similar results have been reported by other scholars (Chung, 2004; Mitchelstein and Boczkowski, 2009). In a Scandinavian context, newspapers seem to be slowly embracing the concept of interactivity (Engebretsen, 2006; Karlsson, 2006). Indeed, media organizations in Sweden as well as elsewhere appear increasingly positive towards augmenting their press title websites with interactive functionalities (Frisk, 2008). Rightfully so, it would seem – an audience-centered study conducted in the USA indicates that more interactive possibilities on newspapers’ websites result in a sense of belonging or commitment towards the specific web page (Shyam Sundar, 2000). Outing (2001), quoted in Deuze (2003), comments that this commitment seems to appear even if the individual user does not choose to partake in using the different interactive features available on the site. Other researchers have found that more interactive features on websites tend to result in positive attitudes towards the site itself (McMillan, 2000a; McMillan et al., 2003). Although these positive effects of interactivity have been contested (as pointed out by McMillan, 2002b), researchers from a variety of disciplines seem to agree that a variety of opportunities for interactivity will lead to an increase in web-surfer satisfaction as well as surfer behavior such as repeat visits, purchasing from the site (where applicable) as well as referring others to it (Aikat, 1998; Kiernan and Levy, 1999; Wu, 1999). All together, research seems to indicate that a website rich with interactive features will lead to more, as well as more active, visitors, as well as a sense of commitment or belonging from those visitors.

Different views on interactivity In 1998, Jensen stated that the concept of interactivity is hard to describe explicitly because of its widespread usage not only in academia, but also in society at large. He argued that our ‘culture has lived out what we might call the interactive turn’, where interactivity permeates all of society, and therefore is ‘watered down’ as an academic and scientific concept (1998: 186). Schultz mirrors these opinions when he refers to the term as ‘inflated’ and as a ‘dull buzzword’ (2000: 205). Recently, Domingo (2008: 686) argued that the concept of interactivity is ‘a bit too elastic’, pointing at ‘the wide range of options referred to under the label of interactivity’. Indeed, interactivity has been ‘often mentioned but seldom operationalized’ (McMillan and Hwang, 2002: 29). Even though efforts have been made towards presenting a clearer definition, more than a decade

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

4

Convergence: The International Journal of Research into New Media Technologies

after Jensen’s article uncertainty still remains regarding what interactivity is and how it might be understood academically (see also Bucy, 2004; Koolstra and Bos, 2009). Judging from the literature available, we can distinguish three main views on interactivity: functional, perceived and process. Rather than a typology of interactivity, these three views represent different approaches that scholars have chosen when pursuing research on interactivity. Studies adopting a functional view often employ content analysis to find ‘specific features that can be identified and categorized as interactive’ (McMillan, 2002a: 165) and mostly define levels of interactivity as the presence or absence of specific features (Song and Zinkhan, 2008: 100). Sundar’s related claim that ‘Interactivity is a message (or medium) attribute, not a user attribute’ (2004: 386) has not gone unchallenged, however. To illustrate, Gerpott and Wanke (2004) opted for the functional view in their study of interactive features of German press titles on the web, stating that this view ‘avoids perceptual distortions included in user-web site interactivity ratings caused by idiosyncrasies of visitors’ (Gerpott and Wanke, 2004: 245). However, as McMillan (2000a) points out, interactivity might well be located ‘in the eye of the beholder’, rather than being a purely functional feature. Simply adding more interactive features does not automatically increase the levels of perceived interactivity (Song and Zinkhan, 2008: 109), that is, how visitors of websites tend to perceive various interactive features (Downes and McMillan, 2000; Kiousis, 2002; McMillan, 2000a; Newhagen and Cordes, 1995). The apparently much less employed process view involves focusing on discourse and the reciprocity between the participants in communicative contexts. For example, Rafaeli and Sudweeks (1997) define interactivity as the ‘relatedness’ of sequential messages in a specific communicative context. Similar definitions and studies have followed (e.g., Kiousis, 2002; Koolstra and Bos, 2009). Most research regarding interactive features of online newspapers has been descriptive and has not attempted to explain why certain publications employ interactive features to higher or lower degrees (Chung, 2007: 43–44). Trying to move beyond this ‘mere counting’ of different interactive features available on these websites (as suggested by Boczkowski, 2002: 277), this article will employ a functional view of interactivity and will also attempt to analyze what factors seem to influence the employment of interactive features. As made clear from the foregoing discussion, the choice of a functional view is not without its limitations. For one thing, by focusing only on the website itself, a functional approach will inevitably exclude the site visitors from scrutiny. However, the focus of this study is on the websites of Swedish newspapers and not on the visitors of these sites. Although it seems clear that the perceived or the process view respectively could provide interesting insights regarding these matters, applying and augmenting the functional view in the way proposed earlier should provide the research community with important insights as to why newspaper websites vary when it comes to their utilization of interactive features.

Chung’s typology of interactive features Regardless of whether one chooses to study interactivity from the functional, perception or process view, the literature identifies several different ways to synthesize different types of interactive features often found on websites. Two such types that frequently appear in the research available are human and medium interactivity (e.g., Bucy, 2004; Chung, 2008; McMillan, 2002b). Human interactivity refers to diverse facilitations of user-to-user-interaction, such as chat rooms and message boards. This dimension of interactivity draws on what is often labeled a sociological definition of the concept (Downes and McMillan, 2000), and largely builds on the ideal of face-to-face

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

5

interaction as the default or standard form of interactivity (e.g., Walther and Burgoon, 1992) and how the inherent ‘conversational ideal’ is used to understand and evaluate the different kinds of interactivity available on websites. Medium interactivity refers to the various ways that visitors to websites are allowed different choice options when browsing them. This type of interactivity regularly includes descriptions of such basic features of websites as different navigational possibilities for the site visitor or ways in which the users are allowed to interact with the content on the site. In the context of news websites, medium interactivity features might include video streams and interactive news graphics. In a recent article, Chung (2008) opted for a perceived view of interactivity when studying patterns of online newspaper readers’ uses of interactive features. Providing empirical data, Chung showed that her respondents not only identified human and medium interactive features, but also combinations of the two. Accordingly, a human–medium interactive feature ‘allows users to express their opinion’ (Chung, 2008: 666) within the confines of the site. Utilizing these kinds of features will allow users to contribute to the site themselves by uploading and annotating their own texts and pictures. In other words, newspaper websites rich with human–medium interactivity are in some ways ‘engaging users as content co-producers’ (Boczkowski, 2002: 278). Common functions include ‘submit stories,’ ‘submit photos,’ ‘submit news tips,’ and letters-to-the-editor features. Features grouped under the medium–human label allow for user customization of site content. This follows what Deuze (2003: 214) has called an adaptive ideal, giving the users control, allowing them to choose what kind of news to consume and when to consume it. Common functions include customized topics, headlines and so on as well as various search and alerts features. This article will employ the suggested typology in order to classify the interactive features found on Swedish newspaper websites. Three main reasons for employing this typology can be discerned: first, it is empirically grounded; second, it was developed specifically with newspaper websites in mind; third, it goes beyond the distinction between human and medium interactivity and provides the research community with a more diverse way of looking at the concept at hand. The four-part typology of interactivity was further elaborated in Chung and Nah’s recent article (2009), and its main features are summarized in Table 1. While the typology introduced here provides a suitable starting point for approaching different types of interactivity in the online newspaper context, we must also attempt to asses what factors appear to have influence on the utilization of the different types. The next section introduces a number of such influences, based on the findings of previous research.

Influences for utilizing interactive features Since the websites studied in this article are all subsidiaries of print newspapers, it seems reasonable to view interactivity as a form of newspaper content. Looking at earlier studies, it can be concluded from the extensive body of research available that newspaper content is influenced by several factors, both internal and external to the media company itself. Building on a literature review of earlier research, this article will employ the following factors as independent variables: newspaper size, media ownership, region and competition within region, length of web presence and number and age of staff. These factors and the motivations for including them as independent variables are described in detail in the following paragraphs.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Table 1. Summary of Chung’s four-part typology of interactive features Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Type of interactivity Human

Human–Medium

Chung (2008: 666) ‘Human interactive features facilitate interpersonal communication online.’ ‘Human–medium interactive features allow users to express their personal opinions.’

Medium

‘Medium interactive features generally allow readers more control or choice options in experiencing news stories.’

Medium–human

‘Medium–human interactive features allow users to customize news to their liking.’

Chung and Nah (2009: 858) ‘Interactive options that promote human interactivity facilitate user-touser mutual communication.’ ‘Human–medium interactive features that allow users to submit customized perspectives and opinions further provide the audience with a sense of ownership.’ ‘Features representing medium interactivity rely on technology to allow users to select and elicit choice options.’ ‘Medium–human interactive features, or features that provide interactive tailoring, allow users to personalize information to their liking.’

Function on site

Example features

Users can communicate with peers and staff.

Chat, discussion forums, ‘email-a-friend’ feature.

Users can contribute with own content.

Email-links to journalists, ‘reader news tip,’ reader blogs, reader news, pictures, etc.

Users can experience content in a variety of ways.

Video streams, news graphics, mobile version of site.

User customization of site looks, content and use.

Customizable content, RSS feeds, email-alerts, SMS alerts.

Larsson

7

Newspaper size When it comes to the characteristics of newspaper content, many studies indicate that factors regarding size matter. Nerone and Barnhurst (2001) studied the websites of US newspapers, finding that larger newspapers offered more interactive features on their websites than their smaller equivalences. Even though Shultz’s (1999) study largely ignored other types of interactivity than human, his results indicate that the size (operationalized as printed newspaper circulation) of the newspaper could be a good indicator of online interactive features. Similarly, Gubman and Greer (1997) found that the circulation of the associated print newspaper was a reliable predictor of website sophistication. Performing longitudinal studies of newspaper websites from 1997 to 2003, Greer and Mensing reported that ‘whereas medium and large newspapers now have equally sophisticated sites, the small newspapers lag behind in every measure’ (2006: 29). This view of ‘strength-in-numbers’ is contrasted by studies that show somewhat opposite results. In their study of the websites of 300 US TV-stations, Chan-Olmsted and Park (2000) found that market size was the least relevant predictor of site contents. Similar results were reported by Lin and Jeffries (2001) in their study of the websites of US newspapers, radio- and television stations. In conclusion, Massey (2000) found that although newspaper circulation correlated positively with the degree of customization (here labeled as medium–human interactivity) available to users, it did not significantly predict other aspects of interactivity. Following Zeng and Li (2006), it nevertheless seems plausible that larger media organizations would offer more interactive features on their websites than smaller ones. As suggested by the studies mentioned earlier, circulation of the parent newspapers will be used as a measure of newspaper size.

Media ownership Since the 1980s, the international media market has been increasingly characterized by ownership concentration (Picard, 1989), often creating media ownership chains. Characteristics of media owners, particularly economic ones, are often regarded as having influences on media content and journalistic professionalism (e.g., Beam, 1993). For online conditions, results indicate that media owner characteristics could be good predictors of website contents (Chan-Olmsted and Park, 2000), as well as studies stating the opposite (Adams, 2007; Schultz, 1999). Nevertheless, it seems relevant to include a variable measuring the size of the media owner. In order to assess the economic strength of owners, the net economic result of newspaper owners will be included in the analysis.

Region and competition within region It seems reasonable to assume that ‘big, national news sites with more money, broader coverage scope . . . would be in a better position to set up a highly interactive news website than local newspapers’ (Zeng and Li, 2006: 144). For Swedish conditions, at least some of the national newspapers seem to give weight to this statement (Hadenius and Weibull, 1999) However, mixed research results are found when it comes to the influence that region of operation appears to have, both supporting (Peng et al., 1999) and rejecting (Dibean and Garrison, 2001) it. In their 2006 study, Zeng and Li (2006) found that the websites of national newspapers offered more of what the authors called content interactivity (here labeled medium and medium–human interactivity) than

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

8

Convergence: The International Journal of Research into New Media Technologies

their local counterparts, but non-significant results when it came to differences concerning communicative interactivity (here human and human–medium interactivity). In order to gauge any differences, the study will include data on whether newspapers operate on a local or national level. There might also be an effect from competition between newspapers operating in the same region. Competition among newspapers has been described as having an enhancing effect on the quality of journalistic output (Lacy, 1989) – a claim that seems plausible also for online web versions of newspapers. For example, Chyi and Sylvie (1998) suggested that development of specific internet-related features (such as interactivity) might be of importance to newspaper sites engaged in regional competition. However, in an early study of 62 television websites, Kiernan and Levy (1999) found little relation between website characteristics and station competition. In order to assess any variation stemming from competition, a variable measuring the number of competitors within each newspaper’s region of publication will be included.

Length of web presence As discussed earlier, there might be a cumulative effect of online longevity on the contents of newspaper websites. Although she stresses that the development of interactive features on these types of sites is scarce, Chung claims that ‘positive accounts of news publications’ implementation of interactivity with increasing sophistication in form and content over time exist’ (2008: 662), a position partly supported by Schultz (2000). As pointed out by Zeng and Li, the internet is a relatively new medium that has gone through substantial changes regarding structure as well as content during its short lifespan (Zeng and Li, 2006: 144). Longer experience in operating and maintaining professional websites might result in more refined content and advanced features. Consequently, a variable measuring the length of online presence for each newspaper’s website will be included.

Number and age of staff The number of website staff might be a valid predictor of site sophistication (Arant and Anderson, 2001). Indeed Chyi and Sylvie (1998: 5) state that ‘maintaining interactive services can be extremely labor intensive’. In a study from the same year, Tankard and Ban (1998) found number of staff to be a significant predictor of certain aspects of internet publishing, although it did not correlate significantly with their measurement of interactivity. Conversely, Zeng and Li (2006) concluded that number of staff was one of the better predictors of interactivity on newspaper websites. These results might suggest that the influence of number of staff has grown more important over time. Addressing the age factor, several studies have shown that younger generations are generally better versed in using interactive features than older people (e.g., Facht, 2008). Younger journalists also tend to be more educated than their older colleagues (Deuze and Dimoudi, 2002), which might have some effect in this regard. Additionally, in her study on the IT usage of Swedish journalists, Balsvik (2004) proposed that age differences should be considered in future studies. All together, these findings indicate that younger journalists will be more inclined towards using and utilizing interactive features on their respective newspaper websites. When it comes to factors relating to website staff, variables measuring the number and age of this staff will be included in the analysis.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

9

Method A comprehensive list of Swedish newspapers was gathered from The Swedish Newspaper Publishers’ Association (dagspress.se, n.d.). The population consisted of 143 websites. Sites that did not offer any journalistic content (N ¼ 35, 25%) or that featured incomprehensible language (N ¼ 2, 1%) were excluded. The final number of sites to be analyzed was 106 (74% of the population. Data regarding newspaper size (operationalized as circulation of parent newspaper), media ownership (operationalized as net result of owner), region (operationalized as local or national region of operation) and competition within region (operationalized as total N of newspapers within each respective region) were gathered from three sources: The Swedish Newspaper Publishers’ Association (dagspress.se, n.d.), a feature in the Swedish journalist’s trade union paper (Frigyes, 2008) and http://allabolag.se/ (allebolag.se, n.d.), a website providing information regarding finances, ownership and so forth for Swedish enterprises. Data regarding length of web presence (operationalized as N of days the newspaper had had a website), number and age of staff (operationalized as N and mean age of web staff) was obtained by means of a small questionnaire sent to the webmasters of the included websites. After two reminders, answers were received from 65 (61.3%) of the newspapers included in the study.

Content analysis Content analysis has historically been used to study ‘fixed’ communication artifacts such as newspaper articles or television segments (Krippendorff, 2004). As shown by McMillan (2000b), the method can also be applied to the internet. This study employed Krippendorf’s terminology of sampling, coding and context units to guide the analysis. Sampling units are defined as ‘units that are distinguished for selective inclusion in an analysis’ (Krippendorff, 2004: 98). This study chose a twofold view of sampling units. First, the start or home pages of the websites were analyzed. Weare and Lin suggest that ‘this strategy works well for research questions that focus on generic dimensions of web content’ such as interactive features, and further that ‘given its attentiongrabbing and organizational roles, a home page is likely to contain many central elements of web design’ (2000: 281). Any interactive feature presented and/or linked to from the home page was coded. Second, more ‘news-specific’ features of interactivity, such as user comments on news stories, were included. In order to find these features, the headlining news story from the start page of each site was analyzed. Interactive features were coded as present or non-present on the sites. Coding units are ‘units that are distinguished for separate description, recording or coding’ (Krippendorff, 2004: 99–101). Coding units, or variables, were created using the four-part typology of interactivity presented earlier. This process was based on features identified in previous research as discussed earlier. All of the websites were also visited before the analysis in order to detect interactive features not identified in previous studies. In total, 34 coding units were recognized; 5 for human interactivity, 14 for human–medium, 10 for medium and 5 for medium–human (see Table 1 for examples of features). For each feature available on the websites, a score of 1 was recorded. Inspired by previous studies (e.g., Gerpott and Wanke, 2004; Massey, 2000), the variables for each type of interactivity were added and divided by their theoretical maximum value, thus creating four indices ranging from the minimal value of 0 to the maximum value of 1. The same calculations were also performed for all interactivity variables in order to create an overall index, encompassing all types.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

10

Convergence: The International Journal of Research into New Media Technologies

As for context units, they are ‘units of textual matter that set limits to be considered in the description of recording units’ (Krippendorff, 2004: 101). Here, context units helped classify the coding units. For example, if a blog was found on one of the websites, it was determined if the blogger was affiliated with the site or a reader, if other readers were invited to start blogging on site and so on – in order to help classify the specific recording units.

Coding procedure While online content is often described as ever changing (e.g., Koehler, 2002), focus here laid on the structural elements of newspaper websites, such as interactivity. Schultz (1999) states that the interactive features of newspaper websites are fairly constant, suggesting that although journalistic content might change, interactive features are more consistent. The coding procedure was finished within a week in January 2009, which could be considered a rapid time period (as recommended by McMillan, 2000b). To gauge reliability, a second coder was trained and recoded a random 20 per cent sample of the sites. Intercoder agreement ranged from .88, a score well above the common threshold of .80 (Krippendorff, 2004) to perfect agreement, all using Holsti’s formula (1969). The overall figure of intercoder agreement was .96.

Results According to Sohn and Lee (2005), interactivity has often been described as a heterogeneous concept in theory, but treated as a homogenous one in empirical analysis, thereby failing to see the differences between the proposed underlying dimensions. Boczkowski comments on this as ‘the lumping together of apples and oranges’ (2002: 276–277) – a seemingly common practice of treating different forms of interactivity as the same when performing analysis. In order to determine any characteristics of or differences between the four types of interactivity used in this study, the results will be presented in a way that makes this distinguishable. In order to test the influence of the factors identified earlier on the different types of interactivity (as well as on the total measure of interactivity), statistical analyses (multiple regression and mean comparison) were performed using the four types and the total as dependent variables respectively. Because these dependent variables did not meet the assumptions necessary for regression (i.e., the variables were not normally distributed), they were logit transformed in order to facilitate analysis. Results of the regression analysis can be found in Table 2. From the literature review presented earlier, we could conclude that larger newspapers would probably keep websites that featured higher levels of interactivity than smaller ones. The results in Table 2 show that the variable measuring newspaper size (circulation) does indeed emerge as a significant predictor for the total measure as well as for three of the specific measures (human, human–medium and medium–human). Although the size-related variable failed to significantly predict medium interactivity, the overall assessment is that size of the newspaper organization does indeed matter for levels of interactivity on their respective websites. Previous research has also made the claim that financially strong media owners would have more resources available to invest in their newspapers, potentially resulting in higher degrees of interactivity on the websites of these newspapers. From Table 2 we can conclude that the net result of owner variable was not a significant predictor in any of the analyses performed. Owners, financially strong or not, appear to have limited influence over these websites – at least when it comes to interactivity.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

11

Table 2. Regression analyses of factors predicting interactivity Total Circulation Net result of owner N of newspapers in region Length of web presence Number of web staff Mean age of web staff

Human

.33* .40* .12 .02 .16 .15 .07 .17 .25 .20 .30* .02 R2 ¼ .42 Adj. R2 ¼ .21 Adj. R2 ¼ .34 R2 ¼ .11

Human–medium

Medium

Medium–human

.32* .01 .23 .02 .20 .31* R2 ¼ .32 Adj. R2 ¼ .23

.17 .04 .08 .07 .50** .08 R2 ¼ .40 Adj. R2 ¼ .31

.36* .08 .03 .11 .28* .10 R2 ¼ .41 Adj. R2 ¼ .33

Standardized Beta values presented, **p < .01, *p < .05

Table 3. Mean comparisons of standardized interactivity scores

National newspaper Local newspaper Overall mean

Total

Human

Human–medium

Medium

Medium–human

.49 (.22) .34 (.17) .35 (.17) eta2 ¼ .05*

.40 (.30) .34 (.24) .35 (.24) eta2 ¼ .04

.49 (.19) .37 (.16) .38 (.16) eta2 ¼ .03

.54 (.28) .29 (.23) .31 (.24) eta2 ¼ .07**

.51 (.25) .32 (.16) .34 (.17) eta2 ¼ .08**

**p < .01, *p < .05, Standard deviations in parenthesis

As discussed earlier, the websites of nationwide newspapers might be expected to offer more interactive features than the websites of their local counterparts. Table 3 shows the results of five mean comparison tests, measuring the differences in means of different interactivity scores between national and local newspapers. From Table 3 we can tell that even though the mean differences of human and human–medium interactivity between national and regional newspapers were not significantly different, differences can be found when it comes to total interactivity (.49 and .34 respectively), as well as for medium (.54 and .29 respectively) and medium–human interactive features (.51 and .32 respectively). In general, the results indicate that national newspapers tend to offer more interactive features on their websites than local newspapers. As previous research provided mixed results regarding the effects of competition on newspaper website performance, a variable assessing competition within each newspapers region (N of newspapers within region) was included in the analysis. As shown in Table 2, this variable did not manage to significantly predict any of the four types of interactive features, nor did it predict the total amount of interactivity. In the Swedish context, competition between newspapers does not appear to have an effect on how online newspapers perform regarding utilization of interactive features. Scholars have suggested that there might be a cumulative effect of longevity when it comes to the working routines of online newspapers. However, the length of web presence variable did not emerge as a significant predictor for any of the five measures under scrutiny here. When it comes to use of interactive features, online longevity does not seem to be an important factor in the Swedish media environment. Maintaining a highly interactive website is frequently described as very labor intensive. Indeed, previous research has suggested that the number of website staff would have a positive effect on

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

12

Convergence: The International Journal of Research into New Media Technologies

the availability of interactive features available on the various websites. From the regression analyses in Table 2, we can tell that even though the number of web staff variable significantly predicts levels of medium and medium–human interactivity, it fails to predict human, human– medium or the total amount of interactivity. Thus, the number of employees working with the website apparently has some effect on the utilization of interactive features. Previous research has suggested that employment of younger staff members would have positive effects on the availability of interactive features. The age variable included in the regression analyses, mean age of web staff, did indeed prove to be a significant predictor of total and human– medium interactivity as shown in Table 2. However, it failed to significantly predict the other three types of interactivity (human, medium and medium–human). Nevertheless, from these results it would seem that hiring younger web staff has at least a partial effect when it comes to their potential influence on the websites with which they work. In sum, it seems that the websites in the study that offer higher levels of total interactivity are those that belong to large, national newspapers with younger web staff. However, the foregoing analyses did identify some interesting discrepancies regarding the predictors of the four types of interactivity. These discrepancies, and the overall results of the study, will be discussed in the next section of the article.

