2006:13

DOCTORA L T H E S I S

Health and Performance in Small Enterprises Studies of Organizational Determinants and Change Strategy

STIG VINBERG

Luleå University of Technology Department of Human Work Sciences, Division of Industrial Work Environment 2006:13|: -1544|: - -- 06 ⁄13 -- 

Doctoral Thesis Department of Human Work Sciences

Health and Performance in Small Enterprises Studies of Organizational Determinants and Change Strategy

STIG VINBERG

Luleå University of Technology and National Institute for Working Life June 2006

Doctoral Thesis 2006:13 ISSN 1402-1544 ISRN LTU-DT--06/13--SE © Stig Vinberg

Printed in Luleå, Sweden by Universitetstryckeriet The cover illustration was designed by Örjan Östlund

Abstract This thesis focuses on relationships between work organizational factors, health and performance outcomes, and change strategies in small enterprises. The reasons for this work are the growing interest in entrepreneurship and small business development, and statements to the effect that small enterprises lack the adequate resources and competences needed for successful management of workplace change processes. Another reason is the fragmented state of small business research about relationships between organizational factors, change strategies and outcomes. The thesis comprises one theoretical study and four empirical studies using a total sample of 118 small enterprises and 50 small public workplaces with a total of 1206 co-workers and leaders. Data was collected through the use of questionnaires, structured interviews and analyses of human resource accounting data. Correlation analyses in Study II showed rather strong relationships between the outcome indicators related to health and performance. The relations between these outcomes and indicators of assumed determinant organizational factors resulted in a rather large number of relationships (Studies I, II and V). A general result is that more strong relationships between determinants and performance were identified than between determinants and health. However, rather strong relations between, on the one hand, indicators of respectful leadership, team spirit, creative work, work intensity demands and restructuring worries and, on the other hand, health were found. Structural analyses in Studies I and II resulted in the identification of six components. Thus, it was possible to group indicators into larger “bundles” which have similarities to research results about larger enterprises. In Study II analyses using multidimensional scaling resulted in a grouping of enterprises with high positions versus enterprises with low positions on two main dimensions – performance and health related dimensions. Study IV showed that studied micro-enterprises attached less priority to goals related to workplace health and work organization compared to the studied public workplaces. The micro-enterprises attached more priority to goals associated with the physical working environment, production and quality development. In Study V two change strategy approaches were studied using a longitudinal design. The results concerning changes of determinants and outcomes, after workplace related interventions were performed, indicate that change processes with a broad strategy, and high employee and management involvement can apply to small enterprises, and help to improve their health and performance. This is partly in line with the discussion in Study III about the need for integrated planning and participative approaches in workplace change processes. The thesis findings point to the importance of organizational factors such as aspects of leadership, psychosocial work environment and learning for small business development. Keywords: Small enterprises, health, performance, change processes, workplace interventions i

II

Sammanfattning Denna avhandling fokuserar på relationer mellan arbetsorganisatoriska faktorer, utfall relaterade till hälsa och prestation/resultat samt förändringsstrategier i små företag. Motiv för arbetet är det växande intresset för entreprenörskap och småföretagsutveckling samt uppgifter om att de små företagen saknar adekvata resurser och kompetens för att framgångsrikt driva arbetsplatsinriktade förändringsprocesser. Ett annat skäl är fragmenterad småföretagsforskning om relationer mellan organisatoriska faktorer, förändringsstrategier och utfall. Avhandlingen omfattar en teoretiskt baserad studie samt fyra empiriska studier om totalt 118 små företag och 50 små offentliga arbetsplatser som totalt inkluderar 1206 medarbetare och ledare. Data samlades in med hjälp av frågeformulär, strukturerade intervjuer och personalekonomiska analyser. Korrelationsanalyser i Studie II visade på ganska starka samband mellan utfall relaterade till hälsa och prestation/resultat. Relationer mellan dessa utfall och indikatorer av antagna organisatoriska determinanter resulterade i ett ganska stort antal samband (Studie I, II och V). Ett generellt resultat är att det är fler starka samband mellan determinanter och prestation/resultat än mellan determinanter och hälsa. Emellertid finns relativt starka samband mellan, å ena sidan, indikatorerna respektfullt ledarskap, laganda, kreativt arbete, arbetskrav och oro för förändringar och, å andra sidan, hälsa. Strukturella analyser i Studie I och II resulterade i att sex komponenter kunde identifieras. Således var det möjligt att gruppera indikatorer i större ”buntar” vilket har likheter med en del resultat avseende större företag. I Studie II resulterade analyser med hjälp av multidimensionell skalning i en gruppering av företag med höga värden respektive låga värden för två huvuddimensioner – en prestation/resultat- och en hälsorelaterad dimension. Studie IV visade att studerade mikroföretag fäster mindre prioritet på mål relaterade till arbetshälsa och arbetsorganisation jämfört med studerade offentliga arbetsplatser. Mikroföretagen fäster mer prioritet på mål som hör ihop med fysisk arbetsmiljö, produktion och kvalitetsutveckling. I Studie V studerades två förändringsstrategier med en longitudinell design. Resultaten avseende förändringar av determinanter och utfall, efter genomförda arbetsplatsrelaterade interventioner, indikerar att förändringsprocesser med bred strategi och hög involvering från medarbetare och ledare kan användas i små företag och bidra till att förbättra deras hälsa och resultat. Detta är till del i linje med diskussioner i Studie III om behoven av integrerad planering och participativa ansatser i arbetsplatsbaserade förändringsprocesser. Resultaten i avhandlingen pekar på betydelsen av organisatoriska faktorer som aspekter av ledarskap, psykosocial arbetsmiljö och lärande för småföretagsutveckling. Nyckelord: Små företag, hälsa, resultat, prestation, förändringsprocesser, arbetsplatsinriktade interventioner

iii

IV II

Preface This doctoral thesis is one result of four research- and development projects in County of Jämtland, Sweden and Trøndelag, Norway, conducted during the period 1997 to 2003. In these projects, several valuable practical results have been achieved of use for small organizations, leaders, co-workers and authorities and for developing questions for researchers. These introductory words expose my belief in that working life researchers have a responsibility to and could benefit from co-operating with practitioners. By participating in these projects, and other workplace development projects, my experience is that sustainable and participative approaches for workplace development are needed. This experience, together with the view that there is a lack of knowledge about how improvements of organizational factors can contribute to health and performance in small firms have been driving forces for the studies in this thesis. Naturally this thinking is motivated by my recognition that small enterprises are of great importance for working life and the economy First of all I want to thank all those in these projects who have contributed to make my studies possible, particularly all co-workers and leaders in the small enterprises and workplaces who took their time to complete extensive questionnaires and answer my many questions. During the last eight years, I have been working at the National Institute for Working Life in Östersund as a ‘development leader’ - an interesting journey that have given me valuable knowledge about key factors and obstacles when an organization is growing. I am grateful towards the institute for supporting my research even when my primary work tasks have been more development oriented. I also want to express my gratitude to the following persons: Karl W Sandberg, Mid-Sweden University, my academic advisor and friend, for guidance and for giving important encouragement at times when it was most needed. Jan Johansson, Luleå University of Technology, for good advice and helpful support concerning all the many practical issues associated with my disputation. Kaj Frick, National Institute for Working Life, Lars-Erik Wolvén, MidSweden University and Gunnar Gelin, statistical advisor, for support in the former part of this work. My friends, and project partners; Claes Malmquist and Johan Larsson, for good advice and creative dialogues after all the tennis sessions and lunches at ‘Tages’. Lena Abrahamsson, Luleå University of Technology, Jörgen Lithander, Swedish Institute for Growth Policy Studies and Mikael Nordenmark, Chatrine v

Höckertin and Lisa Sundin, National Institute for Working Life, for their wise and important comments on the manuscript. All other colleagues at the National Institute in Östersund for encouragement and support. Jonathan Stubbs, for professional guidance and help with the English language and for giving me helpful advice. At last, but of course not least, big thanks to my wonderful children, Anton and Joanna, for their love, encourage and understanding. Frösön, May 4, 2006 Stig Vinberg

vi

VI II

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

Vinberg, S., Gelin, G. & Sandberg, K. W. (2000). Information technology levels, competence development and performance in Swedish small business enterprises. Behaviour & Information Technology, 19(3), 201210.

II.

Vinberg, S. & Gelin, G. (2005). Organizational and health performance in small enterprises in Norway and Sweden. Work: A Journal of Prevention, Assessment & Rehabilitation, 24(3), 305-316.

III.

Sandberg, K. W. & Vinberg, S. (2000). Information technology and learning strategies in small enterprises. Behaviour & Information Technology, 19(3), 221-227.

IV.

Vinberg, S. (2004). Change processes and health outcomes in microenterprises and small public workplaces in rural areas. Finnish Journal of Rural Research and Policy, 12(4), 151-162.

V.

Vinberg, S. (2006). Workplace health interventions in small enterprises – a Swedish longitudinal study. Re-submitted to Work: A Journal of Prevention, Assessment & Rehabilitation.

The articles are reprinted with permission.

vii

Table of Contents Abstract Sammanfattning Preface Appended papers

i iii v vii

Introduction

1

Why study health and performance in small enterprises?

2

Purpose and research questions

6

Definitions and delimitations

7

Theoretical framework Small enterprise definitions and characteristics

9 9

Health and Performance – two outcome concepts

Summary and chapter comments

13 13 14 15 18 21 22 24 24 26 27 29 31

Materials and methods Methodological choices

33 33

Analysis model

33

Empirical base and research design

35 36 37 38 38 40

Health Performance

Healthy work organizations – models and definitions Leadership aspects Psychosocial work environment aspects Learning aspects

Change strategy and workplace interventions Change strategy Workplace interventions Interventions on a general level Small enterprise interventions

Study I Indicators Statistical analysis Studies II and V Indicators – Study II

Quality of data

41 42 42 42 43 43 43

Summary of study results Result overview

45 45

Relationships between determinants and outcomes (Studies I, II, V)

47

Structure of indicators (Study I, II)

49

Structure of enterprise groups (Studies I, II, IV, V)

52

Change strategies and change goals (Studies III, IV)

53

Effects of workplace interventions (Study V)

55

Discussion and conclusions Relationships between determinants and outcomes

59 59

Structure of indicators and enterprise groups

61

Change strategies and change goals

62

Effects of workplace interventions

63

Methodological considerations

64

Conclusions and implications for practitioners and future research

66

References

71

Indicators – Study V Statistical analyses Study IV Indicators Statistical analyses Study III

Appended papers I - V

Introduction This thesis looks at relationships between work organization factors and health and performance outcomes plus how these factors and outcomes relate to different change strategies in small enterprises1. Concepts about ‘healthy work organizations’ formed a base for the orientation of the work reported. These are: • That it is possible to distinguish healthy from unhealthy work systems and • That worker well-being and organizational performance can be fostered by a common set of organizational characteristics. Although this view is attractive, there is a lack of knowledge and understanding about how we can improve both health and performance-related outcomes; especially for small enterprises. A motive for the studies included in the thesis was also to look at and compare research findings from different research areas, which state that small enterprises typically have difficulty carrying out workplace development processes. Limited resources and knowledge about such processes are often described as the primary causes of failure. Hopefully the findings in this thesis will stimulate further research, provide practitioners with information that can contribute to worker and leader health, and help small operations to be more effective. The empirical base for this thesis is from four research and development projects performed during the period 1997 – 2003 in the northern part of Norway and Sweden2. In these projects, several public and private actors cooperated with researchers towards: • Generating networks between small workplaces and • Developing models of workplace change processes and workplace health promotion. The thesis disposition is as follows. The introductory chapter consists of the research area’s background, the purpose, the research questions, the definitions and the demarcations in the study. In Chapter Two the theoretical frame of reference is presented. Areas such as small enterprise definitions and characteristics, health and performance outcomes, healthy work organizations, change strategy and workplace interventions are presented. The emphasis is 1

2

The terms “enterprise”, “firm”, “company” and “business” are used interchangeably to describe a privately owned producer of goods and/or services. The main focus in this thesis is on small enterprises (including micro-enterprises) with less than 50 employees. Names of the projects were: Företagsutveckling genom ökad kompetens och bättre hälsa. Ett pilotprojekt på små och medelstora arbetsplatser (Business development through increased competence and better health. A pilot project in small and medium sized enterprises); Företagsutveckling – utan gränser (Company development – without limits); Ragundaprojektet (the Ragunda-project); Friskare arbetsplatser i Jämtlands län (Healthier workplaces in Jämtland County). 1

partly about organizations on a general level and partly about small companies. In Chapter Three the methodological choices are presented and discussed. The analysis model used and the empirical base and the research design are described. Chapter Four consists of a summary of study results. The results are grouped into the subsections: relationships between determinants and outcomes, structure of indicators, structure of enterprise groups, change strategies and change goals, and effects of workplace interventions. In Chapter Five the findings with respect to the purpose of the thesis and research questions are discussed together with methodological considerations, practical implications and proposals for future research.

Why study health and performance in small enterprises? There are several good reasons to be interested in health and performance in small enterprises. This section presents and discusses these reasons. Historically, there has been a strong faith in the capacity of big enterprises; a faith that big enterprises are the major actors in national and regional economies in Sweden and elsewhere. From the mid 1980s onward, a growing interest in small enterprises as an important actor in job-creation processes, job loss, outsourcing by big enterprises and increasing globalisation begin to develop (Lithander, 2005). According to Henrekson (2001), these changes have influenced general attitudes towards entrepreneurs and small enterprises in Sweden. Found was that between 1978 and 1997 the number of people who consider it important to encourage entrepreneurship and establishment of new small businesses has almost trebled - from 30 to 88 percent. Small enterprises comprise the overwhelming majority of workplaces and employ a sizable proportion of the total workforce in most countries – between one third and one half of total employment depending on the country (Eakin et al., 2000). In Sweden small firms employ more than 50 percent of the total workforce, and seven out of ten vacant jobs are with companies employing fewer than 200 persons (Davidsson et al., 2001; Wiklund, 1998). According to Statistics Sweden (2004), 99.2 percent of private enterprises in Sweden in 2003 were small enterprises (with fewer than 50 employees) and 94.7 percent were microenterprises (with fewer than 10 employees). Corresponding figures for a rural area such as the Swedish County of Jämtland, the major data source of the empirical part of the thesis, are 99.6 and 96.7 percent. The importance of small enterprises in rural areas can also be observed when looking at their contribution to employment; in 2003 employment in small enterprises accounted for 54.4 percent of the jobs in Sweden; for Jämtland it was 70.4 percent (Lithander, 2005). A similar pattern can be seen for rural areas in Norway as in Nord-Trøndelag, the other source of empirical data; where approximately 60 percent of the workforce is employed by small enterprises (Spilling, 2000). 2

The factors described above have made small business research a fast growing speciality over the past decade (Abrahamsson, 2006; Curran & Blackburn, 2001; Davidsson et al., 2001; Johannisson & Landström, 1999). Curran & Blackburn (2001, p. 8) note that “this research area is not a discipline in a conventional academic sense, rather it is an area which can be described as relatively young, with a multi- or cross-disciplinary approach to a degree rare elsewhere in social and business research”. They concluded that small business research is an everrenewable resource and pointed out areas, as for example, stress and knowledge management as relevant topics for future research. In a recent Swedish anthology, Abrahamsson (2006) pointed out that work organization perspectives, and studies concerning processes within an organization have been neglected in small business research. That the majority of small firms do not grow and that many are not interested in pursuing growth is a common observation made by small enterprise researchers and commented on in popular literature. Wiklund (1998) states that rapid-growth firms are not concentrated in specific industries and that it is therefore important to determine which factors affect performance of individual small firms. In particular, to identify internal factors created by small firms themselves. One interesting finding from a Swedish study about managerial willingness for growth (where data was collected from three separate studies over a ten-year period) was that managers judged non-economic concerns like employee well-being to be more important than the possibility of profitability (Wiklund et al., 2003). Although interest in small firms has increased among researchers, the knowledge base for micro-enterprises is scant. One indicator is a search using two major scientific databases (Academic Search Elite and Business Source Elite). The results yielded 300 times more hits for small enterprises than for micro-enterprises3. During the last few years, sickness absence in Sweden and Norway has increased more than in many other European countries (RFV 2002:11). There are no medical facts that explain these dramatic changes (SOU 2002:62). It is a remarkable paradox that the sickness absence figures in small enterprises are much lower than in larger ones (Bornberger et al., 2003; Lindström et al., 2000; SOU 2000:121). Drawbacks are that small enterprises have higher levels of injuries and they do not systematically manage working environments in the same way as larger enterprises (Walters, 2001). Göransson et al. (2002) found that small workplaces have a lower sick-leave frequency rate than larger companies. 3

The search was performed both in Academic Search Elite and Business Source Elite (200510-30). The words “small business” or “small firm” or “small enterprise” gave 39 038 hits. The words “micro business” or “micro firm” or “micro enterprise” or “sole proprietor’s business” or “one-man business” or “very small enterprise” or “micro-sized enterprise” gave 127 hits. The result shows 307 times more hits for the small enterprise group.

3

They contended that this was not a case of the smaller workplaces recruiting less illness-prone individuals and rather that the control and significance felt by small workplace employees caused them to use less sick-leave. Some contrary explanations for these differences in sickness absence and workplace size are that the social control is higher in small enterprises and that it therefore is difficult to be off work (Bornberger et al., 2003). Another is that work content and respectful leadership are more developed in small firms (Eakin et al., 2000). In one recent review about sickness absence in working life, it was stated that knowledge of what factors cause sickness absence is sparse. This review found only one approved study (out of the 600 included in the review), which dealt with relationships between sickness absence and the size of the workplace (SBU, 2003). The concept that the development of a good working environment and work organization is related to increased workplace health and organizational performance is not new. In Swedish governmental investigations, such as the Working Environment Commission report (SOU 1990:49), three main reasons for increased working environment efforts are discussed. First, a human reason that it is unethical to expose individuals to risks that they cannot control. Second, from a business economy perspective - a company with low staff turnover and high employee attendance rates is assumed to have lower costs than those where there is high staff turnover due to work environment problems and high sickness absence rates. Third, a national economic perspective - to reduce costs at a national level. In a Swedish Productivity report (SOU 1991:82), competence development and renewal of work organization development are seen as key components for productivity improvements. The links between working environment development and organizational performance are discussed in several recent governmental investigations; for example, in the government’s plan for renewal of the working environment (Ds 2001:28) and in a report about the development of sickness absence in working life (SOU 2000:121). One indicator that public authorities perceive that small companies have specific needs is that several of these reports have proposed specific measures for working environment development and workplace health in small companies. However, although a working group sponsored by the Commission analyzed measures in small enterprises and made proposals concerning occupational health and safety management issues (Johansson & Johansson, 1990) and large project efforts as the Swedish Working Life Fund (Gustavsen et al., 1995), only a few of all proposals made in the mentioned investigations for improvement of working environment, work organization and workplace health have been used (Olsson, 2003). Similarly, Norwegian, programmes for organizational development and innovation such as ‘Enterprise Development 2000’ (Oscarsson, 1999) and ‘Value Creation 2010’ (Gustavsen, 2002) have mainly focused on medium-sized and large enterprises. 4

Going from the Swedish and Norwegian to a European perspective, there is a European Commission document for a new working environment strategy for the years 2002-2006 (KOM 118/2002). It points out that a safe and secure work environment and a good work organization are factors that have a positive effect upon economy; both generally and for an enterprise. Another example of a European initiative is from the European Network for Workplace Health Promotion. A Network sponsored report about small and medium-sized enterprises (SMEs), associated health and well-being in the workplace with building economic efficiencies in successful businesses (Breucker, 2001). The report also found that the majority of small enterprises paid little attention to questions of occupational health and workplace health promotion. The European Agency for Safety and Health at Work (2004) has proposed specific strategies and measures about promoting health and safety in small and medium sized enterprises, especially for very small businesses. During the last decades many Scandinavian researchers have been focusing on work organization, work environment and change projects in working life (see e.g. Gardell, 1982; Gustavsen, 1995, 2002; Huzzard, 2003; Johansson, 1998; Thorsrud & Emery, 1969). In this research, a major approach has been relationships between working environment and work organization improvements, and organizational health and performance outcomes. Also researchers outside Scandinavia, as Karasek (1998), established that major workplace reorganization could rarely be undertaken with health concerns alone as the goal as productivity is the primary goal of a company’s workplace arrangements. Other researchers with a similar approach are focusing on the concept “healthy work organizations” with the view that organizational factors and working conditions affect both employee health and well-being, and performance, quality and productivity (Murphy & Cooper, 2000; Saintfort et al., 2001; Wilson et al., 2004). Murphy & Cooper (2000) argue for that for a workable healthy work organization model more empirical work is needed to test existing models and to develop new ones. One question that they raise is whether there is a generic healthy work organization model, which is applicable to all workplaces, or if some workplaces require specialized models. They also propose more research into identifying those job and organizational factors which predict both health and performance outcomes. In different research areas, it accepted that knowledge about the effects of workplace interventions is limited and that any new research should use a longitudinal design when working with organizational-level interventions (e.g. Kompier et al., 2000; Källestål et al., 2004; Westgaard & Winkel, 1997). One frequently mentioned topic concerning small firms is that it is difficult for them to carry on development processes due to limited economic resources and lack of knowledge about such processes (Axelsson, 2002; Bostedt et al., 1995; Eakin et al., 2000; Frick et al., 2000; Johansson, 1998). As a consequence, this leads to an 5

interest in research concerning which specific measures are appropriate for different types of small firms. This call for work in this area supports the need for the studies in this thesis about organizational aspects and their influence on the design of workplace change processes. A summary comment is that small business research is highly fragmented and that different research areas have focused on narrow objectives. A reluctance to adopt a holistic perspective on healthy work organizational factors and relationships with different outcomes in small firms is also notable. To achieve more knowledge about how different factors contribute to health and performance in small enterprises then more integrated approaches are needed.

Purpose and research questions The background presented above leads to this description of the general purpose of the thesis; namely to contribute to knowledge about: • relations between health and performance outcomes; • relations between organizational factors such as aspects of leadership, learning, psychosocial working environment, and health and performance outcomes; and • relations between these factors and outcomes, and different change strategies in small enterprises. Additional motivators driving the research projects covered by this thesis were to make this knowledge relevant for developing workplace change processes, and to encourage further research about organizational factors and outcomes related to health and performance in small enterprises. The following research questions are addressed: 1. Are there any relationships between outcomes related to health, sickness absence and performance in small enterprises? 2. What relationships can be found between the outcome indicators of health and performance and indicators of assumed organizational determinants? 3. Is there a bundling tendency for several organizational aspects to co-vary in small firms and, if so, is it possible to find groupings of co-varying indicators and groups of companies? 4. What needs for change could be found concerning work organization and working environment aspects in studied small enterprises? 5. Are there differences concerning changes in organizational determinants and health and performance outcomes when using different types of change strategies? 6. What conclusions for the design of workplace change processes can be drawn from the answers to the above research questions?

6

Definitions and delimitations In the following chapter about theoretical framework the main concepts used in the thesis are discussed and defined. Here, some initial definitions and delimitations are introduced. In this thesis, small enterprises are defined as those with less than 50 employees (including micro enterprises with less than 10 employees) but exclude sole proprietorships. The reason for excluding the latter group is the thesis focus on organizational aspects. The two outcome concepts – health and performance – are looked upon as aggregated concepts at an organizational level, which is in line with models of healthy work organizations. These concepts concern the use of resources and can therefore be seen as related to effectiveness of the firm. The organizational determinants are related to aspects of leadership, psychosocial work environment and learning – these to some extent overlap with those in models of healthy work organizations. Change is simply defined as a wish to make things different, usually better, although the notion of change is extremely broad. In this thesis, change strategy refers to different change strategy types in workplace interventions, including both content and processes. Delimitations of the presented studies are that they are limited to small enterprises and organizations in Sweden and Norway, and the data sets are not representative samples so no statistical inferences as they pertain to broader populations are intended. However, the distribution of firms is not so divergent from the total distribution for small firms in Sweden and Norway.

7

8

Theoretical framework This thesis focuses on relationships between health and performance outcomes, organizational determinants, and change processes in small enterprises. This has been done with a multidisciplinary approach and the following sections do not intend to cover all the studied areas. They do seek to give an overview of key issues regarding small enterprise definitions and characteristics, the two-outcome concepts of health and performance, healthy work organizations, change strategy, and workplace interventions. The overview focuses on organizational factors such as aspects of leadership, learning and psychosocial work environments. These factors are then related to models of healthy work organizations, and health and performance outcomes. Research findings are partly about enterprises and organizations on a general level and partly about small enterprises. Emphasized in the findings are the practical applications of the research the thesis describes – strategies small businesses can use to build a healthier and more effective business.

Small enterprise definitions and characteristics When seeking to define a small enterprise it becomes evident that the concept of “smallness” varies and that there is no single suitable definition. In a way, this is remarkable due to how frequent the concept is used by politicians, practitioners and researchers. For example, Wiklund (1998) described two different ways of defining enterprises. The first approach labelled ‘theoretical’. Here, the criteria for defining a small enterprise would include market share, personalised management, vulnerability to environmental conditions and the manager’s noneconomic objectives. This type of definition is theoretical in the sense that small firms are presumed to be fundamentally different than large firms in these dimensions. The second definition labelled ‘quantitative’. Here, size itself is the criterion for smallness, and quantitative size data regarding sales, employees or equity are normally what is used when classifying firms as large or small. Some researchers try to define small enterprises with a more theoretical and qualitative approach (e.g. Bridge et al., 1998; Julien, 1998) although there also are quantitative elements in these definitions. Bridge et al. (1998) stated that a small business possesses at least two of the following characteristics; (1) independent management, (2) capital and ownership are provided by an individual or a small group, (3) the company mainly operates locally in the same community where the owner and the employees live, and (4) size of the business is small compared with the biggest units within its industry. Julien (1998) proposes the following characteristics to define the concept of a small enterprise; 9

(1) small size based on the number of employees (and/or turnover), (2) management centralization, (3) low level of specialization, (4) an intuitive or informal strategy, (5) an uncomplicated or unorganized internal information system, and (6) simple external information system. The European Commission definition is based on three types of quantitative criterion related to the number of employees and the size of the business in financial terms (turnover and balance sheet) as shown in Table 1. The definition also consists of a qualitative element about its independence with specific directions (see EU, 2003). Table 1. EU Commission’s quantitative criteria for small and medium-sized enterprises. Criterions

Micro-enterprise

Small enterprise

Medium-sized enterprise

Number of employees

< 10

< 50

< 250

Yearly turnover

< 2 million EUR

< 10 million EUR

< 50 million EUR

Yearly balance sheet

< 2 million EUR

< 10 million EUR

< 43 million EUR

The definition indirectly implies that a large enterprise is defined as (1) an enterprise with 250 or more employed or (2) an enterprise with either a yearly turnover of 50 milj EUR or more or a yearly balance sheet of 43 million EUR or more. The workforce size criteria divides the enterprises into micro-enterprises (0-9 employees, small (less than 50 employees) and medium-sized (less than 250 employees). This means that the EU describes small enterprises as having up to 49 employees and that micro-enterprises are a sub-group in the group of small enterprises. The very smallest enterprise size, sole proprietorships, is not separated out in the EU definition. According to the writings in the criterion document, only the first criterion about the number of employees sometimes could be used to facilitate practical routines. In the present thesis, only the workforce criterion is used, and the main focus is on micro and small enterprises (excluding sole proprietorships). The majority of the enterprises studied, with two exceptions, have the same characteristics as those described by Bridge et al. (1998) and Julien (1998). The two exceptions are: ‘low level of specialization’ and ‘mainly operates in the same community’. Some of the studied enterprises are specialized and some operate outside of their local areas. In this text, the term ‘small enterprise’ refers to both micro and small enterprises. Selection of this thesis definition is consistent with Storey’s perspective on building definitions:

10

“[…] in practice, researchers have to tailor their definitions of a small firm according to the particular groups of small firms which are the focus of their interest” (Storey, 1994, p. 16). As a group, small businesses can be looked upon as heterogeneous (Bridge, 1998; Julien, 1998; Storey, 1994; Ylinenpää et al., 2006). For instance, they operate in many lines of business, they differ in their stages of ambition and development, and in their types of leadership. According to a recent Swedish anthology (Ylinenpää et al., 2006), there is a large variation and heterogeneity concerning leadership and management in small enterprises. Although they are heterogeneous, they have distinctive features that differentiate them from the leadership in large enterprises. Storey (1994) states that many small firms occupy ‘niches’ and provide a highly specialized service or product; possibly in a geographically isolated area. This means that that they are vulnerable to shifts in market trends. On the positive side, small business managers can often have closer contact with their customers; making them better at immediately satisfying customer needs (see e.g. Julien, 1998). Another characteristic of small enterprises is that the manager and the owner of the organization is often the same person. This fact means that the leader perceives a strong solidarity with the organization, and that the manager’s and the business’s goals often correspond (Bridge et al., 1998; Hård af Segerstad, 2000). These researchers also noted that the relationship could create a person-oriented culture in the company. That the top-down approach can be attractive in a stable environment if a manager is able to motivate the employees. Leaders of small firms are believed to have difficult working conditions such as high time pressures and heavy workloads (Persson, 1991). They must simultaneously lead people and carry out their trade. Task delegation as a way to shift time demands, an option for leaders of larger enterprises, is not possible (Holmquist, 2006). Some researchers point at the polarized picture of leadership in small firms (e.g. Wilkinson, 1999). According to one view there are harmonious relations between employees and the leaders plus conflicts are unusual and work is flexible: Another view maintains that leadership is authoritarian and working conditions are unsatisfactory with high injury risk and poor physical working environments. When looking at structure, work organization and psychosocial working conditions, small enterprises often lack formal structures plus the vertical and horizontal division of labour is minimal when compared to large organizations (Vossen, 1998). According to Julien (1998), small firms are less hierarchical and more flexible than large organizations due to greater opportunity to vary tasks and not having to perform constant, repetitive work. Abrahamsson (2006), in studies of Swedish small enterprises, found that work organization is “diffuse” and without a conscious and strategic structure. Hierarchical levels are almost non-existent in the smallest companies. However, according to this researcher, 11

although they do not have a formal work organization, there are boundaries and structures concerning e.g. hierarchy, specialization and working groups. In studies of small industrial companies in Sweden and United States, a finding was that they develop and sometimes grow although they have not implemented new management and organizational concepts (Rantakyrö, 2004). That small companies operate in different sectors of business causes there to be differences in learning in response to the differing conditions. Ylinenpää (1999) has in studies of small manufacturing companies found competence development mainly related to the daily work, and that methods as study visits, delegation of work tasks and work rotation are common. In these studies, the companies were divided into “high-performing firms” where the employees are well educated and the company utilise a broader repertoire of methods for developing the internal competence, and “low-performing firms”, which have a “narrow range of competences”. Also, noted was that typical small enterprises do not have the human resource development expertise and other resources for learning which larger organizations enjoy (Hill & Stewart, 2000). Additionally noted was that there are differences in HR practices within the small enterprise category (Kotey & Sheridan, 2004). Concerning working environment and health issues in small enterprises, much research has reported higher injury rates, lower sickness absence figures and important physical working environment problems when compared to larger enterprises (e.g. Bornberger et al., 2003; Walters, 2001). According to Walters (2001), small firms have accidents and exposures that account for an important proportion of the EU’s yearly average of 6,000 fatalities and 10 million occupational injuries and diseases. An observation in a Swedish literature overview (Bornberger-Dankvardt et al., 2005) was that although the sickness absence figures are about half of the figures for larger enterprises, increases from 2003 to 2004 can be noted also among small companies. Some researchers have questioned the simplified view of small enterprises and believe instead that conditions are complex and contradictory (Ram, 1991). For example, familiar circumstances can make open dialogue much more possible. They can also lead to more conflicts between leaders and employees because of constant and close contact. According to Eakin et al. (2000), when comparing their conditions to larger companies, employees in small companies feel they are more able to influence their work, that the content of their work is better and that their leadership is more developed than in larger companies. A contrary view is that because of social control in small enterprises employees go to work despite their ill health (Bornberger-Dankvart et al., 2003). The characteristics mentioned above are open to being criticized as being incomplete, a possibility since heterogeneity characterises small enterprises as a group – in effect, there is always more that can be included when considering small business issues. Small businesses in different development stages that are 12

operating in different sectors and regions have widely different characteristics. The important role small enterprises have for working life and economy, plus because of the somewhat complex character of the relationships described above, it seems constructive to contribute to the study of small enterprises and their characteristics.

Health and Performance – two outcome concepts Health In the studies in the present thesis, health has been measured in several ways which are covered in some of the definition areas that follow. Though organizations are the focus in the studies, aggregated individual data concerning yearly sickness absence and different indicators of inverted psychosocial and musculoskeletal health and self-rated health have been used. This approach of looking at health as an organizational outcome is consistent with models about healthy work organizations. Cox & Thomson (2000) use e.g. a concept of organizational healthiness based on an analogy with individual health in their definition. Working life research has to a great extent been focusing on ill health and its workplace determinants. Efforts to define and operationalise health and to generate knowledge about health promotion conditions at workplaces are less common; possibly because health is a difficult and complex concept. The difference between the widely used negative definition of health as the absence of disease and the World Health Organization’s (WHO, 1948) less specific definition of health as “a state of complete physical, mental and social wellbeing, and not merely the absence of disease or infirmity” have long been recognized (Bowling, 2005). The WHO definition of health has a holistic approach which notes that there is a variation in how health is perceived among individuals. One major question is whether health should be seen as the opposite of disease/ill health or as an independent concept including aspects such as wellbeing and individual resources. In the literature, four main perspectives of health are discussed (Brülde & Tengland, 2003): 1) health as absence of disease, 2) health as well-being, 3) health as balance and 4) health as ability. The first perspective is a biomedical approach, while the other three perspectives have a holistic and a humanistic approach. The view in a combined holistic and humanistic approach is that it is possible to have a disease and still experience health since it is the individual who chooses what to experience. Also a consideration is that it is possible to experience ill health even if a disease cannot be found.

13

According to Wikman et al. (2005), the concept trilogy of “illness”, “disease” and “sickness” has been used to capture different aspects of ill health. Illness is seen as a rather wide concept which is defined as the ill health the person identifies with themselves: it can often be based on self-reported mental or physical symptoms. Disease is defined as a condition that is diagnosed by a medical expert. Sickness concerns the social role a person with a physical or mental disease takes or is given in society. Data concerning sickness absence from work is often used to measure social consequences of a person’s ill health. These several concepts for health and ill health which have resulted in common ways used in scientific literature and in public debate to conceptualise health are (Westerlund, 2005): lack of manifest disease, subjective well-being and health, resources, lack of sickness absence, and health related autonomy such as lack of disability in relation to activities of daily living. According to Hensing et al. (1998), sickness absence data is increasingly being used as an integrated measure of health in the working population even though it’s multi-factorial and influenced not only by the health status of individuals but also by work environments, attitudes and commitment to work. A common way to measure health is also to ask people about their health, either in questionnaires or through interviews (Westerlund, 2005). In a review of how to measure health, Bowling (2005) argued that traditional negatively oriented indices must be complemented with more positively oriented indices, e.g. individuals’ perception of general health or quality of working life. Some of these latter questions and indices have been shown to have good test-retest reliability, and to correlate strongly with other direct or indirect measures of health. Though the relations between health and ill health related measures are unclear, several researchers argue for the importance of combining different measures to cover multiple health aspects (Bowling, 2005; Westerlund, 2005). This line of reasoning has been used in the studies reported on in this thesis. Both aggregated self-assessed health and ill health related measures were used together with aggregated sickness absence figures for measuring organizational outcomes. Performance Although organizational performance is one of the most commonly used outcome variables in the organizational research of today, many different measures of performance are used. Many researchers have discussed overall performance of small enterprises, but little consensus on appropriate small firm performance measures has emerged (Murphy et al., 1996; Wiklund, 1998; Zinger et al., 2001). Greenley (1994) found that many studies used common financial performance measures, composite calculations and subjective measures like perceived performance. In a review of 70 published scientific contributions, Wiklund 14

(1998) found that many studies used growth as the sole performance indicator and that it is difficult to find any single variable that is represented in more than a small fraction of all studies. In an overview of entrepreneurship performance measurement, 60% of the studies used only one or two dimensions of performance, generally without justification for selection (Murphy et al., 1996). Due to the absence of any consensus as described above, several researchers are discussing the need for both hard and soft indicators, and both managementand employee-oriented measures. Murphy et al. (1996) argued that studies should include multiple dimensions of performance where financial measures are used along with non-financial ones and that it is necessary to explicitly describe performance measures used. Rhodes & Butler (2004) state that subjective evaluations of entrepreneurial success do not differ a great deal from objective indicators; they also remove the problem of response bias due to participant failure to answer financial questions. Kotey & Meredith (1997) point out that performance measured by respondents’ ratings on a set of performance measures has been found to have high reliability and validity rates and to reflect a firm’s objective performance. In accordance with the above discussion about the need to use both objective and subjective performance measures, different measures of performance are used in this thesis. These are: profit per capita in each company, leader and employee judgement of ‘perceived lack of efficiency’, ‘workplace adaptability’ and ‘customer satisfaction’, and leader judgement of changes in profit over the last two years, number of customers, number of products, lead-time for services/products and complaints from customers.

Healthy work organizations – models and definitions Healthy work organizations (the terms healthy companies and organizational health have also been used) represent the convergence of research and thinking in several related workplace research areas. According to Wilson et al. (2004) researchers in the human resources/organizational development area have focused on identifying the traits of healthy companies and exploring the characteristics of high-performance work systems. Researchers in the job stress area have long been interested in what job and organizational attributes characterize healthy environments, and attention is shifting toward understanding organizational factors when analyzing the causes of stress within organizations. Interestingly, Wilson et al. (2004) observed that a similar shift can be seen among quality movement researchers, occupational and safety researchers, and health promotion researchers. Although the view that organizational factors and working conditions affect (1) employee health and well-being and (2) performance, quality and productivity is attractive, there is a lack of knowledge and understanding about how both 15

health and performance related outcomes can be improved (Murphy & Cooper, 2000; Sainfort et al., 2001; Wilson et al., 2004). In fact, research on organizational outcomes, such as organizational effectiveness or profits, and employee health or well-being, has proceeded almost independently of one another (Sainfort et al., 2001). That is, studies have tended to show that working conditions and/or organizational characteristics are predictive of either employee or organizational outcomes – only parts of what can go on within enterprises were investigated. Particularly, there is a lack of knowledge about these research directions applied to small enterprises (Lindström et al., 2000). The National Institute for Occupational Safety and Health (NIOSH) proposed the Healthy Work Organization Model as shown below (Sauter et al., 1996). Management Practices Performance Outcomes Organizational Culture/Climate Health and Satisfaction Outcomes Organizational Values Figure 1. Healthy Work Organization Model suggested by NIOSH.

This model suggests that work organization factors and organizational characteristics can influence both organizational outcomes such as effectiveness and performance; as well as employee outcomes such as health and satisfaction. According to Wilson et al. (2004) the term ‘healthy work organization’ is a logical extension of ‘work organization’ and presupposes that it should be possible to distinguish healthy from unhealthy work systems. These researchers define a healthy work organization as one characterized by intentional, systematic, and collaborative efforts to maximize employee well-being and productivity by providing well-designed and meaningful jobs, a supportive social-organizational environment, and accessible and equitable opportunities for career and working life enhancements. In their investigation of models of healthy work organizations Murphy & Cooper (2000) identified six models with a varying level of development. However, there appears to be a convergence of opinion and evidence on at least some important factors in the models. For example, most models emphasized meaningful work which utilizes worker skills, worker autonomy/control, job 16

security, rewards for performance, worker involvement/participation, and safe and healthy physical work environments as necessary components of a healthy work organization. Other important healthy work factors noted in recent research are teamwork and commitment to core values. The organizational health perspective differs from many traditional approaches to reducing occupational stress due to its focus on both employee well-being and organizational performance (Griffin et al., 2000). This perspective points to the reality that having satisfied employees is of little value to organizations unless the employees are performing efficiently and productively. An efficient organization is also of little or no value if it is achieved at the expense of employee well-being. The concept of organizational healthiness is based on an analogy with individual health and is about the nature and viability of organizations as systems (Cox & Thomson, 2000). These authors, suggest a model where staff development, goodness of task and problem solving can affect employee health - through the design and management of work systems/procedures, stress levels and organizational impact on work behavior. The health of employees in unhealthy organizations is assumed to be poor with employees not performing well and showing reduced commitment to work systems. The concepts of healthy work organizations have similarities with Scandinavian research about concepts of quality of working life (see e.g. Gustavsen, 2002; Huzzard, 2003; Johansson, 1999) and sustainable work systems (Docherty et al., 2002), which are related to the socio-technical view of organizational design. According to Huzzard (2003), there appears to be a broad consensus among researchers that it involves a focus on work redesign and all aspects of working life that might conceivably be relevant to worker satisfaction and motivation. This research argues for jobs with a meaningful content that are not alienating, and that allows for greater job satisfaction and influence upon workplace decisions. Such jobs could generate higher-level organizational performance, less sickness absence and reduced employee turnover. Such sustainable work systems present a vision for the future competitive organizations in which human resources are regenerated and allowed to grow (Docherty et al., 2002). In contrast, intensive work systems consume resources generated in the social system of the work environment and are characterized by e.g. diminished work motivation, stress and long-term sickness absence. However, it is not commonly accepted that employee well being and firm performance are positively related, many organizations continue to be managed as there is a conflict between these outcomes (Huzzard, 2003). Also among other Swedish researchers there has been a growing interest in understanding healthy work organizations. Menckel & Österblom (2002, pp. 6061) refer to the description by Nerell & Sandberg (1994) as the basis for the definition of a healthy enterprise: “A culture where the workers experience security and freedom, where they can express themselves without the risk of 17

repression and where all ideas and suggestions are considerable valuable. In this culture the individual members are prestige-less. It is significant that the group can solve the problems it has before it or has been burdened with. It is an open system where decisions are taken and actions are carried out when and where a problem or conflict arises”. Some other Swedish researchers have introduced the concept of ‘development work’ as an important work innovation that could lead to improvements in working conditions while at the same time strengthening quality and performance (Eklund, 2000). Although few researchers have tried to test their conceptualizations of healthy work organizations (Wilson et al., 2004), there are studies; particularly of larger enterprises. Sauter et al. (1996) found in a study of 5000 employees of a large manufacturer that the parts related to management practices, organizational climate and organizational values were significantly related to both perceived organizational effectiveness and perceived stress among the employees. In ‘Healthy Work’ by Karasek & Theorell (1990), four extensive literature reviews show associations between quality of working life factors, job redesign, productivity and health. According to a review of Swedish empirical studies, there is “considerable support for the existence of a positive relationship between quality of working life aspects and various performance measures that we might reasonably expect to enhance competitiveness” (Huzzard, 2003, p.17). In an interesting Finnish study of 217 small and medium-sized enterprises (Lindström et al., 2000) one finding was that supervisory support, positive social climate and job satisfaction significantly correlated with company effectiveness (leader perceived productivity and profitability). Another finding was that factors such as supervisory support, positive social climate, appreciation of one’s work, continuous improvement practices and physical workload showed the strongest correlations with health measures (sickness absenteeism and employee-judged exhaustion symptoms). In the following section, some research results relevant to models of healthy work organizations are described. The organizational determinants used in the thesis - aspects of leadership, psychosocial work environment and learning, and their relationships to outcomes of health and performance are touched upon; with a particular emphasis on small enterprises. Leadership aspects Despite the extensive research on leadership there is a lack of knowledge about how leadership influences organizational performance and employee health; particularly for small enterprises. The limited amount of leadership literature discussing the impact of how leadership affects the health of subordinates is summarized in a Swedish review of leadership research (Nyberg et al., 2005) and in a review of leadership and human resources from Australia (Rahman, 2002). 18

Both conclude that more research on the role of top management in health, safety and well-being among employees is strongly needed. When writing about leadership and efficiency, Andersen (1994) summarizes by saying that leadership research is inconsistent, and that it is absurd to believe that a small number of variables could have causal relations with such different outcomes as productivity, profitability, employee motivation and work satisfaction. This is in line with findings from Ogbonna & Harris (2000) with empirical evidence of a link between leadership style and organizational performance which is mediated by the nature and form of organizational culture. Northouse (2001, p. 3) defines leadership as “a process whereby an individual influences a group of individuals to achieve a common goal”. This definition means that leadership is a process with transactional events that take place between the leader and the employees. Much leadership research has focused on the two-factor model of leadership behaviour with the dimensions of employee and production orientation (Yukl, 2002). Arvonen has, from studies of different populations, found evidence for a third dimension - change orientation. Which includes behaviours such as visionary qualities, creativity, action implementation, and risk taking. One conclusion from this research is that managers in general need to maintain a balance between change, production and employee orientation to be successful. Research studies indicate that relation oriented managers often produce good satisfaction in groups, especially when this is combined with a moderate or high production orientation leadership style (Yukl, 2002). There are studies of how leadership moderates or influences health and organizational performance. In his review of more than 130 field studies of organizational change, Pfeffer (1998) emphasized that leadership with a focus on developing human resource practices is a necessary component for high organizational performance. This corresponds with the view of Cascio (1995), to the effect that leaders today need to integrate a human resource focus with a financial focus - for instance, by developing teamwork, creating conditions where workers have opportunity to make decisions about their jobs and have access to extensive training. That relations-oriented leaders are associated with a relatively good level of satisfaction, less absenteeism and less workforce turnover among employees, than are more structure-oriented and controlling leaders is shown by Bass & Avolio (1994) and Evans & Johnson (1990). McGee et al. (1987) found that individuals who reported both low stress levels and a high degree of job satisfaction also described their leaders as more relation- and structure-oriented whereas individuals who reported both high stress levels and low job satisfaction were less likely to similarly describe their leaders. This is partly in line with results presented by Arvonen (2002) in a study of a Swedish forest company (781 employees answered a questionnaire). Found was that both the relation-oriented and the structure-oriented leaders reduced mental fatigue and psychosomatic load 19

among their co-workers. However, the relationship between change-orientation and mental fatigue had a different orientation. In another Swedish intervention study (Theorell et al., 2001), there was a significant decrease in serum cortisol levels and a significantly improved authority over decisions among employees with managers who had participated in an educational program when compared to a control group. Other Swedish researchers (Aronsson & Lindh, 2004) have shown relations between factors such as leadership support, resources for doing a good job, employee’s experiences of good quality at work and their long-term health (an index of low sickness absence/low sickness presence during a two-year period). These results are in line with those found in a Swedish case study of a small manufacturing company (Lagrosen, 2004). The company has a positive image and has received a major award for work environment issues. A finding in this study was the rather strong relationships between how the employees perceived the organizational values of ‘leadership commitment’ (favourable) and how they perceived their health (favourable). A Swedish study in the County of Jämtland (Larsson et al., 2005) reported that leaders influenced by McGregor’s X hypotheses (‘individuals are passive and lazy’) had lower results concerning employee judged leadership and quality aspects (and to some extent, lower results concerning employee judged health outcomes) when compared to leaders more influenced by McGregor’s Y hypotheses (‘individuals have a will to contribute and perform’). The leadership in small enterprises is often seen as of particular importance, due to company size and leader identification with their company (Beckérus & Roos, 1985; Johannison & Lindmark, 1996; Walters, 2001). Leaders of small firms have been described as facing difficult working conditions with high time pressures and major work demands (Persson, 1991). In a field study of stress among small business owners, Mack (2000) found positive relationships between the small business owner’s locus of control (a characteristic of the “person”), experienced control and the amount of stress. In one of few studies of microenterprises, Matlay (1999) found that leadership in English micro-firms was more informal than in larger enterprises. Work-related problems seemed to a great degree to be solved by informal discussions, and this leadership style contributed to a positive working climate. Westerberg (2001) found in a study of small enterprises that leader tolerance for ambiguity and self-efficacy were of importance for company development. The conclusions from the study were that high values on the two dimensions increased the capability to handle turbulent situations; and fostered growth and profitability. Smallbone et al. (1995) examined the development of 306 small and medium-sized enterprises and found that one of the most important factors for high growth was the commitment of the leader of the company to achieving growth. In a study of 372 small firms, Chaganti et al. (2002) found that the fastest