Discussion While the results of statistical analyses, like those featured in this article, can provide overarching views for different research phenomena, they need to be contextualized with regards to previous research and the specific research setting in order to prove truly meaningful. As such, this section discusses the results presented earlier in relation to previous research as well as in relation to online journalistic practices. The first independent variable (circulation) dealt with the size of the parent newspaper for each website. Even though earlier studies have disputed the link between newspaper size and website contents (e.g., Chan-Olmsted and Park, 2000; Lin and Jeffres, 2001; Massey, 2000), the results presented here suggest that, at least in the Swedish context, online interactive practices are indeed related to the size of the supporting newspaper (as suggested by Nerone and Barnhurst, 2001; Schultz, 1999). Evidently, the claim by Greer and Mensing that ‘small newspapers lag behind’ (2006: 29) seems valid also in the Swedish online media environment, indicating that smaller newspapers might have trouble holding their own when it comes to offering their readers opportunities to interact and contribute to their online versions. While the circulation variable failed to significantly predict employment of medium interactive features, the number of staff variable emerged as the sole significant predictor for this type of interactivity. This suggests that medium interactivity features, such as video streams, news graphics and so on, require the allocation of specific web staff to work exclusively with augmenting and servicing the website in this regard. Related to the size of the newspaper is the size of the owner organization. Concentration of media ownership is increasing in the Swedish newspaper market, and studies have shown that the size of the owner is a suitable predictor of newspaper performance (Beam, 1993) as well as website sophistication (Chan-Olmsted and Park, 2000). The results presented here suggest the opposite, mirroring the conclusions of other previous studies (Adams, 2007; Schultz, 1999). As such, the size of the ‘parent’ organizations appear to have little to do with the online performance of the individual newspapers. Perhaps the explanation for this lack of ‘strength in numbers’ can be found among the theorists of media concentration. These scholars argue that large media owners tend to

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

13

streamline content and focus more on economic revenue than on journalistic quality and on inciting new ways of journalistic practice (Bagdikian, 2004; McManus, 1994). As pointed out by Mitchelstein and Boczkowski (2009: 564), conservative tendencies such as these are often associated with large owners – a claim that apparently holds true also in the era of online journalism. As traditional media organizations find themselves in an increasingly complex economic climate, large media owners might seek to maximize synergy effects, leading to similar or even identical websites for the newspapers in the organization, rather than providing visitors of the various sites with more diverse material and varied features. Even though the effect of owner size was found to be non-significant, newspaper size (measured with the circulation variable) proved to be a significant predictor for total interactivity as discussed earlier. Taken together, this indicates that when it comes to employment of interactive features, the characteristics of the individual newspaper itself seem more important than the characteristics of its owner. Perhaps individual newspapers are not that dependent on their respective owners when it comes to decisions regarding website practices. The fact that the average national newspaper has significantly higher means of total, medium and medium–human interactivity than their local counterparts (as shown in Table 3) was addressed earlier. However, the rather low eta2-values reported for these differences (ranging from .05 to .08) suggest a moderate-at-best effect of national or local affiliation when it comes to the employment of these specific types of interactive features (interpretation guidelines from Cohen, 1988). In other words, even though there are significant differences between national and local newspapers in this regard, these differences are perhaps not as pertinent as might have been expected. While local newspapers tend to employ fewer interactive features than their national counterparts, there appears to be a certain degree of variation at play here. The effect of competition within region on interactivity was found to be non-significant, a result in line with previous studies on web performance of media organizations engaged in competition (Kiernan and Levy, 1999). As more and more media consumption moves online, competition between online media outlets will inevitably increase. The employment of interactive features could probably come to play an important part in creating attractive websites (as suggested by Chyi and Sylvie, 1998), and as such, this might be an interesting opportunity for the practitioner community. Breaking down the results for the human–medium category of interactivity, we can conclude that even though 25 per cent of the newspapers in this study allow readers to contribute news of a more ‘novelty’ character (such as reports from local sporting events, church gatherings or local community events), only 4 per cent of the sites encourage visitors to submit their own full-text news stories. Delving further into the results, we see that even though nearly 40 per cent of the sites feature blogs by non-journalists, only 8 per cent of the sites provide readers with the opportunity to write their own blogs on the newspaper site. Services such as Twingly (twingly.com, n.d.), that allow blog posts on external reader blogs regarding specific articles to appear in association with those articles on the newspaper website, are also being used, although rather moderately (12.3%). In comparison, nearly 90 per cent of the sites featured a ‘news tip-off’ function. Taken together, these results imply that while newspapers are eager to obtain news stories and ideas from their readers, they seem hesitant when it comes to letting the readers write those stories themselves. This apparently conservative attitude towards allowing reader contributions might not be a solid strategy in the long run. At a time when blogging and other ‘citizen journalism’ opportunities are more easily accessible to the general public than ever, allowing users to contribute to the newspaper website might prove to be a better approach, drawing more people to visit as well as revisit the

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

14

Convergence: The International Journal of Research into New Media Technologies

specific site. However, while increased opportunities for reader contributions might be an interesting approach in order to involve the readers in the journalistic process, these features also carry with them potential issues of editorial quality control and need for fact checking (Thurman, 2008). According to the results presented earlier, length of web presence did not emerge as a significant predictor of total interactivity, providing opposing results to the ones presented by earlier studies (Boczkowski, 2004; Chung, 2004; Engebretsen, 2006; Greer and Mensing, 2006; Karlsson, 2006). As rationales for website design come and go quickly (Zeng and Li, 2006), a need for flexibility in this regard is probably important in order to keep up with the latest online developments and trends. The results presented here suggest that relative newcomers to the web are able to ‘leapfrog’ over old and onto new web design paradigms, where focus is placed on employing more interactive features, leaving the pioneers of the medium stuck in old systems difficult to update and harmonize with the latest website design principles. The two final independent variables included in the analysis focused on the website staff. Whereas the number of web staff did not emerge as a significant predictor of total interactivity (but as a significant one for the medium and medium–human subcategories), the age of staff proved significant in predicting total interactivity. Perhaps younger journalists introduce new ideas and perspectives on website management to the newspaper staff rooms, challenging the work routines of their ‘cautious traditionalist’ (Chung, 2007: 53) older colleagues. These results might also be explained by the tendency for younger journalists to be more highly educated than their older colleagues (Deuze and Dimoudi, 2002), or, in a more general sense, that younger generations are often described as being better versed in using the web than older people (Facht, 2008). Furthermore, a sizeable web staff is seemingly not enough to create a site rich with a variety of interactive features. Perhaps larger web staffs need to follow stricter guidelines and cannot easily break the norm of established web-working practices within larger media companies with many employees dedicated to the web. While the results presented in this article show that the number of web staff appears to be important, there are also other forces at play when it comes to predicting the degree to which Swedish newspapers provide interactive features on their websites. As shown in analyses performed in this article, the size variable of circulation and the variable measuring the mean age of web staff emerged as significant predictors of total interactivity. However, for human and human–medium interactive features, only circulation proved to be a significant predictor. For both medium and medium–human interactivity, number of web staff provided significant predictions. The fact that circulation also predicted the medium–human category mirrors the results of Massey (2000), who found that a similar measure correlated positively with the degree of customization features available to site visitors, features grouped under the aforementioned category in this study. Looking at Table 3, we can see that the means for total, medium and medium–human interactivity were significantly higher for national newspaper websites, but not the means for the human and human–medium category. Taken together, these results indicate that while the utilization of features such as chat, discussion forums, reader news tips and so forth (examples of human and human–medium interactive features) depend on the general size and strength of the newspaper (as measured by the circulation variable), websites that feature high levels of medium and medium–human interactive features (such as video streams, interactive news graphics and customizable content) are often associated with national newspapers and require more specific (and perhaps also specially trained) personnel in order to shoot and edit the video to be published on the site, design and implement news graphics features and so on.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

15

The results from this study provide the research community with insights into why newspapers employ interactive features on their websites to higher or lower degrees. Although the study makes an important contribution to this particular field of research, its limitations need to be clearly addressed. This is done in the next and final section of this article. This section also provides suggestions for future research on the topic at hand.

Limitations and suggestions for future research This article has studied different kinds of interactivity available on Swedish newspaper websites, looking specifically at what factors seemingly influence the availability of these features. While approaching this topic with quantitative methods has its merits (and its flaws, for that matter), scholars planning future studies regarding these matters should consider using qualitative methods in order to delve even further into the work rationales of online newsrooms. For example, an ethnographic approach, studying the work routines and rationales of online journalists, might provide the research community with even deeper knowledge of how interactive features such as the ones outlined in this article are thought of and used in the journalistic work environment. Also, from the coefficients reported in Table 3, we can tell that the average Swedish newspaper website offer 35 per cent of human interactivity indicators, 38 per cent of human–medium, 31 per cent of medium and 34 per cent of medium–human interactivity features. Overall, a little more than a third (35%) of the features identified in this study are employed by the average newspaper website. With almost two-thirds of interactive features not being used on average, the employment of new ways to interact with site visitors might prove useful for practitioners such as site operators. In relation to this, one might also pose the question of whether 35 per cent is to be considered a high mean percentage of human interactivity, or if 31 per cent should be judged as a low mean for medium interactivity? An interesting idea for future research might be to perform longitudinal studies on online newspapers, collecting data over several years and studying the development of interactive features. While this study has provided useful results regarding online newspaper use of interactivity, it is nevertheless is a ‘snapshot’; of the Swedish media landscape in early 2009. The media industry is indeed a dynamic one, with mergers and takeovers occurring frequently. Such industry actions should be expected to have an effect on a number of influences for the newspapers. Longitudinal studies, as suggested earlier, could also prove to give us insight into the dependent variables used. For example, it might be interesting to see if age of web staff will remain as a significant predictor as today’s younger journalists grow older and new generations enter the profession. Also, the role of owner organizations might be subject to change in an ever more competitive media environment. Indeed, Wimmer and Dominick state that ‘research is a never-ending process’; (2006: 8), referring to the necessary open-ended nature of perhaps all social science research efforts. These and other questions arise from the results presented here, and could perhaps lead to future research initiatives – some of which have been suggested earlier in this article. Finally, it might be reasonable to ask the question: Does more interactivity always leads to better websites? Even though there seems to be a connection between high levels of interactivity and popularity of websites, the effect of interactivity is frequently described as two-sided. Research suggests that excessive interactivity might hinder user experience, demanding more patience, expertise and cognition of the site visitor, standing in the way of allowing the visitors to understand fully the message provided (Bucy, 2004; Sundar, 2000). In the field of advertising, Liu and Shrum (2009) found that user involvement in the product advertised and internet

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

16

Convergence: The International Journal of Research into New Media Technologies

experience were important factors in explaining how user perceptions of interactivity in online advertisements are formed. Similarly, Sohn et al (2007: 116–117) found that ‘creating fancy and dynamic websites for attracting consumers may return unexpected negative consequences’, suggesting that less than positive effects might be looming unless website operators take time to get to know the preferences of their audiences. Inspired by these results, future research projects focusing on online newspapers should address the question of how much interactivity might be considered ‘enough’ interactivity in the context of newspaper websites – both from the point of view of the producers of such sites, as well as for the visitors. It might also be interesting to study how different newspaper website visitors perceive different opportunities for interaction. Results from studies such as the ones suggested here should be of use to researchers, as well as to media practitioners interested in providing attractive and suitably interactive websites to their visitors. References Adams JW (2007) U.S. weekly newspapers embrace web sites. Newspaper Research Journal 28(4): 36–50. Aikat DD (1998) News on the Web: Usage trends of an on-line newspaper. Convergence 4(4): 94–110. allebolag.se (n.d.) Available at: http://allabolag.se/ (accessed 21 November 2011). Arant MD and Anderson JQ (2001) Newspaper online editors support traditional standards. Newspaper Research Journal 22(4): 57. Bagdikian BH (2004) The New Media Monopoly, Boston, MA: Beacon Press. Balsvik G (2004) Information Technology Users: Studies of Self-Efficacy and Creativity among Swedish Newspaper Journalists. Stockholm: Economic Research Institute, Stockholm School of Economics. Beam RA (1993) The impact of group ownership variables on organizational professionalism at daily newspapers. Journalism Quarterly 70(4): 907–918. Boczkowski PJ (2002) The development and use of online newspapers: What research tells us and what we might want to know. In: Lievrouw L and Livingstone S (eds) The Handbook of New Media: Social Shaping and Consequences of ICTs. London: SAGE, 270–286 Boczkowski PJ (2004) Digitizing the News: Innovation in Online Newspapers, Cambridge, MA: MIT Press. Bucy EP (2004) Interactivity in society: Locating an elusive concept. The Information Society: An International Journal 20(5): 373–383. Carlsson U and Facht U (2007) Medie-Sverige: Statistik och analys. 2007. Go¨teborg: Nordicom-Sverige. Chan-Olmsted S and Park J (2000) From on-air to online world: Examining the content and structures of broadcast TV stations’ web sites. Journalism & Mass Communication Quarterly 77(2): 321–339. Chung DS (2004) Into interactivity? How news websites use interactive features. Paper presented at International Communication Association Annual Convention, New Orleans, LA, 27–31 May. Washington, DC: ICA. Chung DS (2007) Profits and perils: Online news producers’ perceptions of interactivity and uses of interactive features. Convergence 13(1): 43–61. Chung DS (2008) Interactive features of online newspapers: Identifying patterns and predicting use of engaged readers. Journal of Computer-Mediated Communication 13(3): 658–679. Chung DS and Nah S (2009) The effects of interactive news presentation on perceived user satisfaction of online community newspapers. Journal of Computer-Mediated Communication 14(4): 855–874. Chyi HI and Sylvie G (1998) Competing with whom? Where? And how? A structural analysis of the electronic newspaper market. Journal of Media Economics 11(2): 1–18. Cohen J (1988) Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: L. Erlbaum Associates. dagspress.se (n.d.) Swedish Newspaper Publishers’ Association. Available at: http://www.dagspress.se/ (accessed 21 November 2011). Deuze M (2003) The Web and its journalisms: Considering the consequences of different types of newsmedia online. New Media & Society 5(2): 203–230.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

17

Deuze M and Dimoudi C (2002) Online journalists in the Netherlands: Towards a profile of a new profession. Journalism 3(1): 85–100. Dibean W and Garrison B (2001) How six online newspapers use web technologies. Newspaper Research Journal 22(2): 79–94. Domingo D (2008) Interactivity in the daily routines of online newsrooms: Dealing with an uncomfortable myth. Journal of Computer-Mediated Communication 13(3): 680–704. Downes EJ and McMillan SJ (2000) Defining interactivity: A qualitative identification of key dimensions. New Media & Society 2(2): 157–179. Engebretsen M (2006) Shallow and static or deep and dynamic? Studying the state of online journalism in scandinavia. Nordicom Review 27(1): 3–16. Facht U (2008) Nordicom-Sveriges Internetbarometer 2007. MedieNotiser. Available at: http://www.nordicom.gu.se/?portal¼mr&main¼info_publ2.php&ex¼266 (accessed November 2011). Frigyes P (2008) Va¨stmakten vill va¨xa vidare. Journalisten (13 November–3 December): 4–6. Frisk J. (2008) Medier bjuder in engagerade la¨sare. Journalisten (18–31 March): 14–16. Gerpott TJ and Wanke H (2004) Interactivity potentials and usage of German press-title web sites: An empirical investigation. Journal of Media Economics 17(4): 241–260. Gillmor D (2004) We the Media: Grassroots Journalism by the People, for the People. Sebastopol, CA: O’Reilly. Greer J and Mensing D (2006) Evolution of online newspapers: A longitudinal content analysis, 1997–2003. In: Li X (ed.) Internet Newspapers: The Making of a Mainstream Medium. Mahwah, NJ: Lawrence Erlbaum Associates, 13–32. Gubman J and Greer J (1997) An analysis of online sites produced by US newspapers: Are the critics right? In: Education in Journalism and Mass Communication Annual Convention, Newspaper Division, Chicago, IL, USA, 30 July–2 August 1997. Hadenius S and Weibull L (1999) The Swedish newspaper system in the late 1990s. Tradition and transition. Nordicom Review 20(1): 129–152. Holsti OR (1969) Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison-Wesley. Hujanen J and Pietikainen S (2004) Interactive uses of journalism: Crossing between technological potential and young people’s news-using practices. New Media & Society 6(3): 383–401. Jensen JF (1998) Interactivity: Tracking a new concept in media and communication studies. Nordicom Review 19(1): 19. Karlsson M (2006) Na¨tjournalistik: en explorativ fallstudie av digitala mediers karakta¨rsdrag pa˚ fyra svenska nyhetssajter, Lund: Avdelningen fo¨r medie- och kommunikationsvetenskap, Lunds universitet. Kiernan V and Levy MR (1999) Competition among broadcast-related web sites. Journal of Broadcasting & Electronic Media 43(2): 271–279. Kiousis S (2002) Interactivity: A concept explication. New Media & Society 4(3): 355–383. Koehler W (2002) Web page change and persistence – A four-year longitudinal study. Journal of the American Society for Information Science and Technology 53(2): 162–171. Koolstra CM and Bos MJW (2009) The development of an instrument to determine different levels of interactivity. International Communication Gazette 71(5): 373–391. Krippendorff K (2004) Content Analysis: An Introduction to Its Methodology, Thousand Oaks, CA: SAGE. Lacy S (1989) A model of demand for news: Impact of competition on newspaper content. Journalism Quarterly 66(1): 41–48, 128. Lin CA and Jeffres LW (2001) Comparing distinctions and similarities across websites of newspapers, radio stations, and television stations. Journalism & Mass Communication Quarterly 78(3): 555–573. Liu YP and Shrum LJ (2009) A dual-process model of interactivity effects. Journal of Advertising 38(2): 53–68. Martin SE and Hansen KA (1998) Newspapers of Record in a Digital Age: From Hot Type to Hot Link, Westport, CT: Praeger.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

18

Convergence: The International Journal of Research into New Media Technologies

Massey BL (2000) Market-based predictors of interactivity at Southeast Asian online newspapers. Internet Research-Electronic Networking Applications and Policy 10(3): 227–237. McManus JH (1994) Market-Driven Journalism: Let the Citizen Beware? Thousand Oaks, CA: SAGE. McMillan SJ (2000a) Interactivity is in the eye of the beholder: Function, perception, involvement, and attitude toward the website. In: Proceedings of the 2000 Conference of the American Academy of Advertising (ed. MA Shaver), East Lansing, MI: Michigan State University, pp. 71–78. Richmond, VA: American Academy of Advertising. McMillan SJ (2000b) The microscope and the moving target: The challenge of applying content analysis to the World Wide Web. Journalism & Mass Communication Quarterly 77(1): 80–98. McMillan SJ (2002a) Exploring models of interactivity from multiple research traditions: Users, documents, and systems. In: Lievrouw L and Livingstone S (eds) The Handbook of New Media: Social Shaping and Consequences of ICTs. Thousand Oaks, CA: SAGE, 163–182. McMillan SJ (2002b) A four-part model of cyber-interactivity: Some cyber-places are more interactive than others. New Media & Society 4(2): 271–291. McMillan SJ and Hwang JS (2002) Measures of perceived interactivity: An exploration of the role of direction of communication, user control, and time in shaping perceptions of interactivity. Journal of Advertising 31(3): 29–42. McMillan SJ, Hwang JS and Lee G (2003) Effects of structural and perceptual factors on attitudes toward the website. Journal of Advertising Research 43(4): 400–409. Mitchelstein E and Boczkowski PJ (2009) Between tradition and change: A review of recent research on online news production. Journalism 10(5): 562–586. Nerone J and Barnhurst KG (2001) Beyond modernism: Digital design, Americanization and the future of newspaper form. New Media & Society 3(4): 467–482. Newhagen JE and Cordes JW (1995) [email protected]: Audience scope and the perception of interactivity in viewer mail on the Internet. Journal of Communication 45(3): 164. Outing S (2001) News Sites Repeat the Mistakes of the Past: We Still Don’t Recognize the Power of Interactivity. Available at: http://www.editorandpublisher.com/editorandpublisher/features_columns/article_display.jsp?vnu_content_id¼1461161 (accessed 13 october 2009). Pavlik J (2000) The impact of technology on journalism. Journalism Studies 1(2): 229–237. Peng FY, Tham NI and Xiaoming H (1999) Trends in online newspapers: A look at the US web. Newspaper Research Journal 20(2): 52–63. Picard RG (1989) Media Economics: Concepts and Issues. Newbury Park, CA: SAGE. Quinn G and Trench B (2002) Online news media and their audiences: Multimedia content in the digital age. MUDIA. Heerlen, the Netherlands: IST. Rafaeli S and Sudweeks F (1997) Networked interactivity. Journal of Computer-Mediated Communication 2(4): 1–6. Riley PK, Colleen M, Christiansen T, et al. (1998) Community or colony: The case of online newspapers and the web. Journal of Computer-Mediated Communication 4(1): 1–14. Schultz T (1999) Interactive options in online journalism: A content analysis of 100 U.S. newspapers. Journal of Computer-Mediated Communication 5(1): 1–24. Schultz T (2000) Mass media and the concept of interactivity: An exploratory study of online forums and reader email. Media, Culture & Society 22(2): 205–221. Shyam Sundar S (2000) The Internet – multimedia effects on processing and perception of online news: A study of picture, audio, and video downloads. Journalism Quarterly 77(5): 480–499. Sohn DY and Lee BK (2005) Dimensions of interactivity: Differential effects of social and psychological factors. Journal of Computer-Mediated Communication 10(3): 1–16. Sohn DY, Ci C and Lee BK (2007) The moderating effects of expectation on the patterns of the interactivityattitude relationship. Journal of Advertising 36(3): 109–119. Song JH and Zinkhan GM (2008) Determinants of perceived web site interactivity. Journal of Marketing 72(2): 99–113.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

19

Sundar SS (2004) Theorizing interactivity’s effects. The Information Society: An International Journal 20(5): 385–389. Sundar SS (2000) Multimedia effects on processing and perception of online news: A study of picture, audio, and video downloads. Journalism & Mass Communication Quarterly 77(3): 480–499. Tankard J and Ban H (1998) Online newspapers: Living up to their potential? In: Annual Convention of the Association for Journalism and Mass Communication, Baltimore, MD, USA, 5–8 August 1998. Thurman N (2008) Forums for citizen journalists? Adoption of user generated content initiatives by online news media. New Media & Society 10(1): 139–157. Thurman N and Myllylahti M (2009) Taking the paper out of news. Journalism Studies 10(5): 691–708. twingly.com (n.d.) Available at: http://www.twingly.com/ (accessed 21 November 2011). Walther JB and Burgoon JK (1992) Relational communication in computer-mediated interaction. Human Communication Research 19(1): 50–88. Weare C and Lin WY (2000) Content analysis of the World Wide Web: Opportunities and challenges. Social Science Computer Review 18(3): 272–292. Wimmer R and Dominick J (2006) Mass Media Research: An Introduction. Belmont, CA and London: Thomson Wadsworth. Wu G (1999) Perceived interactivity and attitude toward websites. In: Proceedings of the 1999 Conference of the American Academy of Advertising (ed. MS Roberts). Gainesville: University of Florida, USA, March 1999, pp. 254–262. Richmond, VA: American Academy of Advertising. Zeng Q and Li X (2006) Factors influencing interactivity of Internet newspapers: A content analysis of 106 U. S. newspapers’ web sites. In: Li X (ed.) Internet Newspapers: The Making of a Mainstream Medium. Mahwah, NJ: Lawrence Erlbaum Associates, 139–158.

Biography Anders Olof Larsson is a PhD candidate at the Department of Informatics and Media, Uppsala University, Sweden. He is also associated with The Swedish Research School of Management and Information Technology. Larsson’s research has been published in journals such as First Monday and New Media and Society. His research interests include institutional and audience use of online interactivity and social media. Anders Olof Larsson’s website can be found at http://www.andersoloflarsson.se.

Downloaded from con.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Paper II

New http://nms.sagepub.com/ Media & Society Interactive to me − interactive to you? A study of use and appreciation of interactivity on Swedish newspaper websites Anders Olof Larsson New Media Society 2011 13: 1180 originally published online 27 April 2011 DOI: 10.1177/1461444811401254 The online version of this article can be found at: http://nms.sagepub.com/content/13/7/1180 Published by: http://www.sagepublications.com

Additional services and information for New Media & Society can be found at: Email Alerts: http://nms.sagepub.com/cgi/alerts Subscriptions: http://nms.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://nms.sagepub.com/content/13/7/1180.refs.html

>> Version of Record - Nov 14, 2011 OnlineFirst Version of Record - Apr 27, 2011 What is This?

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Article

Interactive to me – interactive to you? A study of use and appreciation of interactivity on Swedish newspaper websites

new media & society 13(7) 1180–1197 © The Author(s) 2011 Reprints and permission: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1461444811401254 nms.sagepub.com

Anders Olof Larsson Uppsala University, Sweden

Abstract Research has indicated that although online interactive features are not used by the visitors of different websites, such features might be appreciated by the visitors. This article examines the use and appreciation of interactive features by visitors on Swedish newspaper websites. Utilizing an online survey focusing on different traits and habits of newspaper website visitors, the study presents a typology of visitor types, characterized by the different ways they use and appreciate interactive features in the online news media context. Although certain types make extensive use of interactivity, the overall results of the survey points towards rather low levels of both use and appreciation. As such, newspaper website visitors might be characterized as ‘slow learners’, taking their time to adapt to the interactive capabilities offered by the online news media. Keywords interactivity, online newspapers, survey, Sweden, visitor types

Introduction As one of the first buzzwords to be associated with the internet, interactivity has often been presumed as an intrinsic quality of the new medium (i.e. Hujanen and Pietikainen, 2004: 388). In the news media industry, practitioners have struggled while moving from an offline to an online context (Fortunati and Sarrica, 2010), often finding it hard to adapt to the opportunities for interacting with their website visitors. Although positive attitudes towards the internet can be found in several newsrooms (Chung, 2007), many journalists Corresponding author: Anders Olof Larsson, Department of Informatics and Media, Box 513, 751 20 Uppsala, Sweden Email: [email protected]

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1181

still appear cautious when dealing with the new medium (Domingo, 2008). Remembering that interactivity could very well be defined as situated in the eye of the beholder (McMillan, 2000), the study presented in this article opts for a perceived view of interactivity, looking at how newspaper website visitors experience the concept. Focusing on Sweden, the aim of this article is to study variations in newspaper website visitor use and appreciation of interactive features. Based on the results of an online survey, the study presents a typology of newspaper website visitor types. In times when newspapers are trying to find ways to reach out to their online audiences, Sweden is an interesting country to study, as it is characterized by both extensive internet usage and high levels of newspaper readership (Carlsson and Facht, 2010). The need for international research on online newspapers has been acknowledged (Chung, 2007; Mitchelstein and Boczkowski, 2009), as has the need to identify factors that predict the use of interactive features (Chung, 2008: 673). This article answers the call from previous research and provides results regarding user attitudes towards interactive features (as suggested by Chung, 2007; Hwang and McMillan, 2002; D. Sohn and Lee, 2005). Furthermore, the identification of visitor types should be helpful to academics and practitioners interested in newspaper websites.

Literature review Use of interactive features Although some have suggested that citizen media initiatives and audience participation on newspaper websites will lead to audience empowerment as well as better journalistic practice (Gillmor, 2004), media organizations often appear reluctant when it comes to letting audience members participate. While audience members may be allowed to take part in online discussions hosted by the newspaper, they are often barred from other stages of news production (Domingo et al., 2008). Thurman (2008) interviewed British journalists and found mostly negative attitudes towards user-generated content. User contributions were seen as problematic not only because of the alleged need for quality control, but also because these new opportunities challenged traditional journalistic norms and practices. Similarly, Domingo (2008) claimed that interactivity has become an uncomfortable myth in the journalistic context. Chung (2007) found comparable attitudes among US news website producers. Her results indicated that although journalists find the migration from paper to web difficult, some respondents expressed an interest in the use of interactive features to involve the readers. Although journalists face ‘a host of issues’ (Singer and Ashman, 2009: 18) when dealing with user interaction, positive attitudes like these also seem to be prevalent in both Swedish and international contexts (Frisk, 2008; Mitchelstein and Boczkowski, 2009; Singer, 2010). A similarly reluctant attitude can be found among the news audience. In Sweden, usage of newspaper website interactivity has been reported at low levels (Bergström, 2008). Indeed, audience unwillingness to participate in the online newspaper context seems prevalent also in other countries (e.g. Hujanen and Pietikainen, 2004).

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1182

new media & society 13(7)

Conceptualizations of interactivity Although several ways of operationalizing interactivity co-exist, three conceptualizations frequently reappear in the literature. While different authors give these conceptualizations different labels, the distinction between functional, perceived or process related views (suggested by Leiner and Quiring, 2008) is suitably encompassing. The functional view conceptualizes interactivity as an attribute of the medium (Sundar, 2004). Interactivity is operationalized as ‘the presence or absence of particular features’ (Song and Zinkhan, 2008: 100). Studies adopting the functional view are often descriptive and employ content analysis in order to find ‘specific features that can be identified and categorized as interactive’ (McMillan, 2002a: 165). As such, more interactive features equals higher levels of interactivity. The functional view is often seen as limited as it ignores site visitor characteristics that might have influence on how these features are perceived (Bucy, 2004). For example, McMillan (2002b) found that respondents in her study did not necessarily classify sites with more opportunities for interactivity as more interactive than sites with smaller amounts of features. Taking these ‘idiosyncrasies of visitors’ (Gerpott and Wanke, 2004: 245) into account, the perceived conceptualization states that simply adding more features to websites does not necessarily make them more interactive. Here, interactivity is seen as a user attribute (Leiner and Quiring, 2008), and questions of why certain interactive features appear to be judged as more or less interactive by various users are central (Quiring, 2009). The process view places its focus on the roles of the participants in interactive situations. Reciprocity between the participants is of interest, and this view defines interactivity as the relatedness of messages in communicative situations (Rafaeli and Sudweeks, 1997). Focusing on how visitors of newspaper websites use and appreciate different interactive features, this article will employ a perceived view of interactivity.

Indirect effects of interactivity One important strand in the research performed on interactivity has been the study of its effects. Even if ‘interactivity effects are still enigmatic’ (Rafaeli and Ariel, 2007: 84) trends can be discerned in the literature available. For example, scholars have identified what might be labeled indirect effects of interactive features – specifically, effects that take place simply by the user registering that an opportunity for interaction is present on the visited site. Such effects have been found to be positive, negative or both. As for positive indirect effects, Chung and Zhao (2004) studied user perceptions of internet advertisements and found that the perceived interactivity of these ads had a significant positive effect on the respondents’ attitudes towards the websites where the ads were placed. Similarly, Deuze noted an indirect effect of the ‘interactive bells and whistles’ available on newspaper websites (2003: 214). Indeed, web surfers seem to have more favorable attitudes towards sites that they perceive to be rich with interactive features (McMillan, 2002b). For newspaper websites, it has been suggested that an increase in interactivity will lead to more visitors and more on-site activity by these visitors (Gerpott and Wanke, 2004).

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1183

There might also be negative indirect effects caused by interactivity. Sites are perceived as bloated or difficult to navigate when too many interactive features are available. Sundar (2004) argues that an abundancy of interactive cues lead to overstimulation of the user and negative evaluations. Sohn et al. found that ‘increasing the level of Web site interactivity may not always yield positive communication outcomes’ (2007: 116). This might also hold true for newspaper sites, as higher degrees of interactivity might not fit in with the schema of newspaper readers (Gerpott and Wanke, 2004). Summing up, perhaps Rafaeli and Ariel are correct in that ‘Interactivity’s effects are curvilinear […] at a certain point, saturation sets in’ (2007: 80). Similarly, Hwang and McMillan (2002: 7) suggested that ‘limitless options might overwhelm consumers’. As such, a careful balance needs to be struck between too high and too low levels of interactivity.

Chung’s typology of interactive features There is currently no one clear definition of what interactivity entails (Koolstra and Bos, 2009). Regardless of conceptualization, choices must be made regarding what kind of interactive features should be examined. As pointed out by Quiring (2009), the term is associated with a number of different features and services. In the academic milieu, two types of interactive features can be distinguished (Stromer-Galley, 2004). First, human interactivity builds on what might be labeled a sociological definition of the term (Downes and McMillan, 2000). It places an emphasis on the conversational ideal of faceto-face communication (Schudson, 1978: 323). Human interactivity features include chat and discussion forums. Second, medium interactivity refers to interaction between a human user and a technical interface (i.e. a web page). Focus here is on user control (Downes and McMillan, 2000). Medium interactive features, in the context of online newspapers, includes the availability of video streams. Chung (2008; 2009) provided empirical data on interactivity in the context of online newspapers. Her respondents identified the two types of interactive features presented above, and also conceived of two combinations of them. These two additional types can be characterized as focused around the concept of web 2.0 – allowing users to take part in creating and shaping the online experiences (O’Reilly, 2005) and to ‘construct and share their own media and information products’ (Harrison and Barthel, 2009: 157). Example features from the four groups are provided in Figure 1 below. Human/medium interactive features allow users to ‘express their opinion’ (Chung, 2008: 666) and promote the engagement of users as co-producers (Boczkowski, 2002; Gillmor, 2004) by allowing them to contribute to the site. Examples include reader news stories, blogging opportunities or reader news tips. Medium/human features follow an adaptive ideal (Deuze, 2003) and allow users to customize site looks, content or use (Chung, 2008). Users might be allowed to personalize the news presented on the website, get access through RSS feeds or easily share site contents on social network services like Facebook or Twitter. This study will employ Chung’s typology of interactive features in order to study if any differences can be discerned regarding how different users perceive and use the four types of interactivity.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1184

new media & society 13(7)

Figure 1. Examples of interactive features. Clockwise from top left: Human (invitation to chat, screenshot from aftonbladet.se), Medium (video stream, screenshot from aftonbladet.se), Medium-Human (share to social network service, screenshot from sydsvenskan.se) and HumanMedium (reader news tip, screenshot from aftonbladet.se). All screenshots were taken from the start pages of the sites on 23 November 2010.

Method Data collection was performed by means of a web-based survey. Before the independent and dependent measurements are disclosed, the first two sections focus on survey design, pre-tests and respondent recruitment.

Survey design and pre-testing The survey was performed using the LimeSurvey application, installed on a university server. Numerous precautions were taken in order to maximize the number of participants and minimize dropouts. For example, the web domain pointing to the survey was clearly associated with the hosting university (Cho and LaRose, 1999) and the university logo was visible on all pages of the survey to indicate academic sponsorship (Porter and Whitcomb, 2005). A progress indicator was used (Crawford et al., 2001) and the

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1185

questions included in the survey were styled according to the suggestions of previous research (e.g. Couper et al., 2004; Reips, 2002). IP address information was logged to control for multiple answers (Chung, 2008). The survey was submitted to two waves of pre-tests. The first wave employed individual interviews with four independent evaluators. The focus was on clarifying any ambiguous formulations or phrasings and on assessing the optimal ordering of questions. The second wave had students at a large Swedish university college testing the survey. In total, 56 out of 82 invited students (68%) took the survey during this second wave. Feedback from the two waves was integrated into the final version of the survey.

Respondent recruitment Respondents were recruited via online advertisements on two of Sweden’s biggest newspaper websites, Aftonbladet (http://www.aftonbladet.se/) and Sydsvenskan (short for Sydsvenska Dagbladet) (http://www.sydsvenskan.se/). When site visitors clicked the invitation to the survey placed on the sites, they were first taken to an introductory page, featuring an informed consent form. The link to start the survey was available at the bottom of this page. In total, 1343 respondents who started the survey also completed it. Table 1 contains descriptive data regarding the respondent sources. Aftonbladet, a nationally distributed tabloid, was founded in 1830 as part of the Social Democratic press. As one of the first newspapers to go online in Sweden, Aftonbladet hosts one of Sweden’s most popular websites (Karlsson, 2006; Tezis, 2008). Sydsvenskan, a broadsheet distributed in southern Sweden, was founded in 1848 and adopts a more conservative stance towards current affairs. The online rendition of Sydsvenskan is one of the most popular newspaper websites in Sweden (KIA-Index, 2010). Both newspapers are best described as general or non-niche in terms of both contents and target audiences – levels of newspaper readership in Sweden are high and not socially determined (Hallin and Mancini, 2004; Tezis, 2008).

Variables The following two sub-sections explain the rationale for including the selected variables. Each section also explains how the different variables are used in the subsequent analyses. Table 1. Data regarding respondent sources Length of survey exposure on site

Daily circulation of newspaper

Unique site visitors/ week during time period

Respondents from site

Aftonbladet

18/3–28/3 (11 days)

5,124,774

940

Sydsvenskan

18/3–3/4 (18 days)

358,600 (February 2010) 121,200 (February 2010)

344,701

403

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1186

new media & society 13(7)

Dependent variables. Data from Chung (2008; Chung and Nah, 2009) and from Larsson (forthcoming) were used to select indicators for each of the four different categories in Chung’s typology. The newspaper websites were also visited beforehand in order to see if any features not previously identified could be found. In total, 14 features were employed as indicators. The respondents were asked to report their level of use and appreciation for each feature, using a series of seven-point Likert-type scales. 28 statements were posed to the respondents – 14 for use and 14 for appreciation. For the variables measuring use, a score of 1 indicated non-usage and a score of 7 indicated frequent use (i.e. ‘On a daily basis’). For example, the respondents were asked how often they took part in chat sessions (example of human interactivity) or how often they contributed to the site with their own news texts or news tips (both indicators used independently as examples of human-medium interactive features). For the variables measuring appreciation, the respondents were asked to judge a series of statements about the indicators. Using seven-point Likert-type scales, a score of 1 indicated disagreement and a score of 7 indicated the opposite. For example, respondents were asked to judge statements such as ‘I enjoy watching video streams when visiting the site’ (example of medium interactivity) or ‘If site visitors are given too much opportunity for individual customization, they might miss out on news of importance or public interest’ (example of mediumhuman interactivity, reverse coded in order to assess respondent attentiveness). These variables will be used in two ways. First, mean comparisons will uncover any statistically significant differences between use and appreciation of the different interactive features. Second, exploratory factor analysis will be utilized in order to find patterns and underlying factors relating to the use and appreciation of interactive features. Independent variables. In order to study variations in user traits (Tremayne, 2005) and how these traits might influence ‘user activation’ (Sundar, 2004: 387) in the context of online newspapers, this study employs the independent variables described below. Demographics. Data on respondent age, income, education etc. are often used in different social scientific settings. However, it seems that the effects of such sociodemographic data on use and perception of online interactivity has not been extensively examined. For example, Sohn and Lee (2005) argued that these kinds of respondent characteristics were not directly relevant for their study of respondent perception of interactivity. Chung (2008) used some of the sociodemographic data collected in her analysis (age, gender), while some were left out (education, income). In order to explore the effects of sociodemographic data on the use and appreciation of interactive features, data regarding respondent gender, age, education and income was collected. Societal engagement. In order to gauge the influences of various forms of societal engagement on use and perception of interactive features, this study employed eight variables measuring various aspects of such engagement. These variables were inspired by previous research (Chung, 2008; Jennings and Zeitner, 2003) and were adapted to the Swedish context. The question posed was: ‘Have you taken part in any of the following activities during the last year?’ For example, respondents were asked about their frequency of involvement in tenants’ associations/housing cooperatives, partaking in

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1187

political demonstrations and frequency of signing petitions and protest lists (on- or offline). For all questions, seven-point Likert-type scales were used, where 1 indicated the lowest score (i.e. ‘never’) and 7 the highest (i.e. ‘10 times or more’). Reliability for the measurements was assessed using Cronbach’s α, with a satisfactory result of .71. Media use. In order to assess the effects of media use, a series of questions were included. First, the respondents were asked ‘For how many years have You used the Internet?’ Second, for frequency of visits to the online newspaper sites, the respondents were asked ‘During an average week, on how many days do You visit the [NAME OF NEWSPAPER] web site?’ Similarly, for intensity of visits, respondents were asked to assess the amount of time spent on the newspaper’s website on an average day that they visited it. Additionally, the questionnaire included a multiple choice question that asked what sections of the website (e.g. news, entertainment, sports etc.) that the respondents tended to visit on an average day of accessing the specific site. Web-oriented Digital Literacy. Following Rafaeli and Ariel, ‘an advanced internet user […] might have a different interpretation and might perform differently with interactive features’ (2007: 82). Internet skill was assessed in two ways. First, a self-reported measurement was employed using a Likert-type scale, where the values 1 and 7 indicated the lowest and highest perceived skill level respectively. Second, Hargittai’s work on Weboriented Digital Literacy was consulted (2005, 2009). Following these studies, familiarity of computer- and internet-related terminology is a suitable proxy for observed internet skill measures. Using a series of seven-point Likert-type scales, the respondents were asked to indicate their level of familiarity with seven internet-related terms. A value of 1 indicated no knowledge of the term, and a value of 7 indicated a very high level of familiarity. This list of seven terms was completed with one bogus term in order to test whether the survey items were ‘simply checked off by respondents in a haphazard manner’ or not (Hargittai, 2009: 131). Following control for the bogus item, a composite measure of Web-oriented Digital Literacy was created by summing and averaging the remaining seven indicators. A Cronbach’s α score of .86 indicated a satisfactory level of reliability for the variable. The independent variables described above are used in two ways. First, descriptive statistics are reported in order to provide an overview of the sample. Second, in order to test the influences of these variables on the factors extracted from the previously described dependent variables, multiple regression analyses are performed.

Results A total of 1804 respondents took part in the survey and 1343 (940 from Aftonbladet, 403 from Sydsvenskan) completed it. The completion rate was 74 percent. In general, a slight majority of the respondents who completed the questionnaire were female (54%), and the mean age for respondents was 47 (SD = 14.64). Table 2 presents the means and standard deviations of the variables measuring use and appreciation of interactive features. The Wilcoxon signed rank test was employed to test whether the differences reported from the mean comparisons were significant or not.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1188

new media & society 13(7)

Focusing first on the column presenting results for the use variables as presented in Table 2, interactive features are used rather infrequently, although there are a few exceptions. The most frequently used feature was video clips, audio clips or image slideshows (M = 4.04, SD = 1.58), allowing the visitors choice options in experiencing news stories. Also, links to similar content appeared to be relatively popular (M = 3.50, SD = 1.64), indicating that visitors fairly often take the opportunity to click through to related material. Reader questions and polls also tend to be used more than other features (M = 3.24, SD = 1.71), as did the search features often available on newspaper websites (M = 2.60, SD = 1.59). However, since the means for these variables are all low, none of the features stand out as exceedingly popular. The differences when compared with the results for the appreciate variables are all highly significant (p = < .001), except for the questions regarding use and appreciation of commenting functionalities (p < .05) and the comparison between use and appreciation of video/audio/slideshows (p = > .05, NS). When comparing the results for use and appreciation, it is apparent that interactive features are more appreciated than used. For example, the contact staff functionalites are the most appreciated ones (M = 4.55, SD = 2.10), albeit not used that often (M = 1.21, SD = 0.58). These results echo the suggestion that users might not want to use the features offered, but that they appreciate the opportunity for use (Deuze, 2003: 214). The overall picture is one of appreciation rather than one of use – and even then, the levels of both use and appreciation must be considered low since the variables were measured on 1- to 7-point scales. Apparently, the availability of interactive features is not highly important for the respondents.

Table 2. Means and standard deviations for variables measuring use and appreciation of interactive features Feature

Use

H – Comment H – Chat H – Email to friend HM – Reader blogs/news HM – Contribute image/video HM – Contribute news tips HM – Contact staff HM – Reader questions/polls M – Video/audio/slideshow M – Links to similar content MH – Share content to SNS MH – News updates MH – Search MH – Customization

Appreciate

Wilcoxon signed-rank test

M

SD

M

SD

Z-value

1.52 1.11 1.53 1.18 1.06 1.11 1.21 3.24 4.04 3.50 1.32 1.26 2.60 1.18

1.13 0.48 1.06 0.73 0.39 0.47 0.58 1,71 1.58 1.64 0.90 0.99 1.59 0.76

3.66 2.73 3.25 3.61 2.78 3.29 4.55 4.11 4.02 4.41 2.36 2.10 4.49 2.66

1.84 1.76 1.92 1.79 1.67 1.60 2.10 1.78 1.82 1.77 1.77 1.54 1.99 1.64

26.93** 27.51*** 27.45*** 31.34*** 28.59*** 31.48*** 32.24*** 16.94*** .964 16.65*** 22.33*** 18.99*** 27.92*** 14.12***

p ! .001, **

p < .01, *

Reported z-values indicate significant differences at the *** respectively.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

p.  .05 levels

Larsson

1189

As discussed earlier, factor analysis and multiple regression analysis will be employed in order to identify user types. First, factor analysis using varimax rotation was performed. By examining the data in this way, we will be able to see if any underlying categories (or factors) emerge from the analyzed variables – factors that can then be tested against the independent variables. All 28 variables measuring use and appreciation of interactive features were used as items in the initial model. Items that cross-loaded on two or more factors were excluded from the final model reported below, as were items with factor loadings lower than .50. As shown in Table 3, the final model included 23 of the initial 28 variables. Results from KMO testing (.86) and Bartlett’s test of sphericity (Sig. < .000) indicated that the data was suitable for factor analysis. Consequently, the 23 items loaded on five factors, altogether accounting for 50.2 percent of the variance. Reliability analysis (using Cronbach’s α) was performed for each of the factors. As seen in Table 3, the α for factors 3, 4 and 5 fall below the often suggested lower limit of .70. However, each of these factors scored close to or above the .60 threshold often suggested for exploratory studies such as this one (Hair, 2010). Also, α is sensitive to the

Table 3. Factor analysis 1 – The Bystander Appreciate H – email article to friend Appreciate H – Chat Appreciate HM – Reader blogs/news Appreciate HM – Contact staff Appreciate M – Links to similar content Appreciate MH – Share content to SNS Appreciate MH – Search Use H – Comment Use H – Chat Use HM – Reader blogs/news Use HM – Contribute image/video Use HM – Contribute news tips Use HM – Contact staff Appreciate H – Comment Use HM – Reader questions/polls Use M – Video/audio/slideshow Use M – Links to similar content Use H – email article to friend Use MH – Share content to SNS Use MH – News updates Appreciate HM – Reader blogs/news Appreciate MH – News updates Appreciate MH – Customization Eigenvalues Variance explained Reliability (Cronbach’s D) Mean inter-item correlation

2 – The Prosumer

3 – The Lurker

4 – The Filter

5 – The Critic

.702 .658 .555 .711 .670 .608 .557 .677 .703 .739 .588 .554 .634 .589 .617 .703 .708 .582 .675 .580

3.46 15.03 .82 .40

2.67 11.61 .70 .31

1.97 8.58 .63 .30

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1.79 7.8 .56 .25

.567 .582 .646 1.66 7.2 .58 .30

1190

new media & society 13(7)

number of items in each factor (Streiner, 2003). Factors 3, 4 and 5 are made up of relatively few items each. In situations like these, reliability can be assessed by the mean inter-item correlation between the factor item (Streiner, 2003). Following previous research (Briggs and Cheek, 1986; Clark and Watson, 1995), mean inter-item correlation values for such items should range from .2 to .4. The reported values all fall within the suggested range. Consequently, the factors were considered reliable for further analysis. The factors were labeled according to their characteristics regarding use and appreciation of interactive features. The labels given were: The Bystander, The Prosumer, The Lurker, The Filter and The Critic. The first factor, The Bystander, is characterized by high scores on variables measuring appreciation, but not use, of interactive features. By contrast, the variables loading on The Prosumer are all measuring use, indicating a visitor who regularly contributes, chats and comments on the site. According to the items that make up The Lurker, this factor is characterized by a rather passive front, enjoying the comments of other visitors but with use variables limited to partaking in reader questions and polls. The fourth factor was labeled The Filter because of the tendency to share site contents with friends, via e-mail or social network sites. Finally, The Critic might be described as an opinionated visitor who appreciates various opportunities for visitor input and influence on the site. Second, multiple regression analysis was employed to test the influences of the independent variables on the five factors. By gauging these influences we take a step beyond the previously reported factor analysis and provide more detail for each of the factors, substantiating the visitor types. Specifically, the factors presented above were used as dependent variables respectively, while the same independent variables were used in all models. Table 4 presents the results of the analyses performed. Turning first to the sociodemographic predictor variables utilized in the analysis (Gender, Age, Education and Earnings), it is clear from Table 4 that at least one of these variables emerged as a significant predictor for all but one of the visitor types (The

Table 4. Multiple regression analysis Independent variables

1 – The Bystander

2 – The Prosumer

3 – The Lurker

4 – The Filter

5 – The Critic

Gender (0 female, 1 male) .092** .114*** .042 .063* .083** Age .078** .048 .022 .058 .077* Education .031 .034 .070 .001 .088** Earnings .076** .030 .037 .002 .038 Years online .029 .001 .003 .003 .005 Self-assessed internet skill .001 .063 .008** .070* .005 Measured internet skill .269*** .190*** .113** .191*** .164*** Civic/Political engagement .113*** .128*** .054 .128*** .054* Days per week .022 .049 .184*** .070* .011 Length of visit per day .067* .127*** .176*** .016* .054 Parts of site visited .127*** .080** .189*** .041 .080** R (R2) .361 (.131) .369 (.136) .428 (.183) .289 (.084) .250 (.063) Standardized Beta values presented. Significance levels are reported at the *** respectively.

p  .001, **

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

p  .01, *

p  .05

Larsson

1191

Lurker). This indicates that the characteristics of this particular type are rather spread among various parts of the population, making it hard to predict The Lurker based on sociodemographic data only. As for the variables focusing on online experience and skill, the first of these (Years online) did not turn out to be a significant predictor for any of the visitor types. This could be related to the high levels of internet penetration in Sweden, as well as to the fact that internet use in Sweden has spread quickly since the mid-1990s. With comparably high levels of internet experience spread throughout the population, variations regarding this characteristic might be hard to uncover. As such, no clear difference could be discerned between those who have used the internet for longer periods of time and those who are relative newcomers to the medium. Moreover, while the Measured internet skill emerges as a highly significant predictor for all of the five visitor types, the Self-assessed internet skill variable only significantly predicts two of the types, The Lurker and The Filter. As stated earlier, civic and political activity are often found to have influences over online activity such as commenting or contributing content. The variable measuring Civic/Political engagement emerged as a significant predictor for all visitor types but The Lurker. Apparently, this visitor type adopts a passive stance in the online as well as the offline environment. Finally, three variables were included that measured the intensity of use. The first of these, Days per week, emerged as a significant predictor for The Lurker and The Filter, indicating that these types tend to visit the newspaper’s sites more frequently than the other types. As for Length of visit per day, the only visitor type not significantly predicted by this variable was The Critic. This type stands out as his rate of visitations and time spent on the site are not easily predicted. The last variable included in the analyses (Parts of site visited) assessed how many parts of the site that the respondent engaged with on a typical day of visiting the website. As shown in Table 4, the variable significantly predicted four of the five visitor types, with The Filter as the exception. This could be interpreted as a more selective attitude of this type towards what parts of the site are visited when the site is accessed. Results indicate that The Filter is a more focused visitor than the other identified types, concentrating his or her visit to selected parts of the site. The main characteristics of the five visitor types can be summarized as follows. The Bystander can be described as a passive visitor of newspaper websites, who does not use interactive features, but who tends to appreciate their presence. The Bystander is a young, internet-savvy female with low earnings characterized by infrequent visits to many different parts of the newspaper site. The arguably most active visitor type, The Prosumer, tends to be a man with high levels of internet skill and societal engagement. He does not visit the site on a daily basis, although when he does, he spends a comparatively large amount of time on the site and tends to visit different sections of it. Rather few of the independent variables employed in this study helped predict The Lurker, a visitor type characterized mainly by passive behavior. This type stands out as the only one not predicted by the variable measuring civic and political engagement. The Lurker is also a very frequent and intense visitor, with highly significant positive values for all variables measuring site visits (Days per week, Length of visit per day, Parts of site visited). The Filter is often a female, quite focused visitor. This could relate to the fact that The Filter

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1192

new media & society 13(7)

tends to use news updates and therefore does not have to visit as many parts of the site in order to get their information as other types might have to. Finally, results indicate that The Critic often is a man, further characterized by his higher age and lower level of education. He appreciates user-generated content and news update services and would like to see more options for user customization of the sites. Following the non-significant results for the variables measuring frequency and intensity of visit (Days per week and Length of visit per day) we can conclude that he does not appear to be a habitually low or high frequency visitor. However, when he chooses to visit newspaper websites, The Critic tends to visit several parts of it, as suggested by the significant result for the Parts of site visited variable. In sum, the results presented above suggest that the identified visitor types take rather different approaches towards the newspaper websites that they frequent. The results also give rise to a number of over-arching questions regarding visitor use and appreciation of interactive features in the newspaper website context. These questions are raised for further discussion in the final section of this article.