20

growing firms were strong in both the consideration and initiation styles of leadership. According to Swedish researchers (Ylinenpää et al., 2006), important leadership characteristics in small enterprises are the holistic perspective, a relationship between ownership and management, high degree of internal security and external insecurity, and large possibilities to use flexibility as a competitive device. Psychosocial work environment aspects Over the last decades much research on the relationship between the large organization psychosocial work environments and health and well-being has shown that factors related to work content, work characteristics, work organization and social relations are of importance for individual health in a wide sense (see e.g. Gardell, 1982; Johnson & Johansson, 1991; Karasek & Theorell, 1990). There is considerably less research in this area concerning small enterprises (Walters, 2001). Eakin & MacEachen (1998) have also reported that occurrence of illness and injury related to social and organizational factors in small enterprises is little known and explored. There is also a lack of knowledge about how psychosocial factors are related to organizational performance, particularly in small firms (Lindström et al., 2000). The work stress literature points out job complexity, autonomy, role ambiguity and work load as factors associated with well-being at work. Perhaps the most influential model within the research on psychosocial work environment and health is the ‘Job demand-control model’ developed by Robert Karasek (Karasek, 1979; Karasek & Theorell, 1990). The model focuses on two dimensions of the work environment: job demands and job control. Job demands refer to the workload and have been operationalized in terms of time pressure and role conflict. Job control, which is sometimes called decision latitude, refers to the person’s ability to control his or her work activities. The ‘strain hypothesis’ in the model states that the most adverse reactions of psychological strain and physical illness are experienced in a ‘high-strain’ job - that is, a ‘high demands-low control’ job. The model also shows that high demands in combination with high control leads to increased learning, motivation and development of skills. Another hypothesis is that personal control can buffer the potentially negative effects of high demands on health and well-being. Although the model often has been used with a focus on individuals, Karasek & Theorell (1990) mean that the model is not limited to the individual level; it can also be applied at the organizational level. Work demands would apply to the organization’s performance level and freedom of action to the organization’s decision-making processes. Despite its wide usage, the model has been criticised. One critique has been that the model is too simplistic. That it does not include factors other than 21

psychological demands and decision latitude. This critique has led to the addition of a social support dimension into the model in which is called the ‘Job demandcontrol-support model’ (Johnson & Hall, 1988). In his meta-analysis of studies on autonomy and participation, Spector (1986) found that high levels of perceived control were associated with high levels of job satisfaction, commitment, involvement, performance and motivation as well as low levels of stress-related symptoms and turnover. In an overview of studies, Jeding & Theorell (1999) referred to the importance of social support from both leaders and co-workers in relationship to well-being at work. They suggested that social support given by the leaders might have a larger impact than support given by co-workers. Studies on relations between individual well-being and various group processes have usually found them positive. Satisfaction with group work and team spirit has also been regarded as indicative of the effectiveness of a group (Sonnentag, 1996). In regard to the psychosocial conditions, some studies have found that small organizations, as compared to large organizations, tend to have employees with higher job satisfaction (Kovach 1978; Zemke 1988). Bornberger et al. (2003) maintain that positive features in Swedish small companies are good team spirit, variation in work tasks, work satisfaction due to high control, a holistic perception of work, short roads to decisions, personal commitment to work and low sickness absenteeism. According to these authors, negative psychosocial factors in small companies are high social control, time pressure, stress, inadequate physical work environments, increased risk of injury, sickness absence, and a lack of personnel and financial resources. In a study of 343 Finnish workplaces, Lindström et al. (2000) found that employees from enterprises with less than ten employees, compared with larger enterprises, had better job control, supervisory support, work climate and appreciation of one’s own work. These authors concluded that small company size, as defined by the number of employees, was a central determinant of good organizational practices and climate. A result was also that there were clear differences in many job and organizational characteristics between branches of industry. This is partly in line with the view of Walters (2001) that comparisons within the small company sector and between small and large enterprises should be made with caution. Learning aspects Most of the existing literature about learning and human resource management (HRM) is based on large firms. Hill & Stewart (2000) also observed that there is much evidence that small enterprises do not have the human resource development expertise and general resources which larger organizations more frequently enjoy. They meant that each small organization is unique and that their 22

HRM activities are essentially informal, reactive and short-term in outlook. According to these researchers it is important to understand why and how small organizations generate their own models of human resource practices and to clarify the links between such practices and performance. The resource-based approach can be seen as related to the concepts of HRM, and a brief introduction of the two follows. The concept of a resource-based approach is founded on the belief expressed by Hamel & Prahalad (1989) that competitive advantage is obtained if a firm can obtain and develop human resources which enable it to learn faster and apply its learning more effectively than its rivals. Barney (1991) contends that sustained competitive advantage stems from acquisition and effective use of bundles of distinctive resources that competitors cannot easily imitate. Unique talents among employees, including superior performance, productivity, flexibility, innovation, and the ability to deliver high levels of personal customer service are ways in which people provide a critical ingredient in developing an organization’s competitive position (Armstrong, 2001). The aim of a resource-based approach is to improve resource capability – achieving strategic fit between resources and opportunities and obtaining added value from the effective deployment of resources. This approach experienced a revival during the 1990´s in the strategic management literature and also in small business research. For example, it appeared in some Swedish theses about small firm growth and performance (Wiklund, 1998), competence development in small firms (Ylinenpää, 1997) and small-firm network organizations (Wallenklint, 2001). The theses just mentioned also stated that the resource-based perspective must be combined with perspectives that concern environmental influences. Simply stating that hard-to-imitate resources are necessary for competitive advantage and that human resources are necessary and especially valuable for competitive advantage does not go far (Boxall, 1996). There exists an increasing body of theoretical and empirical research, which has led to a general consensus that the method used by a firm to manage its workforce can have positive impact on firm performance (e.g. Huselid, 1995; MacDuffie, 1995; Way, 2002). Pfeffer (1994) pointed out that factors like employee participation, job redesign, inclusion of team-based production systems, extensive employee training and performance-contingent incentive compensation are widely seen as the ways to improve the performance of organizations. However, much of this research has excluded small enterprises and there are studies where limited or no correlations between HRM and financial performance, productivity and quality were evident (Guest & Hoque, 1995; Storey, 1999). Delaney & Huselid (1996) stated that substantial uncertainty remains as to how HRM practices affect organizational outcomes, whether some practices have stronger effects than others and whether complements or synergies of such practices can further enhance organizational performance. With respect to learning and competence development inside both large and small organizations, 23

there is little agreement about the nature of learning and relationships between learning and performance in SMEs (Beckar & Gerhart, 1996; Ylinenpää, 1999). For instance, Baldwin et al. (1994) did not find relationships between training and economic results as in the above-mentioned findings of Huselid (1995). Although the research is limited concerning learning aspects in small firms, there are examples of relevant studies. Storey (1999) tried in a study of 314 English small and medium-sized companies to verify the results that Huselid (1995) found in larger enterprises. The results were contradictory, with some relationships between human resource/learning variables and financial performance outcomes. However, there were also other relationships between these variables and outcomes that were not significant. A conclusion was that the studied relationships were weaker then expected. Way (2002) studied high performance work systems (HPWS) in the US small business sector. HPWS is a set of distinct but interrelated HRM practises that are used together to select, develop, retain and motivate a workforce. Way found that HPWS practices were associated with a lower workforce turnover but not with labor productivity. In a Swedish study about competence development in small enterprises (Nilsson, 1996), the results found that education related to changes in work organization and/or job content gave better results than education not related to work organization. Concerning conditions within micro firms, some studies have found that increased company size is associated with more formal training, a move towards hierarchical structures and more administrative processes (de Kok & Uhlaner, 2001; Kotey & Sheridan, 2004). Kotey & Sheridan argue that such processes might be counter-productive for micro firms, and reduce a firm’s flexibility. This is partly in line with a Swedish study about working environment activities related to learning in micro enterprises. The results indicated that the employers did not give priority to more formalised strategies (Axelsson, 2002).

Change strategy and workplace interventions Change strategy According to Aronsson et al. (1995), two factors have to be present to achieve success in change projects: (1) employee influence and participation in the change process, and (2) top management’s active support. Results from research areas such as organizational change (Gustavsen et al., 1995), leadership (Arvonen, 2002), quality development (Axelsson, 2000), stress management (Kompier et al., 2000) and ergonomics (Westgaard & Winkel, 1997) give support to this two-factor idea. Several researchers (e.g. Anttila, 1997: Aronsson, 1995;

24

Håkansson, 1995) point out that change competence is a question of vital importance for coping with change. Researchers put different main emphases on the importance of process and content in workplace change programs. Håkansson (1995) looks at change competence as a relative concept and means that it is about (1) an organization’s ability to handle change processes according to its development level, and (2) employee opportunity to participate in the design of issues as they relate to contents and organization of change processes. Ekman Philips (2002), states that participation, authority, legitimacy and cooperation are central concepts for achieving support around change processes in workplaces. This is consistent with the views of Pasmore (1994) to the effect that change is significantly less likely to succeed when employee participation is restricted to accomplishment of change measures and they are excluded from any planning and creating. Researchers such as Gustavsen (1995) and Ekman Philips (2002) highlight the importance of establishing a ‘development organization’ for working with change processes. They describe this as the creation of arenas for dialogue and constructive discussion. Leader active participation is seen as another important condition for driving change processes with high goal fulfilment (Aronsson, 1995; Gustavsen, 1995). The thesis definition of change strategy follows those of Håkansson and Ingelgård described below. Håkansson (1995) identifies four types of change strategy: the expert project (a project group, considered to be experts, makes suggestions and carries out change), the problem-oriented strategy (still expertorganized, but dealing with several tasks of an urgent nature), the processoriented strategy (broad participation is achieved, and the process is considered more important than the content) and the broad strategy (combine a high process activity with a broad change process, there is a balance between content and process). In her doctoral thesis, she reported that for short-term projects none of the strategies was superior. For projects running over a longer period, a broad strategy, which deals with several extensive problems concurrently while having broad leader/employee participation, gave significantly better results. This is in line with research by Ingelgård (1998), where findings were that: (1) the use of a learning strategy for change (several problem areas are addressed, a large number of employees participate in and have influence on change processes, and the processes are guided by a vision of the future) and (2) high top management involvement (management actively pursuing the change process, management consideration of opinions and different views, having a strong belief about positive outcomes, and have a strong identification with the overall visions of the change process) significantly correlated with economic and quality of working life outcomes. A surprisingly result was that the use of a learning strategy was more strongly related to the outcomes than was top management involvement. This supports the concept that broad employee and leader participation during 25

change processes is essential for there to be success. According to Johansson (1998), a learning-based change strategy has a particular relevance for small enterprises as planning and controlling are the exception in small firms. Too much planning can be an obstacle when change is rapid; formal planning is often resisted. Employee engagement and influence is often improved when using a more informal learning-based strategy. As mentioned in the introductory part of this thesis, different researchers have emphasised that small enterprises have trouble working with change processes due to their limited economic and personnel resources, and their lack of knowledge about such processes. Concerning occupational health and safety change processes, Eakin et al. (2000) argued that an in-depth understanding of the nature of working life in small companies is critical for the development of effective approaches to improved working conditions. According to Walters (2001), there are special problems connected with such approaches in microenterprises as they are entrepreneurial, have minimal organizational structure, have limited resources and must lay-off employees in the event of failure. In contrast to this, Langaa Jensen et al. (2001) observed that leader and worker knowledge in small firms is enough to be able to effectively improve work environments and to cooperate with external resources. Researchers from the workplace health promotion area (Bjørnstad, 2004; Stokols et al., 2002) made similar positive observations about change by pointing out that small firms are employee-dependent, can make quick changes and employee participation can be better than in larger companies. In the study by Håkansson (1995), effects of change programs were studied in a random sample of 336 Swedish workplaces. Some interesting results were that, according to studied effects, there were no clear differences between workplaces of different sizes. A conclusion was that large workplaces do not use personnel resources for working with change processes in larger extent than small and medium-sized workplaces. Other results were that economically successful workplaces were more likely to be effective at implementing work environment improvements. Workplace interventions According to several reviews of workplace intervention research, there is a lack of knowledge about the different effects, barriers and facilitators in workplace interventions, and there is a need for more longitudinal studies with adequate methodological designs (Harden et al., 1999; Kompier et al., 2000; Westgaard & Winkel, 1997). Jeding et al. (1999) concluded in an overview of research studies that there is a lack of knowledge about health effects of workplace intervention programs, particularly concerning small enterprises and measures concerning organizational and psychosocial factors. Despite this identified need, there are 26

relevant research and evaluation studies from different areas presented below; first on a general level and then specifically for small enterprises. Interventions on a general level As mentioned in the introductory chapter, several large projects have been carried out in Norway (Gustavsen, 2002; Oscarsson, 1999) and Sweden (Huzzard, 2003; Johansson, 1999). These were influenced by the socio-technical approach, where some common ideas are that an organization with semi-autonomous group work, great work content and worker influence can contribute to increased productivity and work satisfaction. In several of these projects work organization development, labor-management co-operation and participative change approaches have been used. Although, there have been criticisms (e.g. von Otter, 1997), evaluations have showed several positive effects concerning productivity, worker health, learning aspects and participation in change activities (e.g. Alpenberg & Karlsson, 1996; Gustavsen, 1995, 2002; Oscarsson, 1999). Concerning the economic effects of work environment improvements, Ichniowski et al. (1995) found in a study of 26 American steelworks that workplaces that had invested in work environment and work organization had significantly better productivity and quality outcomes what compared to workplaces with limited investments. In a Swedish before-after study of 22 public workplaces, there was significantly more positive change for perceived psychosocial conditions, self-rated health among employees and biological stress indicators after investments in leader and worker development and work organization (Anderzén & Arnetz, 2005). In a Swedish evaluation of workplace programmes financially supported by the Working Life Fund (Johanson, 1997), a significant reduction in sickness absence and a concomitant increase in productivity were attributed to the programmes. This is in line with results from a project in the County of Jämtland where 178 workplaces (around half were small enterprises) took part in work environment and rehabilitation programmes (Vinberg, 1997). Some results were that about half of the studied individuals perceived improved health and had a lower degree of sickness absence after programme participation. In a Swedish steel industry longitudinal study of participatory work environment interventions there were improved physical and psychosocial working conditions plus an increased quality and efficiency in production (Abrahamsson, 2000). With respect to larger organizations, there are studies which have shown that workplace change processes have positive effects at the organizational level. Arthur (1992) investigated 30 US steel mini-mills and found that steel mills with a “high-commitment” strategy had significantly higher levels of productivity and quality compared with those that had a “low-commitment” strategy. Huselid (1995) found in a study of 968 US firms that when firms increased their “highperformance work practices” there were significant reductions in employee 27

turnover and significant increases in productivity and profits. Particularly, employee skills, organizational structures and employee motivation were found to be important for the outcomes. In another study focusing on 740 firms, Huselid & Becker (1996) created an index of the HRM systems. They found that firms with higher values on this index, other things being equal, had significantly higher levels of firm performance (profit change, changes in market value). A study by Patterson et al. (1997) investigated what factors most influence business performance. Regression analysis showed that job satisfaction explained five percent of the variation between companies in change of profitability and sixteen percent of the variation in productivity. Organizational culture explained ten percent of the variation in profitability and twenty-nine percent of the variation in productivity. Human resource management practices accounted for a nineteen percent variation in profitability and an eighteen percent variation in productivity. The conclusion of this research was that employee commitment and a positive psychological contract are fundamental in the improvement of performance. The link between quality and performance in large enterprises has been studied extensively. Researchers have drawn different conclusions about effects at a company level in quality improvement (Hansson, 2003; North et al., 1998; Nwankwo, 2000). Some researchers cite evidence that effective total quality management (TQM) implementations improve long-term profitability (Hendricks & Singal, 1997; Easton & Jarell, 1998). Hendricks & Singhal (1997) have compared quality award recipients with different control companies. They found that the quality award winners had better financial performance after the companies had implemented TQM. During the implementation periods there were no significant differences between the quality award recipients and control companies. Some researchers have a more pessimistic opinion about the benefit of quality improvements (Bergquist & Ramsing, 1999). A study by Powell (1995) found that most features generally associated with TQM such as quality training, process improvements and benchmarking, do not generally produce competitive advantages. In a critical review of the quality movement by Bejerot & Hasselbladh (2002), the uncritical perspective of the effects of quality development is questioned. These authors point at a study by Kivimäki et al. (1997) which found an increased workload among employees when implementing TQM. A deficiency in many studies is also that they focus only on financial performance and leave out more ‘soft performance measures’ - for example, employee working conditions and health. Quality managements systems are increasingly being integrated in with business managements systems that include health, safety and environmental issues (Frick et al., 2000; Greenan et al., 1997; Wilkinsson & Dale, 1999). There are few studies about the relationships between these areas. In one study, Eklund 28

(1995) found that working environment problems concerning psychologically demanding work and ergonomic tension were related to major increases in the frequency of quality defects. This is consistent with studies of Axelsson (2000) where it was found that ergonomics and quality have many common concerns and that in some cases up to 40% of quality deficiencies were related to ergonomic problems. According to the European Network for Workplace Health Promotion (1997), workplace health promotion (WHP) ideally consists of the combined efforts of employers, employees and society to improve the health and well-being of people in their work environments. This can be achieved through a combination of improving the work organization, improving work environments, promoting active employee participation and the encouragement of personal development. Several researchers argue that participative approaches, which combine individual and organizational level measures in WHP interventions, are necessary (Chu et al., 2000; Cox, 1997; Harden et al., 1999; Källestål et al., 2004). In a systematic review of the effectiveness of health promotion interventions in the workplace (Harden et al., 1999), 110 outcome evaluations were assessed. Most of the programmes targeted individual behaviour; assessment of supporting organizational changes was limited. According to the authors, the findings identified a substantial discrepancy between recommended WHP ‘good practices’ and what was reported by the studied evaluations. One Swedish review of workplace health promotion interventions (Källestål et al., 2004) analysed 36 interventions during 1990 – 2004. Concluded was that WHP interventions could be effective but that only a few of studies reviewed had a focus on the organizational level. This is in line with findings about work stress interventions in Europe, where Kompier et al. (2000) argued that the paucity of organizationlevel interventions forms a barrier to advancements in the reduction of workrelated stressors. Small enterprise interventions The European Network for Workplace Health Promotion (2001) has declared that workplace health and well-being in small and medium-sized enterprises is an essential prerequisite for increasing innovation. The Network’s declaration went on to say that small enterprises have a unique influence on employee health because of the: (1) immediate control a company owner can have on working conditions, (2) family atmosphere common to small and micro-enterprises, (3) greater individual responsibility, and (4) simpler organizational structure. The network proposed that any work with health in small companies should integrate health into daily managerial practices by involving all employees in decisionmaking processes, ensuring a good working atmosphere, recognizing and rewarding good performance, and monitoring improvements – especially those related to work organization. 29

In spite of these laudable statements, workplace health promotion is less developed in small enterprises and existing research on such activities is limited in both the health and business literature (Griffin et al., 2005; Moser & Karlqvist, 2004). According to Griffin et al. (2005), there are several reasons for why small enterprise involvement in health promotion issues is low: they lack the resources and motivation to deal with health issues, there are few organizational mechanisms for communication, they have few in-house resources for occupational health issues, and the perceived lack of evidence of the benefits of health at work interventions may discourage efforts. Contrary to this picture of small enterprises, Stokols et al. (2002) believe that small firms provide a highly advantageous context for promoting health due to their unique social, organizational and environmental attributes. For example they have fewer people to accommodate, fewer administrative costs and higher rates of employee participation in workplace change processes. There are few studies of the effect on health and performance on small firm occupational health. According to Champoux & Brun (2003), the research tends to show that leader isolation, lack of knowledge of a firm’s risks, rights and obligations, and a lack of resources are problems for health and safety within small firms. Several researchers point at the need for external support on how to carry out health and safety change processes in small enterprises (Eakin et al., 2000; Langaa Jensen et al., 2001; Walters, 2001). In addition, Walters (2001) mentioned that empirical evidence from several studies had associated the presence of trade union health and safety representatives with lower injury rates. In a Swedish case study of 20 small manufacturing enterprises (Johansson, 1998), results were that enterprises with successful work environment development activities were characterized by active and engaged employees, a dialogue between leaders and employees, and a positive learning climate. There are contradictory results about occupational health in small firms. For example, Champoux & Brun (2003) found in their study that health and safety activities were practised regularly in studied small firms. Langaa Jensen et al. (2001) found that some firms were characterized by a qualitative staff who could manage issues without having to ask for outside assistance. Another example of studies concerning change programs in small enterprises is a study by Alpenberg & Karlsson (1996) which indicated that in many cases investments for improvement of the working environment created opportunities for increased economic efficiency. They also noted that projects with a comprehensive approach tended to have the greatest potential for successful achievement of goals. The reported results also indicated that subsidies tend to fix starting points and make each project more comprehensive in terms of size and complexity. In a Finnish study of change programs in small and mediumsized enterprises (Lindström et al., 2000), one finding was that more intensively planned interventions tend to have slightly more positive effects on sickness 30

absenteeism and job satisfaction when compared to enterprises having minor or no interventions. A study of 280 small Swedish enterprises that had taken part in competence development programs, showed higher productivity levels and financial performance compared to a control group of 200 enterprises, not taken part in such programs (Bager-Sjögren et al., 2003). Some researchers mean that small organisations have advantages in implementing quality improvements due to their flexible organizational structure, few hierarchical positions and strong organizational culture (Haksever, 1996). Despite these advantages, small firms have been slow to begin to work with quality improvements in a systematic way due to limited resources, lack of financial resources and problem in providing internal training (Lee & Oakes, 1995; Noci, 1996; Nwankwo, 2000). As mentioned above, most of the studies about performance have investigated larger organizations. There are studies about SMEs. Anderson & Sohal (1999) reported on a study of Australian SMEs and noted that, among a number of factors, strong customer focus, quality systems and good information management provided the greatest positive influence on the quality outputs. In a Swedish doctoral thesis focusing on small organisations (Hansson, 2003), comparisons were made between profit-seeking companies that received a quality award and two control groups (i.e competitors and branch indices). According to the author, the findings indicated that there is a link between successful TQM implementation and an advantageous financial performance development (change in sales, change in total assets). This suggested that TQM, as a means for systematic and holistic quality management efforts, should be profitable in the long run.

Summary and chapter comments In summary, much research to date points to the importance of organizational factors when health and performance related outcomes at workplaces are considered. Frequently emphasized is that such relationships may differ according to workplace size and sector. Apparent from the literature is that there is a lack of knowledge about how strong these connections are in different contexts and that small enterprise research is limited and somewhat contradictory. Research focusing on healthy work organizations and workplace health promotion are emerging research areas. Research studies often focus on either health outcomes or performance outcomes; not both. The literature reviewed found few studies where both outcomes have been investigated in tandem. Few concerned small enterprises. However, there is a large body of research concerning various aspects of the concept of a healthy work organization. These are: aspects of leadership, psychosocial work environments and learning. Several studies have confirmed 31

relationships between these and health and performance outcomes. Contradictory results also exist. Much of what this research has in common is that a limited number have investigated small enterprises and there is a lack of knowledge about the strength of the mentioned relationships in this enterprise group. Concerning workplace change processes and change strategy, some consensus exists about the need for participative and holistic approaches. However, few studies have deeply studied the nature of participative processes and how they can be applied to small firms. Researchers from several areas point to problems for small enterprises due to a lack of resources, knowledge and motivation for workplace change processes. Results that contradict this position also exist. Some researchers state that because of their size, small firms have a greater potential to benefit from work environment improvements and that they have change competence for these issues. Concerning research studies of workplace interventions, there are several showing different positive effects - especially for larger workplaces. There is a need for more large and small workplace quantitative and qualitative longitudinal studies over longer time periods. According to some researchers there is also a need to develop methodologies and research design in such studies. Individually related explanations must be complemented with more organizationally focused ones to obtain an understanding of the mechanisms behind health and performance in small firms. As the resource limitations of small enterprises could complicate workplace change processes, there is a need for more integrated models - particularly for ones that are adjusted to conditions in small firms.

32

Materials and methods In this section the methodological choices are discussed together with the used analysis model, empirical base, research design and quality of data in the subsidiary studies. For additional knowledge about the described studies, details may be found in the appended papers.

Methodological choices The empirical base used in this thesis comes from four research and development projects performed during the period 1997 to 2003. In these projects public and private organizations cooperated with researchers. The aims of the work were to generate networks between small workplaces and to develop workplace change process and health promotion models. Naturally, these projects have influenced the methodological choices of this thesis. For example, data collection was used both for the workplace analyses in the practical part of the projects and for the actual research studies. The studies in this thesis used both a quantitative and a qualitative approach. This is consistent with the view of some researchers that the previously harsh polarization between these two approaches has more or less faded out (Alvesson & Sköldberg, 1994). The quantitative approach is often used when the goal is to investigate defined in advance phenomena, distribution in a population or in situations and to investigate whether there are any relations between two or more phenomena. Deduction and exploration are the most common principles of analysis (Starrin & Svensson, 1994). Typical methods include survey methods such as were used in the thesis studies. The qualitative approach uses any kind of research that produces findings that have not been found through statistics or any other type of quantitative method (Strauss & Corbin, 1990). A typical qualitative method can be the use of interviews to collect data. Combinations of methods, as in the actual studies, give the opportunity to achieve a deeper and more detailed knowledge about conditions in the studied small enterprises. To combine survey data, both cross-sectional and longitudinal, with case studies and interviews has been proposed by small business researchers (Curran & Blackburn, 2001; Davidsson, 2004).

Analysis model Drawing upon the literature overview of theoretical approaches and research findings, Figure 2 depicts an explorative analysis model that was constructed to provide a broad, interpretative overview of the five studies presented in the 33

thesis. The model attempts to integrate an important set of organizational factors found to contribute to health and performance outcomes in organizations. The assumed organizational determinants covered aspects associated with concepts of leadership, psychosocial work environment and learning. The indicator of change strategy/processes is introduced into the model to indicate that relations between determinants and outcomes are often mediated by change processes. Organizational determinants Outcomes Leadership Respectful leadership Leader change competence

Health

Process Psychosocial Work Environment

Change Strategy

Work intensity demands Psychosocially demanding work Demands higher than control Job stress problems Data stress Decision latitude - Self-determination Job meaningfulness Team spirit – Social support Restructuring worries Ergonomic problems Structure

Change… …competencies …goals …motivation …strategy Ongoing programs Partners Customer oriented practices Knowledge of workplace goals Efficient use of technology/ IT-level Use of systems

Learning Adequate competence Educational investments Creative work Competence demands Problem solving intensity

Health performance Self-assessed health Musculoskeletal ill health Psychosocial ill health Sickness absence

Performance Organizational performance Profit per capita Workplace adaptability Customer satisfaction Turnover Lead-time No of… …products …customers

Figure 2. Analysis model of the main areas and used indicators in the thesis studies. All indicators presented in the model are linked to empirical analyses in the studies. Although the model seeks to group the indicators for studied main areas, it must be noted that a model is by definition a simplification of a complex reality 34

and there is always a compromise as to whether to include some factors or to leave them out (Ruist, 1990).

Empirical base and research design Study I was part of the project Företagsutveckling genom ökad kompetens och bättre hälsa. Ett pilotprojekt på små och medelstora arbetsplatser (Business development through increased competence and better health. A pilot project in small and medium sized workplaces) (Vinberg et al., 1998). This project was a cooperative project between Försäkringskassan (the National Social Insurance Office), Länsarbetsnämnden i Jämtlands län (the County Labour Board of Jämtland) and Företagarnas riksorganisation (The Swedish Federation of Private Enterprises) between 1997 and 1998. The overall aim was to develop models for increased competence concerning work environment, work organization and health in small enterprises in Jämtland County in Sweden. A group of ten small enterprises from the manufacturing, wood and service industries participated. The number of employees in the companies varied from 5 to 23. The Studies II and V are from another project called Företagsutveckling - utan gränser (Company development - without limits) (Vinberg, 2001). This project was a cooperative effort involving Arbetslivsinstitutet (the Swedish National Institute for Working Life), Nord-Trøndelagsforskning (the Nord-Trøndelag Research Institute), Försäkringskassan and Trygdeetaten (the Swedish and Norwegian social insurance offices), and Länsarbetsnämnden i Jämtlands län (the County Labour Board of Jämtland) between 1999-2001. The overall aim was to generate networks between participating enterprises and public organizations, and to influence change programmes at the company level. In total, 42 Swedish and Norwegian companies (30 Swedish, 12 Norwegian), representing the industries of manufacturing, wood products, construction, retail trade and consulting, participated in Study II. Approximately 70 % of the companies employed 10 to 30 persons. In Study V, ten Swedish companies of the above mentioned 42 companies participated in a before and after research study. The number of employees in these companies varied between 7 and 35. Study IV was part of a project called Ragunda-projektet (the Ragunda-project), which was a cooperative project between Arbetslivsinstitutet (the Swedish National Institute for Working Life), the municipality of Ragunda and the local federation of private enterprises. The aim was to contrast micro-enterprises with small public workplaces and, as an input into developing rural entrepreneurship, shed light on some needs for change in the studied small enterprise group. In total, 66 micro-enterprises and 50 small public workplaces participated in the study. The micro-enterprises, which all had fewer than ten employees, represented the wood processing, manufacturing, retail and service sectors. The small public workplaces participated as part of their involvement in the project 35

Friskare arbetsplatser i Jämtlands län (Healthier workplaces in Jämtland County) (Bildt et al., 2005). These workplaces had 6-49 persons and represented branches such as health care and education. The Swedish enterprises that participated in the projects described above, were jointly recruited by Företagshälsovården (Occupational Health Services), Försäkringskassan (the Swedish National Social Insurance Office), Arbetslivsinstitutet (the Swedish National Institute for Working Life) and Företagarna (the Swedish Federation of Private Enterprises). The Norwegian enterprises, were jointly recruited by Trygdeetaten (the Norwegian National Insurance Office) and Nord-Trøndelagsforskning (the Nord-Trøndelag Research Institute). For all projects, the goal was to obtain a sample of small companies from different sectors; particularly from those sectors that are dominated by small enterprises (Antonsson et al., 2002; Spilling, 2000). The participating public workplaces, were recruited by Försäkringskassan (the Swedish National Social Insurance Office) in cooperation with the municipalities in Jämtland County. The ambition was to recruit a sample of workplaces from different local areas and common public branches. The Studies I, II and IV are based on cross-sectional survey data. Study V is a longitudinal study with two measurement points and Study III is a literature study focusing on developing theoretical aspects about change strategies in small enterprises. Although the studies mainly have a quantitative approach, a complement was the semi-structured interviews used in the Studies I, IV and V. Study I The unit of the study is the enterprise. The data used was collected during the period from September 1997 to February 1998 in a study of 10 small enterprises. A total of 20 leaders and 82 employees participated. In the sample group, 45% were female and 55% were male. Data was collected through individual survey questionnaires that were distributed by the researcher at group meetings, individually filled in and collected at the workplaces. The response rate was 88 percent. During the same period, semi-structured interviews with leaders and coworkers provided information about competition level, organization structure and ongoing change programs. Data concerning economic outcome and economic analysis of human resources was also collected. From human resource accounting analyses (Aronsson & Malmquist, 2002) in each company it was possible to get the profit per capita and educational investments for the year 1997. The individual survey data was aggregated into company means which was used in the analysis. To gain an understanding of physical and psychosocial work environment conditions a questionnaire used in several nationwide surveys was used (Wikman, 1991). In total, 72 questions were used to develop a picture and to 36

estimate the frequency/duration of different problems. An established method for measuring competence aspects was used (Anntila, 1997). Both leaders and coworkers answered a questionnaire in which they rated themselves in areas such as competence for change, work environment, work organization and health. They also rated each other’s ability in the mentioned areas and motivation to increase competence. In total, the employees answered 69 questions and the leaders 81 questions that concerned the area of competence (Vinberg et al., 1998). Indicators In this study, 19 indicators based on substantive considerations and factor analyses were used. The organizational level indicators used in the study were Profit per capita (net profit for the year 1997 in SEK/number of employees including leaders), Flat organization (two or less organizational levels), Educational investments per employee (amount on co-worker investments 1997 in thousands of SEK/number of employees including leaders), On-going change program (if the company was pursuing a change program and/or technical/process change and/or organizational/leadership development) and Information technology level (a composite summation index based on eleven aspects of the use of information technology equipment). The used individual level indicators concerning competence aspects were as follows. The indicator Leader change competence (a summation of answers on fourteen questions) covered leader views of ability concerning customer demands/expectations, developing new techniques/work processes, following and prioritizing worker competence development, developing worker competence due to changes in customer demands and taking initiatives for trying new areas of work. The indicator leader change motivation (a summation of answers on five questions) consists of leader views of willingness to increase their competitive abilities, competence development, finding out/taking care of health and working environment problems, and developing quality and work organization. The indicator co-worker change motivation consists of a summation of answers on five questions, similar to those for leader change motivation. Leader minus coworker change motivation indicator measures the difference between the two prior indicators. The indicator Leader change competence minus co-worker conception of leader change competence measures the difference between mean leader change competence and mean co-worker conceptions of leader change competence (the latter indicator is a summation of answers to the same questions used with the leaders). The individual level indicators concerned work environment conditions and health (ill health). The indicator Job stress problems (five questions) covered whether the work was so stressful that it was difficult to think about other things, whether the work demanded a lot of attention and concentration, whether it was 37

difficult to get help from colleagues in critical work situations, whether the work was monotonous and if part of the work-time included understanding and solving demanding problems. The indicator Self-determination in job (three questions) measured the freedom to decide when to shift work tasks, the opportunity to take part in work design issues and degree of freedom at work. The amount of social support indicator (three questions) covered the ability to get support if tasks were difficult and, in particular, support from leaders and co-workers. The indicator Educational job demands consisted of one item about the need for new knowledge and ease of acquisition. The indicator Problem-solving intensity at work consists of one item about whether part of the working time was spent on understanding or solving demanding problems. The Job meaningfulness indicator (three items) covered whether the work was perceived as dull, meaningless and dissatisfactory. The indicator Ergonomic problems (four questions) measured perceived physical work-load, working in forward-bent and twisted postures and with hands at or above shoulder height. The Psychosocial ill-health indicator (six questions) covered perceived levels of not being able to relax at leisure time, heartburn, tiredness, eye-irritations, sleep difficulties and bad feelings when going to work. The Musculoskeletal ill-health indicator (six questions) covered perceived pain after work concerning the upper back, low back, shoulders and arms, wrists/hands, hips, and feelings of body fatigue. Statistical analysis The analyses, which were based on two-level data with organizational level data for the ten companies and individual level data for the leaders and co-workers, were organized in four steps. First, scales and indicators were constructed based on prior knowledge of the questions used. Then, the resulting indicators were used as input in an explorative principal component analysis. In the third step, regression analyses were carried out with four outcome variables. The variables profit per capita and musculoskeletal ill-health can be seen as health and performance-related indicators. Separate stepwise multiple regressions were made for each outcome indicator with all other non-redundant variables in the factor analysis list as potential independent variables in the first step. The risk for multi-collinearity problems appeared to be low, with low interrelations between explanatory variables and no instances with high VIF-values. In the results, only variables with standardized beta-values significant at the 0.05 level were included. Studies II and V In Study II, where the unit of the study is the enterprise, data was collected from November 1999 to May 2000 from both the Swedish and the Norwegian parts of the project. There were 42 small enterprises in the study which employed a total 38

of 1177 persons in 2000. The gender, age and educational distributions for the individuals were: 43% women, 65% under 40 years of age, 16% in the 50+ age group, 16% with a basic education and 25% with a university education. Data was collected through the use of individual survey questionnaires, distributed, filled in and collected at the workplaces in a similar way as in Study I. However, one difference was that consultants from the social insurance offices and occupational health services assisted the researchers. The total response rate was 84 percent. To gain an understanding of how both leaders and co-workers in the companies perceived work environment conditions, leadership and learning, a questionnaire, the ‘Organization Profile ’, developed by the Stress Management Center (Setterlind, 2004), one shortened version of the Job Content Questionnaire (JCQ) developed by Robert Karasek (Ahlberg-Hultén, 1999), and one questionnaire developed by Nord-Trøndelag research (Sletterød & Lysø, 2001) were used. The questionnaires had a total of 162 items covering a wide range of topics, only some of which were used in this study. In Study V ten Swedish enterprises participated in a longitudinal panel data study. The research design consisted of a combination of quantitative and qualitative measures. Quantitative data was gathered through questionnaires distributed to managers and employees. Qualitative measures were acquired through interviews of managers and examination of accomplished change programs at the studied workplaces. At Time 1 (November-December 1999), before accomplished workplace change programs, 261 individuals in the enterprises received the questionnaire. The response rate was 88 percent or 229 individuals. At Time 2 (April-May 2001), after accomplished workplace change programs, there were 211 individuals in the companies and 69 percent answered the questionnaire. The final panel consisted of 102 individuals (48 percent of the participants from 2001) who responded to the questionnaire on both occasions. This selection made it possible to compare results for the same individual, before and after accomplished workplace interventions. The gender, age and educational distributions for the individuals were: 54% women, 58% under 40 years of age and 16% in the 50+ age group, 12% with a basic education and 23% with a university education. The characteristics of the individuals in the two studied enterprise groups (broad strategy versus expert/problem-based strategy) were similar with respect to gender, age, education and sickness absence levels. In this study, semi-structured interviews were conducted with all managers after their enterprises had completed workplace change programs. The interviews were at the respondents’ workplaces and lasted approximately two hours. The interviews focused on change methodologies used, changes in organizational factors, health and performance, and successful factors and obstacles in the processes. During the interviews the interviewer made notes that were then reviewed as soon as possible after each interview. Follow-up questions by 39

telephone and documents about workplace interventions were other components in the qualitative method approach. Indicators – Study II In this study, 22 indicators based on substantive considerations and factor analyses were used. Thirteen determinant indicators were as follows (see the analysis model in Study II, Figure 2). The knowledge of workplace goals indicator4 consisted of two items that sought to measure how well the employees knew workplace goals. The indicator customer oriented quality practices (seven items) covered feedback on quality to employees, programs for improvements, quality goals and cooperation with customers. The efficient use of technology indicator consisted of two items covering the status and effectiveness of technical equipment. The respectful leadership indicator (nine items) covered leadership consideration, information about changes, ethical issues, and dialogue between leaders and coworkers. The creative work indicator (ten items) covered developing and learning at work, meaningful and stimulating work and variation at work. The questions in the adequate competence indicator (two items) were about resources and competence in relation to work demands. The work intensity demands indicator had eight items about hard and demanding work, stress and time pressure. The questions in the psychosocially demanding work indicator (two items) were about the employees’ perception of the degree of demanding work. The data stress indicator consists of two items about stress related to data problems. In the three questions on the demands higher than control indicator, people were asked to say if their job demands were higher than what they could handle. Three items concerning team functioning and team climate at work measured the team spirit indicator. The decision latitude indicator (four items) measured worker autonomy/control and opportunity to influence work. The indicator restructuring worries had two items about concerns related to possible restructuring and its potential impact upon continued employment. The three outcome indicators (see the analysis model in Study II, Figure 2) were created using nine indicators. The organizational performance indicator is the weighted mean of perceived lack of efficiency (four items), workplace adaptability (three items) and customer satisfaction (seven items) as perceived by the employees. The health performance indicator is the weighted mean of indicators measuring not being in a depressed mood/worries (six items), not having musculoskeletal symptoms (three items), not having sleep problems (two items), perceived physical health (one item) and psychological well-being (one item). The sickness absence indicator (one item) measured sickness absence days per employee over the previous 12-month period. 4

In the following overview the range for the items is 1-5, unless otherwise stated. 40

Indicators – Study V Indicators used in this study were similar to those in Study II. The content of the eight determinant indicators were as follows. The indicator customer oriented quality practices (seven items) covered feedback to employees on quality, programs for improvements, quality goals and cooperation with customers. The respectful leadership indicator (seven items) covered leadership consideration, information about changes, ethical issues, and dialogue between leaders and coworkers. The creative work indicator (six items) measured developing and learning at work, meaningful and stimulating work tasks, and engagement at work. The team spirit indicator (two items) measured team functioning and team climate at work. The work intensity demands had three items about hard and demanding work and stress at work. The data stress indicator consisted of two items about stress related to computer problems. The decision latitude indicator (three items) covered worker autonomy/control and the opportunity to influence work. The indicator restructuring worries had two items about concerns related to possible restructuring and its potential upon continued employment. Three indicators were seen as outcome indicators for organizational health and performance. The indicator self-assessed health consisted of two items that sought to measure physical health and psychological well-being. The indicator workplace adaptability (three items) covered effectiveness and adaptability to external demands and changes. The indicator customer satisfaction (four items) sought to measure judgments about degree of customer satisfaction, quality in products and services, and the ability to respond to customer desires. All indicators were based on questions with a response range of 1-5, with an exception for the indicator decision latitude where the range was 1-4. Some items were reversed so that a high value on an indicator always indicates a more positive value. Both before and after the workplace interventions the managers at studied enterprises answered a questionnaire about organizational aspects. Corresponding to a study by Ingelgård & Norrgren (2001), 18 items were used for forming three indices: learning strategy index (ten items about visions of the change process, how multiple problems were addressed, the number of employees and leaders who took part in and influenced the change process, the planning and improvisation which characterized the change process), a top management involvement index (four questions about manager pursuit of the change process plus consideration of opinions and different views, managerial belief in positive outcomes and strong identification with visions), and quality of working life index (four questions about to what degree the actual change process contributed to an improved working climate, co-operation, involvement and more varied work tasks).

41

Statistical analyses The statistical methods in Study II were based on descriptive data for mean sickness absence for firms in different branches, correlation analyses using Pearson’s correlation (r), and structural analysis. For this study, two different perspectives were applied in the analysis. In one perspective, a distinction between determinants and outcomes by integrating possible determining relations between the main concepts in the Figure 2 model was made. The second perspective took a structural view with the aim of finding groups of variables or indicators that tended to co-vary and form broader aspects or latent dimensions. This second perspective also had as an aim to determine if there were groups of companies with similar positions when using a multidimensional scaling method. In Study V, the statistical analyses consisted of comparisons of means at an individual level for the used determinant and outcome indicators, between the first and second study rounds, as well as between groups of enterprises with different change strategies. Pearson’s product correlation coefficient (r) was used for measures of associations and the t-test for testing differences between groups. The enterprises were classified into two different change strategy types: a broad strategy and an expert/problem-based strategy. The classification was similar to what was used by Håkansson (1995). This broad strategy deals with several extensive problems concurrently, the degree of participation is high and several arenas are used in the development process. The expert/problem-based strategy has a lower degree of workforce participation, is more expert oriented and deals with only a few problem areas. Study IV The unit of the analysis in Study IV was also the enterprise. Interviews with 66 micro-enterprise managers during the spring of 2001 were based on a semistructured guide with questions about markets, resources, learning and the company climate. The respondents also answered a questionnaire on outcomes, organizational factors and change goals (Vinberg, 2004). Nearly all the interviews took place at the managers’ workplaces. Comparisons were made with a group of 50 small public workplaces (with 6-49 persons employed in health care and education) where the leaders answered a similar questionnaire (Bildt et al., 2005). Indicators In this study, outcomes concerning health and performance were studied together with organizational aspects and change processes. The performance outcomes were turnover, profit, number of customers/products, lead-time for services/products and customer complaints. The health outcomes covered areas such as sickness absence and staff turnover. The organizational determinant 42

factors covered areas such as competence demands, responsibilities among employees, number of work tasks and work organizational aspects. Change process aspects as methods/systems for improvements, partners for developing working environment/competence and goals for improvement processes were also studied. Statistical analyses In this study, descriptive data from the leader questionnaires was used. A twosided chi-square test was used to assess result differences between the two workplace groups. Study III The study consists of a theoretical and conceptual discussion about change strategies in small enterprises and its relation to outcomes such as health, quality of working life and performance. The discussion was derived from a literature search in several databases about earlier relevant research concerning small enterprises.

Quality of data As mentioned in the introductory chapter, the presented studies are limited to small enterprises and organizations in Sweden and Norway and the data sets are not representative samples so no statistical inferences as they pertain to broader populations are intended. The results should be seen as mainly explorative due to the strong focus on cross-sectional and quantitative data - even if the design is longitudinal with both quantitative and qualitative measures in Study V. Although there is always a risk of selectivity bias (Heckman et al., 1996), the distribution of firms is rather similar to the total distribution for small firms in Sweden (Antonsson et al., 2002) and Norway (Spilling, 2000). Below follows a summary of some data quality aspects for the empirical studies.

43

Table 2. Data quality aspects for the studies I, II, IV and V. Data quality aspects Type of sample and representativeness

Not a random sample; rather a similar sector distribution to the whole small company population in Sweden and Norway (Antonsson et al., 2002; Spilling, 2000). Possible selectivity bias - a prior interest in organizational change.