Discussion According to Chung (2008), interactivity can be viewed as a continuous phenomenon, ranging from lower to higher forms of interactivity (see also Stromer-Galley, 2004). Higher forms of interactivity would include features that ‘require more effort in that individuals must do more than clicking or selecting in order to actively use them’ (Chung, 2008: 661). As is evident from the results presented here regarding use of interactive features, the respondents tended not to use various higher forms of interactivity such as Human (i.e. commenting on news items or chatting) or Human-Medium (i.e. contributing reader news or blog posts) interactive features. As such, there appears to be little interest in participating in various grassroots journalism efforts (Gillmor, 2004). Although lower forms of interactivity (Medium and Medium-Human interactive features) were used more, the overall tendency is one of non-use rather than one of use. This suggests a rather passive stance as the norm when it comes to use of interactive features – lower features allowing clicking or selecting seem sufficient for most visitors. Focusing on the results regarding appreciation, they show that even though newspaper website visitors do not use interactive features to any higher degree, they do tend to appreciate the presence of such features. Even if these indirect effects (Deuze, 2003; Gerpott and Wanke, 2004) of interactivity are present, especially when compared to the use of the features, levels of appreciation still remain rather low. With results indicating low levels of both use and appreciation of interactive features, the average newspaper website visitor could be described as somewhat jaded and uninterested in the opportunities to interact and contribute provided by the media organizations. In sum, the results suggest that opportunities to interact are rarely acted upon by the visitors. However, the identification of five visitor types has provided some noteworthy insights into how different visitors make use of interactive features. Although two of these types are distinguished by their non-use (The Bystander and The Critic), the three remaining types (The Prosumer, The Lurker and The Filter) are all characterized by various forms of use.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1193

Taking the aforementioned distinction between higher and lower forms of interactivity into account, the kinds of features associated with these visitor types are mostly of the less demanding lower variants. With the exception of The Prosumer, most visitors seem content with keeping more active participation to a minimum, using the ‘bells and whistles’ (Deuze, 2003: 214) but rarely contributing self-authored news texts, blog posts or pictures from news events. In this regard, the visitors could be understood as ‘slow learners’. Perhaps as a result of the Swedish press enjoying strong societal positions in both off- and online contexts (Bergström, 2008), readers see the news reported to them as a finished product, not to be tampered with or augmented by non-journalists. The results presented in this study indicate that news consumers still see themselves as precisely that – consumers. In the Finnish context, Hujanen and Pietikainen (2004) found similar attitudes among their respondents, suggesting that the transition from news recipients to active participants might take longer time than was perhaps expected. According to the findings presented above, these results are valid also in the Swedish context. The identification of newspaper website visitor types, their habits and characteristics, should be of interest to scholars who do research into these and similar matters. The results presented above are also relevant for practitioners in the media industry as well as for other professionals who work in similar areas of online publishing. On the one hand, practitioners might be relieved that rather few of their visitors want to contribute to any greater extent. From the early days of the internet, the media have taken on the role of ‘cautious traditionalists’ (Chung, 2007: 52), when attempting to adapt to the new medium. This wary attitude might derive from what is sometimes described as traditional journalistic schooling (Domingo, 2008; Gillmor, 2004). On the other hand, this status quo of the sender–receiver relationship might be challenged by tendencies that are increasingly emanating from online newsrooms in Sweden (Engebretsen, 2006; Frisk, 2008; Karlsson, 2006) and elsewhere (Chung, 2004; Mitchelstein and Boczkowski, 2009). For example, younger media practitioners seem to take a more open view when it comes to audience participation (Deuze and Dimoudi, 2002; Larsson, forthcoming). As such, the readers might not be alone in being characterized as ‘slow learners’ – it might take time for old and new generations of media professionals to adapt to the new possibilities as well. If media industry professionals are serious about creating more than ‘readers’ playgrounds’ (Ye and Li, 2006) on their websites, they should strive to adapt their sites to fit the needs of the audience. This challenge could be approached with the visitor types identified above in mind. The characteristics associated with these types should be of interest to professionals who strive to create appealing and suitably interactive websites for their visitors. The result that visitors tend to appreciate features more than they actually use them should be interesting in this regard. In a shorter temporal perspective, practitioners might want to draw on the results regarding appreciation of interactive features presented above and augment their sites with these features (following McMillan and Hwang, 2002) in order to please the visitors. As for long-term goals, practitioners need to set up interactive structures on their sites that could attract visitors not interested in interaction to take part in higher forms of interactivity, combining appreciation of these features with actual use. Sohn and Lee (2005) highlighted the fact that although several studies have emphasized the need for highly interactive websites in order to successfully communicate with visitors, few of these studies suggested

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1194

new media & society 13(7)

guidelines for designing such sites. The visitor types, their characteristics and preferences regarding interaction presented in this article should be helpful for online news practitioners planning and developing lively and competitive websites. As online news consumers take part of news and interactive services via a variety of different sites using a multitude of devices (i.e. smart phones, tablets or desktop computers) to access these sites, the study of interactivity in online news becomes more complex. While this article did not take this breadth of options into account, the findings presented here provide the research community with important insights regarding use and appreciation of interactive features in the online newspaper context. To what extent these insights transfer to other emerging devices remains to be studied. Taking these and other limitations into account, this final section of the article goes on to suggest directions for future research on similar topics. First, the results presented above suggest that the Swedish online newspaper audience is hesitant to interact – perhaps because of what might be described as an ‘institutional respect’ towards the media organizations. This suggests the need for comparative studies with countries where the media do not enjoy this position. Previous research could identify suitable countries for comparison, whose media systems could provide contrast to the Swedish context (i.e. Hallin and Mancini, 2004). Second, future studies should also take longitudinal perspectives into account. Studying the phenomenon at handover time will allow the research community to gain insight in how various groups of users change their web behavior and expectations (Leiner and Quiring, 2008: 147). Third, although the employment of survey methodology has several advantages and provides the research community with an overall view of a specific phenomenon, this particular data collection rationale also has its disadvantages. Following Hwang and McMillan, researchers should also consider using various qualitative methods in order to ‘provide insight into how consumers develop attitudes toward Web sites’ (2002: 7). While the results presented in this article provide insights into how online newspaper visitors use and perceive the interactive features offered to them, future use of qualitative data can provide a deeper understanding of what drives site visitors to use interactivity. For example, talk-along procedures, in-depth interviews or focus groups could provide suitable methodological approaches to take the next step to let us gain more knowledge about characteristics, preferences and behaviors of the visitor types. Fourth, while the focus here is upon interactivity, user experience of websites is also dependent on factors like design, usability aspects and the actual contents of the site. As such, future studies should attempt to broaden their scope and study interactivity from multiple perspectives, perhaps focusing on specialized news sites. Finally, even though the suggestion was previously made for practitioners to augment their sites with interactive features according to the preferences of the different visitor types, the curvilinear effect of interactivity mentioned earlier should be kept in mind (Rafaeli and Ariel, 2007: 80). Too many options to interact might obstruct the user experience, resulting in negative feelings towards the site rather than appreciation (Bucy, 2004). As such, scholars should study the effects of different levels of interactivity on different types of visitors in the online news media context. The visitor types identified above should provide a suitable starting point for these and other ideas for future studies.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1195

References Bergström A (2008) The reluctant audience: Online participation in the Swedish journalistic context. Westminster Papers in Communication and Culture 5(2): 60–79. Boczkowski P (2002) The development and use of online newspapers: What research tells us and what we might want to know. In: Lievrouw L and Livingstone S (eds) The Handbook of New Media: Social Shaping and Consequences of ICTs. London: Sage, 270–286. Briggs SR and Cheek JM (1986) The role of factor analysis in the development and evaluation of personality scales. Journal of Personality 54(1): 106–148. Bucy EP (2004) Interactivity in society: Locating an elusive concept. The Information Society 20(5): 373–383. Carlsson U and Facht U (2010) Medie-Sverige 2010. Göteborg: NORDICOM. Cho H and LaRose R (1999) Privacy issues in internet surveys. Social Science Computer Review 17(4): 421–434. Chung DS (2004) Into interactivity? How news websites use interactive features. Paper presented at the ICA annual convention, New Orleans, LA. Chung DS (2007) Profits and perils: Online news producers’ perceptions of interactivity and uses of interactive features. Convergence 13(1): 43–61. Chung DS (2008) Interactive features of online newspapers: Identifying patterns and predicting use of engaged readers. Journal of Computer-mediated Communication 13(3): 658–679. Chung DS and Nah S (2009) The effects of interactive news presentation on perceived user satisfaction of online community newspapers. Journal of Computer-mediated Communication 14(4): 855–874. Chung H and Zhao X (2004) Effects of perceived interactivity on web site preference and memory: Role of personal motivation. Journal of Computer-mediated Communication 10(1). Clark L and Watson D (1995) Constructing validity: Basic issues in objective scale development. Psychological assessment 7(3): 309–319. Couper, MP, Tourangeau R, Conrad FG et al. (2004) What they see is what we get: Response options for web surveys. Social Science Computer Review 22(1): 111–127. Crawford SD, Couper MP and Lamias MJ (2001) Web surveys: Perceptions of burden. Social Science Computer Review 19(2): 146–162. Deuze M (2003) The Web and its journalisms: Considering the consequences of different types of newsmedia online. New Media & Society 5(2): 203–230. Deuze M and Dimoudi C (2002) Online journalists in the Netherlands: Towards a profile of a new profession. Journalism 3(1): 85–100. Domingo D (2008) Interactivity in the daily routines of online newsrooms: Dealing with an uncomfortable myth. Journal of Computer-mediated Communication 13(3): 680–704. Domingo D, Quandt T, Heinonen A et al. (2008) Participatory journalism practices in the media and beyond. Journalism Practice 2(3): 326–342. Downes EJ and McMillan, SJ (2000) Defining interactivity: A qualitative identification of key dimensions. New Media & Society 2(2): 157–179. Engebretsen M (2006) Shallow and static or deep and dynamic? Studying the state of online journalism in Scandinavia. Nordicom Review 27(1): 3–16. Fortunati L and Sarrica M (2010) The future of the press: Insights from the sociotechnical approach. The Information Society 26(4): 247–255. Frisk J (2008) Medier bjuder in engagerade läsare. Journalisten, 18 March. Gerpott TJ and Wanke H (2004) Interactivity potentials and usage of German press-title web sites: An empirical investigation. Journal of Media Economics 17(4): 241–260. Gillmor D (2004) We the Media: Grassroots Journalism by the People, for the People. Sebastopol, CA: O’Reilly.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

1196

new media & society 13(7)

Hair JF (2010) Multivariate Data Analysis: A Global Perspective. Upper Saddle River, NJ: Pearson. Hallin DC and Mancini P (2004) Comparing Media Systems: Three Models of Media and Politics. New York, NY: Cambridge University Press. Hargittai E (2005) Survey measures of web-oriented digital literacy. Social Science Computer Review 23(3): 371–379. Hargittai E (2009) An update on survey measures of web-oriented digital literacy. Social Science Computer Review 27(1): 130–137. Harrison TM and Barthel B (2009) Wielding new media in Web 2.0: Exploring the history of engagement with the collaborative construction of media products. New Media & Society 11(1–2): 155–178. Hujanen J and Pietikainen S (2004) Interactive uses of journalism: Crossing between technological potential and young people’s news-using practices. New Media & Society 6(3): 383–401. Hwang J and McMillan SJ (2002) The role of interactivity and involvement in attitude toward the web site. Paper presented at the Conference of the American Academy of Advertising, Auburn, AL. Jennings MK and Zeitner V (2003) Internet use and civic engagement. Public Opinion Quarterly 67(3): 311–334. Karlsson M (2006) Nätjournalistik: en explorativ fallstudie av digitala mediers karaktärsdrag på fyra svenska nyhetssajter. Lund: Lund University. KIA-Index (2010) Web statistics provided by the Association of Swedish Advertisers. Available at: http://www.kiaindex.net/ Koolstra CM and Bos MJW (2009) The development of an instrument to determine different levels of interactivity. International Communication Gazette 71(5): 373–391. Larsson AO (forthcoming) Interactivity on Swedish newspaper websites: What kind, how much and why. Convergence: The International Journal of Research into New Media Technologies. Leiner DJ and Quiring O (2008) What interactivity means to the user: Essential insights into and a scale for perceived interactivity. Journal of Computer-mediated Communication 14(1): 127–155. McMillan SJ (2000) Interactivity is in the eye of the beholder: Function, perception, involvement, and attitude toward the website. In: Shaver MA (ed.) Proceedings of the 2000 Conference of the American Academy of Advertising. East Lansing, MI: AAA, 71–78. McMillan SJ (2002a) Exploring models of interactivity from multiple research traditions: Users, documents, and systems. In: Lievrouw L, Livingstone S (eds) The Handbook of New Media: Social Shaping and Consequences of ICTs. London: Sage, 163–182. McMillan SJ (2002b) A four-part model of cyber-interactivity: Some cyber-places are more interactive than others. New Media & Society 4(2): 271–291. McMillan SJ and Hwang JS (2002) Measures of perceived interactivity: An exploration of the role of direction of communication, user control, and time in shaping perceptions of interactivity. Journal of Advertising 31(3): 29–42. Mitchelstein E and Boczkowski PJ (2009) Between tradition and change: A review of recent research on online news production. Journalism 10(5): 562–586. O’Reilly T (2005) What is Web 2.0? Design Patterns and Business Models for the Next Generation of Software. Available at: http://www.oreillynet.com/lpt/a/6228 Porter SR and Whitcomb ME (2005) E-mail subject lines and their effect on web survey viewing and response. Social Science Computer Review 23(3): 380–387. Quiring O (2009) What do users associate with ‘interactivity’? A qualitative study on user schemata. New Media & Society 11(6): 899–920.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson

1197

Rafaeli S and Ariel Y (2007) Assessing interactivity in computer-mediated research. In: Joinson AN (ed.) The Oxford Handbook of Internet Psychology. Oxford: Oxford University Press, 71–88. Rafaeli S and Sudweeks F (1997) Networked interactivity. Journal of Computer-Mediated Communication 2(4): 1–6. Reips U-D (2002) Internet-based psychological experimenting: Five dos and five dont’s. Social Science Computer Review 20(3): 241–249. Schudson M (1978) Ideal of conversation in study of mass-media. Communication Research 5(3): 320–329. Singer JB (2010) Quality control. Journalism Practice 4(2): 127–142. Singer JB and Ashman I (2009) ‘Comment Is Free, but Facts Are Sacred’: User-generated content and ethical constructs at the Guardian. Journal of Mass Media Ethics 24(1): 3–21. Sohn D and Lee BK (2005) Dimensions of interactivity: Differential effects of social and psychological factors. Journal of Computer-mediated Communication 10(3). Sohn D, Ci C and Lee BK (2007) The moderating effects of expectation on the patterns of the interactivity-attitude relationship. Journal of Advertising 36(3): 109-119. Song JH and Zinkhan GM (2008) Determinants of perceived web site interactivity. Journal of Marketing 72(2): 99–113. Streiner DL (2003) Starting at the beginning: An introduction to coefficient alpha and internal consistency. Journal of Personality Assessment 80(1): 99–103. Stromer-Galley J (2004) Interactivity-as-product and interactivity-as-process. The Information Society 20(5): 391–394. Sundar S (2004) Theorizing interactivity’s effects. The Information Society 20(5): 385–389. Tezis G (2008) European Media Governance: National and Regional Dimensions. Bristol: Intellect. Thurman N (2008) Forums for citizen journalists? Adoption of user generated content initiatives by online news media. New Media & Society 10(1): 139–157. Tremayne M (2005) Lessons learned from experiments with interactivity on the Web. Journal of Interactive Advertising 5(2). Ye X and Li X (2006) Internet newspapers’ public forum and user involvement. In: Li X (ed.) Internet Newspapers: The Making of a Mainstream Medium. Mahwah, NJ: Lawrence Erlbaum Associates, 243–260.

Anders Olof Larsson is a PhD candidate at the Department of Informatics and Media, Uppsala University, Sweden. He is also associated with The Swedish Research School of Management and Information Technology. His research focuses on organizational and audience use of online interactivity and social media. His website can be found at http:// www.andersoloflarsson.se

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Paper III

“Extended Infomercials” or “Politics 2.0”? A Study of Swedish Political Party Web Sites Before, During and After the 2010 Election

Anders Olof Larsson Department of Informatics and Media Uppsala University andersoloflarsson.se [email protected] Published in First Monday, volume 16, issue 4. Available online at: http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view /3456/2858

Abstract Although many of the initial hopes regarding the effect of the Internet on political engagement and participation have largely gone unfulfilled, it is generally held that the Internet still has a substantial role to play during political election campaigns. Several studies have focused on how the Internet medium is employed for such purposes during the actual election campaign, but rather few (if any) studies have adopted a broader temporal scope, studying the web sites of political parties before, during and after the election period. This paper fills this apparent research gap by presenting such a longitudinal analysis of the web sites of Swedish political parties during the election year of 2010. Starting in January of 2010, these web pages were downloaded on a monthly basis, lasting until the end of the year. By studying the web sites of political parties before, during and after an election campaign, this project will provide scholars as well practitioners with unique insights into how web campaigning rationales seem to develop.

1

Introduction The rise of new technologies has always been seen as playing key parts in shaping the processes of electioneering. From Roosevelt’s “great radio voice” via Kennedy’s television persona to Barack Obama’s apparent social media success (Foot & Schneider, 2006, p. 7; Kalnes, 2009, p. 264; Robert, 2009), the development of technological innovations is frequently described as having significant effects on political campaigning (Alberto & Corinna di, 2009). Scholars have referred to the Internet as the “magic elixir” to raise voter participation in the process of government (Stromer-Galley, 2000, p. 113; Vaccari, 2008a, p. 648). While an online presence is virtually mandatory when on the campaign trail (Druckman, Kifer, & Parkin, 2007, p. 426; Roodhouse, 2009), the early promises of increased political engagement and revitalized democracy through online activity appear to have been highly optimistic and have as such gone largely unfulfilled (Chadwick, 2006). Recent online developments and the increasing popularity of accessible web design concepts such as various Web 2.0 or social media technologies (i.e. Cormode & Krishnamurthy, 2008; O’Reilly, 2005) has yet again raised the question of how web campaigning can be used in order to engage the denizens of the web. While several studies have dealt with the web use of political actors, the majority of these have been performed in US or UK contexts (Chadwick, 2006; S. Coleman, 2005; Coleman, 2004; Schweitzer, 2008). As such, there is a pertinent need to focus on other democracies and the ways that the internet has been used for political purposes there (Strandberg, 2009, p. 851; Vaccari, 2008c, p. 17). In this regard, Sweden is an interesting country to study. From early on, Swedish political parties have maintained an active web presence (Gibson, 2004, p. 96) - perhaps a necessity given the high rate of Internet dissemination and use among Swedish citizens (Facht, 2008). While studies have focused on the use of web 2.0 and similar applications during the height of political campaigning (i.e. a few weeks before a set election date), few studies have focused on how political actors make use of their web sites in a longer temporal perspective. This paper presents a longitudinal study of Swedish political party web sites before, during and after the 2010 Swedish general election. While some studies have suggested that party web sites undergo relatively few changes during political campaigning (i.e. Gibson, Lusoli, & Ward, 2008, p. 20), other studies have presented results that indicate the opposite (i.e. Foot & Schneider, 2006, p. 16; Vaccari, 2008c, p. 8). Based on data continuously collected during a twelve-month period, the paper provides unique insights into the developmental processes that political party web sites undergo during an election year. By analyzing the different features employed by these sites before and after the election, the paper contributes to the ongoing dis2

cussion on whether online technological developments have had any substantial effects on how political actors use their web sites in order to engage with the public (Vaccari, 2008a, p. 648; Zittel, 2009, p. 298), or if the potential for change in web campaigning has been thwarted by what is sometimes referred as traditional or ”top-down” electioneering (Gibson, 2004, p. 103; Kalnes, 2009, p. 251).

Background Political campaigning in the digital age – shift or enhancement? Throughout the 20th century, the printed press, radio broadcasting and television all made their marks on how political actors use technological innovations during election campaigns. During its introductory phase in western democracies, the Internet was often described as having the potential to make a large impact on political campaigning (i.e. Castells, 2001; S Coleman, 2005). Indeed, the thoughts of an “informational democracy” (Castells, 1996, p. 353) facilitated through “postmodern campaigning” (Vaccari, 2008a, p. 648; Zittel, 2009, p. 298) were frequently expressed under a variety of labels. Most early research efforts on the potential of the Internet to rejuvenate contemporary democracy were inspired by the concepts of deliberation and direct democracy often associated with the ideal of the public sphere (Bohman, 1996; Habermas, 1989). The internet appeared to offer the means to realize more inclusive public participation and discussions between political elites and citizens (Rash, 1997). Remembering the claim that periods of technological development tend to invite speculation about radical systemic changes (Kalnes, 2009, p. 251) it nevertheless seems valid to conclude that hopes were high for the Internet to revitalize political interest and discussion among the general public (Jackson & Lilleker, 2009, pp. 233-234). However, the mostly somber assessments (Davis, 2010, p. 747; Vaccari, 2008c, p. 6) of research focusing on that particular time period suggest that the new medium was not employed in the groundbreaking ways first imagined. Rather than providing a ”shift” in political campaigning and communication, the adoption of the Internet by political actors led instead to an ”enhancement” of existing patterns of civic engagement, generally providing information rather than opportunities for participation (Vaccari, 2008a, p. 649; 2008c, pp. 1-2; Ward & Vedel, 2006). Although there are exceptions (Gibson, 2004, p. 96) an apparent ”Burkean, Top-Down communication strategy” (Jackson & Lilleker, 2009, p. 247) where providing ”vehicles for downward dissemination of information rather than recruitment of users’ 3

opinions and the promotion of participation” (Gibson, 2004, p. 103) is evident in research findings from several countries. For example, in his study of the 2003 Finnish parliamentary election, Strandberg (2009) found that most candidates did not choose to campaign online, and that the majority of those who did employed web sites mainly used for information provision rather than promoting participation. The results presented by Bergström (2007) in her study of the 2006 Swedish elections indicated that the Internet contributed to an enhancement of established political practices as discussed earlier. During the 2006 Italian elections, Vaccari described the adoption of Internet tools by Italian parties as ”slow and half-hearted” (2008b, p. 75). In France, the same author found few opportunities for voter participation (Vaccari, 2008c, p. 16). Lilleker and Malagon’s (2010) study from the same election seem to corroborate this result, as they found few opportunities for conversations between electors and elected. Indeed, the web sites of political actors have been described as ”extended infomercials” (Stromer-Galley, 2000) or ”electronic brochures” (Druckman, et al., 2007; Jackson & Lilleker, 2009), indicating that little adaptation to the interactive opportunities of Internet medium has taken place. Some of the reasons for this careful approach towards the new medium are discussed in the next section.

The non-use of internet-specific features The capacity for change towards a more participatory form of electioneering brought forward by the advent of the Internet should not be overemphasized at the expense of continuity (Kalnes, 2009, p. 251). Indeed, each innovation has advantages and drawbacks (Druckman, et al., 2007, p. 425), and this certainly holds true in the digital era of political campaigning. The perhaps biggest drawback of fully utilizing web sites for campaigning purposes is the risk of losing control over the content (Stromer-Galley, 2000, p. 124). While such decentralizing services as chat rooms and discussion boards would be central to user participation, allowing proponents (and possibly also opponents) to engage in on-site discussion might risk ”negative graffiti-style comments” (Jackson & Lilleker, 2009, p. 246) and could distort the intended messages of the campaign (Druckman, et al., 2007, p. 434). The dynamic potential of the Internet does not seem to fully harmonize with the established tenets of political campaigning, which historically have been more focused on provision of information and attempts at persuasion rather than engaging in peer-to-peer dialogue on any larger scale (Jackson & Lilleker, 2009, p. 235). While the potential for information distribution via the Internet is obvious, the risks of allowing visitors to co-create online content are apparently looming (Lilleker & Malagón, 2010, p. 26). Political actors might choose to employ certain features made available by the Inter4

net (i.e. information provision) to higher degrees than other features (i.e. participatory features or features that allow for site visitor contributions). Furthermore, highly dynamic web sites featuring tools for participation require larger staff and continuous efforts throughout a campaign period in order to sort and filter the incoming materials. Indeed, chat rooms, discussion boards and incoming messages, photos and videos require more than ”just one-off software engineering” (Druckman, et al., 2007, p. 430; Vaccari, 2008c, p. 10), demanding allocation of resources that might not always be available to smaller political actors. Recent online developments towards a ”Web 2.0” rationale of web design have led some researchers to the suggestion that the previously described era of ”electronic brochures” could be facing serious challenges. Indeed, in his interviews with senior campaign officials during the 2004 US presidential campaign, Vaccari reported his respondent’s beliefs that the Internet would become increasingly important in future campaign situations (2008a). In the following section, this notion of a matured form of online campaigning, a ”Politics 2.0”, is discussed in conjunction with the implied notion of ”Politics 1.0”.

”Tired old politics 1.0” and ”fresh politics 2.0” While the early experiments with online campaigning seem to have ended in a status quo of ”politics-as-usual” (Margolis & Resnick, 2000), recent online developments including a ”Web 2.0” rationale of web design are frequently described as having the potential to transform online political campaigning. As such, the normalization hypothesis, suggesting that online patterns of political activity will come to resemble those in the offline environment (Hara & Jo, 2007; Margolis, Resnick, & Wolfe, 1999, p. 24), is under challenge yet again from what is often labeled the innovation hypothesis. According to this latter line of thought, the development of web campaigning will lead to new and innovative ways of engaging the electorate, providing a change from typical offline patterns of electioneering (Schweitzer, 2008, p. 450). Where a ”Web 1.0” rationale appears to have been focused on the dissemination of information to broader audiences (Jackson & Lilleker, 2009, p. 233; Kalnes, 2009), its 2.0 successor is often described as placing its focus on principles such as user participation, openness and network effects (O’Reilly, 2005; Pascu, Osimo, Ulbrich, Turlea, & Burgelman, 2007). Although somewhat enigmatic and not entirely easy to define (i.e. Anderson, 2006), the concept has proved to be popular in everyday online activities (i.e. Grossman, 2006) as well as in specifically political contexts. A variety of Internet services and applications, such as the video sharing site YouTube or social networking tools like Facebook or Twitter, are seen as providing new 5

possibilities for politicians to interact with the electorate (Castells, 2007, p. 255). Indeed, incremental use of these and other 2.0 applications by candidates up for election has been reported by researchers (Lilleker & Malagón, 2010, p. 26; Vaccari, 2008a, p. 663). The suggested transition from ”tired old politics 1.0” to ”fresh politics 2.0” (Kalnes, 2009) has been evident in the US political context, where different forms of cyber-campaigning have evolved since the 1996 election cycle (Gibson, 2004, p. 96). Although use of political web sites had become ”virtually ubiquitous” by the 2000 presidential elections (Gibson, 2004, p. 98), the subsequent 2004 election is often pointed to as having spurred a watershed change in online political activities, with web users demanding more than mere information on party politics and talking points from the web sites of the candidates (Foot & Schneider, 2006, p. 11; Vaccari, 2008a, pp. 647-648). According to the Pew Internet and American Life Project, some 75 million Americans used the Internet for campaign-related activities like information gathering or participating as campaign volunteers during 2004. (Rainie, Cornfield, & Horrigan, 2005). Similar indications of heightened demand for online political activities have also been reported from outside of the US context (Jackson & Lilleker, 2009, p. 236). Following the apparent social media success of Barack Obama’s 2008 presidential campaign (Kalnes, 2009, p. 264), it seems as though many political actors have recently woken up to the electoral potential of new internet tools like the ones mentioned earlier. In sum, the recent developments towards a Web 2.0 rationale have yet again raised the hopes for the Internet to play the role as a catalyst for democratic engagement and political deliberation. Although such optimistic hopes have been raised before, many of the attempts at ”politics 1.0” seem to have resulted in mere ”campaign gimmicks” rather than the enhancement of democratic life first imagined (Lilleker & Malagón, 2010, p. 26). With both voters and candidates established and experienced in the online environment, perhaps the stage is set for the next phase in online political campaigning, a ”politics 2.0”, where focus is placed on user participation rather than on information dissemination (Kalnes, 2009, p. 254). The following and final section of this background chapter details the features often employed by political web sites, in both the 1.0 and 2.0 traditions.

Features of political web sites Among the many frameworks and typologies available for the classification and analysis of political web site features, the four-part typology employed by Foot and Schneider (2006) will serve as the starting point for this study. 6

The four types of online functionalities suggested by the authors are labeled informing, involving, connecting and mobilizing. This fourfold and its different components is described in the following sections. The practice of informing entails providing information on party and/or candidate issue positions and plans. In the context of online campaigning, techniques for informing involve providing documentation regarding party and/or candidate history, ideology and talking points (Foot & Schneider, 2006, p. 50 ff; Kann, Berry, Grant, & Zager, 2007; Vaccari, 2008c). The practice of informing has been evident in online campaigning since the earliest attempts at online political activity (Foot & Schneider, 2006, p. 48), providing parties with opportunities to inform and possibly also convert undecided voters into supporters (Bimber & Davis, 2003, p. 48). Candidates must decide how much information to provide, as too much information can overwhelm site visitors, and too little information may leave visitors unsatisfied (Foot & Schneider, 2006, p. 61). Information provision can also be seen as hindering ”strategic ambiguity” (Foot & Schneider, 2006, p. 62; StromerGalley, 2000, p. 125) - practices of non-disclosure that allow politicians to ”becloud their policies in a fog of ambiguity” (Downs, 1957, p. 136) in order to seem appealing to voters of different political persuasions. While practices of informing are commonplace in web campaigning, features that allow the practice of involving are scarce (Robert, 2009). Foot and Schneider define involving as ”facilitating interaction between site visitors and the campaign organization” (Foot & Schneider, 2006, p. 70). While the basic techniques for involving, such as opportunities to contact the campaign are quite common, more advanced features (such as discussion boards or live chats with politicians) are rather uncommon (Vaccari, 2008c, p. 1). Involving visitors in these ways is not risk free. Involving features are laborintensive and larger organizations are needed to sort the incoming contributions (Vaccari, 2008c, p. 10). Indeed, most previous research efforts have found involving practices to be limited to providing contact information (Foot & Schneider, 2006, p. 99). The practice of connecting is associated with facilitating interaction with other online actors and providing context for the campaign (Foot & Schneider, 2006, p. 103). In linking to other actors, the campaign can be described as deeming those actors important in some way (Rogers & Marres, 2000, pp. 16-17). The practice of connecting also provides the visitors of campaign web sites with a way to place the party in a specific cognitive context or frame (Foot & Schneider, 2006, p. 105), making the candidate ”understandable” in relation to other societal actors. As discussed earlier, connecting might not be as straightforward a practice as it first might seem to be. Giving away the attention of the visitor to some other, linked site might not be very desirable in a campaigning context (Foot & Schneider, 2006, p. 125). Simi7

larly, linking also implies a loss of control over the information presented at the linked-to site – information that is potentially not consistent with the candidate’s narrative (Druckman, et al., 2007, p. 434). Finally, Foot and Schneider define the practice of mobilizing as ”using the web to persuade and equip campaign supporters to promote the candidate to others, both online and offline” (Foot & Schneider, 2006, p. 131; see also Kann, et al., 2007). By motivating supporters to take concrete action and promote the candidate or party, site visitors can be made to function as campaign organizations on to themselves (Bimber & Davis, 2003, p. 59). Parties can engage in mobilizing practices by providing site visitors with a number of tools, ranging from ”tell-a-friend” on-site functionalities to supplying campaign materials for on- and offline distribution. Advanced forms of mobilizing might involve parties encouraging their supporters to blog, tweet or act online in some other way in order to spread the party message (Foot & Schneider, 2006, p. 143 f). As mentioned earlier, the main problem with mobilizing techniques can be said to be party loss of control over the message (Stromer-Galley, 2000, p. 111). Indeed, campaigns cannot easily force site visitors to use provided online functionalities in the ways they were originally intended to be used (Foot & Schneider, 2006, p. 155). As a result, parties have apparently not yet taken full advantage of the Internet’s potential for political mobilization (Nixon, Ward, & Gibson, 2003; Russmann, 2009, p. 9), although certain research findings have indicated that this might be subject to change (Vaccari, 2008a, p. 649). While the four-part typology provides an analytical framework for the different kinds of features that are often found on political party web sites, it does not account for the reasons behind the utilization of these different features. In the next section, factors often believed to have influence over the use or non-use of these features is discussed.

Influences for party use of web site functionalities Following Druckman et. al. (2007, p. 436), there are a number of influences on the degree to which political actors employ different functionalities on their web sites. In this chapter, four such influences are described – party size, party status, temporal influence and ideological influence. The effect of party size on utilization of web site features has been studied extensively (Gibson, et al., 2008, p. 18). While some studies have found that minor parties tended to use the web more or in more novel ways than major parties (i.e. Gibson, 2004, p. 102; Schweitzer, 2005) the majority of research done corroborates the view of major parties as providing more features on 8

their sites than their diminutive counterparts (i.e. Bimber & Davis, 2003, p. 27; Kalnes, 2009, p. 263; Margolis & Resnick, 2000; Russmann, 2009, p. 9; Strandberg, 2009, p. 851; Vaccari, 2008c, p. 6). These findings seemingly validate the necessity of vast resources for providing content rich web sites (Jackson & Lilleker, 2009, p. 240). The influence of party status has also been acknowledged. Parties or candidates enjoying incumbency are often found to be “less likely to update their sites” than challengers, as incumbents might feel less pressured to gain repeat visitors (Druckman, et al., 2007, p. 429). The positive influence on web campaigning of being a challenger rather than an incumbent seems to be valid in most studies available (Carlson, 2007; Herrnson, Stokes-Brown, & Hindman, 2007; Herrnson, 2007). Differences regarding web campaigning have also been found between parliamentary and non-parliamentary parties. Schweitzer (2008, p. 457) found that the more established parties (i.e. those already in parliament) in the German context provided more professional web sites than the “fringe” parties struggling for seats in the parliament. Conversely, as fringe parties often lack the resources necessary for publicity in offline environments (Margolis, Resnick, & Levy, 2003), they might be expected to employ various forms of potentially low-cost internet initiatives in order to reach out (Kalnes, 2009, p. 255). As the role of a campaign web site can be expected to evolve during the election year (Foot & Schneider, 2006, p. 16), the temporal influence should be taken into account. In his study of the 2006 Italian elections, Vaccari (2008b) sampled the party web sites three times and found that as election day approached, parties tended to improve their sites in various ways. Lilleker and Malagón (2010) studied the 2007 French presidential elections focused on the final iteration of the party web sites, potentially missing out on any developmental processes that these sites might have undergone earlier during the campaign. Following this, it seems reasonable to examine the temporal influence on web site development during the course of a campaign. Also, by analyzing party web sites in their post-election state, the claim that parties are now operating with web sites as ”permanent structures […] to mobilize resources not only during campaigns, but also outside of them” (Vaccari, 2008c, pp. 11-13) can be scrutinized. Several studies have found various forms of ideological influence, giving credence to the claim “party does matter, although not necessarily in terms of size” (Gibson, et al., 2008, p. 26). During the 2007 French presidential election, Vaccari (2008c) found that right-wing parties tended to provide more information on their web sites than their left-wing counterparts. Conversely, parties to the left tended to employ involving and mobilizing features to higher degrees than parties to the right. Studying the same election, Lilleker and Malagón suggested that these differences could stem from the parties different political platforms, with the right offering a ”top-down party 9

centric form of participation” and the left acting more like a grassroots movement (Lilleker & Malagón, 2010, p. 35). Similar trends were also reported from the 2006 Italian elections (Vaccari, 2008b, p. 76). Morover, researchers have found that environmental parties tend to use the web in more novel ways than other political actors (i.e. Russmann, 2009; Strandberg, 2009), perhaps due to the suggested environmental ”cleanliness” of the Internet medium (Gibson, 2004, p. 109). In order to assess these characteristics, table 1 presents data regarding the main parties competing during the 2010 election. In order to asses party status at the beginning of 2010, data is based on the results of the previous election, held in 2006.

10

Party

Votes (%) 2006 elections

Seats in parliament (total of 349)

Ideological position

Social Democrats (S)

35.2

130

Left

Moderate Party (M)

26.1

97

Right

Centre Party (C)

7.9

29

Centre/Right

Liberal Party (Fp)

7.5

28

Centre/Right

Christian Democrats (Kd)

6.6

24

Right

Left Party (V)

5.8

22

Left

Green Party (Mp)

5.2

19

Centre/Left

Sweden Democrats (Sd)

2.9

0

Right

Feminist Initiative (Fi)

0.7

0

Left

Pirate Party (Pp)

0.6

0

Centre

English translation (Swedish abbreviation) Large (20% >)

Medium (4-19.9%)

Small (< 3.9%)

Table 1. Characteristics of Swedish political parties. Election results gathered from The Swedish Election Authority (2006). Table design inspired by Kalnes (2009).

The parties are grouped according to size, showing considerable variation both when it comes to this factor as well as ideological positions. Party names are translated into English, while the common Swedish abbreviations are shown in parenthesis. The Swedish political system is based on proportional representation, where parties must reach at least 4 % of votes in order to take seat in the parlia11

ment. The system can be likened to what Sartori (1990) labels moderate pluralism, with multiple parties and few centrifugal forces. While the Social Democrats dominated Swedish political life for most of the 20th century, Sweden has been ruled by an alliance of right and center/right parties (M, C, Fp, Kd) since 2006. This collaboration proved successful also for the 2010 elections, as the parties were given ruling mandate for yet another period.

Method Given that the structure of a party web site facilitates a certain range of potential political actions (Schneider & Foot, 2004), quantitative content analysis was employed in order to study the election year development of Swedish political party web sites. While the majority of research performed on such sites appear to be feature-based content analytical efforts (Lilleker & Malagón, 2010, p. 28), the study at hand will take this approach one step further by basing its analysis on longitudinal data collected before, during and after the 2010 Swedish general election. By approaching the party web sites in this novel matter, the study will provide much needed data regarding how these web sites develop during an entire election year.

Data collection While similar studies have employed three- or four-week selection frames (Schweitzer, 2008; Vaccari, 2008b) this study expanded the temporal element and used twelve monthly recurring waves of data selection. Each site was downloaded three levels deep – meaning that the first “start” page as well as the main second and third pages were subject to archiving. Downloading was performed by using a web site archiving tool (SiteSucker for Mac OS X). Where “robot exclusion” warnings made archiving by means of the tool unsuitable, the pages were saved as PDF files, utilizing specialized software (WebSnapper for Mac OS X). In order to capture the final iteration of the party web sites on the day before election day of September 19th as well as the “initial or tentative approaches” (Lilleker & Malagón, 2010, p. 31) downloading was performed on the 18th of each month, starting in January 2010 and ending in December of the same year.

Coding The archived web pages were subject to a feature-based analysis taking its starting point from the four-part typology of web site features suggested by Foot and Schneider (2006) as described earlier. Each monthly version of each party site was evaluated using a coding scheme including 82 features 12

that were coded as present or non-present. Specifically, 39 features were utilized for the informing category, 18 for involving, 16 for connecting and 9 for mobilizing. The features included were based on Foot and Schneider’s original 2006 study and supplemented with features identified by other, similar studies (i.e. Alberto & Corinna di, 2009; Bimber & Davis, 2003; Carlson & Strandberg, 2008; Druckman, et al., 2007; Gibson & Ward, 2000; Hara & Jo, 2007; Jackson & Lilleker, 2009; Kalnes, 2009; Kann, et al., 2007; Kluver, Jankowski, Foot, & Schneider, 2007; Matthew & Kelin, 2009; Russmann, 2009; Smith, 2009; Strandberg, 2009; Stromer-Galley, 2000; Vaccari, 2008a, 2008b, 2008c; Wiklund, 2005; Wright & Street, 2007). The web sites were also visited a priori in order to find any features that had not been identified by previous research. The author served as the main coder. In order to gauge reliability, a second coder was trained and analyzed a randomly selected 20 % sample of sites. Following the recommendation of Lombard et. al. (2002), Krippendorf’s α was used to determine the levels of reliability for the variables. Specifically, the KALPHA macro (constructed by Hayes & Krippendorff, 2007) for SPSS was employed to facilitate this. The values of α ranged between .65 and 1.0, with a mean agreement figure of .95. As such, the coding process was considered reliable. Following previous similar research efforts (Schweitzer, 2008; Vaccari, 2008c), each of the 82 features were coded as either present (given a score of 1) or not (given a score of 0). Next, the variables in each category were added together and divided by their theoretical maximum value, thus creating four indices ranging from the minimal value of 0 to the maximum value of 1. A similar assessment was also performed for the grand total of functions included.

Results In order to assess the longitudinal changes of the party web sites, figure 1 contains a multiple line graph providing an overview of the changes for each of the indices per month over the course of the election year. The overarching view in figure 1 is completed with a more detailed description of the results below.

13

Figure 1. Overview of changes in utilization of Informing, Involving, Mobilizing and Connecting features. For each month, the overall mean is reported. A more detailed account of these changes can be found in appendix 1.

Focusing first on the informing category, the results in figure 1 indicate these features are more common on the web sites of Swedish political parties than any of the other types (All year mean = .6, SD = .12). However, the growth of these features throughout the election year is minimal – the index starts out at .56 (SD = .11) in January, reaches its peak during the election month of September (.64, SD = .14) and then gradually decreases to a score of .6 (SD = .12) at the end of the year. Results for the informing index further indicate that the Pirate Party (PP) and the Sweden Democrats (Sd) used comparably few of these features, while the Christian Democrats (Kd) and the Liberal Party (Fp) appeared to employ informing features more. While the overall mean for the involving index is lower (all year mean = .51, SD = .11) than that for the involving category, this index displays a more dynamic development throughout the examined period. Starting out at a mean level of .46 (SD = .08) in January, it reaches its peak during the election month with a score of .59 (SD = .12). Following September, the mean score for involving decreases to .49 (SD = .11) and remains unchanged for the rest of the election year. As for minimum and maximum values, the Centre Party (C) and the Sweden Democrats (Sd) employed involving features 14

comparably less than their competitors, while these features appeared to be common on the web sites of the Pirate Party (PP) and the Social Democrats (S) throughout the election year. The index for mobilizing starts out in January with a mean score of .39. The standard deviation is higher than for both previous indices (reported at .24), indicating plenty of variation regarding how parties use these particular features. Between January and July, the index increases incrementally and reaches its peak values of .58 (SD = .21) and .59 (SD = .2) for August and September respectively. Following election month, the mobilizing index decreases to .51 (SD = .24). For the remainder of the year, the index remains at levels approaching .5 (November: M = .49, SD = .26; December: M = .48, SD = .26). During the months building up to the election, results indicate that the Center Party (C) and the Left Party (V) used mobilizing features least, while the most ardent users of mobilizing functionalities during the same period were the Moderate Party (M), the Green Party (Mp), the Social Democrats (S) and the Liberal Party (Fp). As for the post election period, lower levels of use are reported for the Center Party (C), the Left Party (V), Christian Democrats (Kd), Sweden Democrats (Sd) and Feminist Initiative (Fi), while the Social Democrats (S) holds the top mobilizing index score (.88) during the five month period of August to December. Finally, the connecting index is similar to the informing index in that it is stable throughout the entire election year. In fact, party use of links appear virtually unchanged during all twelve months. Scores for the index vary between .38 and .39 January to June, while it stays at .39 (SD = .15) for the remainder of the year. Furthermore, the results show that the Sweden Democrats used links rather conservatively (minimum value of .25 from January to April, .06 for the rest of the year). Conversely, linking practices were extensively used by the Christian Democrats (Kd) from January to June, as well as by the Social Democrats from July and throughout the year. In sum, the results presented above indicate that while the scales for informing and connecting functionalities are not subject to much change during the election year, the scales for involving and mobilizing appear more dynamic throughout the covered time period. While the index for connecting remains stable during the entire year, index scores for informing, involving and mobilizing features tend to reach their high point during the election month of September, after which they decrease.