Unit of study, sample size Study I: 10 enterprises, 102 individuals. Study II: 42 enterprises, 988 individuals. Study IV: 66 microenterprises and 50 small public workplaces, 116 leaders. Study V: 10 enterprises, 102 individuals Non-response

12% in Study I, 16% in Study II, 22% in Study IV and 52% in Study V.

Internal non-response

Generally low, ranging from no internal non-response to 5-15 % for used survey questions.

Indicator reliability

In Study I the indicator “leader change competence” is a composite indicator of four scales for which the Cronbach alpha reliabilities are between 0.84 and 0.95. The alpha values for the other competence indicators are in a similar range. For the indicators of determinants in Study II, the alpha values are between 0.55 and 0.91; eight of the thirteen indicators have alpha values above 0.80. The alpha value for organizational performance is 0.77 and for health performance 0.76. The alpha values for the indicators in Study V varied between 0.61 and 0.90.

Questionnaires used

In Study I, the questions are from Wikman (1991) and Anntila (1997). In Studies II and V, the questions are from the Stress Management Center ‘Organizational Profile’ (Setterlind, 2004), the Swedish short version of the JCQ (Job Content Questionnaire) (Ahlberg-Hultén, 1999), and from questionnaires developed by NordTröndelag Research Institute (Sletterød & Lysø, 2001). In Study IV data from a developed workplace survey to the leaders was used (Vinberg, 2004).

44

Summary of study results In this section the results from the five studies are summarized, first in condensed form (below and in Table 3) and after that grouped into five subsections. The subsections are: relationships between determinants and outcomes, structure of indicators, structure of enterprise groups, change strategies/change goals and effects of workplace interventions.

Result overview Study I. The focus was to analyze relations between organizational aspects and health and performance outcomes in ten Swedish small enterprises. In regression analyses, four indicators explained 79% of the company variation in profit per capita and three indicators explained 26% of the musculoskeletal ill health variation. Factor analysis resulted in six components, which explained 71% of the total variance. Study II. Similar relations as in the former study were investigated in a group of 42 Norwegian and Swedish small enterprises. Correlation analysis showed relations between health and performance outcomes, and between these outcomes and several organizational determinant indicators. Factor analysis resulted in six components, explaining 74% of the total variance. Using a structural analysis it was possible to group the companies according to two main dimensions. Study III. Results in this theoretical study were the importance of integrating different theoretical perspectives and the use of a participative learning-based strategy when working with workplace change processes in small enterprises. Study IV. In this study 66 Swedish micro-enterprises were contrasted with 50 small public workplaces. Results showed a lower increase in sickness absence and work demands, and more attention to the development of aspects of work organization in the micro-enterprises. The enterprise group had more productionoriented goals, and used structured systems for work environment to a lesser degree. Study V. In this longitudinal before-after study of ten Swedish small enterprises, several rather strong relationships were found between health and performance outcomes and organizational determinants. Some weaker relationships were also present. Results concerning performed workplace interventions showed that companies using a broad change strategy estimated as higher the indices for learning strategy, top management involvement and quality of working life than did companies using an expert/problem oriented strategy. When comparing the two change strategy enterprise groups, there were significant differences concerning changes in indicator means for six indicators. 45

Table 3. An overview of the five studies. Study

Purpose/empirical base

Main results

Conclusions

I

To analyze relations between health and performance outcomes and organizational aspects, change competence, change processes and IT-levels. The focus is on co-variation and relationships between assumed outcomes and organizational determinants. A group of ten small enterprises including 20 leaders and 82 co-workers participated in the study.

Six components emerged from a principal component analysis. A flat organization, leader change competence and motivation were interrelated and also related to profit in the companies. Leader change competence, a flat structure, competence difference and an on-going program contributed significantly to profit per capita.

The analysis model seem to provide a satisfactory integration of the results but underlying mechanisms must be investigated in future studies.

II

To analyze empirical relationships between outcomes for performance indicators and assumed organizational determinants, and to find out if organizational aspects co-vary and if groups of enterprises are similar concerning indicator patterns. A group of 42 Norwegian and Swedish small enterprises including 988 individuals participated in the study.

Six components emerged from a principal component analysis. Several relationships were found between outcomes of health and organizational performance and organizational determinants. Structural analysis showed that groups of firms had high levels on two dimensions – ‘general health performance’ and ‘high supportive organization’.

Some relations in the analysis model received empirical support. The inter-relatedness of a large number of factors, and that it was possible to find out segments of firms point at the importance of change processes oriented towards multi-purpose goals.

III

The purpose of this literature study was to define and analyze different theoretical perspectives concerning change strategies adapted to small enterprises.

The need for integrated models with different theoretical perspectives. The importance of a learning strategy which involves the employees in small enterprises.

Integrating several theoretical perspectives and using a participative strategy could be related to good quality of working life outcomes.

IV

To analyze and discuss the importance of organizational factors related to health outcomes in micro-enterprises, and shed light on some needs for change in this enterprise group. A group of 66 micro-enterprises and 50 small public workplaces participated in the study.

Lower increase in sickness absence, competence and work demands, and more developing work in the micro enterprises, in comparison with the public workplaces. More production oriented goals and need for improvements in the micro-firms.

Work organizational aspects could be of more importance for health outcomes than formalized systems. Need for improvements in the small firms.

V

To contribute to the workplace change processes knowledge base through a focus on organizational health and performance in small enterprises. A group of 10 small enterprises including 102 individuals participated in the study.

Some strong links between organizational factors and outcomes for health and performance were found. Differences concerning indicator means due to used two change strategy types.

Holistic processes emphasizing both leader and co-worker participation are more likely to contribute to positive changes in health and performance outcomes.

46

Relationships between determinants and outcomes (Studies I, II, V) A summary of the correlation and regression analysis results is given below in Table 4. The table is not to be read as a summary of a systematic investigation made in the same way in all three studies. Studies II and V are the only ones where a systematic investigation was made with all possible relations between the studied determinants and the studied outcomes. In the table, those coefficients that represent the strong or rather strong relations found in Studies I, II and V are noted. The table shows relations between the two outcome indicators - health and performance - and studied determinant indicators. The detailed levels for the coefficients are specified below the table. With health as an outcome indicator there are rather strong positive relationships with the determinant indicators of respectful leadership, decision latitude, job meaningfulness, team spirit, creative work and efficient use of technology. There are rather strong negative relationships between the health outcome indicator and the determinants for work intensity demands, psychosocially demanding work, demands higher than control, job stress problems, data stress and restructuring worries. For performance as an outcome indicator, there are strong or rather strong positive relationships with the determinants for respectful leadership, leader change competence, self-determination-decision latitude, team spirit, flat structure, adequate competence, creative work, customer oriented quality practices, efficient use of technology, ongoing programs and change competencies. There were strong or rather strong negative relationships with indicators related to work intensity demands and stress, and restructuring worries.

47

Table 4. Overview of the correlation and regression analysis results. Organizational determinants and Change strategy/process

Health

Study:

I

II

Leadership Respectful leadership

Outcomes Performance

V

I

○

II

V

●

●

●□

Leader change competence Psychosocial work environment Work intensity demands (reversed) Psychosocially demanding work (reversed) Demands higher than control (reversed) Job stress problems (reversed) Data stress (reversed) Decision latitude-Self determination Job meaningfulness Team spirit-Social support Restructuring worries (reversed) Ergonomic problems Flat structure

○ ○ ○

○

○ ○ ●

○ ○

○

○ ○

○ ○

● ○

● ○

□

□ ○ ○

□

Learning Adequate competence Educational investments Creative work Competence demands Problem solving intensity

○ ○

○

●

○

●

Outcomes

Change strategy/process Customer oriented quality practices Knowledge of workplace goals Efficient use of technology – IT-level Ongoing programs Change competencies/motivation

□

○ □ □

Key to signs used: ● = r ≥ 0.5 and p < 0.01, r means Pearson correlation coefficient ○ = 0.25 ≤ r < 0.5 and p < 0.1 □ = p (absolute value of (stand b)) ≤ 0.05, ‘stand b’ means standardized regression coefficient For Study V: correlations refer either at both or only one of the measuring occasions at Time 1 or 2.

48

In Study I, two health and performance outcomes were used as dependent variables in separate stepwise multiple regression analyses. Four variables explained 79% of the company variation in profit per capita. An increase in leader change competence, a flat organization, a higher leader than co-worker change competence and an ongoing change program contributed significantly to profit per capita at the company level. Three variables explained 26% of the physical ill-health variation among the employees. An increase in enterprise information technology level and more meaningful jobs tend to lower musculoskeletal ill-health while stress problems co-vary with musculoskeletal illhealth. In Study II, the outcome measure organizational performance was strong or rather strongly positively related to customer oriented quality practices, efficient use of technology, respectful leadership, creative work, adequate competence and team spirit. Organizational performance was strongly or rather strongly negatively correlated with work intensity demands, psychosocially demanding work, data stress and demands higher than control. For health performance there were strong or rather strong positive correlations with team spirit and strong or rather strong negative correlations with work intensity demands and restructuring worries. Correlation analysis between the outcome variables showed that there were rather strong relationships in assumed directions between organizational performance, health performance and sickness absence. In Study V, correlation analyses were performed before and after workplace interventions (measured at Times 1 and 2). The health outcome indicator selfassessed health related significantly in the assumed direction with the indicators respectful leadership, creative work, team spirit, work intensity demands and data stress at both measurement points. For customer oriented quality practices there were no significant relationships. For decision latitude and restructuring worries there was a relationship only at Time 2. For the performance outcome indicators – judged workplace adaptability and customer satisfaction – there were strong relationships with the assumed determinants for customer oriented quality practices, respectful leadership, creative work and team spirit at both measuring points. For indicators concerning work intensity demands and data stress the relationships were not significant except for data stress and workplace adaptability at Time 2. The indicator decision latitude correlated significantly with customer satisfaction at Time 2. For restructuring worries the relationships with the outcome variables were significant except for the relationship with customer satisfaction at Time 2.

Structure of indicators (Study I, II) In Studies I and II, results from exploratory factor analysis are presented showing co-variation between indicators in the studies. In Study I 19 indicators were used 49

as inputs and six components emerged which explained 71% of the total variance (Table 5). Table 5. Loadings for 19 indicators on six components. Indicators 1

2

Component 3 4

5

6

High performance organizational aspects

Profit per capita Leader change competence Flat versus more hierarchical organisation Leader minus co-worker changemotivation Leader change motivation

0.826 0.088 0.079 0.156 0.270 -0.218 0.753 0.277 0.222 0.016 0.333 -0.036 -0.748 0.452 0.131 0.161 0.096 -0.004 0.721 0.150 -0.215 0.237 -0.129 0.472 0.606 0.018 -0.313 0.459 -0.215 0.433

Information technology level and competence drive

Educational investments per employee Ongoing program of company change Information technology level

0.093 0.870 0.087 - 0.035 0.154 0.034 -0.104 0.802 -0.170 -0.237 -0.079 -0.017 0.536 0.710 0.111 0.159 0.290 -0.038

Stress and high work demands

Problem solving intensity at work Job stress problems Amount of support Educational job demands Psycho-social ill-health

0.138 -0.082 0.168 0.064 0.315

-0.036 -0.117 -0.098 0.369 0.079

0.823 0.071 0.162 0.201 0.732 0.113 - 0.328 -0.054 0.644 0.183 0.227 0.087 0.426 - 0.014 0.391 0.178 0.406 0.257 -0.331 -0.007

Change propensity

Co-worker change motivation -0.131 -0.165 -0.129 0.879 -0.111 -0.040 Leader change competence – co-worker conception of leader change competence 0.114 -0.005 0.191 0.843 0.227 -0.145 Ergonomic and musculoskeletal problems

Ergonomic problems Musculoskeletal ill-health

-0.115 -0.078 0.087 -0.030 -0.660 -0.022 -0.038 -0.111 0.163 -0.034 -0.642 -0.099

Job meaningfulness and self-determination

Self-determination in job Job meaningfulness

-0.054 0.106 0.063 -0.096 0.064 0.806 0.027 -0.288 0.129 -0.037 0.466 0.630

Note: Loadings > 0.4 are in bold.

The first component is high performance organizational aspects. The factors that are strong indicators of this component are high profit per capita, leader change competence, a flat (in contrast to a more hierarchical) organization, leader change motivation and the difference between leader change motivation and co-worker change motivation. The second component is described as information technology level and competence drive and consists of educational investments per employee, an ongoing program of company change and IT-levels as strong 50

indicators. The variables on the third component are related to stress and high work demands - how much working time is devoted to hard problem solving, a job stress index, an index measuring support and encouragement, a measure of educational job demands and an index measuring psycho-social ill-health. The fourth component is about change propensity with the two high loading indicators of co-worker change motivation, and the difference between the leaders own perception of their change competence and co-worker assessments of their leaders’ change competence. Component five is about ergonomic and musculoskeletal problems. The two variables loading high on component six measure job meaningfulness and self-determination. In Study II a total of 22 indicators were used and six components emerged that explained 74% of the total variance (Table 6). Five indicators clearly belonged to the first component. In a broad sense these seemed to be about high work demands. The second component labelled as performance drivers had three clear indicators which were about efficient use of technology, workplace adaptability, and how well the workplace had developed customer oriented quality practices. The third component got its name from the only indicator with a clear loading - adequate competence. The fourth component was about general health performance. The indicators of the fifth component were mainly about respectful leadership and creativity. The sixth component was assigned the name knowledge of workplace goals.

51

Table 6. Loadings for 22 indicators on six components. Indicators 1

2

Component 3 4

5

6

High work demands

Psychosocially demanding work Work intensity demands Demands higher than control Data stress Sleep problems

-0.864 -0.819 -0.712 -0.668 -0.597

- 0.003 - 0.152 - 0.564 0.180 - 0.010

0.182 -0.009 0.106 0.483 0.535

-0.215 -0.239 -0.308 -0.109 -0.321

-0.145 -0.090 -0.160 0.251 -0.639 -0.068 -0.235 -0.068 -0.230 0.055

-0.018 0.812 0.006 0.370 0.804 -0.367 -0.001 0.762 0.045

0.131 0.343 0.011

0.178 -0.184 0.381 -0.099 0.083 -0.063

0.100 0.191 -0.802

0.023

0.027 -0.164

Performance drivers

Efficient use of technology Workplace adaptability Customer oriented quality practices Adequate competence

Adequate competence General health performance

Physical health Psychological well-being Mean absence days per year Depressed mood-worries Musculoskeletal symptoms

0.039 0.122 0.486 -0.006 -0.100 - 0.207 -0.401 0.133 -0.198 -0.293

0.153 -0.466 0.034 0.259 -0.318

0.834 0.072 0.125 0.702 0.265 -0.024 -0.696 -0.340 0.154 -0.686 0.071 0.442 -0.667 -0.321 -0.308

0.151 0.475 0.017 0.578 -0.449 -0.435 0.468

0.022 -0.372 0.253 -0.169 0.375 -0.369 -0.486

0.096 0.274 0.264 0.290 -0.456 -0.414 0.438

0.804 0.773 0.758 0.710 -0.541 -0.532 0.509

0.303 -0.257 0.063 0.129 0.539 -0.345 -0.396

0.167 - 0.072 -0.005

0.165

0.251

0.863

Respectful leadership and creativity

Decision latitude Respectful leadership Creative work Customer satisfaction Perceived lack of efficiency Restructuring worries Team spirit

-0.038 0.486 0.411 0.534 -0.180 -0.117 0.486

Knowledge of workplace goals

Knowledge of workplace goals Note: Loadings > 0.4 are in bold.

Structure of enterprise groups (Studies I, II, IV, V) The results discussed in this section concern similarities-differences between enterprises while the previous section was about similarities between variables. Figure 6, which may be found in the reprint of Study I, shows an example of an explorative grouping of companies with a tendency for profit per capita to covary with structure and leader change competence and the interaction between these two variables. In Study II a more comprehensive analysis is performed concerning grouping of enterprises using multidimensional scaling. The analysis gives a rather clear structuring in the two main dimensions “supportive and 52

efficient organization” (a combination of the indicators respectful leadership, team spirit, customer satisfaction and workplace adaptability, and in the opposite direction, demands higher than control, perceived lack of efficiency and sleep problems) and “general health performance” (a combination of five health indicators). In Study IV an explorative grouping of micro-enterprises were contrasted with a grouping of small public workplaces. Some of the results were that managers in the micro-enterprise group judged that working tasks, competence demands, responsibilities, overtime and sickness absence had increased less when compared to the judgments of managers from the public workplaces. Other results were that the micro-enterprises seem to have more organizational development aspects in their work and that they were focusing more on a production orientation than on work environment goals. In Study V it was possible to separate the studied small enterprises into one group that used a more broad change strategy and into one group that used a more expert/problem-based strategy. The leaders in the broad strategy enterprise group judged significant higher values on indicators for learning strategy, top management involvement and quality of working life. Altogether, the presented results in this sub-section show that it is possible to group the studied small enterprises by organizational aspects, and health and performance outcomes.

Change strategies and change goals (Studies III, IV) Study III consists, among other aspects, of a theoretical and conceptual discussion of change strategies in small enterprises and its relation to outcomes such as health, quality of working life and performance. The fact that small enterprises are a heterogeneous group together with the changing market environment and shortening of product life cycles necessitates an integrated planning of organizational strategies, technology, and human resources. This need for multifactor, integrated planning increases the complexity of organizational change processes. Except the need to integrate different perspectives, the importance of involving workers in broad change strategy approaches is discussed in the study. The fundamental point as to why a participative approach is deemed necessary is that those affected by change are best able to decide what is in their interest. The chance to participate in change, and thus influence a working environment, benefits the psychosocial climate and enhances quality of the improvements. The use of a learning strategy approach means that many kinds of simultaneous changes in different parts of an organization are included and that this participation could lead to an increasing competence when handling future changes. The learning strategy may lead to an increased quality of proposed solutions because both the leaders and the co-workers are active in the processes. 53

However, there is also a risk that change strategies used are too advanced for an organization and that the maturity required for the chosen change process is lacking. Main points in the study are the importance of integrating different theoretical perspectives and the use of participative learning-based change strategies adapted to specific circumstances in small enterprises. In Study IV, conditions in micro-enterprises are contrasted with those in small public workplaces. Surprisingly, many of the managers in the micro-enterprises estimated that profits, numbers of customers and products have increased over the last few years. There were also differences between the private and public workplaces concerning for instance, competence demands, sickness absence, turnover and quality outcomes. The managers of the micro-enterprises also stated that their employees alternated between work tasks to a great extent and developed existing work tasks by themselves. About half of the managers in the micro-enterprises (45 percent) answered that they had workplace-related competence development; the corresponding figure for the public workplaces was 20 percent. Most of the micro-firms did not use working environment, environment or quality systems. It is interesting to note that they still used quality systems more than did the public workplaces. The managers in the microenterprises stated that customers and professional contacts are important but that they used consultants less frequently than did the public workplaces. One confirmation of this result is that only 15 percent of these enterprises were cooperating with occupational health services in human resource development. In the following two tables (Tables 7 and 8) the managers’ opinions about planned or on-going change processes are presented. Table 7. Goal aspects concerning workplace health and work organization (Percentages of leaders rating goals as dominating/main goal). Goal

Micro-enterprise (N=66)

Public workplaces (N=50)

Improve psychosocial conditions

33

90

Improve vocational rehabilitation

0

70

Leadership development

29

46

Development of team-based work

38

52

Competence investments

38

62

Bold style: p < 0.01; differences between the groups are significant at the 0.01 level

The private workplaces seemed to attach less priority to goals that concerned workplace health and work organization than did the public ones. It is valuable to note that around a third of the micro-enterprises were formulating goals that 54

concerned psychosocial working conditions, leadership, development of teams and competence development (Table 7). Table 8 shows that the micro-enterprise managers attached more priority to goals connected with the tangible working environment, production and quality development. Table 8. Goal aspects concerning working environment, production and quality (Percentages of leaders meaning goals are dominating/main goal). Goal

Micro-enterprise (N=66)

Public workplaces (N=50)

Improve working environment

52

58

New production/technical equipment

53

13

Improve production flows

57

12

Improve quality

77

62

Bold style: p < 0.01; differences between the groups are significant at the 0.01 level

Effects of workplace interventions (Study V) Study V focuses on workplace interventions in ten small enterprises. From the leader questionnaire and the interviews it was possible to group the workplace interventions into two types: broad strategy and expert/problem-based strategy. The leaders in enterprises using a broad strategy in the workplace change processes had higher values on both the learning strategy and top management involvement index. They also judged that quality of working life aspects had improved to a greater extent when compared to responses from leaders of enterprises using a expert/problem-based strategy (all differences are significant using a t-test for independent samples). In Table 9 below the employees’ judgements of changes concerning job and organizational characteristics, and outcomes for health and performance are presented. They are grouped into the two change strategy types.

55

Table 9. Mean changes (M) in job and organizational characteristics and outcomes for health and performance for two change strategy types (n=102). Indicator

Broad strategy (N=31)

Expert/ problembased strategy

M

(N=71) M

t

Sig.

Customer oriented quality practices

-0.09

-0.49

-1.98

0.050

Respectful leadership

-0.02

-0.48

-2.52

0.014

Creative work

0.01

-0.44

-2.86

0.005

Team spirit

0.00

-0.27

-1.36

0.178

Work intensity demands (low)

-0.09

0.03

0.54

0.591

Data stress (low)

0.03

0.15

0.44

0.659

Decision latitude

-0.04

-0.04

0.01

0.991

When considering changes in mean values for the indicators used, there were no Restructuring significant differences within the enterprise group using a broad strategy. Within worries (low) 0.08 -1.34 the group using an expert/problem-based strategy there were -4.73 significant0.000 lower values at Time 2 for all indicators except for-0.43 the indicators-1.29 work intensity Self-assessed health -0.16 0.199 demands, data stress and decision latitude. When comparing mean changes for Workplace adaptability 0.08 -0.40 -2.66 0.009 indicators used by enterprises that adopted a broad strategy with enterprises Customer -2.06 0.042 having an satisfaction expert/problem based0.05 strategy, there are-0.28 significant differences When comparing mean changes for indicators used by enterprises that adopted a broad strategy with enterprises having an expert/problem based strategy there were significant differences concerning customer oriented quality practices, respectful leadership, creative work, restructuring worries, workplace adaptability and customer satisfaction (after performed interventions). From the leader interviews some comprehensive patterns for enterprises using a broad change strategy were identified. There were that several areas such as work organization aspects, leadership development and quality improvements combined those with individual measures concerning health promotion. The managers could be characterized as those who were involved, proactive and not afraid to take a stand. They also had a strong belief in connections between working environment, employee health, effectiveness and quality aspects. Other characteristics were that to a relatively high degree employees participated in the change processes and that several arenas for the development of workplaces exist. These leaders also tended look upon change processes as continuous improvements and that human resource issues must be integrated in with business 56

issues. Although the picture described above points towards several positive aspects, the leaders also identified barriers to long-term improvements. Lack of resources for development work, insufficient competence concerning successful measures and external factors were commonly mentioned obstacles. Studied enterprises that used a more expert/problem-based strategy worked with more narrow change processes and with only a couple of the measured areas. Even if the managers were participating in the processes, they delegated more to external experts and they didn’t have the same clear values concerning connections between human resource development and business outcomes. In some of these enterprises only a few employees participated. In some the strategy was more informative without any real opportunity for employees to influence change processes. It also appeared that the processes were more fragile and easily interruptible in these enterprises.

57

58

Discussion and conclusions In this thesis the main areas of interest are relationships between work organization factors, health and performance outcomes, and workplace change strategies in small enterprises. A purpose of the research projects that is partly covered by this thesis is to make this knowledge relevant for developing workplace change processes in small firms and to encourage further research into organizational factors and workplace health for small business development. In this final section the results of the five studies will be discussed in the light of the purpose and research questions addressed in the thesis. The results will also be related to some theoretical aspects and other research of relevance to workplace development in small enterprises.

Relationships between determinants and outcomes The first question in the thesis was if there are any relationships between outcomes related to health, sickness absence and performance in studied small enterprises. Three of the studies together help to answer this question. In Study II bivariate correlation analyses showed a rather strong correlation between the outcome indicators organizational performance and health performance, and between organizational performance and sickness absence. In Study II, the component respectful leadership and creativity, which consist of some performance measures, correlates rather strongly with the component general health performance. In Study IV, improvements concerning turnover, profits and complaints from customers took place over a two year-period; at the same time judged sickness absence had not increased in the studied micro-enterprises. Taken together these three findings may be interpreted as some evidence of a rather strong relationship between the two outcomes in the thesis analysis model. Much discussion among researchers and practitioners has focused on possible causal relations. One argument is that “richness creates health”- economically well-performing companies allocate resources for investments that lead to personnel health (Håkansson, 1995; Lindström, 2000). Another way of thinking (represented by, for instance, Huzzard, 2003 and Pfeffer, 1998) is that employee and workplace health development leads to more productive employees which in turn leads to better organizational performance. Thirdly, one might also consider the two outcomes to be interdependent. To give an answer as to which of these three possible causal relations is predominate other types of data than what is in this thesis is needed. At the very least, these results do not refute the ideas about a

59

link between health and performance in small firms as discussed in the introduction section and the theoretical framework section. The second question is about what relationships can be found between the outcome indicators of health and performance and indicators of assumed organizational determinants. Although the presented studies do not represent a systematic investigation carried out in the same way in all studies, a comprehensive observation is that several indicators of leadership, psychosocial work environment, learning, and change strategy correlate rather strongly with performance outcomes while there are fewer correlations with health outcomes. However, with respect to health as an outcome variable in Study I, an increase in information technology and more meaningful jobs tend to lower musculoskeletal (physical) ill health, while stress problems co-vary with musculoskeletal ill health. In Study II there were rather strong positive correlations between this outcome indicator and team spirit, and rather strong negative correlations with work intensity related indicators and restructuring worries. In Study V, correlation analyzes showed that self-assessed health related significantly in the assumed direction with the indicators of respectful leadership, creative work, team spirit, work intensity demands and data stress at both measuring points. Some interesting results were that in this paper there was no significant relationship between the health outcome variable and customeroriented quality practices, and that for the indicators decision latitude and restructuring worries a relationship was present only at the second measuring point. Concerning performance as an outcome concept, an interesting result in Study I was that four variables explained a large extent of the variation in profit per capita. These were an increase in leader change competence, a flat organization, a higher leader than co-worker change competence and an ongoing change program. In Study II the outcome indictor organizational performance was strongly or rather strongly related in the assumed direction to several assumed determinants. These were respectful leadership, creative work, adequate competence, team spirit, work intensity demands, psychosocially demanding work, data stress, demands higher than control, customer oriented quality practices and efficient use of technology. For the performance indicators in Study V there were strong relationships with the assumed determinants for customer oriented quality practices, respectful leadership, creative work and team spirit at both measuring points. For indicators concerning work intensity demands, data stress, decision latitude and restructuring worries there were weaker relationships or relationships at only one of the measuring points. A remarkable pattern is that many healthy work organizational indicators have rather strong relations with performance outcomes while there are weaker and fewer relations with health outcomes. This is somewhat contrary when compared 60

with many earlier research studies. Here some further comments are offered concerning this point. Much of the conceptual and empirical research on healthy work indicators has focused on larger companies. However, as has been noted in the literature review, small firms seem to have different patterns than do larger ones. One example of this is the “low sickness absence paradox” mentioned in the introduction section – although small companies have relatively high levels of injuries and small resources for work environment development, the sickness absence levels are much lower than in larger companies. In a similar way, it is possible that the relationships between many healthy work organizational indicators and health outcomes would be lower in small companies. A clarification of this issue would demand more developed studies about mechanisms in this area; this is beyond the scope of this thesis. However, the findings here might be seen as openings for further research. The findings on strong or rather strong relations between indicators of respectful leadership, team spirit and creative work, and health and performance outcomes are compatible with the research into the specific importance of these factors in small companies (Beckérus & Roos, 1985; Bornberger et al., 2003; Vinberg, 2003; Ylinenpää et al., 2006). The findings of rather strong relations between quality indicators and organizational performance are in line with other studies about quality development in small enterprises (Anderson & Sohal, 1999; Garvare, 2002; Hansson, 2003). The finding that there are no strong relations between quality indicators and health outcomes open possibilities for further research in this area. This result might also be seen as partly in line with the views of Axelsson (2000) and Bejerot & Hasselbladh (2002) to the effect that too intensive quality development work may cause stress reactions among employees. The clear difference in the relationship between self-assessed health and restructuring worries from Study V (when comparing the results at both measuring points) is possibly connected to general trends in working life about increased insecurity and organizational change as reported by for instance, Härenstam & Bejerot (2004). This general trend may also be relevant for small enterprises. As a summary comment it might be said that the rather large number of relationships between determinants and performance outcomes is compatible with most of the findings regarding the central areas mentioned in the literature review. That these results apply to small companies here can be described as a contribution to the rather small amount of small business research in these areas.

Structure of indicators and enterprise groups On the third question about if there is a bundling tendency for several organizational aspects to co-vary, the principal component analysis results give an partly affirmative answer. 61

However, since two quite different questionnaires and sets of indicators were used in the two studies a discussion comparing the emerging components in the two papers is difficult. Some similarities can be pointed out. Component 1 in Study II, high work demands, roughly corresponds to component 3 in Study I, stress and high work demands. Component 2 in Study II, performance drivers, has some correspondence to component 1, high performance organizational aspects, and component 4, change propensity, in Study I. Component 4 in Study II, general health performance, roughly corresponds to component 5 in Study I, ergonomic and musculoskeletal problems. Component 5 in Study II, respectful leadership and creativity, roughly corresponds to component 6 in Study I, job meaningfulness and self-determination. As a comprehensive comment one might say that the analysis in both studies shows that indicators tend to group into larger bundles; although the content of these bundles varies due to differences between the sets of input variables. Thus, these results complement studies by, for instance, McDuffie (1995), Whitfield (2000) and Wood & Albanese (1995) – it is noteworthy that the results here for small companies have similarities with earlier research results for larger companies. It is also interesting that the multidimensional scaling results in Study II shows that it is possible to find groups of companies that are similar in the sense that they have similar indicator patterns concerning health and performance related dimensions. The analysis shows that many aspects related to performance – or a ‘supportive and efficient organization’ as the dimension is named in the study – are inter-related and also that several health aspects co-vary to form the dimension called ‘general health performance’. The dimensional analysis leads to a grouping of enterprises and, for instance, an identification of enterprises with high positions on both dimensions versus enterprises with low positions on both dimensions. Also in Study IV, it was possible to find differences between studied micro-enterprises and small public workplaces. In Study V it was possible to group companies by change strategies used. These findings relate closely to recent research by for instance, Ketchen et al. (1997), Lepak & Snell (2002) and Way (2002) on configurations of companies.

Change strategies and change goals The findings regarding the fourth question about the needs for change and change goals in studied small enterprises showed that the studied private workplaces attached less priority to goals related to workplace health and work organization when compared to the studied public ones. Still, it should be noted that around a third of the small enterprises were formulating goals about psychosocial working conditions, development of leadership, team based work and competence. The micro-enterprises attached more priority to goals connected to the physical working environment, production and quality development. The micro62

enterprise managers also rated contacts with consultants as less important and most of the enterprises had no formal relation with any occupational health services; this is consistent with other research about working environment conditions in small enterprises (Antonsson et al., 2002; Bornberger-Dankvardt et al., 2003; Walters, 2001). That micro-enterprises have a more technically oriented change focus and that they do not use structure systems or consultants to a great extent are probable indicators of the need for more integrated, flexible and learning-based models that are adapted to small enterprises. This result is in line with other Swedish studies by Axelsson (2002), Garvare (2002) and Johansson (1998, 2006). An interesting, and in a way a paradoxical result, is that the micro-enterprises have low sickness absence although they say that they are prioritizing productionoriented development much more than measures associated with work organization and working environment. However, in their interviews the microenterprise mangers judged greater emphasis on developing organizational aspects when compared the comments of the public managers. This difference could be one explanation for the low sickness absence figures in the small enterprises. Low levels of sickness absence have been confirmed in both Swedish (Bornberger-Dankvart et al., 2003) and European studies (Walters, 2001). The explanations for these differences in sickness vary among both practitioners and researchers. According to Wilkinson (1999), there is a polarized picture of working conditions in small firms. One view is that there are harmonious relations between employees and leaders. Conflicts are unusual and work is flexible. The opposite view states that the leadership is authoritarian and working conditions are unsatisfactory with high injury risks. Either of these simplified views of small enterprises can be questioned; it is more likely that conditions are complex and contradictory. Therefore it is important in future research to identify the processes and mechanisms around the low sickness absence figures in small enterprises.

Effects of workplace interventions For the fifth question - “Are there differences concerning changes in organizational determinants and outcomes of health and performance when using different types of change strategies” - the results in Study V showed that there were differences between groups of enterprises using a broad strategy in contrast to an expert/problem-based strategy. The managers in the former group estimated higher values on both the learning strategy and top management involvement index. They also judged that quality of working life aspects improved to a greater extent than was judged by leaders of enterprises using an expert/problem-based strategy. When considering changes in mean values for the indicators used, there were significant differences in favour 63

for the broad strategy group that concerned customer oriented quality practices, respectful leadership, creative work, restructuring worries, workplace adaptability and customer satisfaction. However, as it was not possible to identify significant positive changes within the broad strategy group before and after the interventions the findings must be interpreted with caution. It is possible that the enterprises that used a broad strategy had “balanced” a general deterioration in job and organizational characteristics. Researchers from different specialities give support to the idea that more holistic change processes with a high degree of participation by both leaders and employees are more effective (e.g. Gustavsen et al., 1995; Håkansson, 1995; Ingelgård, 1998). The results about differences concerning changes in indicator means between the two change strategy groups is in accordance with a Finnish study of workplace interventions in small enterprises (Lindström et al., 2000) and could be interpreted to mean that to some extent a broader change strategy applies to small enterprises. However, the leader interview results point to common obstacles such as a lack of resources for development work, insufficient competence concerning successful measures and external factors; for example, changes in customer relations and market conditions. These findings are in line with other research concerning change processes in small enterprises (e.g. Eakin et.al, 2000; Johansson, 2006; Vinberg, 2004) that point to the need for integrated and flexible models for workplace development in small enterprises. The findings regarding research questions three, four and five have a bearing on the sixth question that concerns the design of workplace change processes. These findings may give some support to those working with a concept-driven or holistic approach, which integrates several aspects instead of focusing on a particular process in isolation. They also complement, for small companies, McDuffie’s (1995) point that innovative HR practices affect performance not individually but as interrelated elements in an internally consistent HR “bundle” or system. These results also coincide with findings by Gustavsen (1995) about the importance of holistic concept-driven change strategies and change programs. The bivariate results for several relationships between determinants and outcomes also support this holistic view. However, such view must be adapted to the particularly circumstances that exist in small enterprises such as a lack of resources for development work and their strong focus on things as they are at the present.

Methodological considerations Most studies in the small business research only partially focus on organizational factors - data is commonly collected through interviews with managers and/or from performance data. Three of the studies in this thesis are innovative in the 64

sense that they contain responses from both employees and company leaders with relatively high response rates when compared to many other small business studies (Curran & Blackburn, 2001, pp. 69-71). Also, one of the studies has a longitudinal design in which both small business researchers (Davidsson, 2004) and working life researchers (e.g. Kompier et al., 2000; Westgaard & Winkel, 1997) argue for future research. Still, it must be stressed that the data in this thesis is mainly cross-sectional and the results should therefore be interpreted with caution and be seen as mainly explorative. In the thesis, causal language has been avoided when possible and a grouping of indicators into determinants and outcomes and a structural perspective have been applied instead of cause-effect reasoning. It must also be stressed that the collection of companies used in the studies is not a representative sample. However, a goal in all thesis studies has been to recruit companies from different sectors; particularly from those sectors that are dominated by small enterprises (Antonsson et al., 2002; Spilling, 2000). Outcomes are mainly measured by survey questions. The measurement problems, which may have been created by the use of such survey questions, have to some extent been remedied by the rather detailed multivariate indicator constructions reported in this work. In future research, however, financial and other objective performance information as well as health outcome data from registers and human resource accounting systems should be used as a complement. In Study I, there is sometimes a mix of organizational and individual data, and therefore a risk for ecological and/or atomistic fallacies (Diez-Roux, 1998). For instance, when organizational data is used for individual level data analyses the lack of intra-company variation might bias the estimates. Results from some small sensitivity analyses indicated that the possible amount of such bias may be small. However, in forthcoming research with larger data sets it is suitable to use hierarchical multilevel regression methods as well as structural modelling of relations between several outcomes as these methods may be more adequate. In Studies II and V several questionnaires used in other research studies were used to gain an understanding of leaders and co-workers judgements of aspects of leadership, psychosocial working environment, learning, and health and performance outcomes. Based on substantive considerations and factor analyses several indices were constructed to provide a comprehensive picture related to the thesis analyses model. Although there is a risk for several measurement errors due to statistical or more psychological reasons (Wikman, 2005), constructed components and indices are adequate as indicators of studied determinants and outcomes. The reason for using aggregated individual data in some of the thesis studies was that the focus here was on organizational level aspects and not on individual ones (Klein & Kozlowski, 2000). Several researchers also state that subjective measures of health and performance related outcomes are of relevance

65

and related to more objective measures (Bowling, 2005; Kotey & Meredith, 1997). Although the longitudinal design in Study V is a strength and made it possible to follow the same individual before and after workplace interventions, some considerations must be noted. The sample of individuals and firms was small, and it is difficult to know after what time-period different changes and effects will arise. For instance, changes in health status may not be known until several years later (Nutbeam, 2000). Therefore, in future studies it is important to measure changes in indicators at several measuring points during a longer time-period. In Studies I, IV and V, semi-structured interviews with managers and leaders were performed. Nearly all interviews took place at the studied workplaces and were performed by the same researchers. To combine both quantitative and qualitative methods is more and more recommended by researchers studying workplace change processes (e.g. Kompier et al., 2000; Westgaard & Winkel, 1997) and small businesses (Curran & Blackburn, 2001; Davidsson, 2004).

Conclusions and implications for practitioners and future research The conclusions of this thesis and the implications for practitioners and future research can be summarized as follows. The thesis studies have given evidence of links between organizational factors such as leadership, learning, psychosocial working environment and change strategy, and health and performance outcomes in small enterprises. That relationships also exist between different indicators of the two outcome concepts supports the idea that a model of healthy work organizations might be applicable to small enterprises. However, there are also some contradictory results; some of the determinants are more or not significantly related to one of the outcome concepts and the mechanisms beyond these results must be dealt with in future research. The fact that studied micro-enterprises had a more technically oriented change focus (although to some extent they were prioritizing workplace health related aspects and were using systems for working environment issues) points to the need for more integrated and flexible models for workplace development that are adapted to the smaller workplaces. The thesis results about obstacles for workplace change processes support this statement. The findings of rather strong relationships between, on the one hand, aspects of leadership, team spirit, work intensity demands and restructuring worries, and, on the other hand, health and performance outcomes point to the importance of action towards increasing leader change competence and developing work organization aspects in small enterprises. Important when working with such actions is to reflect on if and how the results of good development work aspects

66

might contribute to relatively low sickness absence figures and health outcomes in small enterprises. The results concerning changes of determinants and outcomes due to the use of different change strategies, indicates that multi-component change processes with a broad strategy, and high employee and top management involvement can apply to small enterprises and help to improve their health and performance. The fields covered in the thesis are in need of conceptual development and integrative analysis to guide any specific analyses of a small enterprise. The findings give evidence of possible interest to practitioners working with small firm workplace health promotion and organizational development. The most important finding in this respect seems to be the inter-relatedness of a large number of factors. The fact that it was possible to single out enterprise segments with similar positions on general dimensions gives credibility to change strategies oriented towards multi-purpose change processes. One observation is that both practitioners and researchers underestimate organizational factors and health related issues when discussing entrepreneurial orientation and small business development. Below, some suggestions for future research are presented. First, there is a need for larger studies based on representative probability samples with a multi-level design using both subjective and objective measures (e.g. financial data and register sickness absence data), which investigates relationships between organizational determinants and outcomes related to health and performance in small enterprises. Furthermore, studies with a longitudinal design over longer time-periods might facilitate studies of causal relationships between these determinants and outcomes. Second, there is a need for similar approaches when studying effects of different workplace change processes as a base for developing integrated models applicable to small companies. Combining both quantitative and qualitative methods will probably give more relevant results in such studies. Third, there is a need to study mechanisms beyond why some groups of small enterprises have high values on both health and performance related outcomes, while others do not. These studies could also contribute to model development and advice for practitioners. Forth, there is a need to find out more about how company size itself is a determinant of aspects related to work organization, and health and performance outcomes. Several approaches are possible; study and comparison of different enterprise sizes, study changes when small enterprises grow, and study changes when large enterprises are divided into smaller independent organizational units. Knowledge in this field might also be useful for work organization development within larger enterprises and organizations. Fifth, there is a need to investigate the mechanisms beyond ‘the sickness absence paradox’, i.e. the low levels of sickness absence in small enterprises in 67

comparison with larger workplaces despite the former group’s higher levels of injuries and lack of resources for health and working environment improvements. In the introduction to this thesis one stated ambition concerning the findings was that they should provide practitioners with information that can contribute to worker and leader health, and to help small enterprises to be more effective. Here, in conclusion and from a more pragmatic point of view, some guidelines for those working with workplace development and improvements in small firms are made. • Do not underestimate the importance of organizational factors such as aspects of leadership, psychosocial work environment and learning for small firm performance and growth. Particularly, analyze and find measures as to how to improve leadership, team spirit, creative work, work intensity demand aspects, and leader and co-worker change competence. • Consider that small companies are different concerning working environments, work organizations and human resources; thus develop operation-specific strategies for enhancing health and performance outcomes. Find company in-house models for analyzing improvement needs that suit the specific workplace, prioritize resources and arenas for development, and try to integrate issues related to occupational health and safety, workplace health and rehabilitation, and quality development. These issues must be paid attention to also when working with more technically and production oriented change processes. • Focus on both process and content plus try to use holistic approaches for workplace improvement actions. Of great importance are broad change strategies with high involvement from both leaders and co-workers for planning and creating of measures. It can be critical to combine individual and organizational focused measures for workplace development, and to use support from competent external consultants - although the processes must be driven by individuals and teams inside the company. • To work with workplace change development is a learning process, which in it self can contribute to positive outcomes concerning work environment, health and performance. Although, achieving positive results are an important driving force for further change, it is also important to work with long-term and sustainable processes owing to, as for example, positive health effects do not grow rapidly. • Although the sickness absence is lower in small enterprises than in large ones it is important to develop measures for prevention of ill health and promotion of health for both leaders and co-workers in the smaller workplaces. In this area, support from consultants as e.g. Occupational Health Services for developing relevant health measures and routines for vocational rehabilitation can be useful.

68

•

Due to limited resources for workplace development in small enterprises, not least in rural areas, there is a need for developing networks between small enterprises, and cooperation between these, local authorities, consultants and researchers. The network strategy is probably underestimated for developing of work environment and workplace health in small enterprises, and could be effective also for encouraging firms with lower motivation for and belief in enhancing work environment, work organization and workplace health.

69

70

References Abrahamsson, L. (2000). Production economics analysis of investment initiated to improve working environment. Applied Ergonomics, 31, 1-7. Abrahamsson, L. (2006). Småföretag – hjältestory, familjeaffär eller ren business? [Small Business – Hero’s story, family affair or pure business?]. In H. Ylinenpää, B. Johansson & J. Johansson (Eds.) Ledning i småföretag (pp. 199-230). Lund: Studentlitteratur. Ahlberg-Hultén, G. (1999). Psychological Demands and Decision Latitude within Health Care Work (pp. 7-10). Doctoral Dissertation. Stockholm: Stockholm University. Alpenberg, J. & Karlsson, F. (1996). Arbetsmiljöinvesteringar i mindre företag – projektering och uppfattade effekter utifrån Arbetslivsfondens satsningar [Work environment investments in small enterprises – planning and judged effects according to investments of the Working Life Fund]. Licentiatavhandling. Lund: Lunds universitet. Alvesson, M. & Sköldberg, K. (1994). Tolkning och reflektion. Vetenskapsfilosofi och kvalitativ metod [Interpretation and reflection. The philosophy of science and qualitative method]. Lund: Studentlitteratur. Andersen, J. A. (1994). Ledelse og effektivitet [Leadership and effectiveness]. Doctoral Dissertation. Lund: University Press. Anderson, M. & Sohal, A. S. (1999). A study of the relationship between quality management practices and performance in small business. International Journal of Quality & Reliability Management, 16(9), 859-877. Anderzén, I. & Arnetz, B. (2005). The Impact of a Prospective Survey-Based Workplace Intervention Program on Employee Health, Biologic Stress Markers, and Organizational Productivity. Journal of Occupational and Environmental Medicine, 47, 671-682. Antonsson, A. B., Birgersdotter, L. & Bornberger-Dankvardt, S. (2002). Small enterprises in Sweden. Health and safety and the significance of intermediaries in preventive health and safety. Work and Health 2002:1. Stockholm: National Institute for Working Life. Anttila, M. (1997). Kompetensförsörjning – företagets viktigaste resurs [Creating an environment that fosters competence – The business’s most important resource]. Stockholm: Ekerlids förlag. Armstrong, M. (2001). A Handbook of Human Resource Management Practice. London: Kogan Page Ltd. Aronsson, G. & Lindh, T. (2004). Långtidsfriskas arbetsvillkor. En populationsstudie. [Work conditions among workers with good long term health. A population study]. Arbete och hälsa [Work and Health] 2004:10. Stockholm: Arbetslivsinstitutet. Aronsson, T. & Malmquist, C. (2002). Rehabiliteringens ekonomi [The economy of rehabilitation]. Smedjebacken, Sweden: Bilda Förlag. 71

Aronsson, G., Svensson, L., Leksell, K. & Sjögren, A. (1995). Förändringskompetens. Projektledares erfarenheter från 300 Arbetslivsfondsprojekt [Change competence. Project manager experiences from 300 Swedish National Working Life Fund projects]. Stockholm: Arbetslivsinstitutet. Arthur, J. B. (1992). The Link Between Business Strategy and Industrial Relations Systems in American Steel Mini-mills. Industrial and Labor Relations Review, 45, 488-506. Arvonen, J. (2002). Change, Production and Employees – An integrated model of leadership. Doctorial Dissertation. Stockholm: Stockholm University. Axelsson, J. (2000). Quality and Ergonomics. Towards successful integration. Dissertation No. 616, Linköping Studies in Science and Technology. Linköping: Linköping University. Axelsson, C. (2002). Formalisering som hinder och möjlighet i små företag: en studie av arbetsmiljöarbete ur ett lärandeperspektiv [Hindrances and opportunities in small business due to formalism: A study of work environment research from a learning perspective]. Doktorsavhandling 2002:14. Luleå: Luleå tekniska universitet. Bager-Sjögren, L., Gustafsson, O. & Pettersson, L. (2003). Växtkraft Mål 4 – en effektstudie. Kompetensutveckling i små företag [Growth Policy Goal 4 – a study of the effects of competence development in small businesses]. Rapport A2003:001. Östersund: Institutet för tillväxtpolitiska studier . Baldwin, J., Chandler, W., Lee, C. & Papailiadis, T. (1994). Strategies for Success: A Profile of Growing Small- and Medium-sized Enterprises (GSMEs) in Canada. Ottawa: Statistics Canada. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17, 99-120. Bass, B. M. & Avolio, B. A. (1994). Improving organizational effectiveness through transactional leadership. Thosands Oaks: Sage. Beckar, B. & Gerhart, B. (1996). The impact of human resource management on organizational performance: progress and prospects. Academy of Management Journal, 39, 779-801. Beckéreus, Å. & Roos, B. (1985). Affärer som livsstil – om glädje och tyranni i småföretagandets värld [Businesses as life style – on pleasure and tyranny in the small business world]. Malmö: Liber. Bejerot, E. & Hasselbladh, H. (Eds.) (2002). Kvalitet utan gränser. En kritisk belysning av kvalitetsstyrning [Quality without limits. A critical illustration of quality control management]. Lund: Academica Adacta AB. Bergquist, T. M. & Ramsing, K. D. (1999). Measuring performance after meeting award criteria. Quality Progress, 32(9), 66-72. Bildt, C., Vinberg, S. & Fjell, Y. (2005). Arbetsmiljöutveckling inom offentlig sektor. Friskare arbetsplatser i Jämtlands län [Work environment development in the public sector. Healthier workplaces in Jämtland County]. Arbetsliv i omvandling [Work Life in Transition] 2005:9. Stockholm: Arbetslivsinstitutet.