Discussion Foot and Schneider suggest that “we should not be surprised at the degree to which campaign organizations engage in informing” (2006, p. 63), and the 15

results presented for the informing category here make no challenge to this claim. Indeed, these features are the most commonly used overall, as suggested by the high month-by-month and all year means, as well as by their comparably low standard deviations. An apparent effect of party status can be discerned here, as the lowest levels throughout the year were reported from two non-parliamentary parties (PP and Sd). Both parties campaigned on rather limited platforms, with one calling for a restructuring of copyright law and a ban of government cyber surveillance (PP) and the other for a more restrictive immigrant policy (Sd). Other types of party information appeared to be largely missing from their respective web sites – perhaps the result of a deliberately employed “strategic ambiguity” (Foot & Schneider, 2006, p. 62). An effect of ideological influence is also visible, as both parties competing for the highest informing index score (Kd and Fp) are positioned to the right on the ideological scale. As such, the claims that right-wing parties are more devoted to using informing features seem valid also in the Swedish context (Lilleker & Malagón, 2010, p. 35; Vaccari, 2008c, p. 15). Like the index for informing features, its involving counterpart incrementally increases up until election day – but more so than for the previous scale. A temporal influence can be observed, as the index reaches its top notation during election month after which it immediately decreases. This gives us an indication that parties tend to use involving features primarily in the preelection campaigning phase. Furthermore, some parties appeared more hesitant to utilize involving features. The Centre Party (C) and the Sweden Democrats are located to the right on the ideological scale, and could therefore be expected to employ comparably fewer of these features (as suggested by Kalnes, 2009, p. 258; Vaccari, 2008b, p. 76). Remembering that ”organizational size and resources are key factors” (Gibson, et al., 2008, p. 18), the reported differences could also be due to the fact that both parties are classified as either medium sized or small. The fact that the Left Party (V), a medium sized political actor but politically left oriented, appears to employ rather few involving features (score of .39 for May and June), gives further weight to the argument that party size and status are important predictors of web site sophistication. Influences of party size and ideology can also be traced among the more frequent users of involving features. Indeed, while the Pirate Party (PP) is one of the smallest parties, the Social Democratic party (S) is one of the biggest. The claim by Druckman et al (2007, p. 429) that one party might be more technologically savvy than others seems valid in this regard. This seems reasonable, given the political focus proposed by the Pirate Party. Of note is also the result that the Pirate Party alone reached the highest scores for the involving index during the post election months, perhaps an indication of a more far-reaching approach to online interaction with the electorate. 16

For mobilizing features, the results show a similar pattern of the scale gradually increasing as election month draws closer. The final mean score of the year is considerably higher than the January score – especially when compared with the other indices discussed here. While this indicates that the utilization of mobilizing features are not as clearly tied to campaign periods as their involving counterparts, the high standard deviations for the mobilizing scale suggests plenty of variation regarding to what degree these features are used by the parties. Nevertheless, a temporal influence can be discerned, as the suggested need for parties to utilize mobilization features regardless of whether an election is at hand or not (Vaccari, 2008c, p. 11) is evidently yet to become a priority for political actors (Russmann, 2009, p. 9). Focusing on the monthby-month distribution of minimum values, the results imply an effect of party size, as both the Centre Party (C) and the Left Party (V) are classified as medium sized political actors. A temporal effect is visible also for the month-by-month minimum scales, as three additional parties (Kd, Sd and Fi) report lower scores during the post election phase. Regarding the maximum values for the mobilizing scale, we see a dominance of larger parties like the Moderate Party (M) and the Social Democrats (S), indicating an effect of party size, while medium sized parties like the Liberal Party (Fp) and the Green Party (Mp) provide plenty of these features on their sites – but only before and during the election campaign. Following election month, the maximum score for mobilizing features is held by the Social Democrats for the remainder of the year. Overall, the results for the mobilizing category indicate that party utilization of these features tend to depend more on temporal influence and party size than ideology – with a few exceptions. As a progressive, left wing party, previous research would suggest that the Left Party (V) would have been expected to provide more mobilizing features on their sites (Vaccari, 2008b, p. 76). At the same time, the conservative Moderate Party (M) appear to be embracing the opportunities to mobilize their supporters, although only during the pre-election phase. The same pattern of use is adopted by the medium sized Green Party (Mp), which could be explained by previous research findings indicating that environmentalist or green parties use the web in new and innovative ways - like the employment of mobilizing features (i.e. Russmann, 2009; Strandberg, 2009). Finally, the mean of the index for connecting features remains virtually unchanged throughout the entire year, varying only slightly between .38 and .39. Judging by this, linking practices appear to be the least used functionality on the sites under scrutiny – perhaps a result of a desire to keep the visitors on the site. As expected, there are variations regarding how political parties make use of links on their web sites. The party that employ linking practices the least is the Sweden Democrats. Due to their extreme right wing 17

views on many issues, they can be described as somewhat isolated in Swedish civil society. The results for the connecting index suggest that this outsider status is apparent also in an online context. Looking at what parties employed linking practices the most, the Christian Democrats (Kd) dominate the first half of the year, while the Social Democrats raise to the same index level (.56) during the second half. Taken together with the results for the mobilizing scale, this indicates a restructuring of the Social Democrat web site for the second half of the year – augmenting their site with new functionalities in time for the election campaign. While the concept of postmodern campaign entails “permanent efforts” by politicians in order to maintain long term engagements with the electorate (Vaccari, 2008a, p. 648; Zittel, 2009), the results presented in this paper shows that while some political actors try to maintain such relationships with citizens, most parties choose to concentrate their efforts to engage the voting public to the more intensive campaign period building up to election day. For the most part, the “stop-start” nature of political activity, arranged around election cycles, has made its mark also in the online realm (Gibson, 2004, p. 102). However, as differences relating to party size, parliamentary status and ideological persuasion can be discerned, the aforementioned dichotomy of the internet providing either a shift or enhancement in political campaigning appears unsuitable in order to understand the current situation. Similarly, the notions of normalization and innovation hypothesis both seem somewhat inappropriate. The use of the Internet for political campaign purposes can not be understood through dichotomies – rather, we might be better off by identifying these issues as placed on a scale ranging from more traditional approaches of online campaigning to more progressive ones. Recent studies from Great Britain (Jackson & Lilleker, 2009) as well as from Norway (Kalnes, 2009) have reached similar conclusions, proposing a view of “web 1.5” as a middle way between the two aforementioned extremes in order to understand how these developments play out. As noted above, there are exceptions – some parties tend to be more progressive in their employment of the features discussed in this paper - but this view seems largely feasible also when assessed with the longitudinal data presented above in mind.

References Pepe Alberto, & Gennaro Corinna di, 2009. "Political protest Italian-style: The blogosphere and mainstream media in the promotion and coverage of Beppe Grillo's V-day". First Monday, volume 14, number 12 (December), at 18

http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2740, accessed 3 March 2011. N. Anderson, "Tim Berners-Lee on Web 2.0: Nobody even knows what it means." at Ars Technica - http://arstechnica.com/news.ars/post/200609017650.html, accessed 1 September 2010. Annika Bergström, 2007. “Internetval 2006 - För de redan frälsta,” In: S. Holmberg & L. Weibull (editors). Det nya Sverige : trettiosju kapitel om politik, medier och samhälle : SOM-undersökningen 2006. Gothenburg: SOM-institute, Gothenburg university, pp. 395-404. Bruce A. Bimber, & Richard Davis, 2003. Campaigning online : the Internet in U.S. elections. New York, N.Y.: Oxford University Press. James Bohman, 1996. Public deliberation : pluralism, complexity, and democracy. Cambridge, Mass.: MIT Press. Tom Carlson, 2007. "It's a Man's World? -- Male and Female Election Campaigning on the Internet". Journal of Political Marketing, volume 6, number 1, pp. 41-67. Tom Carlson, & Kim Strandberg, 2008. "Riding the Web 2.0 Wave: Candidates on YouTube in the 2007 Finnish National Elections". Journal of Information Technology & Politics, volume 5, number 2, pp. 159-174. M Castells, 2007. "Communication, power and counter-power in the network society". International Journal of Communication, volume 1, number 1, pp. 238-266. Manuel Castells, 1996. The information age : economy, society and culture. Vol. 1, The rise of the network society / Manuel Castells. Malden, Mass.: Blackwell. Manuel Castells, 2001. The internet galaxy : reflections on the Internet, business, and society. Oxford: Oxford University Press. Andrew Chadwick, 2006. Internet politics : states, citizens, and new communication technologies. New York: Oxford University Press. S Coleman, 2005. Direct representation: Towards a conversational democracy. Institute for Public Policy Research (IPPR): London. S. Coleman, 2005. "New mediation and direct representation: reconceptualizing representation in the digital age". New Media & Society, volume 7, number 2, pp. 177-198. Stephen Coleman, 2004. "Connecting Parliament to the Public via the Internet". Information, Communication & Society, volume 7, number 1, pp. 122. Graham Cormode, & Balachander Krishnamurthy, 2008. "Key differences between Web 1.0 and Web 2.0". First Monday, volume 13, number 6 (June), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2125/ 1972, accessed 3 March 2011. Aeron Davis, 2010. "New media and fat democracy: the paradox of online participation". New Media & Society, volume 12, number 5, pp. 745-761. 19

Anthony Downs, 1957. An economic theory of democracy. New York: Harper & Row. James N. Druckman, Martin J. Kifer, & Michael Parkin, 2007. "The Technological Development of Congressional Candidate Web Sites". Social Science Computer Review, volume 25, number 4, pp. 425-442. Ulrika Facht, 2008. Nordicom-Sveriges Internetbarometer 2007, at http://www.nordicom.gu.se/?portal=mr&main=info_publ2.php&ex=266, accessed 1 August 2010. Kirsten A. Foot, & Steven M. Schneider, 2006. Web campaigning. Cambridge, Mass.: MIT Press. Rachel Gibson, 2004. "Web campaigning from a global perspective". Asia-Pacific Review, volume 11, number 1, pp. 95-126. Rachel K. Gibson, Wainer Lusoli, & Stephen Ward, 2008. "Nationalizing and Normalizing the Local? A Comparative Analysis of Online Candidate Campaigning in Australia and Britain". Journal of Information Technology & Politics, volume 4, number 4, pp. 15-30. Rachel Gibson, & Stephen Ward, 2000. "A Proposed Methodology for Studying the Function and Effectiveness of Party and Candidate Web Sites". Social Science Computer Review, volume 18, number 3, pp. 301-319. Lev Grossman, "Time’s person of the year: You", at http://www.time.com/time/magazine/article/0,9171,1569514,00.html, accessed 1 September 2010. Jürgen Habermas, 1989. The structural transformation of the public sphere : an inquiry into a category of bourgeois society. Cambridge: Polity Press. Noriko Hara, & Youngmin Jo, 2007. "Internet politics: A comparative analysis of U.S. and South Korea presidential campaigns". First Monday, volume 12, number 9 (September), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2621/ 2280, accessed 3 March 2011. AF Hayes, & K Krippendorff, 2007. "Answering the call for a standard reliability measure for coding data". Communication Methods and Measures, volume 1, number 1, pp. 77-89. P. S. Herrnson, A. K. Stokes-Brown, & M. Hindman, 2007. "Campaign politics and the digital divide - Constituency characteristics, strategic considerations, and candidate Internet use in state legislative elections". Political Research Quarterly, volume 60, number 1, pp. 31-42. PS Herrnson, 2007. Congressional elections: Campaigning at home and in Washington: Cq Press. Nigel A. Jackson, & Darren G. Lilleker, 2009. "Building an Architecture of Participation? Political Parties and Web 2.0 in Britain". Journal of Information Technology & Politics, volume 6, number 3, pp. 232-250. Øyvind Kalnes, 2009. "Norwegian Parties and Web 2.0". Journal of Information Technology & Politics, volume 6, number 3, pp. 251-266. 20

Mark E. Kann, Jeff Berry, Connor Grant, & Phil Zager, 2007. "The Internet and youth political participation". First Monday, volume 12, number 8 (August), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1977/ 1852, accessed 3 March 2011. Randolph Kluver, Nicholas Jankowski, Kirsten Foot, & Steven M. Schneider, 2007. The Internet and national elections : a comparative study of Web campaigning. London: Routledge. Darren G. Lilleker, & Casilda Malagón, 2010. "Levels of Interactivity in the 2007 French Presidential Candidates' Websites". European Journal of Commun1ication, volume 25, number 1, pp. 25-42. M. Lombard, J. Snyder-Duch, & C. C. Bracken, 2002. "Content analysis in mass communication: Assessment and reporting of intercoder reliability". Human Communication Research, volume 28, number 4, pp. 587-604. M. Margolis, D. Resnick, & J. Levy, 2003. “Major Parties Dominate, Minor Parties Struggle. US Elections and the Internet,” In R. Gibson, P. Nixon & S. Ward (editors), Political Parties and the Internet: Net Gain?. London, New York: Routledge, pp. 53–69. M. Margolis, D. Resnick, & J. D. Wolfe, 1999. "Party competition on the Internet in the United States and Britain". Harvard International Journal of Press-Politics, volume 4, number 4, pp. 24-47. Michael Margolis, & David Resnick, 2000. Politics as usual : the cyberspace "revolution". London: SAGE. J. Kushin Matthew, & Kitchener Kelin, 2009. "Getting political on social network sites: Exploring online political discourse on Facebook". First Monday, volume 14, number 11 (November), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2645/ 2350, accessed 3 March 2011. P. Nixon, S. Ward, & R. Gibson, 2003. “Conclusions: The Net Change,” In: R. Gibson, P. Nixon & S. Ward (editors). Political Parties and the Internet: Net Gain?. London, New York: Routledge, pp. 234–243. Tim O’Reilly, "What is Web 2.0? Design Patterns and Business Models for the Next Generation of Software", at www.oreillynet.com/lpt/a/6228, accessed 2 September 2010. C. Pascu, D. Osimo, M. Ulbrich, G. Turlea, & J. C. Burgelman, 2007. "The potential disruptive impact of Internet 2 based technologies". First Monday, volume 12, number 3 (March), at http://firstmonday.org/article/view/1630/1545, accessed 3 March 2011. Lee Rainie, Michael Cornfield, & John Horrigan, "The Internet and Campaign 2004", at http://www.pewinternet.org/~/media/Files/Reports/2005/PIP_2004_Campaig n.pdf.pdf, accessed 2 September 2010. Wayne Rash, 1997. Politics on the nets : wiring the political process. New York: Freeman. 21

W. Vaagan Robert, 2009. "Presidential Web sites and the GeorgianRussian War, 8-16 August 2008". First Monday, volume 14, number 9 (September), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2621/ 2280, accessed 3 March 2011. R. Rogers, & N. Marres, 2000. "Landscaping climate change: a mapping technique for understanding science and technology debates on the World Wide Web". Public Understanding of Science, volume 9, number 2, pp. 141163. Elizabeth Anne Roodhouse, 2009. "The voice from the base(ment): Stridency, referential structure, and partisan conformity in the political blogosphere". First Monday, volume 14, number 9 (September), at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2624, accessed 3 March 2011. Uta Russmann, 2009. Web Campaigning in Austria: A Structural Analysis of Austrian Party Web Sites in the 2008 National Elections. Paper presented at the Internet Research 10.0 - Internet: Critical, 7-10 October 2009, Milwaukee, WI. G Sartori, 1990. A typology of party systems. In P. Mair (Ed.), The West European party system (pp. 316–350). Oxford: Oxford University Press. S. M. Schneider, & K. A. Foot, 2004. "The web as an object of study". New Media & Society, volume 6, number 1, pp. 114-122. E. J. Schweitzer, 2005. "Election campaigning online - German party websites in the 2002 National Elections". European Journal of Communication, volume 20, number 3, pp. 327-351. E. J. Schweitzer, 2008. "Innovation or Normalization in E-Campaigning? A Longitudinal Content and Structural Analysis of German Party Websites in the 2002 and 2005 National Elections". European Journal of Communication, volume 23, number 4, pp. 449-470. Aaron Smith, 2009. The Internet's role in campaign 2008. Pew Internet & American Life Project: Pew Research Center. Kim Strandberg, 2009. "Online campaigning: an opening for the outsiders? An analysis of Finnish parliamentary candidates' websites in the 2003 election campaign". New Media & Society, volume 11, number 5, pp. 835854. J. Stromer-Galley, 2000. "On-line interaction and why candidates avoid it". Journal of Communication, volume 50, number 4, pp. 111-132. Swedish Election Authority, 2006. “Election Results from the 2006 Swedish Election”, at http://www.val.se/tidigare_val/val2006/valresultat/index.html, accessed 29 January 2011 Cristian Vaccari, 2008a. "From the air to the ground: the internet in the 2004 US presidential campaign". New Media & Society, volume 10, number 4, pp. 647-665. 22

Cristian Vaccari, 2008b. "Research note: Italian parties' websites in the 2006 elections". European Journal of Communication, volume 23, number 1, pp. 69-77. Cristian Vaccari, 2008c. "Surfing to the Elysee: The Internet in the 2007 French Elections". French Politics, volume 6, number 1, pp. 1-22. S. Ward, & T. Vedel, 2006. "Introduction: The potential of the Internet revisited". Parliamentary Affairs, volume 59, number 2, pp. 210-225. Hans Wiklund, 2005. "A Habermasian analysis of the deliberative democratic potential of ICT-enabled services in Swedish municipalities". New Media & Society, volume 7, number 2, pp. 247-270. S. Wright, & J. Street, 2007. "Democracy, deliberation and design: The case of online discussion forums". New Media & Society, volume 9, number 5, pp. 849-869. Thomas Zittel, 2009. "Lost in Technology? Political Parties and the Online Campaigns of Constituency Candidates in Germany's Mixed Member Electoral System". Journal of Information Technology & Politics, volume 6, number 3, pp. 298 - 311.

23

Appendix 1 – study III Informing M Sd Mi n .56 .11 .41 (PP ) .57 .11 .41 (PP ) .58 .12 .41 (PP ) .6 .13 .41 (PP )

Involving Max M Sd Min

May

.59

.14

Jun

.6

Jul

Jan

Mobilizing M Sd Min

.33 (C)

.61 (PP)

.39

.24

.48 .09

.33 (C)

.41

.22

.49 .1

.33 (C)

.61 (PP, S) .67 (S)

.41

.22

.13 (V)

.52 .1

.39 (C,S d)

.67 (PP, S)

.46

.22

.13 (V)

.33 (Sd )

.77 .52 .1 (Fp)

.39 (Sd, V)

.67 (PP, S)

.5

.25

.13 (V)

.14

.36 (Sd )

.77 .53 .11 (Fp)

.39 (Sd, V)

.72 (S)

.51

.24

.13 (V)

.62

.14

.36 (Sd )

.79 (Kd )

.56 .12

.39 (Sd)

.78 (S)

.51

.24

.13 (V)

Aug

.63

.15

.78 (S)

.58

.21

.25 (V)

.64

.14

.39 (Sd)

.78 (S)

.59

.2

.25 (V)

Oct

.62

.12

.82 .58 .13 (Kd ) .82 .59 .12 (Kd ) .79 .49 .11 (Fp)

.39 (Sd)

Sep

.36 (Sd ) .38 (Sd ) .38 (Sd ) .38 (Sd )

.75 (M,M p) .75 (M,M p) .75 (M,M p) .75 (M,M p, S) .75 (M,M p, S,Fp) .75 (M,M p, S,Fp) .75 (M,M p, S,Fp) .88 (S)

.33 (C)

.67 (PP)

.51

.24

.79 .49 .11 (Fp)

.33 (C)

.67 (PP)

.49

.26

.25 (V,K d,Sd) .25 (V,K d,Sd, C) .25 (V,K d,Sd, C,Fi) .13

Mar Apr

Nov .6

.12

Dec

.6

.12

.38 (Sd )

.79 .49 .11 (Fp)

.33 (C)

.67 (PP)

.48

.26

All year

.6

.12

.33

.82

.33

.78

.49

.23

.51 .11

.13 (C,V ) .13 (C)

Max

.46 .08

Feb

.74 (Kd ) .74 (Kd ) .74 (Kd ) .77 (Fp)

Max

Connecting M Sd Mi n .38 .09 .25 (Sd ,Fi) .39 .09 .25 (Sd ) .39 .09 .25 (Sd ) .39 .09 .25 (Sd )

Max .56 (Kd) .56 (Kd) .56 (Kd) .56 (Kd)

.39

.14

.06 (Sd )

.56 (Kd)

.38

.14

.06 (Sd )

.56 (Kd)

.39

.15

.06 (Sd )

.56 (S,Kd )

.39

.15

.88 (S)

.39

.15

.88 (S)

.39

.15

.88 (S)

.39

.15

.06 (Sd ) .06 (Sd ) .06 (Sd ) .06 (Sd )

.56 (S,Kd ) .56 (S,Kd ) .56 (S,Kd ) .56 (S,Kd )

.88 (S)

.39

.15

.06 (Sd )

.56 (S,Kd )

.88

.39

.12

.06

.56

Table 2. Development of features on party web sites throughout the election year. A score of 0 indicates that none of the features in the specified category were employed. A score of 1 indicates all the features in the specified category were employed. Party abbreviation key: C (Centre Party), Fi (Feminist Initiative), Fp (Liberal Party), Kd (Christian Democrats), M (Moderate Party), Mp (Green Party), PP (Pirate Party), S (Social Democrats), Sd (Sweden Democrats), V (Left Party).

Paper IV

New http://nms.sagepub.com/ Media & Society Studying political microblogging: Twitter users in the 2010 Swedish election campaign Anders Olof Larsson and Hallvard Moe New Media Society published online 21 November 2011 DOI: 10.1177/1461444811422894

The online version of this article can be found at: http://nms.sagepub.com/content/early/2011/11/21/1461444811422894 Published by: http://www.sagepublications.com

Additional services and information for New Media & Society can be found at: Email Alerts: http://nms.sagepub.com/cgi/alerts Subscriptions: http://nms.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> OnlineFirst Version of Record - Nov 21, 2011 What is This?

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Article

Studying political microblogging: Twitter users in the 2010 Swedish election campaign

new media & society 0(0) 1–19 © The Author(s) 2011 Reprints and permission: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1461444811422894 nms.sagepub.com

Anders Olof Larsson Uppsala University, Sweden

Hallvard Moe University of Bergen, Norway

Abstract Among the many so-called microblogging services that allow their users to describe their current status in short posts, Twitter is probably among the most popular and well known. Since its launch in 2006, Twitter use has evolved and is increasingly used in a variety of contexts. This article utilizes emerging online tools and presents a rationale for data collection and analysis of Twitter users. The suggested approach is exemplified with a case study: Twitter use during the 2010 Swedish election. Although many of the initial hopes for e-democracy appear to have gone largely unfulfilled, the successful employment of the internet during the 2008 US presidential campaign has again raised voices claiming that the internet, and particularly social media applications like Twitter, provides interesting opportunities for online campaigning and deliberation. Besides providing an overarching analysis of how Twitter use was fashioned during the 2010 Swedish election campaign, this study identifies different user types based on how highend users utilized the Twitter service. By suggesting a novel approach to the study of microblogging and by identifying user types, this study contributes to the burgeoning field of microblog research and gives specific insights into the practice of civic microblogging. Keywords microblogging, political campaigning, social media, Sweden, Twitter Corresponding author: Anders Olof Larsson, Department of Informatics and Media, Uppsala University, Box 513, 751 20 Uppsala, Sweden Email: [email protected]

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

2

new media & society 0(0)

Introduction More often than not, the introduction of a new communication technology triggers hopes of its democratic impact. As with the internet itself, blogging and microblogging alike have been heralded for their potential for increasing political participation among previously unengaged citizens (e.g. Castells, 2007: 255). Although many initial hopes for e-democracy (Chadwick, 2008; Hilbert, 2009) have gone mostly unfulfilled, the successful employment of the internet during the 2008 Obama US presidential campaign yet again raised voices claiming that so-called social media applications, microblogging services included, provide new opportunities for online campaigning and electorate engagement (Smith, 2009). As such, there is a pertinent need for empirical studies to examine if and how such services contribute to a broadening of participation in public debate, and to what extent it merely serves as yet another arena for already established societal actors. Indeed, previous research has suggested that studies look into microblog use that goes beyond the characterization of ‘interesting novelty’ (Honeycutt and Herring, 2009: 10), and Twitter, given its popularity and status, appears to be an ideal candidate for such studies (Jansen et al., 2009: 2173). Such an aim – to study participation in political debate on Twitter – poses methodological challenges. How can we grasp and make sense of the sudden outbursts of seemingly abundant messages in a seemingly ever-more fragmented debate – constituting something akin to ‘ephemeral communicative spaces’ (Christensen and Christensen, 2008)? This article proposes an approach to studying microblogging as an arena for public political communication. The approach utilizes the potential of online media and netbased computer software to allow for comprehensive collection of data and metadata of large amounts of the actual public debate. As such, the approach facilitates longitudinal, quantitative analyses utilizing statistical methods as well as social network analysis. To illustrate the approach, the article presents findings from a study on Twitter use during the 2010 Swedish general election campaign. As an established democracy with high levels of freedom of speech, high numbers of internet use and ICT penetration (Carlsson and Facht, 2010) as well as high election turnout, the Swedish political context represents a noteworthy case. It seems reasonable to assume that the Swedish election campaign would provide favorable conditions for the employment of a novel internet tool such as Twitter. Data was collected using the YourTwapperKeeper application. In total, 99,832 tweets dealing with the election are analyzed, focusing on the various high-end users in the Swedish political Twittersphere, and the networks that appear between these users. By utilizing social network analyses of a large dataset collected with the aid of emerging online applications, the study contributes to the development of the methodological toolbox for research. In so doing, the article presents findings on Twitter use before, during and after the height of the Swedish election campaign, thereby providing novel insights into practices of civic microblogging.