72

Bjørnstad, O. (2004). Health promoting for small-scaled enterprises. In T. Leino & K. Olsson (Eds.) Workplace Health Promotion – Practice and Evaluation (pp. 39-44). Helsinki: Finnish Institute of Occupational Health. Bornberger-Dankvardt, S., Ohlson, C-G., Andersson, I-M. & Rosén, G. (2005). Arbetsmiljöarbete i småföretag – samlad kunskap samt behov av forskning och utvecklingsinsatser [Work environment management in small enterprises – assembled knowledge and needs for research and development investments]. Arbete och hälsa [Work and Health] 2005:6. Stockholm: Arbetslivsinstitutet. Bornberger-Dankvardt, S., Ohlson, C-G. & Westerholm, P. (2003). Arbetsmiljö- och hälsoarbete i småföretag - försök till helhetsbild [Work environment and occupational health in small business – seeking the full picture]. Arbetsliv i omvandling [Work Life in Transition] 2003:1. Stockholm: Arbetslivsinstitutet. Bostedt, G., Lundberg, L. & Okoutsoglou, A. (1995). Drivkrafter för förändringsarbete rörande arbetsmiljö och arbetsorganisation i privata företag [Driving forces for change process work regarding work environment and organization in private business]. Stockholm: Arbetslivsfonden. Bowling, A. (2005). Measuring Health. A review of quality of life measurement scales. Berkshire: Open University press. Boxall, P. (1996). The strategic HRM debate and the resource-based view of the firm. Human Resource Management Journal, 6(3), 59-75. Breucker, G. (2001). Small, Healthy and Competitive. New Strategies for Improved Health in Small and Medium-Sized Enterprises. Report on the Current Status of Workplace Health Promotion in Small and Medium-Sized Enterprises (SMEs). Essen: Federal Association of Company Health Insurance Funds. Bridge, S., O’Neill, K. & Cromie, S. (1998). Understanding Enterprise, Entrepreneurship and Small Business. Hampshire: MacMillan Press Ltd. Brülde, B. & Tengland, P-A. (2003). Hälsa och sjukdom – en begreppslig utredning [Health and illness – a conceptual analysis]. Lund: Studentlitteratur. Cascio, W. F. (1995). Managing human resources: productivity, quality of work life, profits. New York: McGraw-Hill, Inc. Chaganti, R., Cook, R. G. & Smeltz, W. J. (2002). Effects of Styles, Strategies, and Systems on the Growth of Small Business. Journal of Developmental Entrepreneurship, 7(2), 175-192. Champoux, D. & Brun, J-P. (2003). Occupational health and safety management in small size enterprises: an overview of the situation and avenues for intervention and research. Safety Science, 41, 301-318. Chu, C., Breucker, G., Harris, N., Stitzel, A., Gan, X., Gu, X. & Dwyer, S. (2000). Health-promoting workplaces – international settings development. Health Promotion International, 15(2), 155-167. Cox, T. (1997). Workplace health promotion. Work & Stress, 11(1), 1-5. Cox, T. & Thomson, L. (2000). Organizational healthiness, work-related stress and employee health. In P. Dewe, M. Leiter & T. Cox (Eds.) Coping, health and organizations. London: Taylor & Francis.

73

Curran, J. & Blackburn, R. A. (2001). Researching the Small Enterprise. London: SAGE Publications. Davidsson, P. (2004). Researching entrepreneurship. New York: Springer Science+Business Media, Inc. Davidsson, P., Delmar, F. & Wiklund, J. (Eds.) (2001). Tillväxtföretagen i Sverige [Small Business Growth Firms in Sweden]. Stockholm: SNS förlag. De Kok, J. & Uhlander, L. M. (2001). Organization Context and Human Resource Management in the Small Firm. Small Business Economics, 17, 273-291. Delaney, J. T. & Huselid, M. A. (1996). The impact of human resource management practices on perceptions of organizational performance. Academy of Management Journal, 39(4), 949-69. Diex-Roux, A. V. (1998). Bringing context back into epidemiology: variables and fallacies in multilevel analysis. American Journal of Public Health, 88, 216-222. Docherty, P., Forslin, J. & Shani, A. B. (2002). Creating sustainable work systems. Emerging perspectives and practice (pp. 3-14). London: Routledge. Ds 2001:28. Långsiktig verksamhetsutveckling ur ett arbetsmiljöperspektiv. En handlingsplan för att förnya arbetsmiljöarbetet [Long-term development activity from a work environment perspective: An action plan for the reinvigoration of work environment interventions]. Stockholm: Fritzes förlag. Eakin, J. M. & MacEachen, E. (1998). Health and the social relations of work: a study of the health-related experiences of employees in small workplaces. Sociology of Health & Illness, 20(6), 896-914. Eakin, J. M., Lamm, F. & Limborg, H. J. (2000). International perspective on the promotion of health and safety in small workplaces. In K. Frick, P. Langaa Jensen, M. Quinlan & T. Wilthagen (Eds.) Systematic occupational health and safety management (pp. 227-247). Oxford: Elsevier Science Ltd. Easton, G. S. & Jarell, S. L. (1998). The effects of total quality management on corporate performance: an empirical investigation. Journal of Business, 71(2), 253307. Eklund, J. (1995). Relationships between ergonomics and quality in assembly work. Applied Ergonomics, 26(1), 15-20. Eklund, J. (2000). Development work for quality and ergonomics. Applied ergonomics, 31, 641-648. Ekman Philips, M. (Ed.) (2002). Dialog över etablerade gränser. Om organisationsutveckling i sjukvården [A dialogue concerning going beyond traditional limits. On organizational development in medical care]. Arbetsliv i omvandling [Work Life in Transition] 2002:9. Stockholm: Arbetslivsinstitutet. Evans, V. & Johnson, D. J. (1990). The relationship of principal’s leadership behavior and teacher’s job satisfaction and job related stress. Journal of Instructional Psychology, 17, 11-18. EU, (2003). OJL (Official Journal of the European Union). L124/37-41, 20 maj 2003. Retrieved November, 6, 2004, from http://europa.eu.int/eur-lex/en/oj/.

74

European Agency for Safety and Health at Work, (2004). Promoting health and safety in European Small and Medium-sized Enterprises. Luxembourg: Office for Official Publications of the European Communities. European Network for Workplace Health Promotion, (1997). The Luxembourg Declaration on Workplace Health Promotion in the European Union. (Adopted in Luxembourg on November 1997, updated in June 2005). European Network for Workplace Health Promotion, (2001). The Lisbon Statement on Workplace Health in Small and Medium-Sized Enterprises (SMEs). (Adopted in Lisbon, June 2001). Frick, K., Langaa Jensen, P., Quinlan, M. & Wilthagen, T. (Eds.) (2000). Systematic occupational health and safety management. Oxford: Elsevier Science Ltd. Gardell, B. (1982). Scandinavian research on stress in working life. International Journal of Health Services, 12(1), 31-41. Garvare, R. (2002). Process Management and Sustainable Development in a Quality Perspective – Implementation and Measurement Related to Small and Medium Sized Enterprises. Doctoral thesis 2002:35, Division of Quality Technology and Statistics. Luleå: Luleå University of Technology. Greenan, K., Humphreys, P. & McIvor, R. (1997). The green initiative: improving quality and competitiveness for European SMEs. European Business Review, 97(5), 208-214. Greenley, G. E. (1994). Strategic Planning and Company Performance: An Appraisal of the Empirical Evidence. Scandinavian Journal of Management, 10(4), 383-396. Griffin, M. A., Hart, P. M. & Wilson-Evered, E. (2000). Using employee opinion surveys to improve organizational health. In L. R. Murphy & C. L. Cooper (Eds.) Healthy and Productive Work. An international perspective (pp. 15-36). London: Taylor & Francis. Griffin, B. L., Hall, N. & Watson, N. (2005). Health at work in small and medium sized enterprises. Issues of engagement. Health Education, 105(2), 126-141. Guest, D. E. & Hoque, K. (1995). An assessment and further analysis of the 1990 Workshop Industrial Relations Survey. In D. E. Guest (Ed.) The Contribution of Personnel Management to Organizational Performance. London: IPD. Gustavsen, B. (2002). Mellom tekst og virkelighet: Samarbeid om utvikling mellom bedrifter og forskning [Between text and reality: Co-operation about development between enterprises and research]. Oslo: Norges forskningsråd. Gustavsen, B., Hofmaier, B., Ekman Philips, M. & Wikman, A. (1995). Utvecklingslinjer i arbetslivet och Arbetslivsfondens roll [Lines of development in working life and the role of the Swedish National Fund for Working Life]. Stockholm: SNS Förlag. Göransson, S., Aronsson, G. & Melin, B. (2002). Vilja och villkor för återgång i arbete – en studie an långtidssjukskrivnas situation [Will and conditions for return to work – a study of the situation for individuals on long-term sickness absence]. Bilaga 2:3 i SOU 2002:5, En handlingsplan för ökad hälsa i arbetslivet [An action plan for increased health in working life]. Stockholm: Fritzes.

75

Haksever, C. (1996). Total quality management in the small business environment. Business Horizons, 39(2), 33-41. Hamel, G. & Prahalad, C. K. (1989). Strategic intent. Harvard Business Review, MayJune, pp. 63-76. Hansson, J. (2003). Total Quality Management – Aspects of Implementation and Performance. Investigations with a Focus on Small Organisations. Doctoral thesis 2003:07, Division of Quality & Environmental Management. Luleå: Luleå University of Technology. Harden, A., Peersman, G., Oliver, S., Mauthner, M. & Oakley, A. (1999). A systematic review of the effectiveness of health promotion interventions in the workplace. Occupational Medicine, 49(8), 540-548. Heckman, J. J., Ichimura, H., Smith, J. & Todd, P. (1996). Sources of selection bias in evaluating social programs: An interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation tool. Proceedings of the National Academy of Science (Vol. 93, 13416-13420), USA. Hendricks K. B. & Singhal, V. R. (1997). Does implementing an effective TQM program actually improve operating performance? Empirical evidence from firms that have won quality awards. Management Science, 43(9), 1258-1274. Henrekson, M. (2001). The Entrepreneur and the Swedish Model. In M. Henrekson, M. Larsson & H. Sjögren (Eds.) Entrepreneurship in Business and Research (pp. 139164). Stockholm: Institute for Research in Economic History: Probus förlag. Hensing, G., Alexanderson, K., Allebeck, P. & Bjurulf, P. (1998). How to measure sickness absence? Literature review and suggestion of five basic measures. Scandinavian Journal of Social Medicine, 26(2), 133-144. Hill, R. & Stewart, J. (2000). Human resource development in small organizations. Journal of European Industrial Training, 24, 105-117. Holmquist, C. (2006). Entreprenörskap och ledning [Entrepreneurship and management]. In H. Ylinenpää, B. Johansson & J. Johansson (Eds.) Ledning i småföretag (pp. 25-48). Lund: Studentlitteratur. Huselid, M. A. (1995). The impact of human resource management practices on turnover, productivity and corporate financial performance. Academy of Management Journal, 38(3), 635-672. Huselid, M. A. & Becker, B. E. (1996). Methodological issues in cross-sectional and panel estimates of the human resource-firm performance link. Industrial Relations, 35(3), 400-422. Huzzard, T. (2003). The Convergence of the Quality of Working Life and Competitiveness. A Current Swedish Literature Review. Work Life in Transition 2003:9. Stockholm: National Institute for Working Life. Håkansson, K. (1995). Förändringsstrategier i arbetslivet [Change strategies in working life]. Monograph No. 57, Göteborg: University of Göteborg, Department of Sociology. Hård af Segerstad, S. (2000). Föreställningar om att leda småföretag [Conceptions about small enterprise leadership]. Licentiatavhandling [Licentiate Thesis]. Linköping: Linköpings universitet.

76

Härenstam, A. & Bejerot, E. (2004). Multilevel analyses of organizational change and working conditions in public and private sector. European Journal of Work and Organizational Psychology, 13(3), 305-343. Ichinowski, C., Shaw, K. & Prennushi, G. (1995). The effects of Human Resource Management Practices on Productivity. NBER Working Paper 5333, Cambridge MA. Ingelgård, A. (1998). On Macroergonomics and Learning Strategies in Improving Working Conditions. Ph.D. Dissertation, Göteborg: Göteborg University. Ingelgård, A. & Norrgren, F. (2001). Effects of change strategy and top-management involvement on quality of working life and economic results. International Journal of Industrial Ergonomics, 27, 93-105. Jeding, K., Hägg, G. M., Marklund, S., Nygren, Å., Theorell, T. & Vingård, E. (1999) (Eds.). Ett friskt arbetsliv. Fysiska och psykosociala orsakssamband samt möjligheter till prevention och rehabilitering [A healthy work life. Physical and psychosocial relationships and possibilities for prevention and rehabilitation]. Arbete och Hälsa [Work and Health] 1999:22. Stockholm: Arbetslivsinstitutet. Jeding, K. & Theorell, T. (1999). Arbetets uppläggning och hälsa [Organizing of work and health]. In K. Jeding, G. M. Hägg, S. Marklund, Å. Nygren, T. Theorell & E. Vingård (Eds.) Ett friskt arbetsliv. Fysiska och psykosociala orsakssamband samt möjligheter till prevention och rehabilitering [A healthy work life. Physical and psychosocial relationships and possibilities for prevention and rehabilitation]. Arbete och Hälsa [Work and Health] 1999:22 (pp. 37-53). Stockholm: Arbetslivsinstitutet. Johannisson, B. & Landström, H. (Eds.) (1999). Images of Entrepreneurship and Small Business – Emergent Swedish Contributions to Academic Research. Lund: Studentlitteratur. Johannisson, B. & Lindmark, L. (Eds.) (1996). Företag, Företagare, Företagsamhet [Business, Entrepreneurship, Initiative]. Lund: Studentlitteratur. Johanson, U. (1997). The profitability of investments in work life-oriented rehabilitation. A measurement of perceptions. Personnel Review, 26(5), 395-415. Johansson, B. (1998). En motivationsbaserad analys av arbetsmiljöarbetet i en grupp av mindre tillverkande företag [A motivation based analysis of the working environment activities in a group of small manufacturing enterprises]. Doktorsavhandling 1998:36, Luleå: Luleå tekniska universitet. Johansson, B. (2006). Arbetsmiljöledning i småföretag [Occupational health and safety management in small enterprises]. In H. Ylinenpää, B. Johansson & J. Johansson (Eds.) Ledning i småföretag (pp. 199-230). Lund: Studentlitteratur. Johansson, J. (1999). A Survey of Swedish Work Environmental and Occupational Research during the Twentieth Century. Human Factors and Ergonomics in Manufacturing, 9(4), 343-356. Johansson, J. & Johansson, B. (1990). Arbetsmiljöarbete i de små företagen [Working environment work in small enterprises]. Bilaga A till Arbetsmiljökommissionens betänkande SOU 1990:49 [Supplement A to the Working Environment Commission report]. Stockholm: Allmänna förlaget.

77

Johnson, J. V. & Hall, E. M. (1988). Job strain, workplace social support and cardiovascular disease: A cross-sectional study of a random sample of the Swedish working population. American Journal of Public Health, 78, 1336-1342. Johnson, J. V. & Johansson, G. (1991). Work Organisation, Occupational Health and Social Change. New York: Baywood Publishing. Julien, P-A. (1998). The State of the Art in Small Business and Entrepreneurship. Aldershot: Ashgate Publishing Ltd. Karasek, R. (1979). Job demands, job decision latitude and mental strain: Implications for job redesign. Administrative Science Quarterly, 24, 285-308. Karasek, R. & Theorell, T. (1990). Healthy Work: Job Stress, Productivity and the Reconstruction of Working Life. New York: Basic Books. Karasek, R. (1998). The New Work Organization. Conductive Production and Work Quality Policy. In M. Marmot & E. Ziglio (Eds.) Labor Market Changes and Job Insecurity: A Challenge for Social Welfare and Health Promotion. Copenhagen: WHO/Europe Publication. Ketchen, D. J., Combs, J. G., Russell, C. J., Shook, C., Dean, M. A., Runge, J., Lohrke, F. T., Naumann, S. E., Haptonstahl, D. E., Baker, R., Beckstein, B. A., Handler, C., Honig, H. & Lamoureux, S. (1997). Organizational Configurations and Performance: A Meta-Analysis. The Academy of Management Journal, 40(1), 223240. Kivimäki, M., Mäki, E. & Lindström, K. (1997). Does the Implementation of TQM Change the Well-being and Work-related Attitudes of Health Care Personnel? Study of a Prize-Winning Surgical Clinic. Journal of Organizational Change Management, 10(6), 456-470. Klein, K. J. & Kozlowski, S. W. J. (2000). From Micro to Meso: Critical Steps in Conceptualizing and Conducting Multilevel Research. Organizational Research Methods, 3(3), 211-236. KOM 118/2002. Anpassning till förändringar i arbetsliv och samhälle: en ny arbetsmiljöstrategi för gemenskapen 2002-2006 [Adapting to change in work and society: a new workplace health and safety strategy for the common interest 20022006]. Brussels: Meddelande från Europeiska Gemenskapernas Kommission (Communication from the Commission). Kompier, A. J., Cooper, C. L. & Geurts, A. E. (2000). A multiple case study approach to work stress prevention in Europe. European Journal of Work and Organizational Psychology, 9(3), 371-400. Kotey, B. & Meredith, G. G. (1997). Relationships among Owner/Manager Personal Values, Business Strategies and Enterprise Performance. Journal of Small Business Management, 35(2), 37-64. Kotey, B. & Sheridan, A. (2004). Changing HRM practices with firm growth. Journal of Small Business and Enterprise Development, 11(4), 474-485. Kovach, K. A. (1978). Organization size, job satisfaction, absenteeism and turnover. Washington D C: University Press of America.

78

Källestål, C. (Ed.), Bjurvald, M., Menckel, E., Schærström, A., Schelp, L. & Unge, C. (2004). Hälsofrämjande arbete på arbetsplatser – Effekter av interventioner refererade i systematiska kunskapsöversikter och i svenska rapporter [Workplace health promomotion – Effects reported in systematic literature overviews and in Swedish reports]. Stockholm: Statens folkhälsoinstitut. Lagrosen, Y. (2004). Exploring the Effects of TQM on Employee Health. Journal of Management Systems, 16(3), 1-10. Langaa Jensen, P., Alstrup, L. & Thoft, E. (2001). Workplace assessment: a tool for occupational health and safety management in small firms?. Applied Ergonomics, 32, 433-440. Larsson, J., Vinberg, S. & Wiklund, H. (2005). Leadership values for Quality and Health – How does McGregor’s X and Y Theory Influences Methodologies and Outcomes?. Proceedings Quality Management for Organizational and regional Development, Palermo, Italy, pp. 690-702. Lee, G. L. & Oakes, I. (1995). The ‘pros’ and ‘cons’ of total quality management for smaller firms in manufacturing: some experiences down the supply chain. Total Quality Management, 6, 413-426. Lepak, D. P. & Snell, S. A. (2002). Examining the human resource architecture: The relationships among human capital, employment, and human resource configurations. Journal of Management, 28(4), 517-543. Lindström, K., Schrey, K., Ahonen, G. & Kaleva, S. (2000). The effects of promoting organizational health on worker well-being and organizational effectiveness in small and medium-sized enterprises. In L. R. Murphy & C. L. Cooper (Eds.) Healthy and productive work – an international perspective (pp. 83-104). London: Taylor & Francis. Lithander, J. (2005). Gränslandets ekonomi och entreprenörskap – en studie av småföretagande i glesbygd [The borderland economy and entrepreneurship – a study of small businesses in rural areas]. Licentiate thesis. Arbetsliv i omvandling [Work Life in Transition] 2005:10. Stockholm: Arbetslivsinstitutet. Mack, D. A. (2000). Stress and Success among Small Business Owners: a Field Study of Insurance Agents. Doctorial Dissertation. Arlington, Tx: The University of Texas at Arlington. MacDuffie, J. P. (1995). Human resource bundles and manufacturing performance: organizational logic and flexible production systems in the world auto industry. Industrial and Labour Relations Review, 48, 197-221. Matlay, H. (1999). Employee relations in small firms: A micro-business perspective. Employee Relations, 3, 285-295. McGee, G. W., Goodson, J. R. & Cashman, J. F. (1987). Job stress and job dissatisfaction: Influence of conceptual factors. Psychological Reports, 61, 367-375. Menckel, E. & Österblom, L. (2002). Managing Workplace Health. Sweden meets Europe. Stockholm: Arbetslivsinstitutet. Moser, M. & Karlqvist, L. (2004). Small and medium sized enterprises. A literature review of workplace health promotion. Arbetslivsrapport [Working life report] 2004:17. Stockholm: National Institute for Working Life.

79

Murphy, L. R. & Cooper, C. L. (Eds.) (2000). Healthy and Productive Work – An international perspective. London: Taylor & Francis. Murphy, G. B., Trailer J. W. & Hill, R. C. (1996). Measuring Performance in Entrepreneurship Research. Journal of Business Research, 36, 15-23. Nerell, G. & Sandberg, C. G. (1994). Ledarskap och arbetsorganisation för hälsa, kvalitet och effektivitet [Leadership and work organization for health, quality and effectiveness]. Stockholm: Arbetarskyddsnämnden (the Joint Occupational Safety Council). Nilsson, B. (1996). Företagsutbildning – anpassning och/eller utveckling? [Company education – adaptation and/or development?]. In P-E. Ellström, B. Gustavsson & S. Larsson (Eds.) Livslångt lärande (pp. 203-226). Lund: Studentlitteratur. Noci, G. (1996). Selecting quality based programs in small firms. Small Business Economics, 8, 431-447. North, J., Blackburn, R. A. & Curran, J. (1998). The quality business. Quality issues and smaller firms. London: Routledge. Northouse, P. G. (2001). Leadership: Theory and practice. London: Sage Publications. Nutbeam, D. (2000). Building evidence of effectiveness in health promotion: measures, methods and madness. Abstract presented at 3rd Nordic Health Research Conference, Tampere, Finland, 6-9 September. Nyberg, A., Bernin, P. & Theorell, T. (2005). The impact of leadership on the health of subordinates. Report No. 1:2005, SALTSA – Joint Programme for Working Life Research in Europe. Stockholm: National Institute for Working Life. Nwankwo, S. (2000). Quality assurance in small business organisations: myths and realities. International Journal of Quality & Reliability Management, 17(1), 82-99. Ogbonna, E. & Harris. C. L. (2000). Leadership style, organizational culture and performance: empirical evidence from UK companies. The International Journal of Human Resource Management, 11(4), 766-788. Olsson, E. (2003). Samordnad rehabilitering. Processen som kom av sig [Coordinating rehabilitation. The Process that comes from within]. Stockholm: Hjalmarsson & Högberg Bokförlag. Oscarsson, B. (1999). Benchmarking of Enterprise Development 2000: An impact evaluation and a comparative analysis of programme design. Oslo: The Research Council of Norway. Pasmore, W. (1994). Creating strategic change. Designing the flexible high-performing organization. New York: John Wiley & sons. Patterson, M. G., West, M. A., Lawthom, R. & Nickell, S. (1997). Impact of People Management Practices on Performance. London: Institute of Personnel and Development. Persson, O. (1991). Att leva som småföretagare [Living as a small enterprise owner]. Göteborg: Göteborgs universitet, Psykologiska institutionen. Pfeffer, J. (1998). The Human Equation. Boston: Harvard Business School Press. Powell, A. S. (1995). Global quality: competitive successes and challenges. New York: Conference Board.

80

Rahman, S-U. (2002). Leadership and HR focus in TQM research in Australia: an assessment and agenda. Benchmarking: An International Journal, 9(5), 485-505. Ram, M. (1991). Control and autonomy in small firms: the case of West Midlands clothing industry. Work Employment and Society, 5(4), 601-619. Rantakyrö, L. (2004). Still Searching the Best Way. A multiple-case study in small industrial organizations. Doctoral Thesis 2004:36. Luleå: Luleå University of Technology. RFV Analyserar (RFV 2002:11). Svensk sjukfrånvaro i ett europeiskt perspektiv [Swedish Sickness Absenteeism in a European Perspective]. Stockholm: Riksförsäkringsverket. Rhodes, C. & Butler, J. S. (2004). Understanding Self-Perceptions of Business performance: An Examination of Black American Entrepreneurs. Journal of Developmental Entrepreneurship, 9(1), 55-71. Ruist, E. (1990). Modellbygge för empirisk analys: att se vad som sker i det som synes ske [Model construction for empirical analysis: seeing what is actually taking place]. Lund: Studentlitteratur. Sainfort, F., Karsh, B. T., Booske B. C. & Smith, M. J. (2001). Applying Quality Improvement Principles to Achieve Healthy Work Organizations. Journal on Quality Improvement, 27(9), 469-483. Sauter, S. L., Lim, S. & Murphy, L. R. (1996). Organizational health: A new paradigm for occupational stress research at NIOSH. Occupational Mental Health, 4, 248-254. SCB (Statistics Sweden), (2004). Tabell ur SCB:s företagsregister maj 2004 [Table of Statistics Sweden’s company register May 2004]. Särskilt utdrag, egen bearbetning [Extract - processed by the author]. Setterlind, S. (Ed.) (2004). Den hälsosamma arbetsplatsen [The healthy workplace] (pp. 11-22, 143-183). Karlstad: Stress Management Center. Sletterød, N. A., Lysø, R. (2001). Å grave gull i egen organisasjon. Dokumentasjon fra følgeevaluering av Interregprosjektet ’Bedriftsutvikling uten grenser’ [To dig for gold in one’s one organization. Documentation of the project ’Company development without limits’]. NTF-rapport 2001:1, Steinkjaer: NordTrøndelagsforskning. Smallbone, D., Leigh, R. & North, D. (1995). The characteristics and strategies of high growth SMEs. International Journal of Entrepreneurial Behaviour & Research, 1(3), 44-62. Sonnentag, S. (1996). Work group factors and individual well-being. In M.A. West (Ed.) Handbook of work group psychology (pp. 345-367). Chichester: John Wiley & Sons. SOU 1990:49. Arbete och hälsa. [Work and health]. Betänkande av arbetsmiljökommissionen.. Stockholm: Allmänna förlaget. SOU 1991: 82. Drivkrafter för produktivitet och välstånd [Driving forces for productivity and prosperity]. Produktivitetsdelegationens slutbetänkande. Stockholm: Allmänna förlaget.

81

SOU 2000:121. Sjukfrånvaro och sjukskrivning – fakta och förslag [Absence due to illness and sickleave – fact and proposal]. Slutbetänkande av Sjukförsäkringsutredningen. Stockholm: Fritzes förlag. SOU 2002: 62. Kunskapsläge sjukförsäkringen [Understanding health insurance]. Delbetänkande av utredningen om analys av hälsa och arbete. Stockholm: Fritzes förlag. Spector, P. E. (1986). Perceived control by employees: A meta-analysis of studies concerning autonomy and participation at work. Human Relations, 39, 1005-1016. Spilling, O. R. (2000). SMB 2000 – fakta om små og mellomstore bedrifter i Norge [SMB 2000 – facts about small and mediumsized enterprises in Norway]. Bergen: Fagbokforlaget. Starrin, B. & Svensson P-G. (Eds.) (1994). Kvalitativ metod och vetenskapsteori [Qualitative method and theory of science]. Lund: Studentlitteratur. Statens Beredning för medicinsk Utvärdering – SBU [The Swedish Council on Technology Assessment in Health Care] (2003). Sjukskrivning – orsaker, konsekvenser och praxis. En systematisk översikt [Sickness absence – causes, consequences and practices. A systematic review]. Stockholm: SBU. Stokols, D., McMahan, S. & Philips, K. (2002). Workplace Health Promotion in Small Businesses. In P. O’Donnell (Ed.) Health Promotion in the Workplace (pp. 493540). New York: Delmar Thomson Learning. Storey, D. (1999). Human resource management policies and practices in SMEs in the UK: Does it really influence their performance?. Warwick: The Centre for Small & Medium Sized Enterprises. Storey, D. (1994). Understanding the Small Business Sector. London: Routledge. Strauss, A. & Corbin, J. (1990). Basics of qualitative research. Grounded theory procedures and techniques. Newbury Park: Sage publications. Theorell, T., Emdad, R., Arnetz, B. & Weingarten, A-M. (2001). Employee Effects of an Educational Program for Managers at an Insurance Company. Psychosomatic Medicine, 63, 724-733. Thorsrud, E. & Emery, F. E. (1969). Medinflytande och engagemang i arbetet. Norska försök med självstyrande grupper [Participation and engagement in work. Norwegian efforts with autonomous groups]. Stockholm: Utvecklingsrådet för samarbetsfrågor. Vinberg, S. (1997). Arbetsmiljö- och rehabiliteringsarbete med arbetsplatsen som utgångspunkt. En utvärderingsstudie av AmoRE-projektet i Jämtlands län [The workplace as a starting point for the improvement of working environment and rehabilitation. An evaluation of the ‘AmoRE project’ in the County of Jämtland]. Socialmedicinsk tidskrift, 74(8-9), 400-405. Vinberg, S. (2001). Företagsutveckling – utan gränser. Erfarenheter och resultat från ett interregionalt forsknings- och utvecklingsprojekt [Company development – without limits. Experiences and results from an interregional research- and development project]. Östersund: Arbetslivsinstitutet.

82

Vinberg, S. (2003). Healthy Performance in Small Enterprises. Studies of Organizational Determinants. Licentiate Thesis 2003:33. Luleå: Luleå University of Technology. Vinberg, S. (2004). David möter Goliat – kontraster mellan mikroföretag och offentliga arbetsplatser avseende förändringsaspekter och sjukfrånvaro [David meets Goliat – contrasting microenterprises and public workplaces concerning change aspects and sickness absence]. In L-E. Wolvén & E. Ekstedt (Eds.). Företagande och gemenskap – små företags kultur, samverkan och konkurrens [Entrepreneurship and fellowship – small enterprise culture, co-operation and competition] (pp. 65-81). Arbetsliv i omvandling [Work Life in Transition] 2004:7. Stockholm: Arbetslivsinstitutet. Vinberg, S., Malmquist, C. & Sjödin, L. (1998). Företagsutveckling genom ökad kompetens och bättre hälsa. Ett pilotprojekt på små och medelstora arbetsplatser [Business development through increased competence and better health. A pilot project in small and medium sized workplaces]. Östersund: Försäkringskassan i Jämtlands län. von Otter, C. (1997). Perspektiv på Arbetslivsfonden – analyser och kommentarer [Perspectives on the Swedish Working Life Fund – analysis and comments]. Arbetslivsrapport [Working Life report] 1997:3. Stockholm: Arbetslivsinstitutet. Vossen, R. W. (1998). Relative strengths and weaknesses of small firms in innovation. International Small Business Journal, 16(3), 88-94. Wallenklint, J. (2001). A Managerial Perspective on Small-Firm Network Organizations. An Overview of the Field and a Study of Practice. Licentiate Thesis 2001:46, Luleå: Luleå University of Technology, Division of Industrial Organization. Walters, D. (2001). Health and Safety in Small Enterprises. European Strategies for Managing Improvement. Brussels: P I A-Peter Lang S A. Way, S. A. (2002). High performance work systems and intermediate indicators of firm performance within the US small business sector. Journal of Management, 28(6), 765-785. Westerberg, M. (2001). Vad krävs av småföretagaren i tider av omvälvning? [What are the needs of small businesses in times of upheaval]. In P. Davidsson, F. Delmar & J. Wiklund (Eds.) (2001) Tillväxtföretagen i Sverige. Stockholm: SNS förlag. Westerlund, H. (2005). Health changes in a changing labour market. Doctorial Dissertation, Department of public health Sciences. Stockholm: Karolinska institutet. Westgaard, R. H. & Winkel, J. (1997). Review article, Ergonomic intervention research for improved musculoskeletal health: A critical review. International Journal of Industrial Ergonomics, 20, 463-500. Whitfield, K. (2000). High-Performance Workplaces, Training, and the Distribution of Skills. Industrial Relations, 39(1), 1-25. Wiklund, J. (1998). Small Firm Growth and Performance: Entrepreneurship and Beyond. Doctorial Dissertation. Linköping: Linköping University.

83

Wiklund, J., Davidsson, P. & Delmar, F. (2003). What Do They Think and Feel about Growth? An Expectancy-Value Approach to Small Business Managers’ Attitudes Toward Growth. Entrepreneurship Theory and Practice, 27(3), 247-270. Wikman, A. (1991). Att utveckla sociala indikatorer – en surveyansats belyst med exemplet arbetsmiljö [Development of social indicators – a survey with illustrations from the working environment]. Stockholm: SCB (Statistics Sweden). Wikman, A. (2005). Reliability, validity and true values in surveys. Social Indicators Research, DOI 10.1007/s11205-005-5372-3. Wikman, A., Marklund, S. & Alexanderson, K. (2005). Illness, disease, and sickness absence: an empirical test of differences between concepts of ill health. Journal of Epidemiology and Community Health, 59, 450-454. Wilkinson, A. (1999). Employment relations in SMEs. Employee relations, 21(3), 206217. Wilson, M. G., DeJoy, D. M., Vandenberg, R. J., Richardson, H. A. & McGrath, A. L. (2004). Work characteristics and employee health and well-being: Test of a model of healthy work organization. Journal of Occupational and Organizational Psychology, 77, 565-588. World Health Organization (WHO), (1948). Official Records of the World Health Organization, no. 2, p. 100. Geneva: World Health Organization. Wilkinson, G. & Dale, B. G. (1999). Integrated management systems: an examination of the concept and theory. The TQM Magazine, 11(2), 95-104. Wood, S. & Albanese, M. T. (1995). Can we speak of a high commitment management on the shop-floor? Journal of Management Studies, 32(2), 215-247. Ylinenpää, H. (1997). Managing Competence Development and Acquisition in Small Manufacturing Firms. Doctorial Dissertation 1997:27. Luleå: Luleå University of Technology, Division of Industrial Organization. Ylinenpää, H. (1999). Competence management and small firm performance. In B. Johannisson & H. Landström (Eds.) Images of Entrepreneurship and Small Business – Emergent Swedish Contributions to Academic Research (pp. 217-240). Lund: Studentlitteratur. Ylinenpää, H., Johansson, B. & Johansson, J. (Eds.) (2006). Ledning i småföretag [Management in small enterprises]. Lund: Studentlitteratur. Yukl, G. (2002). Leadership in organizations. New Jersey: Prentice-Hall, Inc. Zemke, R. (1988). Rewards and recognition: yes they really work. Training, 25(11), 4953. Zinger, J. T., LeBrasseur, R. & Zanibbi, L. R. (2001). Factors influencing early stage performance in Canadian microenterprises. Journal of Developmental Entrepreneurship, 6(2), 129-151.

84

Paper I

Information technology levels, competence development and performance in Swedish small business enterprises

Behaviour & Information Technology, Vol 19, No 3, 2000, pp 201-210.

Stig Vinberg, Gunnar Gelin and Karl W. Sandberg

BEH AVIOU R & IN FOR MATION TECHN OLOG Y ,

2000,

VOL .

19,

N O.

3, 201±210

Information technology levels, competence development and performance in Swedish small business enterprises STIG VIN BER G ², G U N NAR G ELIN  and K AR L W. SAN D BER G * ²N ational Institute for Working Life, Ostersund Branch, Sweden; e-mail: [email protected]  Private statistical consultant, Sweden; e-mail: [email protected]

*N ational Institute for Working Life, Ostersund Branch and LuleaÊ U niversity of Technology, Sweden; e-mail: karl.w.sandberg@ niwl.se Abstract. This paper analyses relations between leaders’ and co-workers’ competence, IT-levels and organizational aspects within ten small business enterprises in Sweden. Data from questionnaires and semi-structured interviews were used for statistical data analyses organized in four steps. A result is that IT-levels and change competence go together with higher levels of educational investment, and that companies with high levels on these aspects also tend to have a proneness for joining change programs. Another ®nding is that the higher the rate of IT in a company, the lower the incidence of musculoskeletal problems among employees. The study shows that an increase in leader change competence, a ¯at organization, a higher leader than co-worker change competence, and an ongoing program contribute signi®cantly to pro®t per capita at the company level. The analysis model seems to give a good integration of the results, but underlying mechanisms must be dealt with in future research.

1. Introduction Today’s economies are increasingly based on knowledge, which is codi®ed in systems, manuals and information technology soft/hardware or situated in more macro factors such as organizational learning, communities of practice and organizational cultures. In the Organization for Economic Co-operation and D evelopment (OECD ), the G D P share of knowledge based industries is now more than 50% , up from 45% in 1985. In the N ordic countries, investment in research and development, software and public spending on education is 9 ± 10 % of G D P, a ®gure similar to investment in physical equipment (OECD 1999). Another tendency is that boundaries between organizations are becoming

more diuse, while inter-organizational relations and networks grow in importance (H aÊ kansson and Johansson 1993, Castells 1996, Araujo 1998). These tendencies have led to an increased interest in analysing and developing small and medium-sized enterprises (SM Es), and also to research into the antecedents and consequences of investments in SM E competence development (YlinenpaÈaÈ and H avenga 1997, R antakyroÈ 1998, Stymne et al. 1999), information technology sophistication (Pollard and H ayne 1997, F ink 1998, Silliance et al. 1998, Andersson 1999) and organizational Ð human resource development (Barth 1999, Pfeer 1998). In recent years, many researchers have focused either on `organizational learning’ i.e. studies of the nature of the learning process (Lave and Wenger 1991, H edlund and N onaka 1993, Easterby-Smith and Araujo 1999) or on the `learning organization’, that is action oriented approaches, geared to promoting learning processes inside organizations (G arvin 1993), with little agreement between researchers about the nature of learning within organizations or of facilitating/inhibiting factors. In this article we will concentrate on three learning organization aspects, namely:

· · ·

Leader and co-worker change motivation. Change competence. D ierences between leaders’ and co-workers’ learning processes.

Leaders’ learning has attracted research attention (e.g. Antonacopoulou 1999), and results demonstrate the impact of both individual and organizational factors on

Behaviour & Informatio n T echnolog y ISSN 0144-929 X print/ISSN 1362-300 1 online Ó 2000 Taylor & Francis Ltd http://www.tandf.co.uk/journals

202

S. V inberg et al.

leaders’ learning. Learning seems to be most eective when managers themselves recognise a learning need and when they are motivated to engage in this process (Lloyd 1990). Increasing information technology levels in learning organizations is looked upon as an enabling factor for strengthening productivity, innovation and competitive power (Adler 1992, Anttila 1997, EllstroÈm and N ilsson 1997). Information technology support is both an important condition for higher employee participation in problem solving and product development, and a factor in creating structural change and organizational redesign. H istorically, small enterprises have not adopted IT as much as larger companies, but in the 1990s SM E executives seem to have become more interested in IT systems (Igbaria et al. 1998, Sillince et al. 1998). R egarding H uman R esource M anagement (H R M ) and H uman R esource D evelopment (H R D ) aspects, including health and working environment issues, research has shown clear relationships between these aspects and organizational structures, which permit and stimulate learning and competence development (e.g. Aronsson 1983, H uselid 1995). Comparisons between small and large companies or studies of speci®c conditions in SM Es have been made by Becker and G erhart (1996) , Antonsson et al. (1998), Wagar (1998) and Lindmark (1999). Another related question of interest is whether the so-called `bundling tendency’ (M acD ue 1995) Ð where several organizational factors tend to appear together Ð also exists in small ®rms. In an earlier study (Vinberg et al. 1999), we looked at relationships at company level between competence indicators, and pro®t and health levels in a research project aimed at small business enterprises. Concerning leader change competence and pro®ts we found a strong positive correlation, while there was a negative correlation between leader change competence and learning problems experienced by employees. Co-worker and leader change competence co-varied strongly. In this article, using both company and individual level data, we will develop those initial ®ndings presented in our earlier pilot study.

The following main questions are studied:

·

·

·

Is there a bundling tendency in SM Es? That is, does the co-variation found in studies of large companies, with dierent organizational factors forming integrated `wholes’, also show up among small enterprises? What empirical relationships can be found between outcomes or performance indicators such as pro®t, ill-health, stress, meaningful work and certain determinant factors or `drivers’ such as IT-levels, change competence levels and time given to problem solving at work? What impact might perceptional dierences in competence and motivation between leaders and co-workers have on dierent outcomes?

2.2. Data and study design A group of ten small enterprises from both the industrial and service sectors participated in the study. In total, there were 20 company leaders and 82 coworkers. The leaders’ and co-workers’ competence within certain critical areas, such as competence for change, working environment, work organization, and health was mapped. Both groups answered a questionnaire in which they rated each other’s ability in the above areas mentioned. Indices of basic competence and health were constructed. D ata concerning economic outcome and economic analysis of human resources was also collected in each of the ten companies. Semi-structured interviews with the leaders and co-workers provided information about IT levels, competition levels and organizational aspects. Empirical measures corresponding to the concepts are discussed in Vinberg and M almquist (1998). A summary is also given in the appendix.

2.3. Statistical data analysis The analysis was organized in four steps. F irst, scales and indices were constructed based on earlier knowledge

2. Purpose, data, study design and statistical data analysis 2.1. Purpose This study analyses relations between IT-levels, competence development, change programmes and dierent outcomes in small companies guided by an integrated conceptual model (®gure 1).

Figure 1. Conceptual model of main areas related to workplace change outcomes.

Swedish small business enterprises

change motivation (D IF F M OTI). This component is here labelled High performance organizationa l aspects. The next component is described as information technology level and competence drive and consists of company educational investments per employee (ED U CAPI), an ongoing program of company change (ON G PR OG ) and IT-levels (ITLEVEL) as strong indicators. The variables with high loadings on the third component are related to stress and high work demands: how much working time is devoted to hard problem solving (PR OBSOLV), a job stress index (STR ESSPR ), an index measuring support and encouragement (SU PPOR T), a measure of educational job demands (ED U D EM AN ), and an index measuring psycho-social illhealth (U N H PSYR ). Component four is about change propensity, with two high loading indicators: co-worker change motivation (M OTIVU TV) and the dierence between the leaders’ own perception of their change competence and coworkers’ assessment of their leaders’ change competence (D IF F COM P). Component ®ve is about ergonomic and musculoskeletal problems (ER G OPR and U N H M U SR ). The two variables loading high on component six measure job meaningfulness and self-determination (M EAN IN G and SJALVDET). The principal component analysis resulted in a pattern of input variables, which showed a number of interrelated organizational aspects forming `bundles’ in a similar fashion to results presented by, for example

of the questions used (see appendix). Then the resulting set of questions, indices and scales was used as input to an explorative principal component analysis. The dimensions emerging from this step are discussed in section 3. In the third step, regression analyses were carried out with four outcome variables or performance indicators: pro®t per capita, musculoskeletal ill-health, high stress demands and meaningful work as dependent variables, and relevant indices or single variables as independent variables. The explanatory models which emerged from the application of these steps are presented and commented upon in section 3.2. In the fourth step, scatterplots are used to illustrate and re¯ect on some interesting bivariate relationships.

3. Analysis results 3.1. Interrelationships and bundles The variables and constructed indices used in the study are presented in detail in the appendix. Table 1 shows the results of a principal component analysis with the 19 measures used in the study as inputs. Six components emerge. The factors that are strong indicators of the ®rst component are high pro®t per capita (PR OF CAPI), leader change competence (LEAD CH CO), a ¯at (in contrast to a hierarchical) organization (H I), leader change motivation (LED M O16) and the dierence between leader change motivation and co-worker Table 1.