Background Twitter is often understood as a derivative or miniature version of the regular blog – i.e. a microblog, consisting of ‘short comments usually delivered to a network of associates’

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

3

(Jansen et al., 2009: 2170). By sending short messages – tweets – of up to 140 characters each, Twitter users share these updates to a network of followers. Compared to similar services, the act of following another Twitter user is not automatically reciprocal. A user can follow any number of other users, although the user being followed does not necessarily have to follow back. As with communication apparatuses throughout history (e.g. Winston, 1998), Twitter has been utilized beyond the intended uses. For example, users accommodate alternative forms of use by annotating their tweets with different characters in order to signify a specific form of communication. To allow for conversations, the @ sign is used as a marker of addressivity. For example, posting a message including @USERNAME indicates that the message is intended for or somehow relevant to a specific user. Retweets (RT) refer to the practice of resending a tweet posted by another user. Following the basic typology suggested by Kwak et al. (2010), a tweet can be classified as a Singleton (a statement from a specific user, no @ sign present); a Mention or a Reply (@ sign followed by a user ID) or a Retweet, as mentioned earlier (marked with RT). Tweets can also include hashtags, where the # character is used in conjunction with a word or phrase in order to connect the tweet to a particular theme. This use of the # sign allows users to search the Twittersphere for specific topics of interest and to follow threads of discussion related to those topics. Based on the suggested similarities between blogging and microblogging, the following section outlines the growth of blogging in general, and political blogging in particular. Following this, we present a review of the research that has been performed looking into the diverse uses of Twitter.

Political blogging Blogging has received an ample amount of attention, both in various media outlets (Jones and Himelboim, 2010) as well as among academics (Larsson and Hrastinski, 2011). Remembering that ‘technology is often viewed as a key driver of change in the electoral arena’ (Gibson et al., 2008: 15), researchers have scrutinized the connections between new media and politics throughout history, focusing on print publications in EighteenthCentury America (Warner, 1990) or internet search engines (Halavais, 2009), to mention two temporal extremes, as well as media systems as a whole (e.g. Baker, 2002; Page, 1996). Thus, the extent to which researchers have focused on the political aspects of blogging should come as no surprise (Kerbel and Bloom, 2005; Sweetser Trammell, 2007). Use of blogs for political purposes has been studied from the point of view of the citizen (looking to discuss political matters with peers), as well as from the point of view of the elected official (looking to connect with the electorate). First, politically interested citizens who enter the blogosphere have been labeled ‘technoactivists’ in search of outlets for ‘democratic self-expression and networking’ (Kahn and Kellner, 2004). Blog use has been found to be an important predictor of online political engagement (Gil De Zuniga et al., 2009), and reasons for blogging appear to go beyond more intrinsic motivations. In their survey of top US political bloggers, Ekdale et al. (2010) found that extrinsic motivations for blogging (i.e. providing alternatives to mainstream media outlets or to influence public opinion) were among the top reasons for

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

4

new media & society 0(0)

maintaining a political blog. However, this does not mean that blogs automatically remove every obstacle to political participation. In his study on the potential for blogs to foster democratic discourse in the US context, for instance, Davis concluded that blogs and similar online applications could be problematic as public discussion forums since they were marred by problems like ‘exclusion of others, flaming [and] a great deal of anonymity’ (Davis, 2005: 119). Second, political parties and their candidates have also entered the blogosphere to ‘engage with supporters and the wider public’ (Gibson et al., 2008: 19). Although there are indications that such activities have indeed fostered new channels for politicians to connect with an increasingly jaded electorate, many political actors appear to struggle with their presence on the internet (Vaccari, 2008b: 75), experiencing difficulty in adapting to the new medium. Perhaps due to the ‘tradeoff between information control and interactive engagement’ (Druckman et al., 2007: 428), studies have found that political actors are careful when venturing online, limiting the options for voter co-creation and interaction and making more use of traditional, informing features (Larsson, 2011; Vaccari, 2008a, 2008c). As such, the majority of online action by political actors has been likened to an ‘electronic brochure’ (Jackson and Lilleker, 2009), indicating that little adaptation to the online format have taken place. Although studies have found politicians to apparently mainly use blogs as ‘campaign gimmicks’ (Lilleker and Malagón, 2010: 26), hopes are still held high regarding the use of online services for political purposes (Vaccari, 2008a). Recently, such hopes have been transferred to microblogging services like Twitter.

Twitter While a number of more or less distinctive uses of Twitter have been reported, such as the case of an American student jailed in Egypt who used the service to signal distress, or the messages sent regarding the US Airways plane that crashed into the Hudson river (Arceneaux and Weiss, 2010; Kwak et al., 2010), academic research on Twitter use is at a very early stage. Studies have largely focused on describing various everyday uses of the service. For example, Marwick and boyd (2010) analyzed the various audience management techniques employed by Twitter users, while Java et al. (2007) identified four general categories of Twitter use: daily chatter, i.e. posts regarding daily events and thoughts; conversations using the @ character; sharing information where URLs are distributed via the posts; and reporting news, where ‘users report latest news or comment about current events’ (2007). Similarly, Honeycutt and Herring (2009) employed a grounded theory approach on their sample of tweets and found 12 distinct categories of tweets: about addressee, announce/advertise, exhort, information for others, information for self, metacommentary, media use, opinion, other’s experience, self experience, solicit information and other. Although more finely grained, these 12 categories roughly correspond to the previous four, indicating reliability in the results of previous research. Besides studying the uses of Twitter in a variety of everyday contexts, researchers have identified a variety of professional Twitter uses (Grace et al., 2010). Given the focus of the study presented here, the next section discusses the different varieties of political Twitter use that have been identified by researchers.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

5

Political Twitter use Researchers have studied political Twitter use focusing on either parliamentary or nonparliamentary uses of the service. The majority of existing studies of parliamentary uses of Twitter deal with US conditions. For example, Golbeck et al. (2010) analyzed the contents of over 6000 tweets from Members of the US Congress. The analysis showed that the members tweeted primarily to disseminate information, often providing URLs to news articles about themselves or to their blog posts. These modes of usage seemingly correspond with the Sharing information and Reporting news categories reported by Java et al. (2007). Congress people also reported on their daily activities, although these updates did not provide insights into the political process, nor did they improve transparency. Golbeck et al. label these tweets ‘vehicles for self-promotion’ (2010: 1620), indicating a mode of use more akin to one-way, top-down communication than of actually engaging with the citizenry. While microblogging in general has evolved towards becoming ‘more conversational and collaborative’ (Honeycutt and Herring, 2009: 10), Golbeck et al. (2010) found such use to be limited among the politicians. Similarly, while the practices of retweeting and hashtagging appear to be widespread among general Twitter users, the researchers found only 5 retweets and 344 tweets with hashtags in their sample. These conservative patterns of use could be a result of forced or semi-adoption of the Twitter platform, an unwillingness to participate in the practice of twittering or perhaps simply a lack of knowledge regarding the different possibilities for use that are available (Golbeck et al., 2010). While these studies provide valuable insights into a specific part of the Twittersphere, they are by default limited to give a biased, partial impression. A focus on non-parliamentary uses adds to this impression. As for non-parliamentary uses, the notion of Twitter revolutions in totalitarian countries has been introduced, although the exact contents and effects of these uprisings are heavily disputed. For example, Gaffney studied Twitter use during the 2009 Iran elections by tracking the use of the #IranElection hashtag. Although Twitter helped protesters in Iran and around the world in organizing their efforts, the author claims that ‘it is difficult to say with any certainty what the role of Twitter was’ (Gaffney, 2010). Evgeny Morozov (2009, 2011) is not as coy in his criticisms of what he claims is the hyperbole surrounding Twitter use during the Iran election. Following Morozov, the Iranian Twitter revolution is ‘a myth, dreamed up and advanced by cyber-utopian Western commentators’ (2009: 11). It’s all media hype, the product of a global Twitterati with little or no insight into the actual protests and processes that went down in Teheran. While studies of Twitter use in political hotbeds like Iran yield important information on political microblogging, to further our understanding, more politically stable contexts should be placed under scrutiny as well. Tumasjan et al. (2010) studied tweets related to the 2009 German federal election. While their focus was on Twitter content as a reflection of the offline political landscape, they also note that ‘while Twitter is used as a forum for political deliberation, this forum is dominated by a small number of heavy users’ (Tumasjan et al., 2010: 4, see also Bruns, 2010). Still, their approach, which relies on content analysis, does not allow for more detailed discussion of these users’ behavior. The methodological setup presented here aims to catch a broader spectrum of Twitter use during an intense period of political campaigning, guided by the pertinent

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

6

new media & society 0(0)

question of to what extent and how a service such as Twitter impact on participation in public debate. For this purpose, we employ a broad approach, encompassing both nonparliamentary and parliamentary Twitter users. By utilizing novel methodological tools to study Twitter users in a stable democratic context, the approach should provide the research community with new insights regarding online political activity. Next, the method section explains the rationale for data collection and analysis.

Method As stated earlier, microblogging can be described as a diminutive version of blogging. As the latter of these internet forms has received some attention in scholarly work, our suggested approach draws on a line of studies on blogs and blogging, often referred to as hyperlink network analysis. Such studies are based on social network analysis, defining a node as a website representing a social actor, and the relationship between two nodes as expressed by hyperlinks. Through comprehensive mappings of links, the aim of such studies is to describe networks through measurements of density, centralization, the relation positioning of nodes, and their specific interconnections. We suggest transferring such an approach to Twitter. Such a move entails some basic advantages compared to studies of the blogosphere. While an attempt to map connections between blogs and bloggers depends on crawls across a wide range of domains and technical setups, leading to substantial challenges for data collection (see Moe, 2011 for discussion), Twitter activity takes place within one domain, with a common Application Program Interface (API). Moreover, while the actual meanings of a mention or retweet still need interpretation, such forms of communication are, arguably, more reliable as data for analysis than hyperlinks on blogs automatically collected by crawler software (e.g. Bruns et al., 2010). Our employment of social network analysis is not without precedents. However, when such an approach has been used thus far, as for instance by Kwak et al. (2010), efforts tend to be directed at the entire Twittersphere, providing somewhat abstract and general descriptions. This study provides an analysis of a specific subset of that same online sphere, focusing on one set of use, namely political communication. Delimiting the analysis in this way enables us to provide a more detailed account regarding Twitter user types in a specific case. The aim is to move beyond descriptive statistics and study interaction between users, shedding light not only on the volume and forms of use, but also on who these users are and how they relate (or not) to each other. For these purposes, data collection can be performed by means of a scrape or crawl. A range of technical solutions can be utilized for this purpose. In our instance, we opted for the YourTwapperKeeper application (TwapperKeeper, 2010). YourTwapperKeeper is a free, publicly available online tool that is installed on a user’s server. It allows for downloading and archiving tweets according to a variety of criteria. Specifically, YourTwapperKeeper produces downloadable CSV-files, consisting of extensive lists featuring various information regarding the archived tweets: the message text, user name and id of the sender, user id of the recipient (if message is a reply), language code, client used to send tweet, geographical code, and the time the tweet was created. Analyses can then be performed using statistical software (SPSS) and the open source graph

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

7

Figure 1. Longitudinal distribution of tweets.

visualization manipulation software Gephi (available at http://gephi.org/). In what follows, we illustrate the approach with a case – the 2010 Swedish election campaign.

The case of the 2010 Swedish election campaign As noted by Golbeck et al. (2010: 1618), the congressional calendar has an obvious impact on the activities of elected officials – and, should they be Twitter users, on the contents of their tweets. With the 2010 Swedish Election Day set to Sunday, September 19th, data collection was employed in order to capture the Twitter activity concerning the election one month beforehand. Archiving via YourTwapperKeeper began on August 17th. In order to catch some of the post-election Twitter activity, data collection was aborted on September 22nd, four days after the election. In the weeks of political campaigning leading up to August 17th, the hashtag #val2010 (Swedish for #election2010) emerged as the most commonly used hashtag indicating content relevant to the upcoming election. Following Gaffney (2010), the tagging system employed by Twitter users allows the researcher to quickly identify transmissions of interest. As such, the delimitation for this study to focus on the #val2010 tag seemed to be a feasible approach to data collection. In order to provide an overview of the total sample used in the study at hand, Figure 1 provides a timeline illustrating the distribution of tweets throughout the examined period of August 17th to September 22nd.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

8

new media & society 0(0)

Table 1. Distribution of tweets by type. Type of tweet

N

%

Singleton @ RT Total

60,088 6964 32,780 99,832

60.2 7 32.8 100

The timeline is characterized by a number of spikes, indicating an increase in the frequency of tweets sent at those particular times. These spikes grow visibly larger as election day (September 19th) draws closer, with the largest increase of tweets appearing on Election Day itself. Perhaps not very surprisingly, the spikes visible in Figure 1 can largely be explained by offline events influencing Twitter activity. Besides election day, which sees the largest spike featuring about 50 per cent of the total number of archived tweets, significant spikes occur in conjunction with televised political debates, statements made by key politicians or offline political meetings or rallies. Further overview of the activity under scrutiny is provided in Table 1, which shows the distribution of the total sample by type of messages (either singleton, @ or RT as discussed earlier). Singletons are most common (N = 60,088, 60.2 per cent), followed by RT (N = 32,780, 32.8 per cent) and directed messages (@; N = 6964, 7 per cent). As such, the majority of tweets sent were non-directed messages. While Figure 1 and Table 1 provides us with an overview of the data, they do not reveal more specified information regarding Twitter use during the examined time period. With the distinction between singletons, mentions/replies and retweets in mind, the data were analyzed in order to find users who distinguished themselves according to this categorization of tweets. An examination of how Twitter contributes to a broadening of public debate, and to what extent it merely serves as yet another arena for already established societal actors requires a focus on high-end users. While this inevitably leaves out the denizens of the #val2010 Twittersphere, it allows for careful scrutiny of who the most active users are, and how they approach the capabilities of microblogging services like Twitter. By focusing our analysis on these high-end users, we provide a framework for interpretation of the different user categories evident from the collected data. Table 2 shows how the highest ranking user stands out with more than three times as many singletons (N = 1932) sent than number two on the list (N = 618). Furthermore, the table provides identifying information about the users. The Twitter pages of all identified users were visited in order to determine occupational status of the user, and to see whether or not the user appeared to be using the service under an assumed name or not. Politicians from different camps and levels are represented: Annika Beijbom (Twitter ID AnnikaBeijbom), the third most active user, is a parliamentary candidate for the Liberal party (Folkpartiet). The remaining politicians on the list are both Conservatives (Moderaterna), all below the top level of their party. A second group identified in table 2 consists of political bloggers and journalists – users who to some extent could be

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

9

Table 2. Ten most active singleton tweeters. Twitter ID

N

all_insane blogfia AnnikaBeijbom Nemokrati Pihlblad juditburda vpressfeldt MuzafferUnsal mickep2 skogskant

1932 618 616 550 544 392 352 345 312 312

Political affiliation

Fp (Liberals)

M (Conservatives) M (Conservatives)

Comments Anonymous, political-satirical content Political blogger National Parliamentary candidate Anonymous, author under assumed name Journalist Local Parliamentary candidate University student Part-time politician Journalist Anonymous

considered to be established voices in the Swedish public debate. The remaining users identified in Table 2 are difficult to describe in any greater detail since they hide their identities and do not state any affiliation, political preference, or other information disclosure regarding themselves and their activities. Three users are anonymous, including the top-ranking ‘all_insane.’ This Twitter ID is a pun with a political edge aimed at the conservative coalition (collectively labeled Alliansen [the Alliance]). The two other anonymous users are Skogskant and Nemokrati, the latter of which claims to be an author using Twitter under an assumed name. While descriptive statistics like the ones presented above provide a starting point for analysis, they do not reveal much about the relationships between the different high-end users identified in the data. As mentioned earlier, Twitter can be used to communicate by means of @ messages, as well as to redistribute messages sent by others through the practice of retweeting. In order to analyze these uses we employ social network analysis. First, in order to assess the conversational potential of Twitter, the reciprocity of mention/reply networks (i.e. tweets directed towards specific users by means of the @ character) are taken into account. Utilizing the Gephi software package (Bastian et al., 2009), Figure 2 provides a network analysis of the top @ networks. The figure features a number of nodes, each representing a particular Twitter user. The color of the nodes represents the outdegree of each user – the darker the color, the more @ messages the specific user sent. Node size is dependent on indegree – the larger the node, the more messages were directed towards the specific user. Straight lines between nodes specify unidirectional communication, while curved lines indicate reciprocity in exchanges of messages. The graphical representation presented in Figure 2 allows us to sort the identified high-end users into broad categories of senders (characterized by darker, smaller nodes), receivers (lighter, larger nodes) and sender-receivers (darker, larger nodes). The main users for each of these categories are shown in Table 3. Where Senders are more active in communicating their views to other, specified users, they are not reciprocal in their use of the replies function – a tendency that is

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

10

new media & society 0(0)

Figure 2. Top @ networks. Degree range: 40 >, graph constructed using the Force Atlas layout in Gephi.

Table 3. Categorizations of top @ users. User category

Examples of identified users

Senders

Feministerna (Politician, Feminist party), federley (Politician, Centre party) theamazinghanna (IT professional) mrquispiam, leoerlandsson (Politicans, Pirate party), evalenajansson (Politician, Social democrats), FRA_PR (Anonymous, political-satirical content), parabrahamsson (IT professional), Gotthjarta (Anonymous) hanscjohansson, Beelzebjorn (Politicans, Pirate party), AnnikaBeijbom (Politician, Liberals), Jodsvall (Politician, Liberals), Vysotskij (Left-wing political blogger), blogfia (Non-partisan political blogger), mickep2, Sdopping (journalists), Nemokrati (Anonymous)

Receivers

Sender-receivers

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

11

Figure 3. Top RT networks. Degree range: 40 >, graph constructed using the Force Atlas layout in Gephi.

reversed for users categorized as Receivers. These latter users tend to receive many directed messages, but are not as active in sending messages of their own. Finally, the Sender-receiver can be characterized as a more well-rounded Twitter user with regards to exchanging messages with others. Users in this category tend to be more versatile in their use of the @ character, in that they demonstrate higher levels of sending as well as receiving messages. Such categorizations allow for assessment of different user strategies. For instance, the map (Figure 2) depicts the liberal politician Annika Bejibom as a central node in the network and as a clear Sender-receiver. Through retweets, a singleton can be redistributed in several steps, leading to a disseminating mode of communication. Studies have demonstrated that retweeting is effective also to distribute messages from users with few followers (Kwak et al., 2010). It follows that retweet activity is crucial as a measure of whose views are made important on Twitter. Again using Gephi, Figure 3 provides a network map of RT activity, identifying the high-end users in this regard. Much like in Figure 2, each node in Figure 3 represents a user. The darker the color of the node, the more active the user is at retweeting the messages of others. Users who are often retweeted are identified by larger node sizes. Line styles are interpreted in the same manner as in Figure 2. Applying similar principles as for Figure 2, three distinct user groups can be identified in Figure 3. First, retweeters

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

12

new media & society 0(0)

Table 4. Categorizations of top RT users. User category

Examples of identified users

Retweeters

Annoula64, MrQuispiam (Politicians, Pirate party), dreadnallen (Politician, Feminist party), all_insane (Anonymous, political-satirical content), Nemokrati (Anonymous) piratpartiet (Offical Twitter, Pirate party), miljopartiet (Official Twitter, Green party), nya_moderaterna (Official Twitter, Conservatives), rodgront2010 (Official Twitter, Red–Green parties coalition), federley (Politician, Centre party), TobiasHoldstock (Press secretary for Conservatives), Pihlblad (Journalist) jocke (IT professional) AnnikaBeijbom (Politician, Liberals), beelzebjorn, falkvinge, annatroberg leoerlandsson (Politicians, Pirate party), vpressfeldt (Student), SDopping, emanuelkarlsten, danielswedin (Journalists), nikkelin (IT professional), UlfBjereld (Professor)

Elites

Networkers

are represented by smaller, darker nodes, indicating high activity with regards to disseminating the messages of other users. Second, larger, lighter coloured nodes – indicating popularity in the network – represent elites as their messages tend to be frequently retweeted. Finally, users classified as networkers are distinguished by their tendency to retweet and to be retweeted. As such, their corresponding nodes tend to be larger and darker. Table 4 provides examples of the categorizations for each of the three identified user types. While the findings show clear examples of Retweeters and Elites, the majority of nodes in Figure 3 can be characterized as relatively larger and darker – the characteristics of the more reciprocally oriented Networker user type. Because of the relative frequency with which their tweets are retweeted by other users, and because they also tend to retweet messages originally posted by others, networkers not only enjoy significant standing in the network – they also contribute to the standings of others. For the category of Elites, the user Pihlblad emerges as one of the largest, lightest nodes in Figure 2. Pihlblad is often retweeted but does not retweet to the same extent – a behavior resulting in Elite network status. An opposite behavioral pattern is distinguishable for the category of Retweeters in general and the user Annoula64 in particular. With perhaps the smallest, darkest node in the graph, this user displays a pattern of use characterized by high levels of retweeting the messages of others, while not being retweeted herself. In sum, the results presented above indicate that core users of the #val2010 hashtag employed quite diverse uses and engaged in different varieties of network connections with each other. Moreover, it becomes apparent from the data presented here that many of the highly active users can be said to at least potentially enjoy privileged positions in their respective professional capacities of journalists, politicians, etc.

Discussion The overall temporal distribution of activity found in the case study signals a relationship between Twitter and mainstream media: spikes of activity in tweeting about the

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

13

2010 Swedish election can be linked to either televised debates, or the media coverage of offline events such as political rallies. Similar results were reported by Bruns (2010) in his study on Twitter use during the 2010 Australian election and by Shamma et al. (2010) in their study of Twitter use during televised US political debates. As such, Twitter activity appears to be largely dependent on other mediated events – a microblogging trend that manifests itself not only in the Swedish, but also in the Australian and US political contexts. In purely quantitative terms, Twitter contributes to a broadening of public debate: it constitutes a novel arena for mediated public communication, and the sheer number of tweets – close to 100,000 in the present sample on the Swedish election – testifies to its use. The important question, though, is to what extent Twitter merely serves as another arena for already established societal actors, or rather facilitates a new distribution among public speakers, allowing new voices or perspectives to be heard. The first important observation from the list of high-end users concerns the significance of their volume of activity: while many from the total sample of users contributed tweets about the election, only a minority did so to any large degree. In terms of volume, high-end users constitute a substantial part of the activity. Second, looking at the identity of this minority, the findings indicate that Twitter indeed serves as a new outlet for speakers already belonging to an elite, or at least affiliated with prominent positions in mainstream media or political life in general. The majority of high-end users are politicians or established journalists and bloggers. This main impression should not, however, lead us to ignore the presence of other actors in the part of the Twittersphere studied here. The most conspicuous group consists of anonymous users. While not ruling out the possibility that these accounts – like ‘all_insane,’ Nemokrati and skogskant – may represent established actors, it does signal the potential for outsiders and less conventional voices to speak up via Twitter. Of course, the very fact that one can build a profile and make an impact in public political discussion under pseudonyms is a rather novel phenomenon – that seems to set online media apart from traditional mass media channels. The ways in which anonymous users interact with and relate to other users, and to what extent their communication gets into wider distribution is a clear topic for further scrutiny. A third observation concerns Twitter as an arena for discussion. Overall, this function seems to be employed first and foremost for disseminating and not for dialogue: merely 7 per cent of the messages in the sample were replies. This distribution largely follows the findings of previous research (Honeycutt and Herring, 2009; Kwak et al., 2010; Tumasjan et al., 2010), and indicates that Twitter is used mainly for sending undirected messages, with its potential for conversation and dissemination employed at a much smaller rate. Yet, if we look closer at how modes of use such as conversations and retweeting played out among the top users, a more complex impression emerges. For the @ network, the category of sender-receiver is arguably the largest one – indicating that many of the top users took a more reciprocal approach to Twitter as conversational tool. For the RT network, the category of networker appeared to be most common, indicating an equal stance towards retweeting and being retweeted among the top users. While the category of elites is dominated by the official Twitter accounts of a variety of major Swedish political actors, other, minor political actors (like non-parliamentary parties

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

14

new media & society 0(0)

[Pirate party, Feminist party], political bloggers, etc.) appear more active in both networks. These findings indicate the potential of the Twitter platform as a means of outreach for such minor, partly marginalized actors. In sum, while major political parties and actors appear to have a hard time adapting to the reciprocal nature of @ and RT practices using Twitter, these means of conversation and networking appear to play some part in the use of minor political actors as identified in this study. Finally, and perhaps most importantly, Twitter remains a marginal activity. While 19 per cent of US adult internet users also use Twitter (Fox et al., 2009), the numbers are substantially lower elsewhere. A German survey found 3 per cent of adults to be Twitter users (Busemann and Gscheidle, 2010), and Norwegian statistics show 6 per cent are users of the service (Arnesen and Solheim, 2010). In Sweden, the estimated number of Twitter users varies between 1 per cent and 8 per cent of internet users – a clear minority in the population (Brynolf, 2011; Najafian, 2010). In a recently completed panel study on the use of various forms of social media in the period leading up to the same election under study here, Dimitrova et al. (2011) found that only around 1 per cent of the respondents in their representative sample used Twitter to follow politicians or political actors on a weekly basis. Taken together, our findings indicate that Twitter falls somewhat short of the expectations held by those most optimistic on behalf of the democratic and disruptive potential of new web tools. The present case study cannot make claims as to whether or not the documented Twitter use had any effect on the outcome of the election. From the results presented here, we can provide anecdotal-at-best evidence suggesting that the effect of Twitter use, if there was one, must be considered minimal. Consider for example a high-end user like Annika Beijbom (user name AnnikaBeijbom as identified above). While the analyses showed Beijbom to be a highly active user of Twitter, her efforts regarding online campaigning in this regard apparently did not help her secure a seat in the Swedish parliament. Similarly, while members of the Pirate Party were among the most active Twitter users in our data (with the official party Twitter account present, as well as party affiliated users like hanscjohansson, Beelzebjorn, falkvinge, annatroberg and leoerlandsson), they did not succeed in gaining parliament representation once the ballots were counted. Obviously, Twitter use cannot serve as a lone variable for predicting political success – the processes at work here are far more complex. Future research should take the intricacies of these matters into account. For example, before we can make any claim on the possible effect of Twitter use, we must improve the methods for ascertaining how many internet users are also users of Twitter. Similarly, the degree to which Twitter users post messages needs to be assessed, accounting for active as well as inactive accounts. Future research in the context studied here should approach the daunting task of assessing the degree to which use of Twitter and other social media applications have effects on voter mindsets and in shaping the electoral process. Such research efforts will be of importance as scholars continue to try and make sense of these ongoing developments of public debate. When we study Twitter users, we study advanced internet users and their patterns of dissemination and interaction. Nonetheless, by studying these high-end users or early adopters might enable us to get a glimpse of how use of Twitter and similar services

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

15

might be shaped in the future – and to what extent and how marginalized political voices can make themselves heard on a wider scale.