203

Varimax rotated loadings from a principal component analysis

Variables

Pro®t per capita (PR OFCAPI) Leader change competence (LEAD CH CO) Flat versus more hierarchical organization (H I) Leader minus co-worker change motivation (D IFF M OTI) Leader change motivation (LED M O16) Educational investments per employee (ED UCAPI) Ongoing program of company change (ON G PR OG) Information technology level (ITLEVEL) Problem solving intensity at work (PR OBSOLV) Job stress problems (STR ESSPR ) Amount of support (SUPPOR T) Educational job demands (ED U D EM AN ) Psycho-social ill-health (U N H PSYR) Co-worker change motivation (M OTIVU TV) Leader change competence minus co-worker conception of leader change competence (D IFF COM P) Ergonomic problems (ER GOPR) M usculoskeletal ill-health (U N HM U SR ) Self-determination in job (SJALVD ET) Job meaningfulness (M EAN ING)

Component

Ð

Ð

Ð

Ð

Ð Ð Ð

1

2

3

4

5

0.826 0.753 0.748 0.721 0.606 0.093 0.104 0.536 0.138 0.082 0.168 0.064 0.315 0.131 0.114

0.088 0.277 0.452 0.150 0.018 0.870 0.802 0.710 0.036 0.117 0.098 0.369 0.079 0.165 0.005

0.079 0.222 0.131 0.215 0.313 0.087 0.170 0.111 0.823 0.732 0.644 0.426 0.406 0.129 0.191

0.156 0.016 0.161 0.237 0.459 0.035 0.237 0.159 0.071 0.113 0.183 0.014 0.257 0.879 0.843

0.270 0.333 0.096 0.129 0.215 0.154 0.079 0.290 0.162 0.328 0.227 0.391 0.331 0.111 0.227

Ð

0.660 0.642 0.064 0.466

Ð

0.115 0.038 0.054 0.027

Ð Ð Ð

Ð Ð Ð Ð Ð

0.078 0.111 0.106 0.288

Ð Ð Ð

Ð

0.087 0.163 0.063 0.129

Ð Ð

Ð

Ð Ð Ð Ð

0.030 0.034 0.096 0.037

Ð Ð Ð

Ð

Ð Ð

Ð Ð

6 Ð Ð

Ð Ð Ð

Ð Ð Ð

Ð

0.218 0.036 0.004 0.472 0.433 0.034 0.017 0.038 0.201 0.054 0.087 0.178 0.007 0.040 0.145 0.022 0.099 0.806 0.630

S. V inberg et al.

204

H uselid (1995) and M acD ue (1995). The ®rst three components were particularly interesting. The high performance organizationa l aspects component implies that a ¯at organization, leader change competence and leader change motivation are interrelated and that such factors are also related to high pro®ts, which is in accordance with organizational development research ®ndings compiled by, for example Pfeer (1998: 31 ± 63). Since the main part of this research was carried out in large companies, it is noteworthy that this study has reached similar ®ndings, based on a sample of small enterprises. The information technology level and competence drive component indicate that high information technology levels go together with higher levels of educational investment and that companies of this type also tend to introduce programs of company change. The stress and high work demands component seems to capture important parts of this much debated problem area.

3.2. Regression analyses with four performance indicators as dependent variables In this analysis phase four outcome measures, or performance indicators, were used:

·

· · ·

Pro®t per capita, which can be seen as a goal or performance indicator for company stakeholders or owners and in most cases also for company leaders, especially in small business enterprises, and also, in those with pro®t sharing programs’ for employees. Physical ill-health, an inverted performance indicator for employee health promotion and also in varying degrees for almost all employees. High stress/demands in jobs, an inverted performance indicator to which public attention is rising rapidly as an increasing part of the labour market seems to contain stressful jobs. Meaningful jobs with self-determination, an outcome or performance indicator in visions of `good work’, potentially relevant for all employees.

Separate stepwise multiple regressions (p in = 0.05, p o u t = 0.10) have been made for each indicator with all other nonredundant variables in the factor analysis list as potential independent variables in the ®rst step. The risk for multicollinearity problems seems to be low, with low intercorrelations between explanatory variables and no instances with high VIF -values. The results are summarised in the ®gures below, with standardised beta-values to facilitate comparisons. Only variables

with coecients signi®cant on the 0.05 level are included. The adjusted r2-value, showing the explained share of the total variation in the dependent variable, adjusted for the number of explanatory variables, is shown to the right. In the ®rst analysis, with pro®t per capita as the dependent variable, two more variables were included in steps ®ve and six, but since the interpretation of these models is complicated we have chosen to present a more robust model with four explanatory variables. D etailed computer output is available upon request from the ®rst author². As ®gure 2 shows, four variables explain 79% of the company variation in pro®t per capita. An increase in leader change competence, a ¯at organization, a higher leader than co-worker change competence, and an ongoing change program contribute signi®cantly to pro®t per capita at the company level. Three variables explain 26% of the physical ill-health variation among the employees (see ®gure 3). An increase in enterprise information technology level and more meaningful jobs both tend to lower physical illhealth, while stress problems co-vary with physical illhealth. F igure 4 shows that a large share of the variation in high job demands and stress remain unexplained in this analysis. H owever, leader change competence co-varies with this outcome, so a rough summary of the results in ®gure 2 and ®gure 4 regarding these aspects might be that a high level of leader change competence is

Figure 2.

D eterminants of pro®t per capita.

Figure 3.

D eterminants of physical ill-health.

Swedish small business enterprises associated with both high pro®t levels and high stress levels. As ®gu re 5 shows, four variables explain about 24% of the employee variation in the index measuring selfdetermination and meaningful job, with level of problem solving demands having the largest coecient. A higher leader than co-worker motivation level is positively associated with this outcome, while pro®t per capita and stress problem level are negatively related, indicating that contradictory situations may exist in some ®rms.

205

Current discussions of health problems in the IT-sector, for example on new forms of loading injuries associated with computer work, have tended to focus on particular sectors and neglect the fact that information technology penetrates almost every economic sector. The ®gure shows that in this sample of small ®rms with a broad sectoral variation, the higher their IT-level, the lower is their mean level of musculoskeletal problems. In the absence of more information, the correlation shown above may of course not be interpreted as the result of a direct causal path, from increasing IT-levels to lowering of musculoskeletal problem levels. Instead, it is prob-

3.3. Illustrations and comments on three particular ®ndings H ere we will comment on three relationships of particular interest found in the regression analyses, and also present illustrative scatter-plots (sometimes such plots may give slightly distorted and simpli®ed views compared with the regression results, sometimes they may give a more accurate description). F igure 6 shows that there is a clear relationship between mean leader change competence and pro®t per capita at the company level, and that this co-variation is stronger for companies with a ¯atter structure than for companies with a more hierarchical structure. There thus tends to be an interaction between ¯at-hierarchical structure and leader change competence; when a company has both a ¯at structure and a leadership with a high level of change competence, pro®t levels tend to be higher than the sum of the two eects. F igure 7 illustrates the strong negative relationship between IT-level and level of musculoskeletal problems.

Figure 4.

D eterminants of high job demands and stress.

Figure 5. job.

D eterminants of self-determination and meaningful

F igure 6. Mean leader change competence and pro®t per capita. (The cubes represent ®rms with a ¯atter structure, and the circles ®rms with a more hierarchical structure).

F igure 7.

IT-level and level of musculoskeletal problems.

206

S. V inberg et al.

ably more reasonable to interpret the tendency shown as being due to other organizational and/or technological factors determining both the IT-level and the level of musculoskeletal problems in a company. F igure 8 shows the values for all employees’ perceived levels of self determination and their problem solving intensity at work. It also shows smoothed points (the ®lled circles) according to the equation S = a+ b 1 P+ b 2 P 2 (where S is perceived self determination level and P is problem solving intensity at work). An inverted u-curve emerges, with a maximum for problem solving intensity level around four. A possible interpretation of this is that there may be an `optimal stress level’, here indicated by the level four on the problem solving intensity measure, which may also be in line with some research on negative and positive stress, done in large companies (K arasek and Theorell 1990, D e Jonge and K ompier 1997).

4. Discussion The present study is limited in scale and the results should be interpreted with caution. The main emphasis of the study, guided by an integrated conceptual model, focused on analysing the relationship between IT-levels, leaders’ and co-workers’ change competence and dierent outcomes at both individual and organizational levels. The analysis model seems to provide a satisfactory integration of the results, but underlying mechanisms must be dealt with in more detail in future research.

Since we in this study sometimes mix organizational and individual data, there is a risk for ecological and/or atomistic fallacies, to use the terminology introduced by D iez-R oux (1998). F or instance, when organizational data are used in individual level data analyses, the lack of intra-company variation may bias the estimates. R esults from some small sensitivity analyses indicate that the possible amount of such bias may be small. H owever, in forthcoming research with larger data sets we will use hierarchical multilevel regression methods (Bryk and R audenbush 1992) as well as structural equation modelling of relations between several outcomes, since we believe these methods to be more adequate. M uch of the public discussion in Scandinavia on the possible consequences of information technology on job content and employee health has been about rather sharply demarcated IT-sector companies in the info-com and system development sectors. A general segregated picture is given, with extreme growth, stress and burnout tendencies in the advanced IT-companies and dull, low competence jobs with musculoskeletal problems in the not so advanced companies. In this paper we have taken another perspective, looking on relations between the use of Information Technology and other organizational and competence aspects in small companies outside of the speci®c IT-sectors. Some points coming out of this analysis are brie¯y summarized here.

·

·

·

· Figure 8. Problem solving intensity at work and perceived level of self determination.

A clear result in the principal component analysis (component two) is that information technology levels and competence drive (indicated by the company having an ongoing program) go together with higher levels of educational investments. The high performance organisationa l aspects component implies that a ¯at organizational structure, leader and co-worker change competence and leader change motivation are interrelated, and that these are also related to company pro®ts. These ®ndings are similar to other research on organizational development in large companies (Pfeer 1998, M acD ue 1995). Thus, it seems that at least some high performance aspects operate in an integrated way both in large and small companies. The interaction tendency found, that the correlation between leader change competence and pro®t level is stronger in organizations with a ¯at structure than in organizations with a more hierarchical structure, is also in agreement with some other organizational research, for instance the ®ndings summarized in Pfeer (1998). Another interesting ®nding is that the higher the IT-level in a company, the lower the incidence of musculoskeletal problems among employees. The

Swedish small business enterprises

·

·

mechanisms behind the relationship must be studied in more depth, for instance trough case studies of particular work designs and technologies, before evidence-based conclusions can be made. The relation between perceived levels of self determination and problem solving intensity among co-workers indicates that the relationship has the shape of an inverted u-curve. Thus, there may be an optimal stress level at a medium level of problem solving intensity. We also ®nd it remarkable that as much as 79% of the variation in pro®t level is explained by four aspects which may be said to belong to the `high performance organization bundle’.

The ®ndings of a strong relationship between leader change competence and pro®t, stress, ergonomic tensions, perceived learning problems and health point to the importance of action towards increasing leader change competence, and that one must also study the determinants of leader change competence. D ierences between self-estimated leader change competence and co-worker judgements of this competence co-vary with both working environment factors and health. The mechanisms behind this will be studied in more detail in forthcoming research.

References A D LE R , P. S. 1992, Technology and the Future of W ork (N ew York: Oxford U niversity Press). A N D E R SSO N, J. 1999, Business Process Development and Information T echnology in Small and Medium-sized Companys (G othenburg Sweden: IM IT, report 1999), 107. A N TO N A C O PU LO U , E. 1999, Developing learning managers within learning organizations: the case of three major retail banks, in M . Easterby-Smith, J. Burgoyne and L. Araujo (eds) Organizational L earning and the L earning Organization Ð Developments in T heory and Practice (London: Sage Publications), 217±242. A N TT I LA , M . 1997, KompetensfoÈrsoÈrjning Ð foÈretagets viktigaste process [Competence Supply Ð the Company’s M ost Important Process] (Stockholm: Ekerlids foÈrlag). A N TO N SSO N, A-B., N I LSSO N, M . and H A NSEÂN , O. 1998, Internkontroll i smaÊ foÈretag Ð verklighet och visioner [Internal Control in Small Organizations Ð R eality and Visions] (Stockholm, Sweden: Institutet foÈr Vatten och LuftvaÊ rdsforskning) A R A UJ O, L. 1998, K nowing and learning as networking, M anagement Learning, 29, 317 ± 336. A R O NSSO N, G. 1983, Arbetets krav och maÈnsklig utveckling [The D emands of Work and H uman D evelopment] Socialpsykologisk arbetslivsforskning (Stockholm: Prisma). BA RT H , H . 1999, Barriers to Growth in Small Firms. Licentiate Thesis, LuleaÊ U niversity of Technology, D epartment of Business Administration and Social Science.

207

BE C KE R , B. and G ER H A RT , B. 1996, The impact of human resource management on organizational performance: progress and prospects. Academy of Management Journal, 39, 779 ± 801. BR Y K , A. and R A U D E N BU SH , S. 1992, Hierarchical L inear Models: Applications and Data Analysis Methods (Newbury Park, California: Sage). C A ST EL L S, M . 1996, T he Information Age, V olume 1 ± T he Rise of the Network Society (M alden, USA: Blackwell Publishers Inc). D E J O NG E , J. and K O M P I ER , M. A. J. 1997, A critical examination of the demand-control-support model from a work psychological perspective. International Journal of Stress Management, 4, 235 ± 258. D I EZ -R O U X , A.V. 1998, Bringing context back into epidemiology: variables and fallacies in multilevel analysis. American Journal of Public Health, 88, 216 ± 222. E A ST E RBY -SM I TH , M . and A R A U J O, L. 1999, Organizational learning: current debates and opportunities, in M . EasterbySmith, J. Burgoyne and L. Araujo (eds), Organizational L earning and the L earning Organization Ð Developments in T heory and Practice (London: Sage Publications), 1 ± 21. E L L ST ROÈ M , P. E. and N I L SSO N , B. 1997, Kompetensutveckling i smaÊ och Medelstora FoÈretag [Competence development in small and medium-sized companies] (LinkoÈping Sweden: LinkoÈpings Universitet). F I N K , D . 1998, G uidelines for the successful adoption of information technology in small and medium enterprises. International Journal of Information Management, 18, 243 ± 253. G A RVI N , D . A. 1993, Building a learning organization. Harvard Business Review, 71, 78 ± 84. H E D L U N D , G . and N O N A KA , I. 1993, M odels of knowledge management in the West and Japan, in P. Lorange (ed.). Implementing Strategic Processes (Oxford: Blackwell). H U SE L ID , M . A. 1995, The impact of human resource management practices on turnover, productivity and corporate ®nancial performance, Academy of Management Journal, 38, 635 ± 672. Ê H A K A N SSO N , H. AN D J O H A N SO N , J. 1993, The network as a governance structure: inter-®rm cooperation beyond markets and hierarchies, in G . Grabher (ed.). T he Embedded Firm (London: R outledge), 35 ± 51. I G BA RI A , M ., Z I NA T EL L I, N . and C A VAY E, A.L.M . 1998, Analysis of information technology success in small ®rms in N ew Zealand. International Journal of Information Management, 18, 103 ± 119. K A RA SE K , R . A. and T H EO R EL L , T. 1990, Healthy W ork: Stress, Productivity and the Reconstruction of W orking L ife (New York: Basic Books). L A VE, J. and WE N G E R , E. 1991, Situated L earning: L egitimate Peripheral Participation(N ew York: Cambridge U niversity Press). L I N D M A RK , C. 1999, Beliefs about motivation and work with quality, environment and working environment in small organizations. Licentiate Thesis, LuleaÊ U niversity of Technology, D ivision of Quality Technology and Statistics. L L O Y D , B. (1990), A learning society. Modern Management, 4, 32 ± 34. M A C DU F F IE , J. P. 1995, H uman resource bundles and manufacturing performance: organizational logic and ¯exible production systems in the world auto industry. Industrial and L abour Relations Review, 48, 197 ± 221.

208

S. V inberg et al.

OECD (1999) Science, T echnology and Industry Scoreboard 1999. Benchmarking Knowledge-Based Economies; http:// www.oecd.org//dsti/sti/stat-ana/prod/scorebd_summ.htmb P F EF F ER , J. 1998, T he Human Equation (Boston: Harvard Business School Press). P O L L A RD , C. E. and H A Y NE , S.C. 1997, T he changing face of information system issues in small ®rms. International Small Business Journal, 16, 70 ± 81. R A N T A KY R OÈ , L. 1998, Small ®rms on the way to learning enterprises? L ifelong L earning in Europe, 1, 24 ± 28. SI LL I A NC E , J.A.A., M A C DO N A LD , S., L E F A N G , B. and F R O ST, B. 1998, Email adoption, diusion, use and impact within small ®rms: a survey of U K companies. International Journal of Information Management, 18, 231 ± 242. ST Y M N E , B., STJ E RN BE R G , T., M A GN U SSO N, R A P P, BEL O T TI, T U N AÈ LV, A N D E R SSON AN D N I LSSO N 1999, T he use of information technology in Swedish SMEs. Progress report from 40 ®eld visits. (Stockholm, Sweden: IMIT R apport), pp. 99 W A GA R , T. H . 1998, Determinants of human resource management practices in small ®rms: some evidence from Atlantic Canada, Journal of Small Business Management, 36, 13 ± 23.

VI NBE R G , S., G EL I N , G. and SA ND BE R G , K . S. 1999, M odels for competence and health development in small business enterprises. Proceedings 10th Anniversary Ergonomic Conference, LuleaÊ U niversity of Technology, Sweden. VI NBE R G , S. and M A LM Q UI ST, C. 1998, FoÈretagsutveckling genom kad kompetens och baÈttre haÈlsa paÊ smaÊ arbetsplatser i jaÈmtlands laÈn. [Company D evelopment through Increased Competence and Better H ealth at Small Workplaces in JaÈmtland County] (OÈ stersund, Sweden: M id-Sweden University). WI KM A N , A. 1991, Developing social indicators. A survey approach illustrated with the working environment case, D octoral dissertation in Swedish with an English summary (Stockholm, Statistics Sweden). Y L IN EN P AÈAÈ , H. and H A VEN G A , K . 1997, Competence development in Swedish, South African and R ussian SM Es Ð a study of attitudes and preferences across countries. Journal of Best Papers, Scott W. Kunkel (ed), 42nd W orld Conference of International Council for Small Business, San Francisco.

Appendix: Variable de®nitions and data collection methods The study is two-level, using both organizational level data for the ten companies and individual level data for the leaders and co-workers in these companies. R eversals have been made when appropriate so that all indices go in the same direction. A. Direct organization measures: pro®t level, IT -level, ¯at structure, educational spending and change activity Some data for the organization level variables were collected through company visits, interviews with leaders and accountancy ®gures. The variables used in this article are: Profit per capita (PR OF CAPI): Information technology level (ITLEVEL):

F lat versus more hierarchical (H I): Educational outlays per capita (ED U CAPI): Ongoing change program (ON G PR OG ):

(net profit for the year 1997 in SEK /number of employees including leaders Composite summation index based on eleven indicators. One point is added if company has . . . F ax PC H and/pocket computer/s M ini or larger computer system e-mail Local area network/similar system Intranet Internet Own home page Computer system for customer contacts/data base profiling K nowledge management system 0 if company has two levels (leaders versus employees/teams), 1 if company has more levels. (Amount on co-worker education 1997 in thousands of SEK ) / number of employees including leaders. 1 if the company was pursuing a change program during 1998-99 (involving competence development and or technical/process change and/or organisational/leadership development), 0 else.

Swedish small business enterprises

209

B. Company means: measures of competence, leadership and motivation An established questionnaire, which has been applied in several, mainly large, European companies (see Anttila 1997), was used to get company level measurements of competence, leadership and change motivation. Partly separate versions of the questionnaire were used for leaders and co-workers respectively, which made it possible to create indices of leader change competence and leader change motivation, to compare the leaders’ conception of their own change competence with co-worker conceptions of leader change competence on the company level, and to create separate indices of leader and co-worker change motivation. These individual level data have been aggregated to get mean values for the company. Variables of this type that are used in this article are listed below. All answers were given on a scale like this : low < - ability - > high 1 2 3 4 5 6 7 8 9 10 Leader change competence index (LEAD CH CO)

D ifference between leader conception of change competence and co-worker conceptions of leader change competence (D IF F COM P) Leader change motivation index (LED M O16) Co-worker change motivation index (M OTIVU TV) D ifference between leader change motivation index and co-worker change motivation index (D IF F M OTI)

A summation of answers on fourteen questions. Example: H ow do you view your ability to create readiness for change by following changes in customer demands/expectations and if needed adapt your work accordingly? H ow do you view your ability to take part in developing work processes and yourself by taking initiatives for trying new areas of work ? In the co-worker questions on leader conceptions typically `your’ was changed to `your leaders’. M ean leader change competence Ð index of mean co-worker conceptions of leader change competence. The latter index is a summation of answers on similar questions, as described above.

A summation of answers on five questions. Example: State to what degree you want to increase your ability to create a readiness for change. N ot at all < - - > To a high degree 1 2 3 4 5 6 7 8 9 10 A summation of answers on five questions, similar to the leader change motivation ones. D ifference between the two indices above.

The questions in the leader change versus the co-worker conceptions competence indices were very similar, so inclusion of the co-worker conceptions index would have created redundancy in the principal component analysis. The questions in the two motivation indices were rather similar, but with some dierences in wordings, so the dierence is more approximate and thus both the indices and the dierence are included in the principal component analysis.

210

S. V inberg et al.

C. Individual level measures: work conditions, psychosocial and ergonomic aspects, ill-health A standard questionnaire created by Anders Wikman (1991) at Statistics Sweden and used in several nation-wide surveys gave individual level data on the following aspects used in the article. We refer to this book for details, and just state the indexes here, with some examples. Psycho-social ill-health index (U N H PSYR ) M usculo-skeletal ill-health index (U N H M U SR ) Ergonomic problems index (ER G OPR ) Stress problem index (STR ESSPR ) Job competence demands (ED U D EM AN ) Problem solving intensity at work (PR OBSOLV) Amount of support given (SU PPOR T) Perceived level of self-determination in job (SJALVD ET) Perceived job meaningfulness (M EAN IN G )

An index with six questions. An index with six questions. An index with four questions. An index with five questions. D oes your job demand recurrent education/learning (information and/or instructions) for you to be able to manage it ? Is part of the working time spent on understanding or solving demanding problems? Almost the whole time, About 3/4 of the time, A little (say 1/10 of the time), N ot at all. An index with three questions. Exa mple: If the work tasks feel hard, do you have a possibility to get advice or help? (Always, M ost of the time, N ot most of the time, N ever). An index with three questions. Exa mple: D oes it happen that you are taking part in decisions regarding how to organize your work (for instance what is to be done, how it is to be done or which persons to co-operate with)? (Always, M ost of the time, N ot most of the time, N ever). An index with three questions in a battery measuring different perceptions of work. Example: H ow do you perceive (look upon) your work: N either/nor D ull work 1 2 3 4 5 Changing Very meaningless 1 2 3 4 5 Very meaningful

Paper II

Organizational and health performance in small enterprises in Norway and Sweden

Work: A Journal of Prevention, Assessment & Rehabilitation, Vol 24, No 3, 2005, pp 305-316.

Stig Vinberg and Gunnar Gelin

305

Work 24 (2005) 305–316 IOS Press

Organizational and health performance in small enterprises in Norway and Sweden Stig Vinberga and Gunnar Gelinb a

¨ National Institute for Working Life, Studentplan 3, S-831 40 Ostersund, Sweden Tel.: +46 63 55 13 07, Fax: +46 63 55 13 20, E-mail: [email protected] b Daphne Research

Abstract. This article focuses on relationships between organizational factors such as leadership, learning, psychosocial work environment and quality aspects as they relate to organizational and health performance outcomes in 42 small enterprises in Norway and Sweden. A rather explorative analysis model was created using indicators that were based on theoretical concepts from a literature review and questionnaire data, concerning 988 employees and leaders. These indicators were then used for correlation analysis. The enterprise is the unit of analysis. Some strong links between organizational factors and organizational and health performance were found. There was also a strong relation between health performance and sickness absence. Using structural analyses, a structure containing six general components, and strong interrelationships between some indicators of organizational performance and health performance were found. It was also possible to position the enterprises according to two general dimensions. As a general guideline for action, the results tend to support the perspective that positive organizational development is related to health performance and a lowering of absence due to sickness. The findings support the strategies of practitioners who use a concept-driven or holistic approach that integrates several facets of workplace development. Keywords: Organizational performance, sickness absence, organizational development, leadership, learning, psychosocial work environment

1. Introduction Considerable research in the Nordic countries is focusing on the identification of key workplace health determinants. The most important reason for this research emphasis is that healthier workplaces are increasingly seen as the most effective counteractant to rising sick leave rates and the increasing number of employees needing vocational rehabilitation. This article focuses on Norwegian and Swedish small enterprise 1 organizational and health performance and the influences of organizational factors such as leadership, learning, psychosocial work environments and quality development. The approach is explorative, and we also study interrelations between different organizational aspects 1 The terms “enterprise”, “firm” and “company” are used interchangeably to describe a privately owned producer of goods and/or services. The main focus in this study is on small firms with less than 50 employees.

and how similar firms may group together on general dimensions. A general assumption is that worker wellbeing and organizational effectiveness can be fostered through a common set of job and organizational design characteristics [37]. The primary reason for selecting this focus is that both researchers and practitioners working in the area of workplace development can benefit from a greater understanding of the different factors that contribute to organizational and health performance in small firms. A second reason is that over the last decade researchers and policy makers have come to recognize that sound small enterprise organizational performance helps to ensure stable (or increasing) employment levels [8,12, 46]. However, the explanations as to how small enterprises of different company sizes contribute to the overall employment picture differ between studies [53]. A third reason is the paradox that hazardous working conditions and injury rates in small enterprises are higher at the same time that their sick leave rates are lower

1051-9815/05/$17.00 © 2005 – IOS Press and the authors. All rights reserved

306

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

than in larger enterprises [14,50]. Finally, we note that small firms have not been as active as larger ones in organizational development and occupational health and safety issues, usually because they lack the expertise, time and economic resources [14,32].

2. Theoretical approaches and previous research This section provides a brief overview of some theoretical approaches and research concerning healthy work in small firms. The goal of this overview is rather modest – to list some research on links between such organizational factors as the psychosocial work environment, leadership, learning and quality development, and outcomes. Identified as important is the need for integration and analyses of “bundles” and configurations of organizational aspects that are related to the healthiness of small firm work environments. In their discussion of models of healthy work organizations, Murphy and Cooper [37] argue that there appears to be a convergence of opinion and evidence on some of the critical factors that tend to create a healthy work organization. They do qualify their observations by observing that additional empirical work is needed to test existing models and develop new ones. For example, most models identify as necessary components: meaningful work which utilizes worker skills, worker autonomy/control, job security, rewards for performance, worker involvement/participation, and safe and healthy physical work environments. Other important healthy work factors found as a part of recent research are teamwork and commitment to core organizational values. The healthy work organization concept is related to the concept of ‘development work’, which some Swedish researchers describe as an important work organization innovation that leads to improvements in work and working conditions while at the same time strengthening quality and performance [15,16]. The extensive research on the relationship between the psychosocial work environment and health/wellbeing shows that factors related to work content, work characteristics, work organization, and social relations are of importance for individual health in a wide sense (see e.g. [19,24,25]). More specifically, literature on work-related stress highlights that certain levels of job complexity, autonomy, role ambiguity and workload usually are associated with well-being at work. With respect to psychosocial conditions of work environments, several studies have found that employees in small organizations tend to have higher levels of job sat-

isfaction than do those in large organizations [28,57]. Evidence of workplace health effects and the condition of the physical work environment and its relationship to workplace size has infrequently been a topic of study. The small amount of studies indicate that “at the very least the situation in small enterprises is no better than in their larger counterparts” [50, p. 99]. In Sweden, sickness absence has increased dramatically from 1997 to 2002. While levels are higher for public sector employees (compared with the private sector) the rate of increase for small and larger enterprises is also of considerable concern [3]. In a study of 343 Finnish workplaces, Lindstro¨ m et al. [32] found that firms with over 50 employees had more sickness related absenteeism than did those in smaller enterprises. This is similar to what was found in Swedish investigations [9]. There are several studies on how leadership moderates or influences health and organizational performance. In his review of more than 130 field studies about organizational change, Pfeffer [41] emphasized that a frequent finding was that leadership that focuses on developing human resource practices stimulated a higher level of organizational performance. Cascio [10] pointed out that today’s leaders need to integrate a human resource focus with a financial focus; developing teamwork, creating conditions where workers can make decisions about their jobs, and extensive training for workers and leaders are examples of this strategy. Research studies indicate that employees in workplaces with relation-oriented managers often experience job satisfaction [56], low levels of sickness absence [4] and low levels of stress [17]. In a study of 372 small firms, Chaganti et al. [11] found that the fastest growing firms had both strong consideration and initiation leadership styles. Smallbone et al. [47] examined the development of 306 SMEs (small and mediumsized enterprises) and found that one of the most important factors for high growth was the commitment of the company leader to achieving growth. Among organizational researchers there has been much focus on learning and competence development in both large and small organizations. However, there is little agreement concerning the nature of learning and relationships between learning and performance in SMEs [7,55]. Huselid [23] found a positive relationship between training and economic results at the company level while Baldwin [6], in a similar study, did not find such a relationship. Swedish research has shown that in a working environment that encourages learning and competence development employees also tend to experiences worker health and well-being [16,19]. Hill

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

and Stewart [22] describe each small organization as unique with learning activities that are essentially informal, reactive and short-term in outlook. Our comprehensive view is consistent with much empiricallybased research in this area which maintains that there is a positive relation between learning and performance that also relates to worker health and organizational effectiveness. Small firms have been slow to systematically adopt quality improvements [29,38] due to limited resources, lack of financial means and the frequent inability to initiate internal training [21]. The link between quality and performance has been studied rather extensively and researchers have drawn different conclusions about the effect of quality improvement at the company level [39,40]. Quality management systems are increasingly being integrated with business management systems that include work health, safety and environmental practices [54]. Axelsson [5] has found that work environment processes and quality development processes have many commonalities and that more than 50% of quality improvement efforts in some cases are related to the work environment. According to Greenan et al. [20], a trend can be seen that environmental and work environment issues are becoming increasingly important in both large and small organizations. This tendency leads to a greater emphasis on the integration of quality, environmental and work environment systems [3]. There is a great need for integration and systematisation of the different areas discussed above. Two approaches seem fruitful for this purpose – the analysis of bundles or groups of interrelated organizational aspects (see e.g. MacDuffie [33]) and the structural analysis of configurations or organizational types [13, 23,34–36,45]. In the organizational research literature concepts like “high commitment management”, “high involvement work practices”, “high-performance work practices” and “configurations” seem to be used interchangeably. A recurrent observation in these literature is that several organizational aspects tend to co-vary and form “bundles”, and that organizations with similar values on these interrelated aspects seem to cluster into distinct groups. Overviews relating to the Swedish situation are given by e.g. leGrand et al. [30] and Sandberg [43]. In an overview of current research, Ketchen et al. [26] found that an organization’s performance is partially explained by its configuration. An important reference for the present study is Wood and Albanese [52] in which evidence is presented that organizational factors rather than factors exogenous to the

307

organization were significantly related to the level of “high commitment management”. In two studies reported earlier we have looked at relationships at the company level between competence and health indicators and economic performance in small firms [48,49]. Concerning leader change competence (leader ability to implement and manage change) and profits we found a strong positive correlation while there was a negative correlation between leader change competence and learning problems experienced by employees. Concerning health indicators we found that musculoskeletal symptoms were related to job content, leadership and credit worthiness ratings in the firms studied.

3. Research design and analysis 3.1. Research design The findings and theoretical research described above form a background for the analyses of data derived from an ongoing research- and development project of SMEs in Sweden (J a¨ mtland county) and Norway (Nord-Trøndelag county). The data reported here comes from the first measurement wave of a comparative longitudinal study of 42 SMEs (30 Swedish and 12 Norwegian) that employed a total of 1177 individuals in 2000. Data was collected through questionnaires distributed to all employees and leaders at the worksites (988 responses, 84% response rate), public records and interviews with leaders. This article is based on 22 indicators constructed from the questionnaire data. The set of companies participating in the study was the result of recruitment choices made by the Occupational Health Services, the social insurance offices of both countries and the Swedish National Institute for Working Life. The goal was to get a sample with small companies from key sectors – ICT, retail trade, manufacturing, wood products, food processing, construction, customer support services, technical development and consulting. The gender, age and educational distributions for the companies are: 43% women, 65% under 40 years of age and 16% in the 50 + age group, 16% with a basic education and 25% with a university education. There may have been a tendency to recruit companies with, for instance, more or less interest in and/or resources for workplace development but it has not been possible to obtain relevant data for a discussion of such or other selectivity issues. It must be stressed

308

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

95% CI Mean absence days

30

20

10

0

-10

N =10 7

4 10 5

1

5

e ic rv s se m al s t e on s y rs a l Pe n i c , IT c h ter T e en s lc t al l t a n C s u i o n ps o n c t ho C tru , s s e o n ad C l tr ing ai tur et R fac u an M Fig. 1. Mean sickness absence for firms in different branches.

that the collection of companies used in the study is not a representative probability sample, so no inferences to a larger population are intended. For instance, p-values for regression or correlation coefficients are meant to be used as summary indicators of strengths of relationships. Figure 1 shows that the mean sickness absence days per employee over the last 12 months in the studied companies were between four and ten days in different sectors. These rates were similar to that for the whole small business sector in Sweden [9]. 3.2. Explorative analysis model and indicators In Fig. 2 the rather explorative analysis model for the relations between assumed determinants and outcomes (the three indicators to the right) is presented. Relationships covered in the literature mentioned in the section about theoretical approaches were incorporated into this model. For the outcomes we assume positive relations between organizational performance and health performance, and negative relations between organizational performance and mean sickness absence, and between health performance and mean sickness absence. Regarding the relations between determinants and outcomes we do not consider the research fields as developed enough to permit specific assumptions for all possible relations. We ask the reader to see this model as one of a rather limited set of similar models.

The indicators knowledge of workplace goals, customer oriented quality practices and efficient use of technology are associated with the concept of quality development commonly used in literature. The indicator respectful leadership measures “good” leadership aspects and the indicators creative work and adequate competence relate to such aspects of learning. Several indicators in the model are related to psychosocial work environment: work intensity demands, psychosocially demanding work, data stress, demands higher than control, team spirit, decision latitude and restructuring worries. Continuous lines are used for assumed positive relationships and broken lines for assumed negative relationships in Fig. 2. The 22 indicators described below were developed on the basis of substantive considerations and factor analyses. Of these, thirteen indicators in the model are seen as determinants (to the left in the model). On the right side of the model, the remaining nine indicators make up the three general areas that are depicted in the figure. These nine indicators focus on outcomes. Three of the nine indicators form the organizational performance concept: this outcome indicator is the mean of the indicators perceived lack of efficiency (four items), workplace adaptability (three items) and customer satisfaction (four items). 2 Five additional indicators form the health performance concept: this outcome indicator is the mean of indicators measuring not being in a depressed mood-worries (six items), not having musculoskeletal symptoms (three items), not having sleep problems (two items), perceived physical health (one item) and psychological well-being (one item). The last concept, mean sickness absence is measured by a single item, mean sickness absence days per employee over the last 12 months. The questionnaire used [44] contained a total of 162 items covering a wide range of topics. Only some of the 162 items were used in this study. The co-workers’ and leaders’ perception of their work environment and job and organizational characteristics as well as various aspects of health and organizational performance were the focus of this study. The content of the 13 indicators to the left side of Fig. 2 that act as determinants is as follows: The knowledge of workplace goals indicator consists of two items3 that seek to measure how well the employees 2 A detailed documentation of indicator construction can be obtained from the first author. 3 In the following overview the range for each item is 1–5, unless stated otherwise.

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

309

Knowledge of workplace goals Costumer oriented quality practices Efficient use of technology Respectful leadership

Organizational performance

Creative work

Adequate competence Work intensity demands Psychosocially demanding work

Health performance

Data stress

Sickness absence Demands higher than control

Team spirit

Decision latitude Restructuring worries

Fig. 2. Analysis model for the empirical analyses. (Continuous lines are used for assumed positive relationships and broken lines for assumed negative relationships).

know workplace goals. The indicator customer oriented quality practices (seven items) covered feedback on quality to employees, programs for improvements, quality goals and cooperation with customers. The efficient use of technology indicator consists of two items covering the status and effectiveness of technical equipment. The respectful leadership indicator (nine items) covered leadership consideration, information about changes, ethical issues and dialogue between leaders and co-workers. The creative work indicator (ten items) covered developing and learning at work, meaningful and stimulating work and variation at work. The questions in the adequate competence indicator (two items) were about resources and competence in relation to work demands. The work intensity demands indicator had eight items about hard and demanding work, stress and time pressure. The questions in the psychosocially demanding work indicator (two items) were about the employees’ perception of the degree of demanding work. The data stress indicator consists of two items about stress related to data problems. In

the three questions on the demands higher than control indicator, people were asked to say if their job demands were higher than what they could handle. The team spirit indicator was measured by three items concerning team functioning and team climate at work. The decision latitude indicator (four items) measured worker autonomy/control and possibility to influence work. The indicator restructuring worries had two items about concerns related to possible restructuring and its potential impact upon continued employment. From the individual indicator data an aggregate data file was constructed with indicator means for the 42 firms in the project. The reason for this is that the focus of this article is on organizational level aspects [27]. 3.3. Research questions and analysis The following questions are addressed: 1. What empirical relationships can be found between outcomes for performance indicators such

310

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

as organizational performance, health performance, sickness absence and assumed determinant organizational factors presented in the model in Fig. 2? 2. Is there a bundling tendency in small firms? That is, do several organizational aspects co-vary? 3. Is it possible to find groupings of co-varying indicators and more specifically is it possible to find groups of companies that are similar in the sense that they have similar indicator patterns? The presentation of the analyses is organized in three subsections. Subsection 4.1 shows and comments on a correlation analysis. In Subsection 4.2 a structural analysis of co-variation between the 22 indicators are shown and commented upon. Here we include all indicators, both those that are parts of the outcome indicators in the determinants-outcome model (Fig. 2) and those that are used in that model as determinants. In Subsection 4.3 a grouping of companies is presented and discussed. We applied two rather different approaches to this analysis. In one approach we make a distinction between determinants and outcomes, integrating possible determining relations between the main concepts in the model, using correlation analysis. Since we have cross-sectional data we cannot test if causal hypotheses are empirically supported without introducing several external assumptions. Thus, the results from this approach are mainly exploratory that suggest refined proposals for further research. The second approach takes a structural view, with one aim being to identify groups of variables or indicators that tend to co-vary to form broader aspects/latent dimensions. A second aim was to determine if there are groups of companies with similar patterns. The exploratory methods used were component analysis and multidimensional scaling.

4. Analysis results 4.1. Determinants and outcomes – indicators of organizational performance, health performance and sickness absence In keeping with the model presented in Fig. 2 above we studied the relationships between those indicators treated as determinants (the 13 indicators on the left side of the figure) and the three right-side outcome indicators: organizational performance, health performance and sickness absence.

Table 1 below shows the simple pair-wise correlations between the determinants and the outcomes. For sickness absence there are strong or rather strong negative correlations with health performance, organizational performance and team spirit, and a rather strong positive correlation with restructuring worries. For organizational performance there are strong or rather strong positive correlations with health performance, customer oriented quality practices, efficient use of technology, respectful leadership, creative work, adequate competence and team spirit, and strong or rather strong negative correlations with sickness absence, work intensity demands, psychosocially demanding work, data stress and demands higher than control. For health performance there are strong or rather strong negative correlations with sickness absence, work intensity demands and restructuring worries, and rather strong positive correlations with organizational performance and team spirit. For two of the determinants, knowledge of workplace goals and decision latitude the correlations with all three outcomes are low. All strong or rather strong correlations have the assumed sign. 4.2. Structural analysis to explore groups of indicators Several aspects covered in this article may be viewed by different professions as falling within their domain of interest. Too, their perspectives on ordering or grouping arrangements may vary. For instance, if ergonomists were asked to describe which main groups the indicators used here could be seen as belonging to they would likely create groupings different from groupings that might be created by practitioners of work psychology, organizational behaviour, quality movement, strategic human resource management or organizational learning. This absence of conceptual coherence across disciplines [41] would lead to assessing as to whether there is empirical support for various integrated models or combinations of models. Such a model testing approach is outside the scope and size of this article. Instead we chose an exploratory approach; aiming first at finding out which general groups of indicators emerge and in Section 4.3 which dimensions emerge for a grouping of the workplaces. The 22 indicators described above were initially created based on substantive considerations and factor analyses. Here, a second-order principal components analysis was carried out to determine which dimensions emerged. It seems reasonable to assume that such di-

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

311

Table 1 Correlations between all indicators and the outcome indicators Sickness absence − −

Organizational performance −0.342 0.026

Health performance −0.492 0.001

Organizational performance

−0.342 0.026

− −

0.333 0.031

Health performance

−0.492 0.001

0.333 0.031

− −

0.023 0.886

0.037 0.814

0.239 0.127

Customer oriented quality practices

−0.128 0.420

0.344 0.026

−0.070 0.660

Efficient use of technology

−0.057 0.720

0.464 0.002

0.009 0.954

Respectful leadership

−0.245 0.118

0.782 0.000

0.170 0.283

Creative work

−0.095 0.551

0.381 0.013

0.163 0.302

Adequate competence

−0.010 0.950

0.356 0.021

0.072 0.648

Work intensity demands

−0.061 0.699

−0.366 0.017

−.294 0.059

Psychosocially demanding work

−0.031 0.846

−0.312 0.044

−0.302 0.052

Data stress

−0.050 0.752

−0.369 0.016

−0.198 0.208

0.158 0.318

−0.656 0.000

−0.277 0.076

Team spirit

−0.312 0.045

0.726 0.000

0.364 0.018

Decision latitude

−0.135 0.399

0.217 0.173

0.038 0.812

Restructuring worries

0.294 0.059

−0.274 0.079

−0.290 0.062

N =

42

Sickness absence

Knowledge of workplace goals

Demands higher than control

42

42

Pearson correlation coefficients are in the first row and p-values in the second row for each indicator.

mensions would be more basic or general and thus also more useful in a model discussion. Table 2 below shows the six components that emerged when using standard criteria. Five indicators were clearly belonging to the first component which, in a broad sense, seemed to be about high work demands. The next component had three clear indicators, which were about efficient use of technology, the adaptability level of the workplace, and how well the workplace had developed customer oriented quality practices. A common aspect for much of the content of these indicators was that they covered practices or routines with an aim to increasing performance. We called this component performance drivers. The third component got its name from the only indicator with a clear loading:

adequate competence. The fourth component was very similar to the health performance indicator described in Section 3.2 and we gave it the name general health performance. The indicators on the fifth component were mainly about respectful leadership and creativity so we labelled the component respectful leadership and creativity. Finally, we assigned the sixth component the name knowledge of workplace goals. These component names are to be seen as preliminary interpretations for use in concept development. We did not use these results in the analysis linked to the determinantsoutcomes model presented above. Rather, these results gave a complementary picture of structural relations between indicators. We return to this point in the final section of the paper.

312

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden Table 2 Loadings for 22 indicators on six components Structure Matrix Psychosocially demanding work Work intensity demands Demands higher than control Data stress Sleep problems

1 −0.864 −0.819 −0.712 −0.668 −0.597

2 −0.003 −0.152 −0.564 0.180 −0.010

3 0.182 −0.009 0.106 0.483 0.535

4 −0.215 −0.239 −0.308 −0.109 −0.321

5 −0.145 −0.160 −0.639 −0.235 −0.230

6 −0.090 0.251 −0.068 −0.068 0.055

Efficient use of technology Workplace adaptability Customer oriented quality practices

−0.018 0.370 −0.001

0.812 0.804 0.762

0.006 −0.367 0.045

0.131 0.343 0.011

0.178 0.381 0.083

−0.184 −0.099 −0.063

0.100

0.191

−0.802

0.023

0.027

−0.164

Physical health Psychological well-being Mean absence days per year Depressed mood-worries Musculoskeletal symptoms

0.039 0.486 −0.100 −0.401 −0.198

0.122 −0.006 −0.207 0.133 −0.293

0.153 −0.466 0.034 0.259 −0.318

0.834 0.702 −0.696 −0.686 −0.667

0.072 0.265 −0.340 0.071 −0.321

0.125 −0.024 0.154 0.442 −0.308

Decision latitude Respectful leadership Creative work Customer satisfaction Percieved lack of efficiency Restructuring worries Team spirit

0.151 0.475 0.017 0.578 −0.449 −0.435 0.468

−0.038 0.486 0.411 0.534 −0.180 −0.117 0.486

0.022 −0.372 0.253 −0.169 0.375 −0.369 −0.486

0.096 0.274 0.264 0.290 −0.456 −0.414 0.438

0.804 0.773 0.758 0.710 −0.541 −0.532 0.509

0.303 −0.257 0.063 0.129 0.539 −0.345 −0.396

0.167

−0.072

−0.005

0.165

0.251

0.863

Adequate competence

Knowledge of workplace goals

Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Loadings with an absolute value > 0.4 are in bold. Since we have used an oblique rotation method a measure of total explained variance cannot be obtained; using a varimax rotation with six components assumed, the total explained variance is 74.8 percent.

The oblique rotation method (delta =0.5) allows for correlation between the components. These were summarized in Table 3; from the loadings matrix above it was also directly evident that, for instance, the three dimensions of high work demands, general health performance and respectful leadership and creativity were interrelated in the sense that some indicators have high loadings on two or all of these components. 4.3. Structural analysis to explore groups of companies Here, instead of grouping the variables we used multidimensional scaling 4 to explore the structure of the units. We were asking which groups of similar companies emerged and we also interpreted the underlying di4 Alscal in SPSS 11.1 was used with interval metrics assumed when creating the proximity matrix. The fit indicated by Young’s S-stress measure was 0.232 (R2 = 0.801) for two assumed dimensions, 0.165 (R2 = 0.862) for three assumed dimensions and 0.120 (R2 = 0.913) for four assumed dimensions. The first two dimensions are quite similar in the three solutions and we use the two-dimensional solution here since it is easier to interpret although the methodological advice would be to use a solution with more dimensions assumed.

mensions on which the companies were grouped. The input is a proximity matrix with similarity coefficients between the 41 companies (one company had missing data), created from the same 22 indicators used in the principal component analysis above. In this analysis the result is a matrix with coordinates for the companies in two dimensions. As a guide for interpreting these dimensions, a correlation analysis was performed (see Appendix). This showed that the first dimension correlates most strongly with the indicators respectful leadership, team spirit, customer satisfaction and workplace adaptability, and in the opposite direction with demands higher than control, perceived lack of efficiency and sleep problems. The second dimension correlated strongly with the indicators loading high on the general health performance dimension in the former analysis. As a summary interpretation we named the first dimension “supportive and efficient organization”, and the second was assigned the same name as in the former analysis – “general health performance” – since the indicators coincide. Figure 3 above shows the companies’ positions on these two dimensions. The five companies in the lower left part with high values on both dimensions (“high-

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

313

Table 3 Correlations between the six components and sickness absence

High work demands (INVERT) Performance drivers Adequate competence General health performance Respectful leadership and creativity Knowledge of workplace goals Mean number of absence days last year

High work demands 1 0.113 −0.356 0.471 0.402 −0.303 −0.179

Performance drivers 0.113 1 0.064 0.257 0.359 −0.149 −0.217

Adequate competence −0.356 0.064 1 −0.057 −0.060 0.184 0.024

General health perf 0.471 0.257 −0.057 1 0.387 −0.062 −0.696

Respectful leadership, creativity 0.402 0.359 −0.060 0.387 1 0.125 −0.348

Knowledge of workpl goals −0.303 −0.149 0.184 −0.062 0.125 1 0.114

Mean # of absence days. . . −0.179 −0.217 0.024 −0.696 −0.348 0.114 1

bold = Correlation is significant at the 0.05 level (2-tailed). bold and underline = Correlation is significant at the 0.01 level (2-tailed).

high”) and the five companies in the upper right part with low values on both dimensions (“low-low”) are marked in the diagram. The companies in each group tended to have similar values on several of the separate indicators belonging to the dimensions. As an illustrative example, we compared the three groups “High-high”, “Low-low” and the large intermediary group in an analysis of variance. Altogether, significant (p < 0.05) differences were found between the groups on 11 indicators: creative work, team spirit, decision latitude, perceived customer satisfaction, psychological well-being, physical health, restructuring worries, perceived lack of efficiency, depressed-mood-worries, musculoskeletal symptoms and mean number of absence days last 12 months. These results in a way follow from how the analysis was done. The main point here is not that we found differences, but that this analysis shows that many aspects related to organizational performance and health are, in a broad sense, inter-related. The fact that it is possible to position similar companies on these general dimensions gives some support to the ideas discussed in Section 2 about the existence of “configurations” of companies.