Conclusion Twitter research is at a very early stage. One interesting sign of its immaturity is the conceptual relegation of its use as a subcategory, or derivative, of blogging, as microblogging. There clearly are similarities between the two, with our references to blog hyperlink analysis illustrating one such link on the level of research. However, the differences between blogs and Twitter are also substantial, not least from the perspective of the reader. A blog is traditionally defined as a frequently updated website consisting of chronologically dated entries presented in reverse order (e.g. Bruns and Jacobs, 2006). While reverse chronology might still describe the reading of tweets, many differences are also apparent. As a stream of messages from a multitude of sources, often related and tagged according to specific themes or subjects and not necessarily presented on a webpage, Twitter shows a number of differences when compared to blogs. This observation points to the need to acknowledge not only the autonomous qualities and features of Twitter, but also to the need for rethinking our definitions of blogs in light of Twitter, alongside the uptake of RSS and related feed applications. While previous research efforts have mostly tried to categorize the content of tweets, or provide large-scale metrics derived from general Twitter API data, the study presented here provides a different approach to the field of Twitter research. We have suggested a novel way of collecting and analyzing Twitter data, contributing to the development of online methodological tools for the study of new media. To illustrate the value of such an approach, we have applied our suggested rationale on a specific case, providing insights into the activities of high-end Twitter users during an election campaign. While Java et al. (2007) as well as Honeycutt and Herring (2009) proposed detailed categorizations of Twitter content, the analysis presented here suggests broader categories of Twitter users for @ messaging (Senders, Receivers, and Sender-receivers) and retweeting (Retweeters, Elites, Networkers), respectively. It is our belief that the user categories identified here could be of interest to similar, future research endeavors. For further research into online political activity, the validity of these user types could be tested in other socio-geographical contexts. It should also be of relevance to see if the roles of users identified in the different categories in this study emerge also in other contexts. Specifically, it should be of interest to see if established politicians take on the role of senders regardless of electoral environment, and if non-parliamentary actors (such as the Pirate party in this study) tend to appear as networkers also outside of the context under scrutiny here. Moving beyond political Twitter use, the categories presented here could be of use also in other thematic contexts. While the approach to studying Twitter presented here has some clear merits, it is not without its limitations. In the following, these limitations are assessed in combination with further ideas for future research efforts. A first set of limitations has to do with hashtags. As studies have showed, in Twitter communication, hashtags are dynamic entities, and often contested (Hickman, 2010). Hashtag identification can be difficult – especially pertaining to a specific topic and as

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

16

new media & society 0(0)

the discussion unfolds. In our illustrative case study, although val2010 remained stable and dominant throughout the period of study in the present case, it is safe to assume that the data did not cover all relevant tweets. Some messages used other hashtags, misspelled them, or left them out altogether. Even those who used the val2010 tag when composing, a singleton, say, criticizing a political candidate on a televised debate, might not employ the tag for a subsequent reply message that results from the original singleton. However, even though the use of a specific hashtag will entail biases, the resulting data set does, compared to other instances of off- and online mediated communication, provide a basis for a comprehensive analysis. A second limitation has to do with context. While an analysis of Twitter use can yield detailed insights into the practices of public communication on one specific online arena, the ramifications of the findings should, in the next step, be made subject to comparisons. This could entail comparison with Twitter use in other social settings, or comparison with other forms of online communication – say blogging, or discussion forums – in the same setting. Such comparisons would enable a better understanding of the weight and impact of the patterns of use as identified in this study. A broader scope applied to new contexts will help further this developing research area. The approach suggested here focuses on structural aspects of Twitter use. Through our analysis, we map out the basis for different Twitter user types. Although the work presented here provides the research community with a point of departure for further inquiry into these matters, it does not, however, take the step into qualitative research efforts. While such in-depth analysis is arguably beyond the scope for the focus of this particular article, more qualitative ways of approaching data like these could be feasible for future research efforts. For example, individual Twitter users of high interest (perhaps identified through an analysis similar to the one presented above) and their patterns of use could be studied. Another interesting aspect might be to focus on retweeted messages. While an analysis such as the one presented here provides an overview on who is retweeting who, the contents of these messages, as well as Twitter messages in general, are of obvious interest. By taking the next step and moving beyond the structural, overarching analytical approach employed here, future research will be able to provide interested scholars with a deeper understanding of the practices of Twitter use. Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References Arceneaux N and Weiss AS (2010) Seems stupid until you try it: Press coverage of Twitter, 2006–9. New Media & Society 12(8): 1262–1279. Available at: nms.sagepub.com (consulted 19 May 2010). Arnesen M and Solheim E (2010) TNS Gallup Interbuss q2 Bruk av sosiale medier. Available at: http://nrkbeta.no/2010/08/13/30-minutter-facebook-daglig/tns-gallup-interbuss-q2-brukav-sosiale-medier/ (consulted 12 October 2011). Baker EC (2002) Media, Markets, and Democracy. Cambridge, MA: Cambridge University Press.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

17

Bastian M, Heymann S and Jacomy M (2009) Gephi: An open source software for exploring and manipulating networks. Paper presented at the Third International ICWSM Conference, San Jose, CA. Bruns A (2010) Key events in Australian (micro-)blogging during 2010. Paper presented at ECREA 2010, Hamburg. Available at: http://www.slideshare.net/Snurb/key-events-in-australianmicroblogging-during-2010 (consulted 13 October 2011). Bruns A, Burgess J, Highfield T, Kirchhoff L and Nicolai T (2010) Mapping the Australian networked public sphere. Social Science Computer Review 29(3): 277–287. Available at: ssc. sagepub.com (consulted 7 April 2011). Bruns A and Jacobs J (2006) Introduction. In Bruns A and Jacobs J (eds) Uses of Blogs. New York, NY: Peter Lang, 1–9. Brynolf H (2011) Twitter Census: En rapport om Twitter i Sverige. Stockholm: Intellecta Corporate. Busemann K and Gscheidle C (2010) Web 2.0: Nutzung steigt – Interesse an Aktiver Teilhabe sinkt. Media Perspektiven (7–8): 359–368. Carlsson U and Facht U (2010) Medie-Sverige: statistik och analys. 2010. Gothenburg: NORDICOM-Sweden. Castells M (2007) Communication, power and counter-power in the network society. International Journal of Communication 1(1): 238–266. Chadwick A (2008) Web 2.0: New challenges for the study of e-democracy in an era of informational exuberance. I/S: Journal of Law and Policy for the Information Society 4(3): 9–42 Christensen M and Christensen C (2008) The after-life of Eurovision 2003: Turkish and European social imaginaries and ephemeral communicative space. Popular Communication 6(3): 155–172. Davis R (2005) Politics Online: Blogs, Chatrooms, and Discussion Groups in American Democracy. New York, NY: Routledge. Dimitrova DV, Shehata A, Strömbäck J and Nord L (2011) The effects of digital media on political knowledge and participation in election campaigns: Evidence from panel data. Paper presented at the ICA annual conference, Boston, MA. Druckman JN, Kifer MJ and Parkin M (2007) The technological development of congressional candidate web sites. Social Science Computer Review 25(4): 425–442. Ekdale B, Namkoong K, Fung TKF and Perlmutter DD (2010) Why blog? (then and now): Exploring the motivations for blogging by popular American political bloggers. New Media & Society 12(2): 217–234. Fox S, Zickuhr K and Smith A (2009) Twitter and Status Updating, Fall 2009. Available at: http://www.pewinternet.org/Experts/~/link.aspx?_id=6C747837133C4A54A4D0351E26834 78B&;_z=z (consulted 13 October 2011). Gaffney D (2010, 26 and 27 April) #iranElection: Quantifying online activism. Paper presented at WebSci10: Extending the Frontiers of Society On-Line. Raleigh, NC. Gibson RK, Lusoli W and Ward S (2008) Nationalizing and normalizing the local? A comparative analysis of online candidate campaigning in Australia and Britain. Journal of Information Technology & Politics 4(4): 15–30. Gil De Zuniga H, Puig-I-Abril E and Rojas H (2009) Weblogs, traditional sources online and political participation: An assessment of how the internet is changing the political environment. New Media & Society 11(4): 553–574. Golbeck J, Grimes JM and Rogers A (2010) Twitter use by the US Congress. Journal of the American Society for Information Science and Technology 61(8): 1612–1621. Grace JH, Zhao D and boyd d (2010) Microblogging: What and how can we learn from it? In: Proceedings of the 28th International Conference on Human Factors in Computing Systems, 10–15 April 2010. Atlanta, GA: ACM.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

18

new media & society 0(0)

Halavais A (2009) Search Engine Society. Cambridge, MA: Polity Press. Hickman J (2010) What’s the hash tag? Folksonomy, brand, and control: Organising and owning conversations on Twitter. Paper presented at ECREA 2010, Hamburg. Hilbert M (2009) The maturing concept of e-democracy: From e-voting and online consultations to democratic value out of jumbled online chatter. Journal of Information Technology & Politics 6(2): 87–110. Honeycutt C and Herring SC (2009) Beyond microblogging: Conversation and collaboration via Twitter. Paper presented at the 42nd Hawaii International Conference on System Sciences. Waikoloa, Big Island, Hawaii. Jackson NA and Lilleker DG (2009) Building an architecture of participation? Political parties and Web 2.0 in Britain. Journal of Information Technology & Politics 6(3): 232–250. Jansen BJ, Zhang MM, Sobel K and Chowdury A (2009) Twitter power: Tweets as electronic word of mouth. Journal of the American Society for Information Science and Technology 60(11): 2169–2188. Java A, Song X, Finin T and Tseng B (2007) Why we twitter: Understanding microblogging usage and communities. In: Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 Workshop on Web Mining and Social Network Analysis. San Jose, CA: ACM. Jones J and Himelboim I (2010) Just a guy in pajamas? Framing the blogs in mainstream US newspaper coverage (1999–2005). New Media & Society 12(2): 271–288. Kahn R and Kellner D (2004) New media and internet activism: From the ‘Battle of Seattle’ to blogging. New Media & Society 6(1): 87–95. Kerbel MR and Bloom JD (2005) Blog for America and civic involvement. Harvard International Journal of Press-Politics 10(4): 3–27. Kwak H, Lee C, Park H and Moon S (2010) What is Twitter, a social network or a news media? In: Proceedings of the 19th International Conference on World Wide Web. Raleigh, NC: ACM. Larsson AO (2011) “Extended infomercials” or “Politics 2.0”? A study of Swedish political party Web sites before, during and after the 2010 election. First Monday 16(4). Larsson AO and Hrastinski S (2011) Blogs and blogging: Current trends and future directions. First Monday 16(3). Lilleker DG and Malagón C (2010) Levels of interactivity in the 2007 French Presidential candidates’ websites. European Journal of Communication 25(1): 25–42. Marwick AE and boyd d (2010) I Tweet honestly, I Tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society 13(1): 114-133. Available at: nms. sagepub.com (consulted 7 July 2010). Moe H (2011) Mapping the Norwegian blogosphere: Methodological challenges in internationalizing Internet research. Social Science Computer Review 29: 313–326. Available at: ssc. sagepub.com (consulted 23 May 2011). Morozov E (2009) Iran: Downside to the “Twitter revolution”. Dissent 56(4): 10–14. Morozov E (2011) The Net Delusion: The Dark Side of Internet Freedom. New York, NY: Public Affairs. Najafian M (2010) Det digitala livet. Microsoft Research Reports. Stockholm: Microsoft. Page BI (1996) Who Deliberates? Mass Media in Modern Democracy. Chicago, IL and London: University of Chicago Press. Shamma D, Kennedy L and Churchill E (2010) Tweetgeist: Can the Twitter timeline reveal the structure of broadcast events. Paper presented at the 2010 ACM Conference on Computer Supported Work. Savannah, GA. Smith A (2009) The Internet’s Role in Campaign 2008. Pew Internet & American Life Project. Washington, DC: Pew Research Center.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

Larsson and Moe

19

Sweetser Trammell KD (2007) Candidate campaign blogs – Directly reaching out to the youth vote. American Behavioral Scientist 50(9): 1255–1263. Tumasjan A, Sprenger TO, Sandner PG and Welpe IM (2010) Predicting elections with Twitter: What 140 characters reveal about political sentiment. Paper presented at the Fourth International AAAI Conference on Weblogs and Social Media. Washington, DC: George Washington University. TwapperKeeper (2010) Your TwapperKeeper – Archive your own Tweets. Available at: http:// your.twapperkeeper.com/ (consulted 13 October 2011). Vaccari C (2008a) From the air to the ground: The Internet in the 2004 US presidential campaign. New Media & Society 10(4): 647–665. Vaccari C (2008b) Research note: Italian parties’ websites in the 2006 elections. European Journal of Communication 23(1): 69–77. Vaccari C (2008c) Surfing to the Elysee: The Internet in the 2007 French Elections. French Politics 6(1): 1–22. Warner M (1990) The Letters of the Republic. Publication and the Public Sphere in Eighteenthcentury America. Cambridge, MA: Harvard University Press. Winston B (1998) Media, Technology and Society – A History: From the Telegraph to the Internet. New York, NY: Routledge.

Anders Olof Larsson is a PhD candidate at the Department of Informatics and Media, Uppsala University, Sweden. He is also associated with The Swedish Research School of Management and Information Technology. His research focuses on how organizations, societal institutions and their respective audiences make use of online interactivity and social media. His web site can be accessed at http://www.andersoloflarsson.se Hallvard Moe is associate professor at the Department of Information Science and Media Studies, University of Bergen. His PhD thesis from 2009 analyzed European public service broadcasters on the internet. He has recently undertaken a project funded by the Norwegian Research Council on how online media participation transforms the public sphere.

Downloaded from nms.sagepub.com at Uppsala Universitetsbibliotek on March 20, 2012

The Swedish Research School of Management and Information Technology MIT The Swedish Research School of Management and Information Technology (MIT) is one of 16 national research schools supported by the Swedish Government. MIT is jointly operated by the following institutions: Blekinge Institute of Technology, Gotland University College, Jönköping International Business School, Karlstad University, Linköping University, Linnaeus University Växjö, Lund University, Mälardalen University College, Stockholm University, Örebro University, IT University of Göteborg, and Uppsala University, host to the research school. At the Swedish Research School of Management and Information Technology (MIT), research is conducted, and doctoral education provided, in three fields: management information systems, business administration, and informatics.

DISSERTATIONS FROM THE SWEDISH RESEARCH SCHOOL OF MANAGEMENT AND INFORMATION TECHNOLOGY Doctoral theses (2003- ) 1. Baraldi, Enrico (2003), When Information Technology Faces Resource Interaction: Using IT Tools to Handle Products at IKEA and Edsbyn. Department of Business Studies, Uppsala University, Doctoral Thesis No. 105. 2. Wang, Zhiping (2004), Capacity-Constrained Production-Inventory Systems: Modelling and Analysis in both a Traditional and an E-Business Context. IDA-EIS, Linköpings universitet och Tekniska Högskolan i Linköping, Dissertation No. 889 3. Ekman, Peter (2006), Enterprise Systems & Business Relationships: The Utilization of IT in the Business with Customers and Suppliers. School of Business, Mälardalen University, Doctoral Dissertation No 29. 4. Lindh, Cecilia (2006), Business Relationships and Integration of Information Technology. School of Business, Mälardalen University, Doctoral Dissertation No 28. 5. Frimanson, Lars (2006), Management Accounting and Business Relationships from a Supplier Perspective. Department of Business Studies, Uppsala University, Doctoral Thesis No. 119. 6. Johansson, Niklas (2007), Self-Service Recovery. Information Systems, Faculty of Economic Sciences, Communication and IT, Karlstad University, Dissertation KUS 2006:68. 7. Sonesson, Olle (2007), Tjänsteutveckling med personal medverkan: En studie av banktjänster. Företagsekonomi, Fakulteten för ekonomi, kommunikation och IT, Karlstads universitet, Doktorsavhandling, Karlstad University Studies, 2007:9.

8. Maaninen-Olsson, Eva (2007), Projekt i tid och rum: Kunskapsintegrering mellan projektet och dess historiska och organisatoriska kontext. Företagsekonomiska institutionen, Uppsala universitet, Doctoral Thesis No. 126. 9. Keller, Christina (2007), Virtual learning environments in higher education: A study of user acceptance. Linköping Studies in Science and Technology, Dissertation No. 1114. 10. Abelli, Björn (2007), On Stage! Playwriting, Directing and Enacting the Informing Processes. School of Business, Mälardalen University, Doctoral Dissertation No. 46. 11. Cöster, Mathias (2007), The Digital Transformation of the Swedish Graphic Industry. Linköping Studies in Science and Technology, Linköping University, Dissertation No. 1126. 12. Dahlin, Peter (2007), Turbulence in Business Networks: A Longitudinal Study of Mergers, Acquisitions and Bankruptcies Involving Swedish IT-companies. School of Business, Mälardalen University, Doctoral Thesis No. 53. 13. Myreteg, Gunilla (2007), Förändringens vindar: En studie om aktörsgrupper och konsten att välja och införa ett affärssystem. Företagsekonomiska institutionen, Uppsala universitet, Doctoral Thesis No. 131. 14. Hrastinski, Stefan (2007), Participating in Synchronous Online Education. School of Economics and Management, Lund University, Lund Studies in Informatics No. 6. 15. Granebring, Annika (2007), Service-Oriented Architecture: An Innovation Process Perspective. School of Business, Mälardalen University, Doctoral Thesis No. 51. 16. Lövstål, Eva (2008), Management Control Systems in Entrepreneurial Organizations: A Balancing Challenge. Jönköping International Business School, Jönköping University, JIBS Dissertation Series No. 045. 17. Hansson, Magnus (2008), On Closedowns: Towards a Pattern of Explanation to the Closedown Effect. Swedish Business School, Örebro University, Doctoral Thesis No. 1. 18. Fridriksson, Helgi-Valur (2008), Learning processes in an inter-organizational context: A study of krAft project. Jönköping International Business School, Jönköping University, JIBS Dissertation Series No. 046. 19. Selander, Lisen (2008), Call Me Call Me for some Overtime: On Organizational Consequences of System Changes. Institute of Economic Research, Lund Studies in Economics and Management No. 99. 20. Henningsson, Stefan (2008), Managing Information Systems Integration in Corporate Mergers & Acquisitions. Institute of Economic Research, Lund Studies in Economics and Management No. 101. 21. Ahlström, Petter (2008), Strategier och styrsystem för seniorboende-marknaden. IEI-EIS, Linköping universitetet och Tekniska Högskolan i Linköping, Doktorsavhandling, Nr. 1188. 22. Sörhammar, David (2008), Consumer-firm business relationship and network: The case of ”Store” versus Internet. Department of Business Studies, Uppsala University, Doctoral Thesis No. 137. 23. Caesarius, Leon Michael (2008), In Search of Known Unknowns: An Empirical Investigation of the Peripety of a Knowledge Management System. Department of Business Studies, Uppsala University, Doctoral Thesis No. 139.

24. Cederström, Carl (2009), The Other Side of Technology: Lacan and the Desire for the Purity of Non-Being. Institute of Economic Research, Lund University, Doctoral Thesis, ISBN: 9185113-37-9. 25. Fryk, Pontus, (2009), Modern Perspectives on the Digital Economy: With Insights from the Health Care Sector. Department of Business Studies, Uppsala University, Doctoral Thesis No. 145. 26. Wingkvist, Anna (2009), Understanding Scalability and Sustainability in Mobile Learning: A Systems Development Framework. School of Mathematics and Systems Engineering, Växjö University, Acta Wexionesia, No. 192, ISBN: 978-91-7636-687-5. 27. Sällberg, Henrik (2010), Customer Rewards Programs: Designing Incentives for Repeated Purchase. Blekinge Institute of Technology, School of Management, Doctoral Dissertation Series No. 2010:01. 28. Verma, Sanjay (2010), New Product Newness and Benefits: A Study of Software Products from the Firms’ Perspective, Mälardalen University Press, Doctoral Thesis. 29. Iveroth, Einar (2010), Leading IT-Enabled Change Inside Ericsson: A Transformation Into a Global Network of Shared Service Centres. Department of Business Studies, Uppsala University, Doctoral Thesis No. 146. 30. Nilsson, Erik (2010), Strategi, styrning och konkurrenskraft: En longitudinell studie av Saab AB, IEI-EIS, Linköpings universitet och Tekniska Högskolan i Linköping, Doktorsavhandling, Nr. 1318. 31. Sjöström, Jonas (2010), Designing Information Systems: A pragmatic account, Department of Informatics and Media, Uppsala University, Doctoral Thesis. 32. Numminen, Emil (2010), On the Economic Return of a Software Investment: Managing Cost, Benefit and Uncertainty, Blekinge Institute of Technology, School of Management, Doctoral Thesis. 33. Frisk, Elisabeth (2011), Evaluating as Designing: Towards a Balanced IT Investment Approach, IT University, Göteborg, Doctoral Thesis. 34. Karlsudd, Peter (2011), Support for Learning: Possibilities and Obstacles in Learning Applications, Mälardalen University, Doctoral Thesis. 35. Wicander, Gudrun (2011), Mobile Supported e-Government Systems: Analysis of the Education Management Information System (EMIS) in Tanzania, Karlstad University, Doctoral Thesis. 36. Åkesson, Maria (2011), Role Constellations in Value Co-Creation: A Study of Resource Integration in an e-Government Context, Karlstad University, Doctoral Thesis. 37. Nfuka, Edephonce N. (2012), IT Governance in Tanzanian Public Sector Organisations, Department of Computer and Systems Sciences, Stockholm University, Doctoral Thesis. 38. Larsson, Anders Olof (2012), Doing Things in Relation to Machines: Studies on Online Interactivity, Department of Informatics and Media, Uppsala University, Doctoral Thesis. Contact  person:    Professor  Pär  Ågerfalk,  Director  of  MIT,  Uppsala  University      [email protected]     Address:    The  Swedish  Research  School  of  Management  and        Information  Technology,  Department  of  Business  Studies,        Uppsala  University,  Box  513,  751  20  Uppsala          http://www.forskarskolan-­‐mit.nu/mit/