5. Concluding discussion The research comments above imply that we assume the existence of several relationships, which have been summarized in the model in Fig. 2 (Section 3.2) from the perspective of determinants and outcomes. We also applied a more open perspective linked to the structural analysis. The results from the different approaches used are summarized below. According to the correlation analysis all three relations specified in the model between the outcome indicators received empirical support. The strong negative relation between health performance and sickness ab-

Fig. 3. The firms grouped in the two main dimensions of the multidimensional analysis.

sence means that the higher health performance a company has, the lower its sickness absence tends to be. Similarly, the rather strong negative relation between organizational performance and sickness absence also means that the higher organizational performance a company has the lower its sickness absence tends to be. The relation between organizational performance and health performance is rather strong and positive, meaning that firms with good organizational performance tend to have good health performance. We want to point out that neither in the model nor in the analyses do we have a basis for saying something about the causal direction between these concepts or indicators. However, some recent research overviews indicate that it is probable that health performance is a determinant of organizational performance [25,42].

314

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

Regarding the relations between outcomes and determinants, for organizational performance, ten of the twelve assumed relationships with the determinants were strong or rather strong and in the assumed direction. For two determinants, knowledge of workplace goals and decision latitude the correlations were low, indicating the absence of a relation. Regarding decision latitude, this concept belongs to the so-called demandcontrol model [25] and maybe it would be more relevant to use a more complex concept that also takes demand measures into account. In this respect it is interesting to note that the indicator demands higher than control has a strong relation with organizational performance in the assumed direction. Regarding the absence of a clear relation between organizational performance and knowledge of workplace goals, we have no explanation and leave this for further research. With health performance as an outcome indicator there were strong relationships with only two of the assumed ten determinants. However, three indicators related to the demand-control model have rather low p-values – work intensity demands, psychosocially demanding work and demands higher than control. This might indicate that with a more sophisticated use of these three indicators, a higher correlation with health performance might emerge. The remaining low correlations that are contrary to what was assumed in the model might lead to revisions of the model and a deeper discussion about the assumptions regarding determinants of health performance. Here, it is of interest to note somewhat comprehensively that the more general organizational performance concept has a strong relation with health performance while many of the more specific determinants do not correlate clearly. Regarding sickness absence it is assumed in the model that only the two other outcome indicators have relations with this outcome indicator. However, in the analysis two determinants – team spirit and restructuring worries – also had rather high correlations with sickness absence. These results might also lead to revisions of the model. As a general comment, it is noteworthy that a large number of determinants have strong or rather strong relations with organizational performance, while there are only two determinants with strong relations with health performance. This might indicate that the determinants of health performance in small enterprises differ from those in larger enterprises. It is also noteworthy that the indicators respectful leadership and team spirit correlate strongly (r > 0.7) with organizational performance. This may be seen as giving support for

the ideas of the importance of “co-worker respectfulness” for organizational performance. Regarding the structural analyses, two main results from the component analysis merit further comment here: a clear structure emerged with six components, and there were also significant and interesting covariations between some of these components. The indicators grouped with the four components called “performance drivers”, “high work demands”, “general health performance” and “respectful leadership and creativity” give a distinct grouping of what might be seen as more fundamental aspects. It appears possible that the two remaining components, “adequate competence” and “knowledge of workplace goals”, could be conceptually combined under a broad “learning” label. Such a structure could be seen as a potentially fruitful starting point for further conceptual and empirical investigations. The other contribution from this analysis concerns the findings about relations between general components. Those labelled “performance drivers”, “high work demands”, “general health performance” and “respectful leadership and creativity” were rather strongly interrelated. This, together with the result concerning grouping of firms on the two dimensions as described in Section 4.3 relates to recent research by, for instance, Way [51] and Lepak and Snell [31] about configurations of companies. This finding points to the potential fruitfulness of approaches that combine certain groupings. The present study is limited in scale and uses crosssectional data, and the results should therefore be interpreted with caution and be seen as mainly explorative. Outcomes are measured by survey questions. Measurement problems, which may have been created by the use of such questions, might to some extent have been remedied by the rather detailed multivariate indicator construction work. However, future research will need to incorporate financial and other performance information as well as health outcome data from registers and other systems as a means to more strongly validate the findings. We have tried to avoid a causal language and instead used a grouping of indicators into determinants and outcomes. We also applied a structural perspective and structural analysis methods instead of relying on cause-effect reasoning. In summary, this study has identified probable links between healthy work organizational factors and outcomes related to organizational and health performance in small firms. The fields covered here are in much need of conceptual development and integrative analysis to

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden

guide more specific analyses. Using different methods we have also found some evidence of interest for practitioners working with workplace health promotion and organizational performance development in small enterprises. The most important finding in this respect seems to be the inter-relatedness of a large number of factors. This may give some support to those working with concept-driven or holistic approaches by integrating several aspects instead of focusing on any particular process in isolation. The fact that it was possible to single out segments of “high-high” firms and “low-low” firms gives credibility to change strategies oriented to multi-purpose change processes.

[3]

Appendix

[8]

Correlations between indicators and dimensions obtained in a two dimensional ALSCAL-solution

[9]

Psychosocially demanding work Work intensity demands Demands higher than control Data stress Sleep problems Efficient use of technology Workplace adaptability Customer oriented quality practices

[4]

[5]

[6]

[7]

DIM1 0.472 0.496 0.848 0.378 0.504

DIM2 0.243 0.087 −0.160 0.130 0.261

[10]

−0.394 −0.719 −0.297

0.294 0.069 0.257

[12]

Adequate competence

−0.277

0.098

Physical health Psychological well-being Mean number of absence days last year Depressed mood-worries Musculoskeletal symptoms

−0.181 −0.503 0.387 0.277 0.329

−0.819 −0.654 0.469 0.712 0.481

Decision latitude Respectful leadership Creative work Customer satisfaction Perceived lack of efficiency Restructuring worries Team spirit

−0.373 −0.839 −0.551 −0.787 0.662 0.372 −0.817

0.188 0.238 0.170 0.145 0.091 0.234 −0.026

Knowledge of workplace goals

−0.101

−0.208

Bold and underline = correlation is significant at the 0.01 level (2-tailed). Bold = correlation is significant at the 0.05 level (2-tailed).

[11]

[13]

[14]

[15] [16]

[17]

[18]

References [1]

[2]

H. Aguinis, Statistical power problems with moderated multiple regression in management research, Journal of Management 2 (1995), 1141–1158. L.S. Aitken and S.G. West, Multiple regression: Testing and interpreting interactions, Sage, Thousand Oaks, 1991.

[19]

315

A.B. Antonsson, L. Birgersdotter and S. BornbergerDankvardt, Small enterprises in Sweden. Health and safety and the significance of intermediaries in preventive health and safety, Work and Health 2002:1, National Institute for Working Life, Stockholm, 2002. J. Arvonen, Change production and employees. An integrated model of leadership, doctoral dissertation, Stockholm University, 2002. J. Axelsson, Quality and Ergonomics. Towards successful integration, Dissertation No. 616, Link¨oping Studies in Science and Technology, Link¨oping University, 2000. J. Baldwin, W. Chandler, C. Lee and T. Papailiadis, Strategies for Success: A Profile of Growing Small- and Medium-sized Enterprises (GSMEs) in Canada, Statistics Canada, Ottawa, 1994. B. Beckar and B. Gerhart, The impact of human resource management on organizational performance: progress and prospects, Academy of Management Journal 39 (1996), 779– 801. D. Birch, The Job Generation Process. Final Report to Economic Development Administration, MIT Program on Neighbourhood and Regional Change, Cambridge, Mass., 1979. S. Bornberger-Dankvardt, C.G. Ohlson and P. Westerholm, Arbetsmilj¨o- och h¨alsoarbete i småf¨oretag - f¨ors¨ok till helhetsbild, (in Swedish). (Work environment and healthy work in small businesses – seeking the full picture). Arbetsliv i omvandling 2003:1, Arbetslivsinstitutet, Stockholm, 2003. W.F. Cascio, Managing human resources: productivity, quality of work life, profits, McGraw-Hill, Inc., New York, 1995. R. Chaganti, R.G. Cook and J.S. Wayne, Effects of styles, strategies, and systems on the growth of small business, Journal of Developmental Entrepreneurship 7(2) (2002), 175–192. P. Davidsson, F. Delmar and J. Wiklund, eds, Tillv¨axtf¨oretagen i Sverige, (in Swedish), (Small Business Growth Firms in Sweden), SNS f¨orlag, Stockholm, 2001. G.G. Dess, S. Newport and A.M.A. Rasheed, Configuration research in strategic management – key issues and suggestions, Journal of Management 19(4) (1993), 775–795. J.M. Eakin, F. Lamm and H.J. Limborg, International perspective on the promotion of health and safety in small workplaces, in: Systematic occupational health and safety management, K. Frick, P.L. Jensen, M. Quinlan and T. Wilthagen, eds, Elsevier Science Ltd, Oxford, 2000. J. Eklund, Development work for quality and ergonomics, Applied ergonomics 31 (2000), 641–648. P.-E. Ellstr¨om and H. Kock, Continuous improvements of work processes. A strategy for itegrating individual and organisational learning, in: Proceedings of the International Conference on TQM and Human Factors – Towards Successful Integration, J. Axelsson, B. Bergman and J. Eklund, eds, Link¨oping University, Link¨oping, 15–17 June, 1999, pp. 311– 317. V. Evans and D.J. Johnson, The relationship of principal’s of leadership behavior and teacher’s job satisfaction and job related stress, Journal of Instructional Psychology 17 (1990), 11–18. Y. Ganzach, I. Saporta and Y. Weber, Interaction in linear versus logistic models: A substantive illustration using the relationship between motivation, ability and performance, Organizational Research Methods 3(3) (2000), 237–253. B. Gardell, Scandinavian research on stress in working life, International Journal of Health Services 12(1) (1982), 31–41.

316 [20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35] [36] [37]

S. Vinberg and G. Gelin / Organizational and health performance in small enterprises in Norway and Sweden K. Greenan, P. Humphreys and R. McIvor, The green initiative: improving quality and competitiveness for European SMEs, European Business Review 97(5) (1997), 208–214. U. Hellsten and B. Klefsj¨o, Some efforts in Sweden to stimulate TQM work in small enterprises, Proceedings of the 10th Annual National Quality Management Conference, San Diego February, 1998, pp. 25–27. R. Hill and J. Stewart, Human resource development in small organizations, Journal of European Industrial Training 24 (2000), 105–117. M.A. Huselid, The impact of human resource management practises on turnover, productivity and corporate financial performance, Academy of Management Journal 38 (1995), 635– 672. J.V. Johnson and G. Johansson, Work Organisation, Occupational Health and Social Change, Baywood Publishing, New York, 1991. R. Karasek and T. Theorell, Healthy Work: Job Stress, Productivity and the Reconstruction of Working Life, Basic Books, New York, 1990. D.J. Ketchen, J.G. Combs, C.J. Russell, C. Shook, M.A. Dean, J. Runge, F.T. Lohrke, S.E. Naumann, D.E. Haptonstahl, R. Baker, B.A. Beckstein, C. Handler, H. Honig and S. Lamoureux, Organizational configurations and performance: A meta-analysis, Academy of Management Journal 40(1) (1997), 223–240. K.J. Klein and J. Kozlowski, From micro to macro: critical steps in conceptualizing and conducting multilevel research, Organizational Research Methods 3(3) (2000), 211–236. K.A. Kovach, Organization size, job satisfaction, absenteeism and turnover, University Press of America, Washington DC, 1978. G.L. Lee and I. Oakes, The ‘pros’ and ‘cons’ of total quality management for smaller firms in manufacturing: some experiences down the supply chain, Total Quality Management 6 (1995), 413–426. C. LeGrand, R. Szulkin and M. Tåhlin, eds, Sveriges arbetsplatser: organisation, personalutveckling, styrning, (in Swedish), (2nd ed.), (Swedish workplaces: organization, personnel development and leadership). SNS, Stockholm, 1996. D.P. Lepak and S.A. Snell, Examining the human resource architecture: The relationships among human capital, employment, and human resource configurations, Journal of Management 28(4) (2002), 517–543. K. Lindstr¨om, K. Schrey, G. Ahonen and S. Kaleva, The effects of promoting organizational health on worker well-being and organizational effectiveness in small and medium-sized enterprises, in: Healthy and productive work – an international perspective, L.R. Murphy and C.L. Cooper, eds, Taylor & Francis, London, 2000, pp. 83–104. J.P. MacDuffie, Human resource bundles and manufacturing performance: organizational logic and flexible production systems in the world auto industry, Industrial and Labour Relations Review 48 (1995), 197–221. A.D. Meyer, A.S. Tsui and C.R. Hinings, Configurational approaches to organizational Analysis, Academy of Management Journal 36(6) (1993), 1175–1195. R. Miles and C. Snow, Organizational strategy, structure and process, McGraw-Hill, Tokyo, 1978. D. Miller, Configurations revisited, Strategic Management Journal 17(7) (1996), 505–512. L.R. Murphy and C.L. Cooper, Models of healthy work organizations, in: Healthy and productive work – an international

[38] [39] [40]

[41]

[42] [43]

[44]

[45]

[46] [47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56] [57]

perspective, L.R. Murphy and C.L. Cooper, eds, Taylor and Francis, London, 2000, pp. 1–11. G. Noci, Selecting quality based programs in small firms, Small Business Economics 8 (1996), 431–447. J. North, R.A. Blackburn and J. Curran, The quality business. Quality issues and smaller firms, Routledge, London, 1998. S. Nwankwo, Quality assurance in small business organisations: myths and realities, International Journal of Quality & Reliability Management 17(1) (2000), 82–99. J. Pfeffer, Barriers to the advance of organizational science – paradigm development as a dependent variable, Academy of Managenment Review 18(4) (1993), 599–620. J. Pfeffer, The Human Equation. Harvard Business School Press, Boston, 1998. Å. Sandberg, ed., Ledning f¨or alla. Om perspektivbrytningar i f¨oretagsledning, (in Swedish), (3rd ed.), (Leadership for All. About Changing Perspectives in Business Leadership) SNS, Stockholm, 1997. S. Setterlind and G. Larsson, The stress profile: A psychosocial approach to measuring stress, Stress Medicine 11 (1995), 85–92. M.A. Sheppeck, J. Militello and H.R. Strategic, configurations and organizational Performance, Human Resource Management 39(1) (2000), 5–16. D. Storey, Understanding the Small Business Sector, Routledge, London, 1994. D. Smallbone, R. Leigh and D. North, The characteristics and strategies of high growth SMEs, International Journal of Entrepreneurial Behaviour & Research 1(3) (1995), 44–62. S. Vinberg, G. Gelin and K.W. Sandberg, Information technology levels, competence development and performance in Swedish small business enterprises, Behaviour & Information Technology 19(3) (2000), 201–210. S. Vinberg, G. Gelin and K.W. Sandberg, ICT development, job content and profit levels in small and medium sized enterprises, Proceedings NES 2000, At the gateway to Cyberspace – ergonomic thinking in a new millennium, Trondheim, October, 2000. D. Walters, Health and Safety in Small Enterprises, European Strategies for Managing Improvement P I A-Peter Lang S A, Brussels, 2001. S.A. Way, High performance work systems and intermediate indicators of firm performance within the US small business sector, Journal of Management 28(6) (2002), 765–785. S. Wood and S. Albanese, Can we speak of a high commitment management on the shop-floor? Journal of Management Studies 32(2) (1995), 215–247. J. Wiklund, Small Firm Growth and Performance: Entrepreneurship and Beyond, Dissertation, Link¨oping University, Sweden, 1998. G. Wilkinson and B.G. Dale, Integrated management systems: an examination of the concept and theory, The TQM Magazine 11(2) (1999), 95–104. H. Ylinenp¨aa¨ , Competence management and small firm performance, in: Images of Entrepreneurship and Small Business – Emergent Swedish Contributions to Academic Research, B. Johannisson and H. Landstr¨om, eds, Studentlitteratur, Lund, 1999, pp. 217–240. G. Yukl, Leadership in organizations, Prentice-Hall, Inc, New Jersey, 2002. R. Zemke, Rewards and recognition: yes they rally work, Training 25(11) 49–53.

Paper III

Information technology and learning strategies in small enterprises

Behaviour & Information Technology, Vol 19, No 3, 2000, pp 221-227.

Karl W. Sandberg and Stig Vinberg

BEH AVIOU R & IN FOR MATION TECHN OLOG Y ,

2000,

VOL .

19,

N O.

3, 221±227

Information technology and learning strategies in small enterprises K AR L W. SAN D BER G and STIG VIN BER G N ational Institute for Working Life, LuleaÊ U niversity of Technology, Sweden; e-mail: {K arl.W.Sandberg, Stig.Vinberg}@niwl.se

Abstract. The aim of this paper is to de®ne and explain information technology strategic dispositions and competence development within small enterprises. Small enterprises are a heterogeneous group of ®rms and for that reason, a universal explanation of strategic dispositions and learning strategies makes it advisable to draw upon a diverse range of methods, practices and approaches. `Competence ergonomics’ that fosters learning, empowerment and participation oers the ¯exibility within which to frame this strategy description. Introduction of information technology into small enterprises is treated as being of strategic importance to the future growth and survival of a ®rm. M ethods and theories now being used in The Swedish N ational Institute of Working Life’s (Arbetslivsinstitutet) Balanced Information Technology Project are those described. The Project is located in the area of OÈ stersund, Sweden where small industries have the need to adopt information technology as a means of promoting continuing competitiveness.

1. Introduction The changing market environment and the shortening of product life cycles requires not only new innovative products but also a permanent organizational adjustment within small companies (M ertins et al. 1993). These time constraints and an evolving environment necessitates an integrated planning of organizational structures, technology and human resources. This integrated planning increases the complexity of organizational change processes. To shorten planning times and eorts, the involvement of skilled employees with their practical experience is necessary. These abilities to plan and the knowledge of quality of working life circumstances cannot be presupposed but have to be *Present address, N ational Institute for Working Life, LuleaÊ University of Technology, Sweden ; e-mail: {K arl.W.Sandberg, Stig. Vinberg}@ niwl.se

mediated for employees. Therefore, there is a need for training methods that are adapted to adults and support an active learning process. On the other side, these methods must focus on the process of organisational change, introduction of new technologies, and quality of working life standards. F or these reasons, research has demonstrated the need for design of models for competence development and IT-strategies for learning in small enterprises (Vinberg and Sandberg 1999).

2. IT strategies in small enterprises Information technology is of great importance to economic development in a national as well as a regional perspective. It has, however, been noted that small enterprises are not using information technology to the same extent as larger ®rms (Lorentzon 1994). These factors give rise to the problem ®eld of this study: the diusion of information technology among small enterprises. As information technology is considered a strategic resource for organisations, diusion of technology will be considered as a consequence of the use of information technology. Studies of why small enterprises adopt information technology have focused on operational issues rather than strategic ones (D oudikis et al. 1994). The focus of such studies has also been more empirical than theoretical in their nature. The following discussion accounts for a number of perspectives on strategy and their implications for IT strategies. The concept of strategies can be described as plans, patterns, positions or perspectives. Adoption of information technology in small enterprises is not initiated by managers within the ®rm, but by external

Behaviour & Informatio n T echnolog y ISSN 0144-929 X print/ISSN 1362-300 1 online Ó 2000 Taylor & Francis Ltd http://www.tandf.co.uk/journals

222

K. W . Sandberg and S. V inberg

actors (D abinett and G raham 1994). Perceptions of this technology, the situation, the managers and individual characteristics constitute a base for the strategic disposition of information technology in small enterprises. To explore the inter-relationship of each change element, it is bene®cial to examine the connections among these four change elements to see if there are any signi®cant relationships between elements. As shown in ®gure 1 there are four pairs of relationships. These are:

· · · ·

Technology coherence: the strategic integration of management processes and IT. Organizational coherence: the strategic integration of structure and human actors. Inner infrastructure: the strategic ®t between management processes and human actors. Outer infrastructure: the strategic ®t between IT and structure.

Identifying these pairings facilitated understanding the importance on how to align change elements within the overall `organization’s strategy’ element.

An ability to cope with information overload has also been described as an important competitive advantage (Lin et al. 1993). While costs are relatively easy to determine bene®ts have been hard to de®ne (M cLeod 1993). Attribute perceptions of information technology play an important role when ®rms evaluate technology in comparison with other approaches. Studies have shown a low degree of correlation between information technology and organizational changes (R obey and Azavedo 1994). R elative advantage has also been hard to distinguish; in both empirical studies on compatibility (M oore and Benbasat 1991) and with perceived compatibility, a concept with several dimensions (R amiller 1994).

3. Shortcomings of traditional ergonomics To improve an organization’s ergonomic situation many factors have to be considered. F or example, resources have to be allocated in the form of money, knowledge, assessment, planning, and ®nally implemen-

Figure 1. General model for the impact of IT strategy and learning.

Information technology and learning strategies tation. Traditionally, ergonomists’ activities have focused on the design of speci®c jobs as well as related human system interfaces and work environments, in other words, a focus on the micro-levels of the systems. H endrick (1995) identi®es three major causes of the shortcomings of traditional ergonomics:

· · ·

It has been too technology-centred. There has been a leftover approach, in which whatever the machine cannot do is left over to the operator. There has been a failure to integrate sociotechnical issues.

N ot all ergonomic improvements are put into practice due to various in¯uences. One may be the lack of acceptance of ergonomic improvements among employees. An illustration of this point is as follows: in a workplace, an overhead crane was installed to move heavy boxes from one pallet to another. It had been designed and installed by the technical group at the corporation’s headquarters. The investment was sizeable. H owever, the crane was never utilised, since, according to the employees, it was too time-consuming and impractical to use. There was also fear that the boxes would break open and that the contents (bottles) would then break due to poor cardboard box quality (IngelgaÊ rd 1998). In this case, the overhead crane was an attractive solution from the designer’s point of view, but a failure in practice. The employees continued to work as they had before: without the crane. This case illustrates the importance of involving the workers in the design and creation of a tool that suits the special conditions at this workplace, e.g. lifting and moving boxes of poor quality. Table 1.

4. Ergonomics and the in¯uence of socio-technical theories There is a close connection between the sociotechnical design of the work organization and ergonomics as ergonomics is concerned with the implementation of both technical and social subsystems. F rom an ergonomic standpoint the technical subsystems are equipment, tools, workplace design, environmental design and software. The social subsystems of ergonomics are job design, training, culture, management style and communication (Joyce 1994). Systems need to be designed jointly for both technical and organizational needs to be optimized. The goal of this integration is to maximize the acceptance and eective use of technology within organizations and to minimize potentially negative impacts. The result is usually a greater assurance of optimal system functioning and eectiveness, including productivity, safety, comfort, intrinsic employee motivation, and quality of working life. The socio-technical systems design approach has also been criticised for the diculty in de®ning such terms as `joint optimization’, `autonomy’, `social systems’ and `well designed’ (H ackman and Oldham 1980). The overall conclusion from IngelgaÊ rd’s (1998) research was that ergonomics may be enhanced using theories of change, speci®cally a learning strategy for change, an acceptance of non linearity, an iterative approach, empowerment of employees, a broad change strategy as well as interdisciplinary and management support. In this circumstance, empowerment is de®ned as an active participation in a change process of organisation, technology, work processes and quali®cation. The empowerment of human actors is understood as `the ability to act individually and in co operation with

Impact of information technology upon jobs.

Objective

Information technology system as control

Information technology system as tools

Job changes

D eskilled Paced Structured M onitored R epetitive Isolated D issatisfaction Controlled M achine stress H igher productivity Control mechanisms In¯exibility

Enriched U npaced U nstructured N ot monitored Varied Social contact Satisfaction Empowerment N o machine stress Similar productivity Quality and initiative F lexibility

Human responses Organisational eect

223

K. W . Sandberg and S. V inberg

224

others in order to control work processes, to in¯uence the structures positively and to improve the performances of a socio technical system’ while considering the well being of the individual (Butera 1995). Table 1 shows the eects of dierent information technology strategies at workplaces.

5. Learning and participation The interest seen during the 1990s in the introduction and implementation of information technology has led to an interest in organizational design and management. H endrick (1995) and Antalovits (1994) acknowledge that there is a strong need for human centred and user participation approaches to design in traditional ergonomics. Participative ergonomics is a growing ®eld of ergonomics and organizational design and management. The accomplishment of permanent participative ergonomics requires the empowerment of people to make decisions as well to realize and evaluate those decisions. Some of the main focuses of participative ergonomics are organizational and psychosocial work environment, human resource development and human competence, labour market and society in change, front line technologies and well being and health (Bradley and H endrick 1994). M any dierent issues have to be addressed in the implementation of technical change to the working conditions for many people today. Eason (1990) gives the following list of concurrent events:

· · · · ·

Installing and testing the technical system. D esigning the local workplace. Training and supporting the employees. Changing the organization. Ensuring acceptance of change.

The fundamental point of why a participative approach is necessary is that only those who will be aected by the change can decide what is in their best interest. Participation may be an important factor for learning at the workplace. The chance to participate and thus in¯uence the working environment bene®ts the psychosocial climate, leading to an enhanced quality of the improvements. Traditional teaching that is largely a presentation of facts cannot serve the aims of empowerment. F or introduction of IT into an organization, learning by doing (with readily available support) is the more appropriate strategy for the acquisition new skills. Learning is always situated in a context, be it in school, at work or at home. In business organizations, training is often conducted in order for the employees to improve

their competence, that is, their ability to perform some kind of task, or in the words of Lyttkens (1994) , our ability to realise our intentions. H owever, there are several aspects of competence. The organization may ask for some kind of formal competence: you have to be quali®ed. Yet, the actual task may demand more or less of what is ocially required. Also, the employee may have a certain amount of competence, but perhaps does not make full use of it. EllstroÈm (1992) makes a distinction between adjustment competence and development competence. The former is the competence an individual develops in order to be able to perform his/ her present tasks where the goals are pre-de®ned, whereas the latter is the competence needed for the individual to be able to in¯uence his/her work and life. Training may take place on the job, in special locations within the work area, or at a totally dierent place. D ierent kinds of methods may be used such as mentoring, guided delegation, job rotation, discussion, and analysis of critical incidents, experiment, and problem solving (G albraith and Zelenak 1991, Watkins 1989). As a result of strong in¯uence from cognitive psychology, learning is a major topic in humancomputer interaction research. Theories have been developed with respect to how learning is taking place, and also with respect to what kinds of knowledge the learning might result in and what kind of content is of interest to learn. Environmental factors (trainer, technology, organization and users) help set the goals of the training. The type of technology (new and dierent) and the organizational context (educational institution) also help set the focus and design of a training programme. The design of training materials (the manual), while keeping the above goals and environmental factors in mind, consists of choosing the content, medium (reference manual, tutorial, online, video, etc.), the type of interaction (listening/viewing, questioning, interacting or reading) and the style (organization, format, consistency and conventions). If we want to improve practice in some way, we must, ®rst of all, acquire a more comprehensive understanding of the systems development process, including not only technical aspects, but also behavioural and sociological aspects. This holds true whether the aim is higher pro®ts, technical eciency, quality of working life or democracy at work. (Bansler and H avn 1991).

6. Integrating strategy for learning and competence ergonomics The competence ergonomics approach is not so dierent from the learning strategy where several topics

Information technology and learning strategies for change are dealt with at the same time (Sandberg and Vinberg 1999). Integrating a learning strategy and competence ergonomics could perhaps be one way of enhancing implementation of ergonomic systems design. The pattern employees typically follow when alternating between workplace learning and distance learning (e.g. web based learning) in their work environment is illustrated in ®gure 2. As is suggested by the illustration, the capacity to maintain a higher workload is likely to be greater with workplace learning. N ewly acquired skills may be used immediately and ease of alternating between work and learning is likely to be less disruptive. N ew technology brings more and more knowledgebased elements into industry and the public infrastructure. K nowledge may now be looked upon as oering the greatest value for enhanced development and competitiveness. Learning must take place at the individual level, as well as the team level. This requires a cross-disciplinary approach, and a culture that fosters co-operative learning. To maintain development and to be ahead, individuals within an organization need relevant education and lifelong learning thereafter. One step is to realise that the era of the separated `production factory’ and `education factory’ has passed. Enterprises and educational institutions have in common a most important long-term issue: how to develop the knowledge of the enterprise or institution so that competitiveness will be maintained in a rapidly changing marketplace where level of competence is rising and quality requirements are tightened. The role of research is to support the organisation’s own development and change. A learning strategy as advocated by N orrgren (1996) includes many kinds of simultaneous change in dierent parts of an organization and an increasing competence

Figure 2. Learning and work.

225

in handling future changes. By an iterative approach, employees are able to experiment and then draw conclusions based on actions performed and their relation to organisational goals. In this way, an acceptance of change is also gained. This learning strategy strives to inspire a feeling of security in participants in the change process. In this way, process control is not in the hands of a few initiated persons, which may lead to a lessened acceptance of change among other employees. The learning strategy may lead to an increased quality of proposed solutions because the leaders of the change process are active in the sense that they can positively intervene if there are, for example, con¯icts among groups or individual employees concerning the goals of any change. On an individual level, personality factors such as previous experiences of change, interest in and commitment to the job, as well as interest in personal development in connection with the job, may be important determinants of change. H aÊ kansson (1995) found that the most successful organizations were those who had chosen an appropriate level of change strategy, in the sense that they had the right experience and knowledge. She found that some organizations used change strategies that were too advanced for the organization: the maturity required for the chosen change process was lacking. If these change strategies are considered as a `toolbox’ model where those most suited to a particular organization’s level of maturity are used, the following would most likely be used:

· · ·

participative approach taking into account the history acceptance of non-linearity

K. W . Sandberg and S. V inberg

226

· · · · ·

an iterative approach the empowerment of all employees at all levels a broad change strategy interdisciplinary support management support

In this way, learning strategy could, besides experts, also involve the workforce in co-designing ergonomic changes. Eason (1995), argued that the employees should take part in the design process, and that the ergonomist be involved as a facilitator, helping the user to establish what is in the user’s best interest. This proposal compliments learning strategy, as advocated by N orrgren (1996) . Elsewhere, from the theories of group psychology and organisational psychology, many concepts may be incorporated into competence ergonomics. F or example, one general impression is that eective people get allocated more and more work, and that in general they complain about this. F rom this, it is then interesting to note that the notion of stress has become a concern and a problem for the organizations too. This may be related not only to an increased general workload but also to demands stemming from new ways of performing and organising work, notably project work and organisational change processes. Joining one project after the other may in the long run become overly stressful for the individual, especially for sought after project leaders. U sing an action research perspective with the goal to change this situation seems to be an important future research topic.

7.

Discussions

If there is not a good match between system and organization, this also aects the success of the learning environment. Organizational development should be based on a competent use of technology resulting in a general and continuing increase in competence. The theories and models presented here have not been developed and used in frame of competence ergonomics and learning. This also indicates that there is a need for further development of the concept of learning and competence ergonomics. That is especially signi®cant for small enterprises. The traditional approaches to work design, discussed at the begining of this paper, assume that most people can work productively and eciently on routinised, standardised jobs. Behavioural approaches, on the other hand, suggest that work eectiveness will be enhanced if employees have jobs that are not simple and repetitive Ð that are, instead, complex, meaningful, and challenging. And the socio-technical systems approach accepts the major premises of the behavioural approaches, but goes beyond them in asserting the importance of group

relationships and organisation environment transactions in establishing eective work systems. Our approach to the design and redesign of work, the approach that will guide the models and prescriptions for change now being adopted and tested in businesses in the area around OÈ stersund, Sweden by the Balanced Information Technology Project under the sponsorship of The N ational Institute for Working Life (Arbetslivsinstitutet), is something of a hybrid of the behavioural and systems approaches. Its major features can be summarised as follows: we draw heavily on one of the behavioural approaches to work design, job characteristics theory, in specifying the attributes of jobs that create conditions for positive work motivation. In addition, we ®nd that increasing the levels of responsibility, meaningfulness, and feedback that are built into jobs often can enhance work motivation. It has been shown that those companies that invest in educational programs for those that are left on the job have a higher potential for growth in the long run (Cascio 1991, 1995). These results indicate that though organizations are pressured to economise, the importance of involving and creating possibilities for employees to learn and grow is still important. Perhaps even more so in the near future, as the importance of knowledge seems to increase rapidly in the labour market (EllstroÈm 1996). F inally, it should be noted that a basic assumption underlying the work presented in this report is that the design of learning environments for users of computer systems can be enhanced, and that in order to accomplish this an expanded framework must be used. This paper has contributed to such an expansion, by developing a pedagogical design framework integrating several knowledge areas. The results can be used in the further development of theory and practise something needed where, for example, there are growing knowledge areas such as the introduction of new information technology into organizations. F uture papers that describe the outcomes of the application of this model in the OÈ stersund area are planned. References A N TA L O VI TS, M . 1994, Changing education and training in ergonomics. Proceeding of the 12th T riennial Congress of the International Ergonomics Association (August 15-19) (Toronto: IEA), 1, 48 ± 51. BA N SLE R , J. P., and H A VN , E. 1991, The nature of software work: systems development as a labour process, in P. Van den Besselaar, A. Clement, and P. JaÈrvinen (eds.) Information system, work and organisational design: Proceedings of the IFIP T C9/W G9.1 W orking Conference on Information System, W ork and Organization (July 10 ± 13) (N orth-Holland, Amsterdam), 145 ± 153.

Information technology and learning strategies BR A DL E Y , G ., E., and H E N D R I C K , H .W. 1994, Preface, in Bradley, G . E. and Henrik, H. W. (eds.) Human Factors in Organizational Design and Management IV , (North-H olland, Amsterdam), pp. v. BU T ER A , F . 1995, Quality of Working Life Criteria and Empowerment in R e-Engineering and Continuous Improvement of N etwork Organizations Supported by Information Technology: A Conceptual F ramework for the Qualit Project, Innovative W ork Organization in Europe and North America. Proceedings of the 4th European Ecology of W ork Conference, Shankill, Co. (May 9 ± 12). (European F oundation for the Improvement of Living and Working Conditions, Dublin), 22 ± 64. C A SC I O, W. F . 1991, Costing Human Resources: T he Financial Impact of Behavior in Organizations, 3rd edition (Boston: PWS-K EN T Publishing Company). C A SC I O, W. F . 1995, Managing Human Resource: productivity, quality of work life, pro®ts, 4th edition (N ew York: M cG raw H ill). D A BI NE TT , G . and G R A H A M , S. 1994, Telematics and industrial change in Sheeld, U K. Regional Studies, 28(6), 605 ± 617. D O U D I KI S, G . I., SM IT HSO N , S. and L Y BERE A S, T. 1994, Trends in information technology in small businesses. Journal of End User Computing. 6(4), 15 ± 25. E A SO N, K. D . 1990, N ew systems implementation, in Wilson, J. R . and Corlett, E. N. (eds) Evaluation of Human W ork: A Practical Ergonomics Methodology (London: Taylor & F rancis). E A SO N, K ., D . 1995, U ser centred design: for users or by users? Ergonomics, 388, 1667 ± 1673. E LL STROÈ M , P. E. 1992, Kompetens, Utbildning och L aÈrande i Arbetslivet. Problem, Begrepp och T eoretiska Perspektiv [Competence, training and learning in working life. Problems, concepts and theoretical perspectives] (Stockholm: Publica). E LL STROÈ M , P. E. 1996, Arbete och L aÈrande [Working and learning] (Solna: Arbetslivsinstitutet). G A L BRA I TH , M . W. and Z EL EN A K , B. S. 1991, Adult learning methods and techniques, in M . W. G albraith (ed.) Facilitating adult learning: a T ransactional Process (K rieger, M alabar), 103 ± 133. H A C KM A N , J. R . and O LD H A M , G . R. 1980, Work R edesign (R eading: Addison Wesley). H AÊ K A N SSO N , K. 1995, FoÈraÈndringsstrategier i Arbetslivet. M onograph from the D epartment of Sociology. (G oÈteborg: U niversity of GoÈteborg). H E N D R I CK , H . W. 1995, F uture directions in macroergonomics, Ergonomics, 388, 1617 ± 1624. I N G E LG AÊ R D , A. 1998, On macroergonomics and learning strategies in improving working conditions. Thesis (D epartment of Psychology, G oÈteborg U niversity). J O Y C E, M . 1994, Ergonomic management: integrating a macroergonomics plan into existing corporate cultures. Proceeding of the 12th T riennial Congress of the International Ergonomics Association (August 15 ± 19) (IEA: Toronto) 6, 13 ± 15.

227

Ê and C LA R K , L S. 1993, Information L I N , B. VA SSAR , J A technology strategies for small businesses. Journal of Applied Business Research, 9(2), 25 ± 29. L O R EN TZ O N , S. 1994, AnvaÈ ndning av Informationsteknik i FoÈretag Ð lokala och Regionala KonkurrensfoÈrutsaÈttningar [U se of information technology in company Ð local and regional conditions for competition] (GoÈteborgs: H andelshoÈgskolan of G oÈteborgs Universitet). L Y TT K E N S, L. 1994, Kompetens och Individualisering [Competence and individualization]. Agenda 2000 Rapport 8. Ds 1994:18 (Stockholm: Utbildningsdepartementet) M CL E O D , R .J. 1993, M anagement information systems: a study of computer-based information systems, 5th ed. (New York: M acmillan). M ER T IN S, K ., A L BR E C HT , R. and E D E L ER , H . 1993, M anufacturing philosophy for the new european F actory, in I. A. Pappas and I. P. Tatsiopoulos (eds) Advances in Production Management Systems: Proceedings of the IFIP T C5/W G5.7 Fifth International Conference on Advances in Production Management Systems Ð APMS `93 (Amsterdam: NorthH olland), 31 ± 38. M O O RE , G . C. and BE N BA SA T, I. 1991, Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192 ± 222. N O RR G R E N , F . 1996, F oÈraÈndringsledning ett aktionsforskningsprojekt vid arbetsvetenskapliga kollegiet/IMIT i G oÈteborg [M anagement change and action research with ergonomics] in G . Ekvall (ed.) NavigatoÈr och InspiratoÈr (Lund: Studentlitteratur). R A M I LL ER , N . C. 1994, Perceived compatibility of information technology innovations among secondary adopters: towards a reassessment. Journal of Engineering and T echnology Management, 11(1), 1 ± 23. R O BE Y, D . and A Z EVE DO , A. 1994, Cultural analysis of the organisational consequences of information technology. Accounting, Management & Information T echnology, 4(1), 22 ± 37. SA N D BE R G , K. W. and VIN BE R G , S. 1999, Competence ergonomics and learning in enterprises. Proceedings of the 10th Y ear Anniversary of M.Sc.Ergonomics International Conference (October 29±30) (Sweden: LuleaÊ University of Technology), 166 ± 172. VI NBE R G , S., and SA N D BER G , K. W. (1999). M odels for competence and health development in small business enterprises. Proceedings of the 10th Y ear Anniversary of M.Sc.Ergonomics International Conference (October 29 ± 30) (Sweden: LuleaÊ U niversity of Technology), 178 ± 184. W A TK IN S, K . E. 1989, Business and industry, in S. B. M erriam and P. M . Cunningham (eds) Handbook of Adult and Continuing Education (San F rancisco: Jossey Bass), 422 ± 435.

Paper IV

Change processes and health outcomes in microenterprises and small public workplaces in rural areas

Finnish Journal of Rural Research and Policy, Vol 12, No 4, 2004, pp 151-162.

Stig Vinberg

&/53&13&/&634)*1

$IBOHF
QSPDFTTFT

BOE
IFBMUI
PVUDPNFT
JO
NJDSPFOUFSQSJTFT
BOE
TNBMM
QVCMJD
XPSLQMBDFT
JO
SVSBM
BSFBT
4ĥĚĘ
7ĚğēĖģĘ

5

IF
NBJO
QVSQPTF
PG
UIJT
QBQFS
JT
UP
BOBMZ[F
BOE
EJTDVTT
UIF
JNQPSUBODF
PG
PSHBOJ[BUJPOBM
GBDUPST
GPS
FYQMBJO JOH
UIF
EJĊFSFODFT
JO
TJDLOFTT
BCTFODF
CFUXFFO
NJDSPFOUFSQSJTFT
BOE
TNBMM
QVCMJD
XPSLQMBDFT
JO
B
SVSBM
BSFB
JO
OPSUIFSO
4XFEFO
"OPUIFS
GPDVT
BT
BO
JOQVU
GPS
EFWFMPQJOH
SVSBM
FOUSFQSFOFVSTIJQ
QPMJDJFT
JT
UP
TIFE
MJHIU
PO
TPNF
OFFET
GPS
DIBOHF
QSPDFTTFT
JO
UIF
NJDSP FOUFSQSJTFT
TUVEJFE
3FTVMUT
PG
RVFTUJPOOBJSFT
BOE
JOUFSWJFXT
XJUI
NBOBHFST
PG

NJDSP
FOUFSQSJTFT
BSF
QSFTFOUFE
BOE
DPNQBSFE
XJUI
UIPTF
GSPN

QVCMJD
XPSLQMBDFT
ɨF
mOEJOHT
QPJOU
UP
B
TIBSQFS
JODSFBTF
JO
TJDLOFTT
BCTFODF
JO
UIF
TNBMM
QVCMJD
XPSLQMBDFT
EVSJOH
SFDFOU
ZFBST
XIJDI
JT
JO
MJOF
XJUI
HPW FSONFOUBM
JOWFTUJHBUJPOT
BOE
SFTFBSDI
SFTVMUT
UIBU
TIPX
B
DMFBS
QPTJUJWF
SFMBUJPOTIJQ
CFUXFFO
XPSLQMBDF
TJ[F
BOE
TJDLOFTT
BCTFODF
ɨF
GBDU
UIBU
NJDSPFOUFSQSJTFT
VTF
TUSVDUVSFE
TZTUFNT
UP
B
MFTTFS
FYUFOU
DBO
CF
TFFO
BT
B
SFTVMU
PG
UIF
BQQMJDBUJPO
PG
B
NPSF
MFBSOJOHCBTFE
TUSBUFHZ
JO
UIFTF
FOUFSQSJTFT
*U
JT
BMTP
QPTTJCMF
UIBU
NJDSP
FOUFSQSJTFT
NBZ
CF
JOWFTUJOH
JO
IVNBO
DBQJUBM
BOE
UIF
XPSLJOH
FOWJSPONFOU
XJUIPVU
VTJOH
TUSVDUVSFE
TZTUFNT
"O
JOUFSFTUJOH
mOEJOH
JT
UIBU
UIF
NBOBHFST
PQJOJPOT
SFHBSEJOH
DIBOHFT
SFMBUFE
UP
FDPOPNJD
PVUDPNFT
OVNCFS
PG
DVTUPNFST
BOE
PSHBOJ[B UJPOBM
BTQFDUT
UIF
XPSL
TIPX
B
NPSF
QPTJUJWF
QJDUVSF
UIBO
UIBU
PGUFO
EJTDVTTFE
JO
DPOOFDUJPO
XJUI
NJDSP
FOUFSQSJTFT
JO
SVSBM
BSFBT
/FFET
GPS


/FX
3VSBM
1PMJDZ


JNQSPWFNFOUT
SFMBUFE
UP
DPNQBOZ
OFUXPSLT
DPPQFSBUJPO
CFUXFFO
DPNQBOJFT
BOE
UIF
MP DBM
BVUIPSJUJFT
SFDSVJUNFOU
DPPQFSBUJPO
XJUI
SFTFBSDIFST
BOE
DPNQFUFODF
EFWFMPQNFOU
DBO
OFWFSUIFMFTT
CF
JEFOUJmFE
"
DPODFJWBCMF
DPODMV TJPO
JT
UIBU
BTQFDUT
SFMBUFE
UP
XPSL
PSHBOJ[BUJPO
BSF
PG
NPSF
JNQPSUBODF
GPS
IFBMUI
PVUDPNFT
JO
NJDSPFOUFSQSJTFT
UIBO
NPSF
GPSNBMJ[FE
BEKVTU NFOU
PG
UIF
XPSLJOH
FOWJSPONFOU


*OUSPEVDUJPO
BOE
QVSQPTF ɨF
DPOTJEFSBCMF
JODSFBTF
JO
TJDLOFTT
BCTFODF
JO
4XFEFO
EVSJOH
UIF
MBTU
GFX
ZFBST
IBT
SFBDIFE
B
MFWFM
UIBU
UISFBUFOT
HPWFSONFOU
mOBODFT
BOE
UIF
XFMGBSF
TFSWJDFT
DBVTJOH
UIF
HPWFSONFOU
UP
QSPQPTF
TQFDJmD
BDUJPOT
UP
CSFBL
UIJT
VOEFTJSBCMF
USFOE
#VEHFUQSPQPTJUJPOFO
 
/FXTQBQFS
BSUJDMFT
UFMM
VT
BCPVU
HPWFSONFOU
QSPQPTBMT
UP
GPSDF
FNQMPZFST
UP
QBZ

QFSDFOU
PG
UIF
TJDL OFTT
CFOFmUT
JO
UIF
OFBS
GVUVSF
BOE
UP
JOTJTU
UIBU
FNQMPZFST
XIP
BSF
OPU
JOWFTUJOH
JO
SFIBCJMJUB UJPO
JNQSPWFNFOUT
UP
UIF
XPSLJOH
FOWJSPO NFOU
BOE
PDDVQBUJPOBM
IFBMUI
TFSWJDFT
TIPVME
QBZ

QFSDFOU
PG
UIF
TJDLOFTT
CFOFmUT
½TUFS TVOETo1PTUFO
 
ɨF
CBDLHSPVOE
UP
UIFTF
QSPQPTBMT
MJFT
JO
UIF
GSFRVFOUMZ
FYQSFTTFE
WJFXT
UIBU
IFBMUI
QSPCMFNT
BNPOH
JOEJWJEVBMT
NBZ
CF
DPOOFDUFE
XJUI
FYDFTTJWFMZ
EFNBOEJOH
XPSLJOH
DPOEJUJPOT
ɨF
MFWFM
PG
TJDLOFTT
BCTFODF
DMFBSMZ
JODSFBTFT
XJUI
UIF
TJ[F
PG
UIF
XPSLQMBDF
BOE
JT
IJHIFS
JO
UIF
QVCMJD
UIBO
JO
UIF
QSJWBUF
TFDUPS




&/53&13&/&634)*1

"DDPSEJOH
UP
POF
PG
UIF
NBOZ
HPWFSONFOUBM
JO WFTUJHBUJPOT
406




MFWFMT
BSF
UXP
UP
GPVS
UJNFT
MPXFS
JO
4XFEJTI
NJDSPFOUFSQSJTFT
UIBO
JO
MBSHFS
FOUFSQSJTFT
BOE
PSHBOJ[BUJPOT
&YQMBOBUJPOT
UIBU
BSF
TPNFUJNFT
NFOUJPOFE
BSF
UIBU
FNQMPZFFT
JO
TNBMM
FOUFSQSJTFT
IBWF
NPSF
FOHBHFNFOU
JO
UIFJS
XPSL
BOE
UIBU
UIF
SFMB UJPO
CFUXFFO
QSFTFODF
BU
XPSL
QSPEVDUJPO
BOE
TFDVSJUZ
PG
FNQMPZNFOU
JT
DMFBSFS
UIBO
JO
MBSHFS
PSHBOJ[BUJPOT
%FTQJUF
UIF
FYUFOU
BOE
DPOTF RVFODFT
PG
UIJT
EFWFMPQNFOU
POF
SFDFOU
SFWJFX
PG
SFTFBSDI
IBT
TUBUFE
UIBU
PVS
LOPXMFEHF
PG
UIF
GBDUPST
UIBU
DBVTF
JU
JT
MJNJUFE
4#6
 
*U
JT
SFNBSLBCMF
UIBU
UIJT
SFWJFX
GPVOE
POMZ
POF
BQ QSPWFE
TUVEZ
PVU
PG

JEFOUJmFE
XIJDI
EFBMU
XJUI
SFMBUJPOTIJQT
CFUXFFO
TJDLOFTT
BCTFODF
BOE
UIF
TJ[F
PG
UIF
XPSLQMBDF 3FTFBSDI
JOUP
TNBMM
FOUFSQSJTFT
IBT
CFFO
B
GBTUHSPXJOH
TQFDJBMUZ
EVSJOH
UIF
QBTU
EFDBEF
CPUI
JO
4XFEFO
BOE
JO
PUIFS
&VSPQFBO
DPVOUSJFT
%BWJETTPO
FU
BM

+PIBOOJTTPOo-BOETUSÚN

8JLMVOE
 
"NPOH
UIF
SFBTPOT
GPS
UIJT
JOUFSFTU
JT
UIF
GBDU
UIBU
TNBMM
XPSLQMBDFT
DPNQSJTF
UIF
PWFSXIFMNJOH
NBKPSJUZ
PG
XPSL QMBDFT
BOE
FNQMPZ
B
TJ[BCMF
QSPQPSUJPO
PG
UIF
UPUBM
MBCPVS
GPSDF
JO
NPTU
DPVOUSJFT
o
CFUXFFO
POF
UIJSE
BOE
POF
IBMG
&BLJO
FU
BM
 
%F TQJUF
FYUFOTJWF
SFTFBSDI
PVS
LOPXMFEHF
PG
IPX
PSHBOJ[BUJPOBM
GBDUPST
JOnVFODF
IFBMUI
QFSGPSN BODF
JO
TNBMM
FOUFSQSJTFT
JT
MJNJUFE
XIJDI
JT
SF NBSLBCMF
JO
WJFX
PG
UIF
TUBUFNFOU
JO
UIF
FBSMJFS
#PMUPO
SFQPSU

UIBU *O
NBOZ
SFTQFDUT
UIF
TNBMM
mSN
QSPWJEFT
B
CFUUFS
FOWJSPONFOU
GPS
UIF
FNQMPZFF
UIBO
JT
QPTTJCMF
JO
NPTU
MBSHF
mSNT
"MUIPVHI
QIZTJDBM
XPSLJOH
DPO EJUJPOT
NBZ
TPNFUJNFT
CF
JOGFSJPS
JO
TNBMM
mSNT
NPTU
QFPQMF
QSFGFS
UP
XPSL
JO
B
TNBMM
HSPVQ
XIFSF
DPNNVOJDBUJPOT
QSFTFOU
GFXFS
QSPCMFNT
UIF
FN QMPZFFT
JO
UIF
TNBMM
mSN
DBO
FBTJMZ
TFF
UIF
SFMBUJPO
CFUXFFO
XIBU
UIFZ
BSF
EPJOH
BOE
UIF
PCKFDUJWFT
BOE
QFSGPSNBODF
PG
UIF
mSN
BT
B
XIPMF
8IFSF
NBOBHFNFOU
JT
NPSF
EJSFDU
BOE
nFYJCMF
XPSLJOH
SVMFT
DBO
CF
WBSJFE
UP
TVJU
UIF
JOEJWJEVBM
/P
EPVCU
NBJOMZ
BT
B
SFTVMU
PG
UIJT
UIF
UVSOPWFS
PG
TUBĊ
JO
TNBMM
mSNT
JT
WFSZ
MPX
BOE
TUSJLFT
BOE
PUIFS
LJOET
PG
JOEVTUSJBM
EJTQVUF
BSF
SFMBUJWFMZ
JOGSFRVFOU

3FTFBSDI
JT
BMTP
GSBHNFOUJ[FE
XIFO
JU
DPNFT




UP
NJDSPFOUFSQSJTFT
XIJDI
BSF
NPSF
OVNFSPVT
JO
SFMBUJWF
UFSNT
JO
SVSBM
BSFBT
TVDI
BT
+ÊNUMBOE
$PVOUZ
UIBO
JO
UIF
NPSF
QPQVMBUFE
QBSUT
PG
4XFEFO
ɨF
UXP
QFSTQFDUJWFT
JOUSPEVDFE
BCPWF
DPO TUJUVUF
UIF
NBJO
QVSQPTF
PG
UIJT
QBQFS
XIJDI
JT
UP
BOBMZ[F
BOE
EJTDVTT
UIF
JNQPSUBODF
PG
PSHBOJ [BUJPOBM
GBDUPST
GPS
FYQMBJOJOH
UIF
EJĊFSFODFT
JO
TJDLOFTT
BCTFODF
CFUXFFO
NJDSPFOUFSQSJTFT
BOE
TNBMM
QVCMJD
XPSLQMBDFT
JO
B
SVSBM
BSFB
"OPUIFS
GPDVT
BT
BO
JOQVU
JOUP
EFWFMPQJOH
SVSBM
FOUSFQSF OFVSTIJQ
QPMJDJFT
JT
UP
TIFE
MJHIU
PO
TPNF
OFFET
GPS
DIBOHF
JO
UIF
NJDSPFOUFSQSJTFT
TUVEJFE
.BOBHFST
PG
NJDSPFOUFSQSJTFT
BSF
DPOUSBTUFE
XJUI
NBOBHFST
PG
QVCMJD
XPSLQMBDFT
XJUI
SFTQFDU
UP
UIFJS
WJFXT
PO
PSHBOJ[BUJPOBM
PVUDPNFT
XPSL
PSHBOJ[BUJPO
GBDUPST
BOE
UIF
HPBMT
PG
DIBOHFT
ɨF
BSUJDMF
JT
PSHBOJ[FE
JO
UIF
GPMMPXJOH
XBZ
'JSTU
BO
PWFSWJFX
PG
FBSMJFS
SFTFBSDI
JT
QSFTFOUFE
UPHFUIFS
XJUI
TPNF
UIFPSFUJDBM
BTQFDUT
ɨFO
B
EFTDSJQUJPO
PG
UIF
EBUB
DPMMFDUJPO
NFUIPET
BOE
XPSLQMBDFT
JT
HJWFO
ɨF
SFTVMU
QBSU
DPNQBSFT
UIF
NJDSPFOUFSQSJTFT
XJUI
UIF
QVCMJD
TFDUPS
XPSLQMBDFT
BOE
UIF
BSUJDMF
FOET
XJUI
B
EJTDVT TJPO
BOE
TPNF
QSPQPTBMT
GPS
GVSUIFS
SFTFBSDI


-JUFSBUVSF
PWFSWJFX
BOE
TPNF
UIFPSFUJDBM
BTQFDUT *U
IBT
CFFO
DPODMVEFE
JO
TFWFSBM
4XFEJTI
JOWFTUJ HBUJPOT
EVSJOH
UIF
MBTU
GFX
ZFBST
UIBU
XPSLQMBDFT
CFMPOHJOH
UP
MPDBM
BVUIPSJUJFT
BOE
DPVOUZ
DPVO DJMT
IBWF
UIF
IJHIFTU
MFWFMT
PG
TJDLOFTT
BCTFODF
%T


406

 
4JODF
UIFSF
BSF
MPDBM
WBSJBUJPOT
JO
TJDLOFTT
BCTFODF
CFUXFFO
XPSLQMBDFT
JU
JT
TUBUFE
JO
UIFTF
JOWFTUJHBUJPOT
UIBU
EJĊFSFODFT
JO
UIF
XPSLJOH
FOWJSPONFOU
BOE
PSHBOJ[BUJPOBM
GBDUPST
DPOTUJUVUF
BO
JNQPS UBOU
FYQMBOBUJPO
.BOZ
TUVEJFT
IBWF
GPVOE
TJH OJmDBOU
DPOOFDUJPOT
CFUXFFO
XPSL
FOWJSPONFOU
GBDUPST
BOE
JOEJWJEVBM
IFBMUI
'PS
FYBNQMF
UIFSF
JT
B
HSFBUFS
SJTL
PG
TJDLOFTT
BCTFODF
BNPOH
NBOVBM
XPSLFST
UIBO
BNPOH
TBMBSJFE
FNQMPZFFT
BOE
UIF
SJTL
JT
IJHI
GPS
JOEJWJEVBMT
EPJOH
EF NBOEJOH
QIZTJDBM
BOE
QTZDIPTPDJBM
XPSL
4[àDT
FU
BM
 
ɨF
JNQPSUBODF
PG
EJĊFSFODFT
JO
XPSL
PSHBOJ[BUJPO
IBT
CFFO
QPPSMZ
FYQMPSFE


/FX
3VSBM
1PMJDZ


&/53&13&/&634)*1

IPXFWFS
8FTUFSIPMNo.BSLMVOE
 
*O
B
'JOOJTI
TUVEZ
EFBMJOH
XJUI
FĊFDUT
PG
TUBĊ
DVUT
JO
B
MPDBM
BVUIPSJUZ
,JWJNÊLJ
FU
BM

JU
XBT
QPJOUFE
PVU
UIBU
TJDLOFTT
BCTFODF
JODSFBTFE
BGUFS
UIF
PSHBOJ[BUJPOBM
DIBOHF
BOE
UIBU
UIJT
JODSFBTF
QSPCBCMZ
XBT
EFQFOEFOU
PO
B
MPXFSJOH
PG
UIF
FNQMPZFFT
JOnVFODF
BU
XPSL
BOE
HSFBUFS
QIZTJDBM
BOE
NFOUBM
CVSEFOT
*O
B
TUVEZ
PG

4XFEJTI
MPDBM
BVUIPSJUJFT
4[àDT
FU
BM

UIF
SFTVMU
XBT
UIBU
UIF
EFHSFF
UP
XIJDI
UIF
FNQMPZ FFT
XPSSJFE
BCPVU
PSHBOJ[BUJPOBM
DIBOHFT
XBT
DMPTFMZ
DPSSFMBUFE
XJUI
UIF
OVNCFS
PG
QFPQMF
PO
UIF
TJDL
MJTU
3FTFBSDI
JOUP
UIF
SFBTPOT
CFIJOE
TJDLOFTT
BCTFODF
DBO
CF
WJFXFE
GSPN
EJĊFSFOU
QFSTQFD UJWFT
BOE
JU
JT
QSPCBCMZ
OPU
QPTTJCMF
UP
DPNCJOF
JOGPSNBUJPO
GSPN
EJĊFSFOU
BSFBT
PG
SFTFBSDI
JOUP
B
HFOFSBM
UIFPSZ
4#6
 
4UFFST

3IPEFT

QSPQPTF
B
QSPDFTT
NPEFM
JO
XIJDI
BC TFOUFFJTN
DBO
CF
TFFO
BT
B
DPNCJOBUJPO
PG
UIF
XPSL
TJUVBUJPO
QFSTPOBM
USBJUT
BOE
UIF
TPDJBM
FOWJSPONFOU
JNQMZJOH
UIBU
QSFTFODF
BU
XPSL
JT
B
GVODUJPO
PG
NPUJWBUJPO
BOE
BCJMJUZ
3FTFBSDI
SFMBUFE
UP
XPSLJOH
NPSBMF
IBT
CFFO
GPDVTJOH
PO
UIF
EFWFMPQJOH
PG
SVMFT
TP
UIBU
BCTFODF
JT
TFFO
BT
B
HSPVQ
QIFOPNFOPO
.BSUPDDIJP
 
8JUI
SFTQFDU
UP
XPSLJOH
DPOEJUJPOT
,BSBTFL

OPUFE
UIBU
UIF
XPSL
EFNBOET
BOE
DPO USPM
BSF
DFOUSBM
EJNFOTJPOT
GPS
UIF
FNQMPZFFT
NPUJWBUJPO
MFBSOJOH
BOE
IFBMUI
BOE
UIBU
UIJT
NPEFM
DBO
BMTP
CF
BQQMJFE
BU
BO
PSHBOJ[BUJPOBM
MFWFM
$SJUJDJTN
PG
UIF
NPEFM
IBT
SFTVMUFE
JO
UIF
BEEJUJPO
PG
B
TPDJBM
EJNFOTJPO
UP
UIF
NPEFM
SFTVMUJOH
JO
UIF
A+PC
%FNBOE$POUSPM4VQQPSU
NPEFM
+PIOTPOo)BMM
 
*O
NPSF
PSHBOJ [BUJPOBMMZ
GPDVTFE
NPEFMT
GBDUPST
TVDI
BT
NFBO JOHGVM
XPSL
XIJDI
VUJMJ[FT
UIF
XPSLFST
TLJMMT
XPSLFS
BVUPOPNZDPOUSPM
KPC
TFDVSJUZ
SFXBSET
GPS
QFSGPSNBODF
XPSLFS
JOWPMWFNFOUQBSUJDJQB UJPO
UFBNXPSL
BOE
TBGF
BOE
IFBMUIZ
QIZTJDBM
XPSLJOH
FOWJSPONFOUT
BSF
TFFO
BT
JNQPSUBOU
GPS
FNQMPZFFT
IFBMUI
.VSQIZo$PPQFS
 
4PNF
PG
UIFTF
MBUFS
NPEFMT
BMTP
TVHHFTU
UIBU
BQBSU
GSPN
CFJOH
PG
JNQPSUBODF
GPS
FNQMPZFFT
IFBMUI
TVDI
GBDUPST
DPVME
DPOUSJCVUF
UP
PSHBOJ [BUJPOBM
FĊFDUJWFOFTT

/POF
PG
UIF
NPEFMT
EFTDSJCFE
BCPWF
UBLF


/FX
3VSBM
1PMJDZ


BDDPVOU
PG
UIF
TJ[F
PG
UIF
XPSLQMBDF
BMUIPVHI
JU
JT
PGUFO
TBJE
UIBU
XPSL
JO
TNBMMFS
FOUFSQSJTFT
JT
PSHBOJ[FE
NPSF
JOGPSNBMMZ
XJUIPVU
CVSFBV DSBUJD
TUSVDUVSFT
"YFMTTPO
 
4PNF
BVUIPST
QPJOU
PVU
UIBU
MFBSOJOH
QSPDFTTFT
JO
TNBMM
FOUFS QSJTFT
BSF
JOUFHSBUFE
JOUP
UIF
EBJMZ
XPSL
BOE
UIBU
AUBDJU
LOPXMFEHF
JT
NPSF
JNQPSUBOU
UIBO
GPSNBM
FEVDBUJPO
:MJOFOQÊÊ
 
"DDPSEJOH
UP
#PSOCFSHFS%BOLWBSEU
FU
BM



TNBMM
mSNT
GBWPVS
B
HPPE
UFBN
TQJSJU
MPX
BCTFODF
MFWFMT
CFDBVTF
PG
TJDLOFTT
WBSJBUJPO
JO
XPSLJOH
UBTLT
HPPE
XPSL
TBUJTGBDUJPO
PO
BDDPVOU
PG
B
HSFBUFS
JOnVFODF
BU
XPSL
BOE
NPSF
QFSTPOBM
FO HBHFNFOU
ɨFTF
SFTVMUT
BSF
JO
MJOF
XJUI
B
'JOO JTI
TUVEZ
-JOETUSÚN
FU
BM

XIJDI
TIPXT
UIBU
FNQMPZFFT
JO
NJDSPFOUFSQSJTFT
FYQFSJFODF
NPSF
DPOUSPM
PWFS
UIFJS
XPSL
CFUUFS
TVQQPSU
GSPN
UIFJS
NBOBHFST
B
CFUUFS
XPSLJOH
DMJNBUF
BOE
NPSF
BEFRVBUF
JOGPSNBUJPO
UIBO
UIPTF
JO
MBSHFS
FOUFSQSJTFT
ɨFSF
BSF
BMTP
OFHBUJWF
GBDUPST
JO
TNBMM
FOUFSQSJTFT
IPXFWFS
TVDI
BT
QIZTJDBM
QSPCMFNT
XJUI
UIF
XPSLJOH
FOWJSPONFOU
SJTLT
PG
JOKVSZ
IJHI
TJDLOFTT
QSFTFODF
BOE
B
MBDL
PG
QFSTPOBM
BOE
FDPOPNJD
SFTPVSDFT
GPS
JOWFTUNFOU
JO
UIF
XPSLJOH
FOWJSPONFOU
#PSOCFSHFS%BOL WBSEU
FU
BM
  0OF
SFDFOU
JOWFTUJHBUJPO
'ÚSFUBHBSOB

OFWFSUIFMFTT
OPUFT
UIBU
FNQMPZFFT
JO
TNBMMFS
mSNT
EP
OPU
DPNF
UP
XPSL
XIFO
UIFZ
BSF
TJDL
BOZ
NPSF
UIBO
PUIFST
BOE
UIBU
UIFZ
XPSL
NPSF
PWFSUJNF
IBWF
HSFBUFS
PQQPSUVOJUJFT
UP
EFDJEF
PO
UIFJS
QBDF
PG
XPSL
BOE
UBLF
EFDJTJPOT
BCPVU
TUFQQJOH
VQ
UIFJS
XPSL
ɨF
MPX
MFWFMT
PG
TJDL OFTT
BCTFODF
IBWF
CFFO
DPOmSNFE
JO
CPUI
4XFE JTI
#PSOCFSHFS%BOLWBSEU
FU
BM

BOE
&VSPQFBO
TUVEJFT
8BMUFST
 
0OF
FYQMBOB UJPO
mUT
JO
XJUI
UIF
BCPWF
JEFBT
UIBU
FNQMPZFFT
JO
TNBMM
FOUFSQSJTFT
IBWF
NPSF
QPTTJCJMJUJFT
UP
JOnVFODF
UIFJS
XPSL
UIF
DPOUFOU
PG
UIFJS
XPSL
JT
CFUUFS
BOE
MFBEFSTIJQ
JT
NPSF
EFWFMPQFE
&BLJO
FU
BM
 
"
DPOUSBEJDUPSZ
FYQMBOBUJPO
JT
UIBU
TPDJBM
DPOUSPM
JT
IJHIFS
JO
UIFTF
FOUFSQSJTFT
BOE
UIBU
FNQMPZFFT
HP
UP
XPSL
GPS
UIJT
SFBTPO
EFTQJUF
IFBMUI
QSPCMFNT
#PSOCFSHFS%BOLWBSEU
FU
BM
  8JMLJOTPO

BMTP
QPJOUFE
PVU
UIF
QPM BSJ[FE
QJDUVSF
PG
XPSLJOH
DPOEJUJPOT
JO
TNBMM




&/53&13&/&634)*1

FOUFSQSJTFT
"DDPSEJOH
UP
POF
WJFX
UIFSF
BSF
IBSNPOJPVT
SFMBUJPOT
CFUXFFO
FNQMPZFFT
BOE
MFBEFST
DPOnJDUT
BSF
VOVTVBM
BOE
XPSL
JT
nFY JCMF
CVU
BDDPSEJOH
UP
BOPUIFS
UIF
MFBEFSTIJQ
JT
BVUIPSJUBSJBO
BOE
XPSLJOH
DPOEJUJPOT
BSF
VOTBUJTGBDUPSZ
XJUI
IJHI
JOKVSZ
SJTLT
BOE
B
CBE
QIZTJDBM
FOWJSPONFOU
ɨVT
TPNF
SFTFBSDIFST
IBWF
RVFTUJPOFE
UIF
TJNQMJmFE
WJFX
PG
TNBMM
FOUFSQSJTFT
BOE
IBWF
DMBJNFE
JOTUFBE
UIBU
DPO EJUJPOT
BSF
DPNQMFY
BOE
DPOUSBEJDUPSZ
3BN
 
'BNJMJBS
DJSDVNTUBODFT
DBO
JODSFBTF
UIF
QPTTJCJMJUJFT
GPS
B
NPSF
PQFO
EJBMPHVF
GPS
FY BNQMF
CVU
UIFZ
DBO
BMTP
MFBE
UP
NPSF
DPOnJDUT
CFUXFFO
MFBEFST
BOE
HSPVQT
PG
FNQMPZFFT
*O
POF
PG
B
GFX
TUVEJFT
PG
PSHBOJ[BUJPOBM
BTQFDUT
PG
NJDSPFOUFSQSJTFT
.BUMBZ

TIPXFE
UIBU
MFBEFSTIJQ
JO
&OHMJTI
NJDSPmSNT
XBT
NPSF
JOGPSNBM
UIBO
JO
TNBMM
PS
MBSHFS
FOUFSQSJTFT
8PSLSFMBUFE
QSPCMFNT
TFFNFE
UP
B
HSFBU
EF HSFF
UP
CF
TPMWFE
CZ
JOGPSNBM
EJTDVTTJPOT
BU
UIF
XPSLQMBDF
BOE
UIJT
MFBEFSTIJQ
TUZMF
DPOUSJCVUFE
UP
B
QPTJUJWF
XPSLJOH
DMJNBUF
"
4XFEJTI
TUVEZ
PG
XPSL
FOWJSPONFOU
SPVUJOFT
JO
NJDSPFOUFS QSJTFT
"YFMTTPO

DPODMVEFE
UIBU
UIFZ
XFSF
JOGPSNBMMZ
PSHBOJ[FE
GBNJMJBS
BOE
nFYJCMF
CVU
BMTP
UIBU
UIFTF
DPOEJUJPOT
DPVME
PCTUSVDU
NPSF
GPSNBMJ[FE
BUUFNQUT
BU
JNQSPWJOH
UIF
XPSLJOH
FOWJSPONFOU
BOE
SFEVDF
FNQMPZFFT
QPTTJCJMJUJFT
GPS
JOnVFODJOH
TVDI
JNQSPWFNFOUT
8IFO
JU
DPNFT
UP
DIBOHF
QSPDFTTFT
SFMBUFE
UP
DPNQBOZJOUFSOBM
GBDUPST
UIFSF
BSF
TFWFSBM
TUVEJFT
UIBU
TIPX
QPTJUJWF
FĊFDUT
BU
BO
PSHBOJ [BUJPOBM
MFWFM
%FMBOFZo)VTFMJE

1GFĊFS

&BTUPOo+BSFMM
 
.BOZ
PG
UIFN
BSF
GPDVTFE
PO
MBSHFS
FOUFSQSJTFT
BOE
PSHBOJ[BUJPOT
PS
PO
TNBMM
PSHBOJ[BUJPOT
UIBU
BSF
OFWFSUIFMFTT
MBSHFS
UIBO
NJDSP
POFT
*U
JT
PGUFO
TBJE
UIBU
TNBMM
FOUFSQSJTFT
IBWF
B
MBDL
PG
SFTPVSDFT
BOE
NPUJWB UJPO
BNPOH
UIFJS
NBOBHFST
BOE
XPSLFST
BT
GBS
BT
DIBOHF
QSPDFTTFT
BU
XPSL
BSF
DPODFSOFE
&BLJO
FU
BM



CVU
UIFSF
BSF
TPNF
TUVEJFT
GPDVTJOH
PO
TNBMM
mSNT
XIJDI
SFQPSU
QPTJUJWF
FĊFDUT
PG
DIBOHF
QSPDFTTFT
BU
UIF
DPNQBOZ
MFWFM
MFBE JOH
UP
RVBMJUZ
JNQSPWFNFOUT
"OEFSTPOo4PIBM

)BOTTPO



XPSLJOH
FOWJSPONFOU
EFWFMPQNFOU
"YFMTTPO

"MQFOCFSHo,BSMT TPO

7JOCFSH

PS
DPNQFUFODF
EFWFM



PQNFOU
#BHFS4KÚHSFO
FU
BM 
ɨFSF
BSF
BMTP
SFTFBSDIFST
XJUI
B
DSJUJDBM
WJFX
PG
DIBOHF
QSPDFTTFT
JO
TNBMM
FOUFSQSJTFT
0OF
&OHMJTI
TUVEZ
4UPSFZ

TVDDFFEFE
JO
TIPXJOH
TPNF
DPOOFDUJPOT
CFUXFFO
DPN QFUFODF
EFWFMPQNFOU
BOE
FDPOPNJD
PVUDPNFT
JO
TNBMM
mSNT
CVU
UIF
SFTVMUT
XFSF
XFBLFS
BOE
DPOUSBEJDUPSZ
SFMBUJWF
UP
UIF
TJUVBUJPO
JO
MBSHFS
PSHBOJ[BUJPOT
XIJMF
PUIFS
TUVEJFT
IBWF
PCTFSWFE
UIBU
NPEFMT
GPS
MFBSOJOH
BOE
DPNQFUFODF
EFWFM PQNFOU
BEBQUFE
UP
MBSHF
PSHBOJ[BUJPOT
DBOOPU
FBTJMZ
CF
USBOTQPTFE
UP
TNBMM
FOUFSQSJTFT
)JMM 4UFXBSU
 

4JODF
TNBMM
FOUFSQSJTFT
BSF
PGUFO
DIBSBDUFS J[FE
CZ
B
EFNBOEJOH
QIZTJDBM
FOWJSPONFOU
XJUI
BO
JODSFBTFE
SJTL
PG
BDDJEFOUT
TPNF
SFTFBSDIFST
IBWF
CFFO
JOUFSFTUFE
JO
QSFWFOUJWF
NFBTVSFT
BU
TVDI
XPSLQMBDFT
"DDPSEJOH
UP
8BMUFST



UIFSF
BSF
TQFDJBM
QSPCMFNT
DPOOFDUFE
XJUI
TVDI
NFBTVSFT
JO
NJDSPFOUFSQSJTFT
TJODF
UIFZ
BSF
FO USFQSFOFVSJBM
BOE
IBWF
B
MPX
EFHSFF
PG
PSHBOJ[B UJPOBM
TUSVDUVSF
MJNJUFE
SFTPVSDFT
BOE
VODFSUBJO
UFSNT
PG
FNQMPZNFOU
&BLJO
FU
BM

QSP QPTF
PO
UIF
CBTJT
PG
BO
PWFSWJFX
PG
DIBOHF
TUSBU FHJFT
VTFE
JO
TNBMM
mSNT
JO
&VSPQFBO
DPVOUSJFT
UIBU
JU
JT
OFDFTTBSZ
UP
BDIJFWF
B
EFFQFS
VOEFS TUBOEJOH
PG
TPDJBM
BOE
PSHBOJ[BUJPOBM
GBDUPST
JO
UIFTF
mSNT
JO
PSEFS
UP
EFWFMPQ
NPSF
HFOFSBMJ[FE
NPEFMT
GPS
JNQSPWFNFOUT
JO
UIF
XPSLJOH
FOWJ SPONFOU
$POUSBSZ
UP
UIJT
TVHHFTUJPO
IPXFWFS
+FOTFO
FU
BM

TUBUF
UIBU
MFBEFST
BOE
XPSL FST
LOPXMFEHF
JT
BEFRVBUF
GPS
JNQSPWJOH
UIF
XPSLJOH
FOWJSPONFOU
BOE
GPS
DPPQFSBUJOH
XJUI
FYUFSOBM
SFTPVSDFT
0OF
4XFEJTI
EJTTFSUBUJPO
+PIBOTTPO

BSHVFT
GPS
B
MFBSOJOHCBTFE
DIBOHF
TUSBUFHZ
JOTUFBE
PG
B
QSPHSBNNBUJD
POF
JO
TNBMM
FOUFSQSJTFT
TJODF
QMBOOJOH
BOE
DPOUSPM MJOH
BSF
VOVTVBM
JO
TNBMM
mSNT
UPP
NVDI
QMBO OJOH
DBO
CF
BO
PCTUBDMF
XIFO
UIF
DIBOHFT
UBLF
QMBDF
RVJDLMZ
UIFSF
JT
SFTJTUBODF
UP
BO
JODSFBTFE
EFHSFF
PG
GPSNBMJ[BUJPO
BOE
UIF
FNQMPZFFT
FO HBHFNFOU
BOE
JOnVFODF
DBO
CF
JNQSPWFE
XIFO
VTJOH
B
MFBSOJOHCBTFE
TUSBUFHZ




*O
TVNNBSZ
HPWFSONFOUBM
JOWFTUJHBUJPOT
BOE
SFTFBSDI
EVSJOH
UIF
MBTU
GFX
ZFBST
QPJOU
UP
UIF
JNQPSUBODF
PG
PSHBOJ[BUJPOBM
GBDUPST
XIFO
IFBMUI
PVUDPNFT
BSF
EJTDVTTFE
BOE
FNQIBTJ[F


/FX
3VSBM
1PMJDZ


&/53&13&/&634)*1

UIBU
TVDI
PVUDPNFT
EJĊFS
EFQFOEJOH
PO
UIF
TJ[F
BOE
TFDUPS
PG
UIF
XPSLQMBDF
ɨFSF
JT
B
MBDL
PG
LOPXMFEHF
BCPVU
IPX
TUSPOH
UIFTF
DPOOFD UJPOT
BSF
JO
EJĊFSFOU
DPOUFYUT
BOE
SFTFBSDI
JT
MJNJUFE
BOE
DPOUSBEJDUPSZ
DPODFSOJOH
XIZ
TNBMM
FOUFSQSJTFT
IBWF
MPXFS
TJDLOFTT
BCTFODF
MFWFMT
UIBO
MBSHFS
PSHBOJ[BUJPOT
.Z
PQJOJPO
JT
UIBU
JOEJWJEVBM
SFMBUFE
FYQMBOBUJPO
NPEFMT
NVTU
CF
DPNQMFNFOUFE
XJUI
NPSF
PSHBOJ[BUJPOBMMZ
GPD VTFE
POFT
JO
PSEFS
UP
PCUBJO
BO
VOEFSTUBOEJOH
PG
UIF
NFDIBOJTNT
CFIJOE
PDDVQBUJPOBM
IFBMUI
QSPCMFNT
FTQFDJBMMZ
JO
TNBMM
mSNT
ɨFSF
JT
TPNF
VODFSUBJOUZ
BT
UP
IPX
DIBOHF
QSPDFTTFT
JOnVFODF
UIF
PVUDPNFT
JO
TNBMM
FOUFSQSJTFT
BOE
UIFSF
BSF
TFWFSBM
SFBTPOT
XIZ
TJNQMJmFE
BOE
NPSF
nFYJCMF
NPEFMT
BSF
OFFEFE




%BUB
BOE
SFTFBSDI
NFUIPET 3BHVOEB
JT
B
TQBSTFMZ
QPQVMBUFE
NVOJDJQBM JUZ
PG


JOIBCJUBOUT
TJUVBUFE
JO
UIF
FBTUFSO
QBSU
PG
+ÊNUMBOE
$PVOUZ
-JLF
PUIFS
SVSBM
BS FBT
UIJT
SFHJPO
JT
DIBSBDUFSJ[FE
CZ
IJHI
VO FNQMPZNFOU
PVUNJHSBUJPO
PG
ZPVOH
QFPQMF
BOE
FĊPSUT
UP
mOE
TUSBUFHJFT
GPS
TVSWJWBM
BOE
EFWFMPQNFOU
$WFULPWJD}
 
%FTQJUF
IFBWZ
FDPOPNJD
TBWJOHT
QSPHSBNNFT
UIF
NVOJDJQBMJUZ
IBT
TUBSUFE
VQ
QSPKFDUT
GPDVTJOH
PO
MPDBM
HSPXUI
DPNQFUFODF
BOE
DPNQBOZ
EFWFMPQNFOU
WPO
#FSHNBOO7JOCFSH
  %VSJOH
UIF
TQSJOH
PG

B
QSFMJNJOBSZ
BOBMZTJT
PG
DPNQBOZ
TUSVDUVSFT
JO
UIF
NVOJDJ QBMJUZ
XBT
NBEF
CBTFE
PO
B
SFHJTUFS
EFWFMPQFE
CZ
MPDBM
BVUIPSJUZ
FYQFSUT
XIJDI
TIPXFE
UIBU
UIFSF
XFSF
B
UPUBM
PG

FOUFSQSJTFT

XJUI
POF
QFSTPO
FNQMPZFE

XJUI
o
QFSTPOT
BOE

XJUI
NPSF
UIBO

QFSTPOT
*O
UPUBM
UIF
UFO
MBSHFTU
DPNQBOJFT
IBE

QFSTPOT
FN QMPZFE
BOE
UIF
TNBMM
DPNQBOJFT

QFSTPOT
ɨJT
TUSVDUVSF
NFBOT
UIBU
UIFSF
BSF
UISFF
HSPVQT
PG
FOUFSQSJTFT
JO
3BHVOEB
NJDSPFOUFSQSJTFT
o
FNQMPZFFT


TNBMM
FOUFSQSJTFT
o
FN QMPZFFT
BOE
NFEJVNTJ[FE
FOUFSQSJTFT
o
FNQMPZFFT 

"
mSTU
TBNQMF
PG

FOUFSQSJTFT
XBT
DIPTFO
JO
DPPQFSBUJPO
XJUI
DPNQBOZ
FYQFSUT
JO
UIF
NVOJDJQBMJUZ
ɨF
BNCJUJPO
XBT
UP
PCUBJO
B


/FX
3VSBM
1PMJDZ


SFQSFTFOUBUJWF
TBNQMF
PG
UIF
UISFF
HSPVQT
UIBU
JODMVEFE
FOUFSQSJTFT
GSPN
EJĊFSFOU
CSBODIFT
PG
BDUJWJUZ
ɨF
mSTU
TBNQMF
PG

mSNT
EFDSFBTFE
UP

FOUFSQSJTFT
EVF
UP
XJOEJOH
VQ
QSPCMFNT
FUD
BOE
B
UPUBM
PG

JOUFSWJFXT
XJUI
NBOBHFST
XFSF
QFSGPSNFE
EVSJOH
TQSJOH

JNQMZJOH
B
SFTQPOTF
SBUF
PG

QFS
DFOU
 
ɨF
JO UFSWJFXT
XFSF
CBTFE
PO
B
TFNJTUSVDUVSFE
HVJEF
XJUI
RVFTUJPOT
BCPVU
NBSLFUT
SFTPVSDFT
MFBSO JOH
BOE
UIF
DPNQBOZ
DMJNBUF
ɨF
SFTQPOEFOUT
BMTP
BOTXFSFE
B
RVFTUJPOOBJSF
PO
PVUDPNFT
PSHBOJ[BUJPOBM
GBDUPST
BOE
DIBOHF
HPBMT
/FBSMZ
BMM
UIF
JOUFSWJFXT
UPPL
QMBDF
BU
UIF
NBOBHFST
XPSLQMBDFT ɨF
SFTVMUT
GPS

NJDSPFOUFSQSJTFT
BSF
QSF TFOUFE
IFSF

FNQMPZJOH
o
QFSTPOT

FNQMPZJOH
o
QFSTPOT
BOE

FNQMPZJOH
o
QFSTPOT 
ɨF
QSFEPNJOBOU
CSBODIFT
PG
UIF
FDPOPNZ
SFQSFTFOUFE
BSF
XPPE
QSPDFTTJOH
NBOVGBDUVSJOH
BOE
UIF
SFUBJM
BOE
TFSWJDF
TFD UPS
$PNQBSJTPOT
BSF
NBEF
XJUI
B
HSPVQ
PG

QVCMJD
XPSLQMBDFT
XJUI
o
QFSTPOT
FNQMPZFE
JO
IFBMUI
DBSF
BOE
FEVDBUJPO


XIFSF
UIF
MFBEFST
BOTXFSFE
B
TJNJMBS
RVFTUJPOOBJSF


3FTVMUT ɨJT
TFDUJPO
QSFTFOUT
RVBOUJUBUJWF
BOE
RVBMJUB UJWF
SFTVMUT
GPS
UIF

NJDSPFOUFSQSJTFT
UPHFUIFS
XJUI
DPNQBSJTPOT
XJUI
UIF
RVBOUJUBUJWF
SFTVMUT
GPS
UIF

TNBMM
QVCMJD
TFDUPS
XPSLQMBDFT
5P
QSPWJEF
BO
PQJOJPO
PO
UIF
EFWFMPQNFOUBM
QPT TJCJMJUJFT
GPS
UIF
XPSLQMBDFT
UIF
MFBEFST
BTTFTT NFOUT
PG
TPNF
PG
UIF
PSHBOJ[BUJPOBM
PVUDPNFT
BSF
TIPXO
BOE
UIFJS
PQJOJPOT
BCPVU
WBSJPVT
BT QFDUT
PG
XPSL
PSHBOJ[BUJPO
MFBSOJOH
BOE
DIBOHF
TUSBUFHJFTHPBMT
BSF
QSFTFOUFE
ɨFTF
BTQFDUT
DBO
CF
SFMBUFE
UP
UIF
PSHBOJ[BUJPOBM
PVUDPNFT
UIPVHI
UIF
UIFPSFUJDBM
WJFXQPJOUT
PVUMJOFE
BCPWF




$IBOHFT

JO
PSHBOJ[BUJPOBM
PVUDPNFT
*U
DBO
CF
TFFO
GSPN
5BCMF

UIBU
SFMBUJWFMZ
NBOZ
PG
UIF
NJDSPFOUFSQSJTF
NBOBHFST
FTUJNBUF
UIBU
UVSOPWFS
QSPmUT
BOE
OVNCFST
PG
DVTUPNFST
BOE
QSPEVDUT
IBWF
JODSFBTFE
EVSJOH
UIF
MBTU
GFX




&/53&13&/&634)*1

)PX
EP
ZPV
KVEHF
UIBU
UIF
GPMMPXJOH
BTQFDUT
IBWF
DIBOHFE
GSPN
UIF
ZFBS

UP
UIF
ZFBS

QFSDFOUBHFT
PG
FOUFSQSJTFTXPSLQMBDFT
XIJDI
BOTXFSFE
UIF
RVFTUJPO
XJUI
UIF
BMUFSOBUJWFT
iEFDSFBTFEw
BOE
iJODSFBTFEw








.JDSPFOUFSQSJTFT

/


%FDSFBTFE

*ODSFBTFE


5VSOPWFS
1SPmU
/VNCFS
PG
DVTUPNFST
/VNCFS
PG
QSPEVDUT
$PNQFUFODF
EFNBOET
BNPOH
FNQMPZFFT
3FTQPOTJCJMJUJFT
BNPOH FNQMPZFFT
/VNCFS
PG
XPSL
UBTLT
4UBĊ
UVSOPWFS
0WFSUJNF
4JDLOFTT
BCTFODF
-FBEUJNF
GPS
TFSWJDFTQSPEVDUT
$PNQMBJOUTEJTTBUJTmFE DVTUPNFST



1VCMJD
XPSLQMBDFT
/

%FDSFBTFE
*ODSFBTFE


























  















































    


































#PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJmDBOU
BU
UIF

MFWFM


5BCMF

$IBOHFT
JO
PSHBOJ[BUJPOBM
PVUDPNFT
JO
UIF
NJDSPFOUFSQSJTFT
BOE
TNBMM
QVCMJD
XPSL QMBDFT

ZFBST
"CPVU
POF
UIJSE
PG
UIF
NBOBHFST
SFQPSUFE
UIBU
XPSLJOH
UBTLT
DPNQFUFODF
EFNBOET
BOE
SFTQPOTJCJMJUJFT
IBE
JODSFBTFE
BNPOH
UIF
FN QMPZFFT
XIJMF
UIF
UJNF
SFRVJSFE
GPS
QSPEVDJOH
QSPEVDUT
BOE
TFSWJDFT
IBE
EFDSFBTFE
JO
B
mGUI
PG
UIF
mSNT
.BOZ
PG
UIF
MFBEFST
JO
UIF
QVCMJD
XPSLQMBDFT
BMTP
SFQPSUFE
UIBU
UIF
OVNCFST
PG
DVTUPNFST
BOE
TFSWJDFT
DPNQFUFODF
EFNBOET
BOE
SFTQPOTJCJMJUJFT
BNPOH
UIF
FNQMPZFFT
IBE
JODSFBTFE
BOE
B
IJHIFS
QSPQPSUJPO
KVEHFE
UIBU
JOEJDBUPST
SFMBUFE
UP
XPSL
MPBE
DPNQFUFODF
EF NBOET
TJDLOFTT
BCTFODF
BOE
QSPEVDUJPO
UJNFT
IBE
SJTFO


0SHBOJ[BUJPOBM
BTQFDUT
PG
XPSL
DPNQFUFODF
BOE
TZTUFNT ɨF
NBOBHFST
PG
UIF
NJDSPFOUFSQSJTFT
TUBU FE
UIBU
UIF
FNQMPZFFT
BMUFSOBUF
CFUXFFO
XPSL
UBTLT
UP
B
HSFBU
FYUFOU
BOE
EFWFMPQFE
FYJTUJOH
XPSL
UBTLT
CZ
UIFNTFMWFT
"MUPHFUIFS
UIF
NJDSP
FOUFSQSJTFT
TFFN
UP
IBWF
IBE
NPSF
EFWFMPQJOH




PSHBOJ[BUJPOBM
BTQFDUT
JO
UIFJS
XPSL
UIBO
UIF
QVCMJD
XPSLQMBDFT
5BCMF
 


"CPVU
IBMG
PG
UIF
NJDSPFOUFSQSJTFT

QFS DFOU
BOTXFSFE
UIBU
UIFZ
IBE
XPSLQMBDFSFMBUFE
DPNQFUFODF
EFWFMPQNFOU
MFUUJOH
FNQMPZFFT
USZ
OFX
XPSL
UBTLT
BOE
OFX
QBSUT
PG
UIF
CVTJOFTT


XIFSF
UIF
DPSSFTQPOEJOH
mHVSF
GPS
UIF
QVCMJD
XPSLQMBDFT
XBT

QFSDFOU
"T
TFFO
JO
5BCMF

NPTU
PG
UIF
NJDSPmSNT
EP
OPU
VTF
XPSLJOH
FOWJSPONFOU
FOWJSPONFOU
PS
RVBMJUZ
TZTUFNT
CVU
JU
JT
JOUFSFTUJOH
UIBU
UIFZ
TUJMM
VTF
RVBMJUZ
TZTUFNT
BOE
B
KVTUJOUJNF
BQQSPBDI
NPSF
UIBO
UIF
QVCMJD
XPSLQMBDFT
EP
0OF
DBO
BMTP
OPUF
UIBU
NBOZ
PG
UIF
MBUUFS
HSPVQ
VTF
XPSLJOH
FOWJSPO NFOU
TZTUFNT
'JWF
QFSDFOU
PG
UIF
NJDSPFOUFSQSJTFT
XFSF
JNQMFNFOUJOH
BO
FRVBMJUZ
QPMJDZ
BOE

QFSDFOU
B
ESVH
QPMJDZ
XIFSF
UIF
DPSSFTQPOEJOH
mHVSFT
GPS
UIF
QVCMJD
XPSLQMBDFT
XFSF

QFSDFOU
BOE

QFSDFOU
SFTQFDUJWFMZ
ɨF
NBOBHFST
JOUFSWJFXFE
JO
UIF
NJDSP FOUFSQSJTFT
TUBUF
UIBU
DPPQFSBUJPO
XJUI
PUIFS


/FX
3VSBM
1PMJDZ


&/53&13&/&634)*1

*T
XPSL
PSHBOJ[FE
TP
UIBU

QFSDFOUBHFT
PG
MFBEFST
BOTXFSJOH
UIF
RVFTUJPO
JO
UIF
BċSNBUJWF 



.JDSP
FOUFSQSJTFT


1VCMJD
XPSLQMBDFT /
/

&NQMPZFFT
BMUFSOBUF
CFUXFFO
EJĊFSFOU
XPSL
UBTLT





 &NQMPZFFT
BSF
DPOUJOVPVTMZ SFDFJWJOH
OFX
XPSL
UBTLT


 &NQMPZFFT
BSF
QVUUJOH
IJHIFS
EFNBOET
PO
FYJTUJOH
XPSL


 &NQMPZFFT
EFWFMPQ
UIFJS
PXO XPSL
UBTLT


 &NQMPZFFT
EFWFMPQ
OFX XPSL
UBTLT





#PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJm DBOU
BU
UIF

MFWFM


JNQPSUBOU
CVU
UIBU
UIFZ
VTF
DPOTVMU BOUT
MFTT
PGUFO
0OF
DPOmSNBUJPO
PG
UIJT
SFTVMU
JT
UIBU
POMZ

QFSDFOU
PG
UIFTF
FOUFSQSJTFT
VTF
PDDVQBUJPOBM
IFBMUI
TFSWJDFT
GPS
IVNBO
SFTPVSDF
EFWFMPQNFOU


(PBMT
BOE
SFRVJSFNFOUT
GPS
DIBOHF
QSPDFTTFT

JO
UIF
XPSLQMBDF

ɨF
GPMMPXJOH
UXP
UBCMFT
5BCMFT

BOE

QSFTFOU
JNQPSUBOU
HPBMT
GPS
DIBOHF
QSPDFTTFT
JO
QMBOOFE
PS
POHPJOH
EF WFMPQNFOU
XPSL

ɨF
QSJWBUF
XPSLQMBDFT
TFFN
UP
BUUBUDI
MFTT
QSJPSJUZ
UP
HPBMT
SFMBUFE
UP
XPSLQMBDF
IFBMUI
BOE
XPSL
PSHBOJ[B 5BCMF


0SHBOJ[BUJPOBM
BTQFDUT
PG
XPSL UJPO
UIBO
EP
UIF
QVCMJD
POFT
5BCMF



CVU
JU
TIPVME
CF
OPUFE
UIBU
BSPVOE
DPNQBOJFT
JT
NPSF
BEWBODFE
XIFO
JU
DPNFT
UP
B
UIJSE
PG
UIF
NJDSPFOUFSQSJTFT
BSF
GPSNVMBUJOH
QSPEVDU
EFWFMPQNFOU
NBSLFUJOH
BOE
DPNQF HPBMT
DPODFSOJOH
QTZDIPTPDJBM
XPSLJOH
DPOEJ UFODF
EFWFMPQNFOU
UIBO
SFHBSEJOH
NBOBHFNFOU
UJPOT
MFBEFSTIJQ
EFWFMPQNFOU
PG
UFBNT
BOE
BOE
PDDVQBUJPOBM
IFBMUI
JTTVFT
ɨF
SFTQPOEFOUT
DPNQFUFODF
EFWFMPQNFOU
*U
JT
OPU
TP
TVSQSJTJOH
PQJOJPOT
PO
XIBU
QBSUOFST
BSF
JNQPSUBOU
GPS
UIBU
UIF
NJDSPFOUFSQSJTF
NBOBHFST
BUUBDI
NPSF
EFWFMPQJOH
UIF
XPSLJOH
FOWJSPONFOU
BOE
DPN QSJPSJUZ
UP
HPBMT
DPOOFDUFE
XJUI
UIF
XPSLJOH
QFUFODF
BSF
TIPXO
JO
5BCMF



FOWJSPONFOU
QSPEVDUJPO
BOE
RVBMJUZ
EFWFMPQ ɨF
NBOBHFST
JO
UIF
NJDSPFOUFSQSJTFT
TUBUF
NFOU
5BCMF
 
UIBU
DVTUPNFST
BOE
QSPGFTTJPOBM
DPOUBDUT
BSF
ɨF
JOUFSWJFXFFT
SFQSFTFOUJOH
NJDSPFOUFS QSJTFNBOBHFS
SFQPSUFE
UIBU
UIFZ
IBE
EJċDVMUJFT
JO
SFDSVJUJOH
DPNQFUFOU
DP "SF
UIF
GPMMPXJOH
NFUIPET
VTFE
QFSDFOUBHFT
PG
MFBEFST
TBZJOH
XPSLFST
"CPVU
B
UIJSE
PG
UIFNNBO UIBU
UIFZ
BSF
VTFE
B
MPU  BHFS
IBE
B
QPTJUJWF
BUUJUVEF
UPXBSET
UIF
FNQMPZNFOU
PċDF
BMUIPVHI
UIFZ


.JDSPFOUFSQSJTFT

1VCMJD
XPSLQMBDFT

/

/

XPVME
IBWF
MJLFE
NPSF
TUSBUFHJD
XPSL
UP
CF
EPOF
JO
SFMBUJPO
UP
SFDSVJUNFOU
8PSLJOH
FOWJSPONFOU
TZTUFN


 JTTVFT
"SPVOE
IBMG
PG
UIF
FOUFSQSJTFT
&OWJSPONFOU
TZTUFN
BSF
DPPQFSBUJOH
XJUI
VQQFS
TFDPOE *40



 BSZ
TDIPPMT
BOE
B
TNBMMFS
QSPQPSUJPO
2VBMJUZ
TZTUFN
*40



 
QFSDFOU
BSF
DPPQFSBUJOH
XJUI
VOJ 2VBMJUZ
EFWFMPQNFOU
52.



2VBMJUZ
DJSDMFT


 WFSTJUJFT
JO
NBUUFST
PG
QSPEVDUJPO
BOE
+VTUJOUJNF
TZTUFN


 RVBMJUZ
EFWFMPQNFOU
GPS
FYBNQMF
/FBSMZ
BMM
PG
UIF
NBOBHFST
MPPLFE
#PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJm PO
DPNQFUFODF
EFWFMPQNFOU
BT
WFSZ
DBOU
BU
UIF

MFWFM
JNQPSUBOU
BOE
UXFOUZ
QFS
DFOU
5BCMF

.FUIPET
VTFE
GPS
JNQSPWFNFOUT
JO
UIF
XPSLJOH
IBE
JNQMFNFOUFE
DPNQFUFODF
QSP FOWJSPONFOU
FOWJSPONFOU
BOE
RVBMJUZ HSBNNFT
JO
UIFJS
DPNQBOJFT
/POF


/FX
3VSBM
1PMJDZ




&/53&13&/&634)*1

UPST
JO
B
GBJSMZ
SFQSFTFOUBUJWF
TBNQMF
PG
NJDSPFOUFSQSJTFT
JO
B
SVSBM
BSFB
BOE
PO
DPNQBSJTPOT
XJUI
B
HSPVQ
PG
TNBMM
QVCMJD
XPSLQMBDFT
PG
EJĊFSFOU
UZQFT
ɨF
DPNQBSJTPOT
BSF
NBEF
JO
PSEFS
UP

.JDSPFOUFSQSJTFT

1VCMJD
XPSLQMBDFT DPOUSBTU
DPOEJUJPOT
JO
UIF
UXP
HSPVQT


/

/

PG
XPSLQMBDFT
*O
UFSNT
PG
UIF
NBJO
QVSQPTF
4VQQMJFST


 XIJDI
JT
UP
BOBMZ[F
BOE
EJTDVTT
UIF
$VTUPNFST



JNQPSUBODF
PG
PSHBOJ[BUJPOBM
GBDUPST
1SPGFTTJPOBM
DPOUBDUT


 5FDIOJDBM
DPOTVMUBOUT


 GPS
FYQMBJOJOH
UIF
EJĊFSFODFT
JO
TJDL 0SHBOJ[BUJPOBM
DPOTVMUBOUT


 OFTT
BCTFODF
CFUXFFO
NJDSPFOUFSQSJT 0UIFS
XPSLQMBDFT
JO
TJNJMBS

FT
BOE
TNBMM
QVCMJD
XPSLQMBDFT
JO
B
PSHBOJ[BUJPOT



 SVSBM
BSFB
UIF
SFTVMUT
QPJOU
BU
B
NPSF
-PDBM
DPNQBOZ
OFUXPSLT


 NBSLFE
JODSFBTF
JO
TJDLOFTT
BCTFODF
1VCMJD
TVQQPSU


 JO
UIF
TNBMM
QVCMJD
XPSLQMBDFT
EVS #PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJm JOH
MBTU
GFX
ZFBST
ɨJT
SFTVMU
JT
JO
MJOF
DBOU
BU
UIF

MFWFM
XJUI
HPWFSONFOUBM
JOWFTUJHBUJPOT
BOE
SFTFBSDI
SFTVMUT
TIPXJOH
B
DMFBS
QPTJ 5BCMF

1BSUOFST
GPS
EFWFMPQJOH
UIF
XPSLJOH
FOWJSPO UJWF
SFMBUJPOTIJQ
CFUXFFO
XPSLQMBDF
NFOU
BOE
DPNQFUFODF TJ[F
BOE
TJDLOFTT
BCTFODF
ɨF
GBDU
UIBU
UIF
NJDSPFOUFSQSJTFT
VOMJLF
UIF
QVCMJD
XPSLQMBDFT
KVEHFE
JO
HFOFSBM
(PBM

.JDSPFOUFSQSJTFT

1VCMJD
XPSLQMBDFT UIBU
UIFSF
XFSF
HSFBUFS
OVNCFST
PG

/


/

EFWFMPQNFOUBM
GBDUPST
BU
XPSL
BOE
OPU
TVDI
ESBNBUJD
DIBOHFT
JO
XPSL
5P
JNQSPWF
QTZDIPTPDJBM
DPOEJUJPOT


5P
JNQSPWF
WPDBUJPOBM
SFIBCJMJUBUJPO

 EFNBOET
NBZ
CF
PG
JNQPSUBODF
GPS
-FBEFSTIJQ
EFWFMPQNFOU


 FYQMBJOJOH
UIF
EJĊFSFODFT
JO
IFBMUI
%FWFMPQNFOU
PG
UFBNCBTFE
XPSL


 PVUDPNFT
4PNF
PSHBOJ[BUJPOBMMZ
GPD $PNQFUFODF
JOWFTUNFOUT



VTFE
FYQMBOBUPSZ
NPEFMT
GPS
PDDVQB #PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJm UJPOBM
IFBMUI
BOE
TJDLOFTT
BCTFODF
BOE
BMTP
SFTFBSDI
SFQPSUT
UIBU
QPJOU
DBOU
BU
UIF

MFWFM
UP
HPPE
XPSL
PSHBOJ[BUJPO
GBDUPST
JO
5BCMF

(PBMT
PG
IFBMUI
BOE
XPSL
PSHBOJ[BUJPO
QSPKFDUT
TNBMM
FOUFSQSJTFT
MFOE
TVQQPSU
UP
UIJT
BU
UIF
XPSLQMBDF
QFSDFOUBHFT
PG
MFBEFST
DMJNJOH
UIBU
MJOF
PG
SFBTPOJOH
UIFTF
HPBMT
BSF
EPNJOBOUNBJO
POFT

*U
JT
QBSBEPYJDBM
JO
B
XBZ
IPXFWFS
PG
UIF
FOUFSQSJTFT
IBE
SFDFJWFE
TVQQPSU
GSPN
UIBU
NJDSPFOUFSQSJTFT
IBWF
MPX
TJDL TQFDJmD
&6
QSPHSBNNFT
IPXFWFS
ɨF
NBKPSJUZ
OFTT
BCTFODF
mHVSFT
TJODF
UIFZ
TBZ
UIBU
UIFZ
PG
UIF
NJDSPFOUFSQSJTF
NBOBHFST
IBWF
B
QPTJ BSF
QSJPSJUJ[JOH
QSPEVDUJPOPSJFOUFE
EFWFMPQ UJWF
BUUJUVEF
UPXBSET
MPDBM
BVUIPSJUZ
DPNQFUFODF
NFOU
NVDI
NPSF
UIBO
NFBTVSFT
DPODFSOFE
XJUI
DFOUFST
BOE
UPXBSET
CFUUFS
EFWFMPQFE
DPNQBOZ
PDDVQBUJPOBM
IFBMUI
BOE
UIF
XPSLJOH
FOWJSPO OFUXPSLT



NFOU
ɨF
OVNCFST
PG
NJDSP
FOUFSQSJTFT
IBWJOH
XPSLJOH
FOWJSPONFOU
TZTUFNT
BOE
QPMJDJFT
GPS
%JTDVTTJPO
BOE
SFnFDUJPOT

EFBMJOH
XJUI
HFOEFS
EJTDSJNJOBUJPO
BOE
ESVHT
XFSF
NVDI
MPXFS
UIBO
JO
UIF
HSPVQ
PG
QVCMJD
PO
UIF
SFTVMUT XPSLQMBDFT
BOE
UIF
NBOBHFST
PG
UIF
NJDSP ɨJT
TUVEZ
GPDVTFT
NBJOMZ
PO
PSHBOJ[BUJPOBM
GBD FOUFSQSJTFT
TUBUFE
UIBU
DPOUBDUT
BOE
DPPQFSBUJPO
)PX
JNQPSUBOU
BSF
UIF
GPMMPXJOH
QBSUOFST
DPODFSOJOH
JOGPSNB UJPO
BOE
JEFBT
GPS
EFWFMPQJOH
UIF
XPSLJOH
FOWJSPONFOU
BOE
DPNQFUFODF
QFSDFOUBHFT
PG
MFBEFST
BOTXFSJOH
UIF
RVFTUJPO
XJUI
UIF
BMUFSOBUJWFT
PG
DPOTJEFSBCMFDSVDJBM
JNQPSUBODF



/FX
3VSBM
1PMJDZ


&/53&13&/&634)*1

NBO
DBQJUBM
BOE
UIF
XPSLJOH
FOWJ SPONFOU
XJUIPVU
VTJOH
TUSVDUVSFE
TZTUFNT
"
DPODFJWBCMF
DPODMVTJPO
UP
CF
SFBDIFE
GSPN
UIF
QSFTFOU
5P
JNQSPWF
UIF
XPSLJOH
FOWJSPONFOU




SFTVMUT
JT
UIVT
UIBU
BTQFDUT
SFMBUFE
5P
JOUSPEVDF
OFX
QSPEVDUJPO UP
XPSL
PSHBOJ[BUJPO
GBDUPST
BSF
PG
UFDIOJDBM
FRVJQNFOU

 5P
JNQSPWF
QSPEVDUJPO
nPXT

 NPSF
JNQPSUBODF
GPS
IFBMUI
PVU 5P
JNQSPWF
RVBMJUZ

 DPNFT
JO
NJDSPFOUFSQSJTFT
UIBO
JT
NPSF
GPSNBMJ[FE
BUUFOUJPO
UP
UIF
#PME
TUZMF
Q


EJĊFSFODFT
CFUXFFO
UIF
HSPVQT
BSF
TJHOJmDBOU
XPSLJOH
FOWJSPONFOU

BU
UIF

MFWFM
8JUI
SFHBSE
UP
UIF
TFDPOE
QVS QPTF
PG
UIJT
TUVEZ
UP
TIFE
MJHIU
PO
5BCMF

(PBMT
DPOOFDUFE
XJUI
UIF
XPSLJOH
FOWJSPONFOU
QSPEVDUJPO
BOE
RVBMJUZ
QFSDFOUBHFT
PG
MFBEFST
JOEJDBUJOH
UIF
OFFET
GPS
DIBOHF
QSPDFTTFT
BO
UIBU
FBDI
JT
UIF
EPNJOBOUNBJO
HPBM

JOUFSFTUJOH
mOEJOH
JT
UIBU
UIF
MFBE FST
PQJOJPOT
PG
DIBOHFT
SFMBUFE
UP
PWFS
QSPEVDUJPO
JTTVFT
XFSF
NPSF
JNQPSUBOU
UP
FDPOPNJD
PVUDPNFT
OVNCFST
PG
DVTUPNFST
BOE
UIFN
UIBO
DPOUBDUT
PWFS
IFBMUI
BOE
XPSLJOH
BTQFDUT
PG
XPSL
PSHBOJ[BUJPO
TIPX
B
NVDI
NPSF
FOWJSPONFOU
JTTVFT
ɨFZ
BMTP
SBUFE
DPOUBDUT
QPTJUJWF
QJDUVSF
UIBO
UIBU
PGUFO
EJTDVTTFE
JO
XJUI
DPOTVMUBOUT
BT
MFTT
JNQPSUBOU
BOE
NPTU
PG
DPOOFDUJPO
XJUI
NJDSPFOUFSQSJTFT
JO
SVSBM
BSFBT
UIF
FOUFSQSJTFT
IBE
OP
GPSNBM
SFMBUJPO
XJUI
BOZ
0OF
SFTVMU
UP
FNFSHF
GSPN
UIF
JOUFSWJFXT
JT
UIF
PDDVQBUJPOBM
IFBMUI
TFSWJDF
B
mOEJOH
XIJDI
JT
OFFE
GPS
JNQSPWFNFOUT
JO
DPNQBOZ
OFUXPSLT
JO
MJOF
XJUI
TPNF
PUIFS
SFQPSUT
"OUPOTTPO BOE
GPS
EFWFMPQJOH
DPPQFSBUJPO
CFUXFFO
DPN 4DINJEU
 
"
TJNJMBS
QBUUFSO
BQQFBST
JO
UIF
QBOJFT
BOE
MPDBM
BVUIPSJUJFT
SFDSVJUNFOU
DPPQ DBTF
PG
QSJPSJUJ[FE
HPBMT
GPS
QMBOOFE
PS
POHPJOH
FSBUJPO
XJUI
SFTFBSDIFST
BOE
DPNQFUFODF
4JODF
DIBOHF
QSPDFTTFT
JO
UIBU
UIF
QVCMJD
XPSLQMBDFT
NBOZ
MFBEFST
QPJOU
UP
UIF
JNQPSUBODF
PG
DPN BSF
QSJPSJUJ[JOH
HPBMT
DPOOFDUFE
XJUI
QTZDIP QFUFODF
BNPOH
UIF
FNQMPZFFT
BOE
UIF
QPTJUJWF
TPDJBM
DPOEJUJPOT
WPDBUJPOBM
SFIBCJMJUBUJPO
BOE
FĊFDUT
PG
DPNQFUFODF
EFWFMPQNFOU
QSPHSBNNFT
DPNQFUFODF
EFWFMPQNFOU
XIJMF
UIF
NJDSPFO #BHFS4KÚHSFO
FU
BM



JU
JT
SFNBSLBCMF
UFSQSJTFT
BSF
QSJPSJUJ[JOH
HPBMT
DPODFSOFE
XJUI
UIBU
OPOF
PG
UIF
FOUFSQSJTFT
TUVEJFE
IFSF
IBE
RVBMJUZ
QSPEVDUJPO
nPXT
BOE
UIF
XPSLJOH
FO DBSSZ
UISPVHI
BOZ
QSPHSBNNFT
SFMBUFE
UP
UIF
WJSPONFOU
&6
HPBM
 
"MUIPVHI
UIF
SFTVMUT
PG
UIJT
TUVEZ
ɨF
GBDUT
UIBU
NJDSPFOUFSQSJTFT
IBWF
B
NPSF
BSF
EFTDSJQUJWF
JU
JT
IPQFE
UIBU
UIFZ
NBZ
TFSWF
UFDIOJDBMMZ
PSJFOUFE
DIBOHF
GPDVT
BOE
UIBU
UIFZ
BT
POF
JOQVU
UP
B
EJTDVTTJPO
PG
JNQSPWFNFOUT
EP
OPU
VTF
TUSVDUVSFE
TZTUFNT
BSF
QSPCBCMZ
BOE
SVSBM
FOUSFQSFOFVSTIJQ
QPMJDJFT
TVJUBCMF
GPS
FYQSFTTJPOT
PG
UIF
OFFE
GPS
NPSF
JOUFHSBUFE
NJDSPFOUFSQSJTFT BOE
nFYJCMF
NPEFMT
UIBU
BSF
BEBQUFE
GPS
TNBMM
.Z
PCTFSWBUJPO
JT
UIBU
CPUI
QSBDUJUJPOFST
BOE
FOUFSQSJTFT
3FTFBSDI
JOUP
SFMBUJPOTIJQT
CFUXFFO
SFTFBSDIFST
VOEFSFTUJNBUF
PSHBOJ[BUJPOBM
GBD RVBMJUZ
EFWFMPQNFOU
BOE
XPSLJOH
FOWJSPONFOU
UPST
BOE
PSHBOJ[BUJPOBM
IFBMUI
XIFO
EJTDVTTJOH
"YFMTTPO

BOE
UIF
DPOUSBEJDUPSZ
SFTVMUT
FOUSFQSFOFVSJBM
PSJFOUBUJPO
BOE
TNBMM
CVTJOFTT
PCUBJOFE
SFHBSEJOH
UIBU
FĊFDUT
PG
DPNQFUFODF
EFWFMPQNFOU
ɨF
TQFDJmD
DPOEJUJPOT
QSFWBJM EFWFMPQNFOU
JO
TNBMM
CVTJOFTTFT
4UPSFZ

JOH
JO
UIFTF
CVTJOFTTFT
QMBDF
EFNBOET
PO
NPSF
:MJOFOQÊÊ

MFOE
TPNF
TVQQPSU
UP
UIJT
JOUFHSBUFE
NPEFMT
XIFO
JU
DPNFT
UP
UIF
XPSLJOH
DPODMVTJPO
ɨF
GBDU
UIBU
NJDSPFOUFSQSJTFT
VTF
FOWJSPONFOU
RVBMJUZ
BOE
UIF
VTF
PG
OFUXPSLT
TUSVDUVSFE
TZTUFNT
UP
B
MFTTFS
FYUFOU
DBO
BMTP
CF
UP
JOTQJSF
DIBOHF
QSPDFTTFT
BU
UIF
XPSLQMBDF
TFFO
BT
B
SFTVMU
PG
UIF
BQQMJDBUJPO
PG
B
NPSF
MFBSO GPS
FYBNQMF
ɨFSF
DPVME
BMTP
CF
HPPE
SFBTPOT
JOHCBTFE
TUSBUFHZ
JO
UIFTF
FOUFSQSJTFT
*U
JT
BMTP
GPS
MBSHFS
XPSLQMBDFT
UP
TUVEZ
UIF
DJSDVNTUBODFT
QPTTJCMF
UIBU
NJDSPFOUFSQSJTFT
NBZ
JOWFTU
JO
IV FYJTUJOH
JO
NJDSPFOUFSQSJTFT
XJUI
UIF
QVSQPTF
(PBM




/FX
3VSBM
1PMJDZ


.JDSPFOUFSQSJTFT

1VCMJD
XPSLQMBDFT
/

/



&/53&13&/&634)*1

PG
MFBSOJOH
UIJOHT
UIBU
DBO
CF
BQQMJFE
UP
UIFJS
PXO
TJUVBUJPO
UPP .PSF
BQQMJFE
SFTFBSDI
XPVME
CF
OFDFTTBSZ
UP
HBJO
B
CFUUFS
LOPXMFEHF
PG
NFDIBOJTNT
CFZPOE
UIF
GBDU
UIBU
NJDSPFOUFSQSJTFT
EFTQJUF
UIF
DMFBS
OFFET
GPS
DIBOHF
IBWF
BDIJFWFE
HPPE
SFTVMUT
SF HBSEJOH
DIBOHFT
JO
TJDLOFTT
BCTFODF
DPNQFUFODF
EFNBOET
BOE
TPNF
FDPOPNJD
PVUDPNFT
BOE
JU
XPVME
BMTP
CF
PG
JNQPSUBODF
UP
QBZ
NPSF
BUUFO UJPO
UP
DPNQBOZ
TJ[F
BOE
PSHBOJ[BUJPOBM
GBDUPST
XIFO
EFWFMPQJOH
FYQMBOBUPSZ
NPEFMT
GPS
XPSL QMBDF
IFBMUI
ɨF
GBDU
UIBU
UIFSF
BSF
TUVEJFT
UIBU
QPJOU
PVU
SFMBUJPOTIJQT
CFUXFFO
PSHBOJ[BUJPOBM
GBDUPST
IFBMUI
BOE
PSHBOJ[BUJPOBM
QFSGPSNBODF
-JOETUSÚN

7JOCFSH

FNQIBTJ[FT
UIF
OFFE
UP
MPPL
NPSF
EFFQMZ
JOUP
UIF
SFMBUJPOT
CFUXFFO
UIFTF
WBSJBCMFT
"T
8JMLJOTPO

BSHVFT
NPSF
SFTFBSDI
JT
OFFEFE
JOUP
PSEFS
UP
BTDFSUBJO
XIFUIFS
DPNQBOZ
TJ[F
JO
JUTFMG
JT
B
EF UFSNJOBOU
PG
XPSL
PSHBOJ[BUJPO
PSHBOJ[BUJPOBM
IFBMUI
BOE
PSHBOJ[BUJPOBM
QFSGPSNBODF

/05&4 

"
NJDSPFOUFSQSJTF
JT
B
DPNQBOZ
UIBU
FNQMPZT
MFTT
UIBO
UFO
QFSTPOT

ɨF
EFmOJUJPO
BMTP
JODMVEFT
TPNF
FDPOPNJD
DSJUFSJB
&6
 


4$#
XXXTDCTF
  
ɨFTF
XPSLQMBDFT
QBSUJDJQBUFE
JO
UIF
QSPKFDU
})FBM UIZ
QVCMJD
XPSLQMBDFT
JO
+ÊNUMBOE
$PVOUZ}
JO
o


5VSOPWFS
BOE
QSPmU
XFSF
OPU
SFMFWBOU
GPS
UIF
QVCMJD
XPSLQMBDFT

3&'&3&/$&4 "OEFSTPO
.o4PIBM
"
4

"
TUVEZ
PG
UIF
SFMBUJPO TIJQ
CFUXFFO
RVBMJUZ
NBOBHFNFOU
QSBDUJDFT
BOE
QFSGPSNBODF
JO
TNBMM
CVTJOFTT
*OUFSOBUJPOBM
+PVS OBM
PG
2VBMJUZ

3FMJBCJMJUZ
.BOBHFNFOU
 


o "OUPOTTPO
"#o4DINJEU
-

4NÌGÚSFUBH
PDI
GÚSFUBHTIÊMTPWÌSE
o
TLB
CFSHFU
LPNNB
UJMM
.V IBNNFE
FMMFS
.VIBNNFE
UJMM
CFSHFU
*7-
SBQQPSU
#
4UPDLIPMN
*7-
4WFOTLB
.JMKÚJOTUJUVUFU




"#
*O
4XFEJTI  "YFMTTPO
$

'PSNBMJTFSJOH
TPN
IJOEFS
PDI
NÚK MJHIFU
J
TNÌ
GÚSFUBH
&O
TUVEJF
BW
BSCFUTNJMKÚBS CFUF
VS
FUU
MÊSBOEFQFSTQFLUJW
%PLUPSTBWIBOEMJOH

-VMFÌ
UFLOJTLB
VOJWFSTJUFU
*O
4XFEJTI  "YFMTTPO
+

2VBMJUZ
BOE
&SHPOPNJDT
5PXBSET
TVDDFTTGVM
JOUFHSBUJPO
%JTTFSUBUJPO
/P

-JOLÚ QJOH
4UVEJFT
JO
4DJFODF
BOE
5FDIOPMPHZ
-JOLÚQJOH
6OJWFSTJUZ "MQFOCFSH
+o,BSMTTPO
'

&O
VUWÊSEFSJOH
BW
"S CFUTMJWTGPOEFOT
TBUTOJOHBS
QÌ
TNÌGÚSFUBH
$FOUSVN
GÚS
4NÌGÚSFUBHTVUWFDLMJOH
)ÚHTLPMBO
J
7ÊYKÚ
*O
4XFEJTI  #BHFS4KÚHSFO
-o(VTUBGTTPO
0o1FUUFSTTPO
-

7ÊYULSBGU
.ÌM

o
FO
FĊFLUTUVEJF
,PNQFUFO TVUWFDLMJOH
J
TNÌ
GÚSFUBH
3BQQPSU
"
½TUFSTVOE
*OTUJUVUFU
GÚS
UJMMWÊYUQPMJUJTLB
TUVEJFS
*O
4XFEJTI  WPO
#FSHNBOOo8JOCFSH
.-

"UUSBLUJWJUFU
PDI
TBNWFSLBO
QÌ
MPLBM
OJWÌ
o
UJMMWÊYUCZO
5SÌOHTWJLFO
J
KÊNGÚSFMTF
NFE
#JTQHÌSEFO
¯TFEB
PDI
(OPTKÚ
*
<UFEU
&o8PMWÏO
-&
SFE 
3FMBUJPOTCZH HBOEF
GÚS
FLPOPNJTL
VUWFDLMJOH
'SÌO
JEÏFS
PN
FLPOPNJTL
VUWFDLMJOH
UJMM
MPLBMU
VUWFDLMJOHTBSCFUF
J
/PSSMBOET
JOMBOE
"SCFUTMJW
J
PNWBOEMJOH

4UPDLIPMN
"SCFUTMJWTJOTUJUVUFU
TJE
o
*O

4XFEJTI  #PMUPO
+
&

3FQPSU
PG
UIF
$PNNJUUFF
PG
*ORVJSZ
PO
4NBMM
'JSNT
$NOE
-POEPO
).40 #PSOCFSHFS%BOLWBSEU
4o0IMTPO
$
(o8FTUFSIPMN
1

"SCFUTNJMKÚ
PDI
IÊMTPBSCFUF
J
TNÌGÚSF UBH
o
GÚSTÚL
UJMM
IFMIFUTCJME
"SCFUTMJW
J
PNWBO EMJOH

4UPDLIPMN
"SCFUTMJWTJOTUJUVUFU
*O

4XFEJTI  #VEHFUQSPQPTJUJPOFO

3FHFSJOHFOT
QSPQPTJUJPO

'ÚSTMBH
UJMM
TUBUTCVEHFU
GÚS

mOBOTQMBO
TLBUUFGSÌHPS
*O
4XFEJTI  $WFULPWJƥ
"

%FO
ɨBJMÊOETLB
QBWJMKPOHFOT
GSBN WÊYU
o
FUU
LVMUVS
PDI
GÚSOZFMTFQSPKFLU
*O
<UFEU
&o8PMWÏO
-&
SFE 
3FMBUJPOTCZHHBOEF
GÚS
FLP OPNJTL
VUWFDLMJOH
'SÌO
JEÏFS
PN
FLPOPNJTL
VU WFDLMJOH
UJMM
MPLBMU
VUWFDLMJOHTBSCFUF
J
/PSSMBOET
JO MBOE
"SCFUTMJW
J
PNWBOEMJOH


4UPDLIPMN
"SCFUTMJWTJOTUJUVUFU
4JE
o
*O
4XFEJTI  %FMBOFZ
+
5o)VTFMJE
.
"

ɨF
JNQBDU
PG
IV NBO
SFTPVSDF
NBOBHFNFOU
QSBDUJDFT
PO
QFSDFQUJPOT
PG
PSHBOJ[BUJPOBM
QFSGPSNBODF
"DBEFNZ
PG
.BOB

/FX
3VSBM
1PMJDZ


&/53&13&/&634)*1

HFNFOU
+PVSOBM
 


QQ
o
%T

&UU
GÚSÊOEFSMJHU
BSCFUTMJW
QÌ
HPUU
PDI
POU
6UWFDLMJOHFO
BW
EFO
TUSFTTSFMBUFSBEF
PIÊMTBO

4UPDLIPMN
'SJU[FT
*O
4XFEJTI  %BWJETTPO
1o%FMNBS
'o8JLMVOE
+
SFE
5JMM WÊYUGÚSFUBHFO
J
4WFSJHF
4UPDLIPMN
4/4
GÚSMBH
*O
4XFEJTI  &BLJO
+
.o-BNN
'o-JNCPSH
)
+

*OUFSOB UJPOBM
QFSTQFDUJWF
PO
UIF
QSPNPUJPO
PG
IFBMUI
BOE
TBGFUZ
JO
TNBMM
XPSLQMBDFT
*O
'SJDL
,o+FOTFO
1
-o2VJOMBO
.o8JMUIBHFO
5
FET
4ZTUFNBUJD
PDDVQBUJPOBM
IFBMUI
BOE
TBGFUZ
NBOBHFNFOU
0Y GPSE
&MTFWJFS
4DJFODF
-UE &BTUPO
(
4o+BSFMM
4
-

ɨF
FĊFDUT
PG
UPUBM
RVBMJUZ
NBOBHFNFOU
PO
DPSQPSBUF
QFSGPSNBODF
BO
FNQJSJDBM
JOWFTUJHBUJPO
+PVSOBM
PG
#VTJOFTT
 


QQ
o &6

0ċDJBM
+PVSOBM
PG
UIF
&VSPQFBO
6OJPO
-

NBK

IUUQFVSPQBFVJOUFVS MFYFOPK  'ÚSFUBHBSOB

%F
GSJTLB
TNÌGÚSFUBHFO
*O
4XFEJTI 

,BSBTFL
3

+PC
EFNBOET
KPC
EFDJTJPO
MBUJUVEF
BOE
NFOUBM
TUSBJO
*NQMJDBUJPOT
GPS
KPC
SFEFTJHO
"ENJ OJTUSBUJWF
4DJFODF
2VBSUFSMZ

QQ
o ,JWJNÊLJ
.o7BUIFSB
+o1FOUUJ
+o'FSSJF
+
&

'BDUPST
VOEFSMZJOH
UIF
FĊFDU
PG
PSHBOJ[BUJPOBM
EPXOTJ[JOH
PO
IFBMUI
PG
FNQMPZFFT
-POHJUVEJOBM
DPIPSU
TUVEZ
#SJUJTI
.FEJDBM
+PVSOBM

QQ
o -JOETUSÚN
,o4DISFZ
,o"IPOFO
(o,BMFWB
4

ɨF
FĊFDUT
PG
QSPNPUJOH
PSHBOJ[BUJPOBM
IFBMUI
PO
XPSLFS
XFMMCFJOH
BOE
PSHBOJ[BUJPOBM
FĊFDUJWFOFTT
JO
TNBMM
BOE
NFEJVNTJ[FE
FOUFSQSJTFT
*O
.VSQIZ
-3o$PPQFS
$-
FET 
)FBMUIZ
BOE
QSPEVDUJWF
XPSL
o
BO
JOUFSOBUJPOBM
QFSTQFDUJWF
-POEPO
5BZ MPS

'SBODJT
QQ
o .BSUPDDIJP
+
+

ɨF
FĊFDUT
PG
BCTFODF
DVMUVSF
PO
JOEJWJEVBM
BCTFODF
)VNBO
3FMBUJPOT

QQ
o .BUMBZ
)

&NQMPZFF
SFMBUJPOT
JO
TNBMM
mSNT
"


)BOTTPO
+

5PUBM
2VBMJUZ
.BOBHFNFOU
o
"TQFDUT
PG
*NQMFNFOUBUJPO
BOE
1FSGPSNBODF
*OWFTUJHBUJPOT
XJUI
B
'PDVT
PO
4NBMM
0SHBOJTBUJPOT
%PDUPSBM
UIFT

NJDSPCVTJOFTT
QFSTQFDUJWF
&NQMPZFF
3FMBUJPOT

QQ
o .VSQIZ
-
3o$PPQFS
$
-

)FBMUIZ
BOE
QSPEVD UJWF
XPSL
o
BO
JOUFSOBUJPOBM
QFSTQFDUJWF

-POEPO
5BZMPS

'SBODJT


JT

%JWJTJPO
PG
2VBMJUZ

&OWJSPONFOUBM
.BOBHFNFOU
-VMFÌ
6OJWFSTJUZ
PG
5FDIOPMPHZ

)JMM
3o4UFXBSU
+

)VNBO
SFTPVSDF
EFWFMPQNFOU
JO
TNBMM
PSHBOJ[BUJPOT
+PVSOBM
PG
&VSPQFBO
*OEVTU

1GFĊFS
+

ɨF
)VNBO
&RVBUJPO
#PTUPO
)BSWBSE
#VTJOFTT
4DIPPM
1SFTT 3BN
.

$POUSPM
BOE
BVUPOPNZ
JO
TNBMM
mSNT
UIF
DBTF
PG
8FTU
.JEMBOET
DMPUIJOH
JOEVTUSZ
8PSL


SJBM
5SBJOJOH

QQ
o +FOTFO
1
-o"MTUSVQ
-oɨPGU
&

8PSLQMBDF
BTTFTTNFOU
B
UPPM
GPS
PDDVQBUJPOBM
IFBMUI
BOE
TBGFUZ
NBOBHFNFOU
JO
TNBMM
mSNT
"QQMJFE
&SHPOPNJDT


&NQMPZNFOU
BOE
4PDJFUZ
 


QQ
o 4#6

4KVLTLSJWOJOH
o
PSTBLFS
LPOTFLWFOTFS
PDI
QSBYJT
&O
TZTUFNBUJTL
MJUUFSBUVSÚWFSTJLU
4UPDL IPMN
4UBUFOT
CFSFEOJOH
GÚS
NFEJDJOTL
VUWÊSEFSJOH
*O
4XFEJTI 


QQ
o
+PIBOOJTTPO
#o-BOETUSÚN
)
FET

*NBHFT
PG
&OUSFQSFOFVSTIJQ
BOE
4NBMM
#VTJOFTT&NFSHFOU
4XFEJTI
$POUSJCVUJPOT
UP
"DBEFNJD
3FTFBSDI
-VOE
4UVEFOUMJUUFSBUVS +PIBOTTPO
#

&O
NPUJWBUJPOTCBTFSBE
BOBMZT
BW
BSCFUTNJMKÚBSCFUFU
J
FO
HSVQQ
BW
NJOESF
UJMMWFSLBOEF
GÚSFUBH
%PLUPSTBWIBOEMJOH

*OTUJUVUJPOFO
GÚS
BSCFUTWFUFOTLBQ
-VMFÌ
UFLOJTLB
VOJWFSTJUFU
*O
4XFEJTI  +PIOTTPO
+
7o)BMM
&
.

+PC
TUSBJO
XPSLQMBDF
TPDJBM
TVQQPSU
BOE
DBSEJPWBTDVMBS
EJTFBTF
"
DSPTT TFDUJPOBM
TUVEZ
PG
B
SBOEPN
TBNQMF
PG
UIF
4XFEJTI
XPSLJOH
QPQVMBUJPO
"NFSJDBO
+PVSOBM
PG
1VCMJD
)FBMUI

QQ
o

/FX
3VSBM
1PMJDZ


406

4KVLGSÌOWBSP
PDI
TKVLTLSJWOJOH
o
GBLUB
PDI
GÚSTMBH
4MVUCFUÊOLBOEF
BW
TKVLGÚSTÊLSJOHTVUSFEOJO HFO
4UPDLIPMN
'SJU[FT
*O
4XFEJTI  406

)BOEMJOHTQMBO
GÚS
ÚLBE
IÊMTB
J
BSCFUTMJWFU
4UPDLIPMN
'SJU[FT
*O
4XFEJTI  4UFFST
3
.o3IPEFT
4
3

,OPXMFEHF
BOE
4QFDVMBUJPO
"CPVU
"CTFOUFFJTN
*O
(PPENBO
1
4o"ULJO
3
4
FET


"CTFOUFFJTN
4BO
'SBODJTDP
+PTTFZ#BTT
QQ
o 4UPSFZ
%

)VNBO
SFTPVSDF
NBOBHFNFOU
QPMJDJFT
BOE
QSBDUJDFT
JO
4.&T
JO
UIF
6,
%PFT
JU
SFBMMZ
JOnVFODF
UIFJS
QFSGPSNBODF
8BSXJDL
ɨF
$FOUSF
GPS
4NBMM

.FEJVN
4J[FE
&OUFSQSJTFT 4[àDT
4o)FNTUSÚN
½o.BSLMVOE
4

0SHBOJTB



&/53&13&/&634)*1

UPSJTLB
GBLUPSFST
CFUZEFMTF
GÚS
MÌOHB
TKVLTLSJWOJOHBS
J
LPNNVOFS
"SCFUF

)ÊMTB

4UPDLIPMN
"SCFUTMJWTJOTUJUVUFU
*O
4XFEJTI  7JOCFSH
4

)FBMUIZ
1FSGPSNBODF
JO
4NBMM
&O UFSQSJTFT
4UVEJFT
PG
0SHBOJ[BUJPOBM
%FUFSNJOBOUT
-JDFOUJBUF
ɨFTJT

%JWJTJPO
PG
*OEVTUSJBM
8PSL
&OWJSPONFOU
-VMFÌ
6OJWFSTJUZ
PG
5FDIOP MPHZ 8BMUFST
%

)FBMUI
BOE
4BGFUZ
JO
4NBMM
&OUFSQSJTFT
&VSPQFBO
4USBUFHJFT
GPS
.BOBHJOH
*NQSPWFNFOU
#SVTTFMT
1
*
"1FUFS
-BOH
4
" 8FTUFSIPMN
1o.BSLMVOE
4

4USBUFHJFT
GPS
PD DVQBUJPOBM
IFBMUI
SFTFBSDI
JO
BO
DIBOHJOH
&VSPQF
8PSL
BOE
)FBMUI

4UPDLIPMN/BUJPOBM
*OTUJUVUF
GPS
8PSLJOH
-JGF 8JLMVOE
+

4NBMM
'JSN
(SPXUI
BOE
1FSGPSNBODF
&OUSFQSFOFVSTIJQ
BOE
#FZPOE
%JTTFSUBUJPO
4FSJFT
/P

+ÚOLÚQJOH
*OUFSOBUJPOBM
#VTJOFTT
4DIPPM 8JMLJOTPO
"

&NQMPZNFOU
SFMBUJPOT
JO
4.&T
&NQMPZFF
3FMBUJPOT
 


QQ
o 8JMLJOTPO
(o%BMF
#
(

*OUFHSBUFE
NBOBHF NFOU
TZTUFNT
BO
FYBNJOBUJPO
PG
UIF
DPODFQU
BOE
UIFPSZ
ɨF
52.
.BHB[JOF
 


QQ
o :MJOFOQÊÊ

)

.BOBHJOH
$PNQFUFODF
%FWF MPQNFOU
BOE
"DRVJTJUJPO
JO
4NBMM
.BOVGBDUVSJOH
'JSNT
%JTTFSUBUJPO

%JWJTJPO
PG
*OEVTUSJBM
0SHBOJ[BUJPO
-VMFÌ
6OJWFSTJUZ
PG
5FDIOPMPHZ ½TUFSTVOET1PTUFO

%ÌMJHB
BSCFUTHJWBSF
GÌS
CFUBMB
NFS
TKVLMÚO

T

*O
4XFEJTI 



/FX
3VSBM
1PMJDZ


Paper V

Workplace health interventions in small enterprises – a Swedish longitudinal study

Re-submitted to Work: A Journal of Prevention, Assessment & Rehabilitation, 2006 Stig Vinberg

Workplace health interventions in small enterprises – a Swedish longitudinal study

Stig Vinberg National Institute for Working Life SE-831 40 Östersund, Sweden Tel: +46 63 55 13 07, Fax: +46 63 55 13 20 E-mail: [email protected]

Abstract This article has a two-fold approach. First, it investigates relationships between work organizational factors, and health and performance outcomes. Second, it compares two change strategy approaches in workplace health interventions by studying changes of these factors and outcomes. The sample consisted of ten Swedish small enterprises including 102 individuals, who answered a before and after questionnaire about organizational factors and outcomes. The leaders were interviewed and answered a questionnaire about performed workplace health interventions. Statistical methods used were reliability tests, correlation analyses and t-tests. Results indicate rather strong links between indicators of respectful leadership, creative work and team spirit, and the outcome indicators self-assessed health and judged workplace adaptability in association with customer satisfaction. The results concerning changes of determinants and outcomes (after workplace health interventions) showed significant differences between enterprises using a broad change strategy and those using an expert/problembased strategy with the former having more favourable results. The leader interview results also point at obstacles concerning workplace change processes as lack of resources, insufficient competence and influence of external factors. The study results suggest that work organizational factors and integrated models for workplace health intervention are of importance for health and performance outcomes in small enterprises. Key words: Small enterprises, health, performance, workplace health interventions, change strategy

1. Introduction During the last few years there has been a considerable increase in sickness absence in Sweden. According to several researchers, this increase is connected to deterioration of psychosocial work conditions, stress related problems and increased organizational change [19,40]. Although small enterprises (defined as those with fewer than 50 employees) have lower sickness absence figures than larger organizations [3,39], there are several reasons for acquiring greater knowledge about how organizational factors and workplace change processes contribute to health and performance outcomes in small firms. Researchers from areas such as working environments [6,39], quality development [36,13] and competence development [35] observe that small enterprises have limited economical and personal resources and a lack of knowledge about how to carry out such change processes. In other words, exactly what causes small organizations to respond differently from larger ones is not yet clearly understood. Although research has shown that small firm growth is very important for creating new jobs [33,43], there are also clear indications that many small business managers deliberately refrain from exploiting opportunities to expand their firms. In a Swedish study, where data was collected during three separate studies over a ten-year period, one interesting finding was that managers judged non-economic concerns, like employee well-being, to be more important than the possibility of profitability [44]. The researchers concluded that the results suggest that small firms relate differently to work atmosphere issues than do larger organizations. Considering these facts, it is remarkable that small business research is focusing so little on how change strategy and work organization factors are related to organizational health and performance outcomes. Despite the small interest in this area, research about small enterprises has been a fast growing speciality during the past decade in Sweden and throughout Western countries [10,22]. It is also remarkable that so few research studies use a longitudinal design for evaluating workplace change processes in work life in general [27,41], and with a focus on small organizations [28]. 1.1 Purpose and research questions The perspectives described above constitute a basis for the study purpose; which is to contribute to the workplace health intervention knowledge base by looking at work organizational factors, and health and performance outcomes in small enterprises. The following research questions were addressed: 1. Which empirical relationships can be found between outcome indicators (of health and performance) and assumed organizational determinants? 2. Are there differences concerning effects on the mentioned outcome indicators and organizational determinants for workplace interventions using different change strategies? 3. What obstacles and success factors can be found concerning workplace health interventions in small enterprises?

1

2. Theoretical aspects and literature overview Based upon the described-above purpose and research questions, a brief summary of theoretical aspects and knowledge from earlier research are presented here on organizational health, change strategies and workplace change interventions. The review includes organizations on a general level, as well as small enterprises. 2.1 Organizational health Theories and models of psychosocial working conditions and health (or ill-health) are based on empirical studies of a traditional industrial labour market at an individual level [19]. One example is the well-known demand-control-support model [24,25]. This model postulates that the most adverse health effects are expected for a combination of high psychological demands, low control and low social support. Several researchers also emphasize the importance of focusing on the organizational level [19] and on promoting health, i.e. with high quality workplace change processes, improved work organization and leadership support [30]. In a Swedish study, a finding was that there were clear relationships between long-term health (measured by an index of low sickness absence/low sickness presence during a two-year period) and leadership support, sufficient resources for doing a good job and employee’s experience of good quality at work [4]. A similar concept is the variable excellent health as introduced by Mackenbach et al. [29]. They define excellent health as a presence of a very good selfassessment of health and conclude that knowledge about determinants for health is less developed than for ill-health. At an organizational level, Murphy & Cooper [31] conclude that there appears to be a convergence of opinion between researchers regarding at least some of the factors important for organizational health. These factors refer to the notion that worker wellbeing and organizational effectiveness can be fostered with a common set of job characteristics. Examples of these characteristics are meaningful work which utilizes worker skills, worker autonomy/control, job security, rewards for performance, worker involvement/participation, team-work, commitment to core values, and safe and healthy physical work environments. In many respects, the description of a healthy work organization represents the convergence of research and thinking from several related areas of workplace research - although very few researchers have tried to test their conceptualisations of a healthy work organization [42]. The concept of a healthy work organization is in line with concepts on ‘development work’ - which are designed to lead to improvements in working conditions and strengthening of quality, effectiveness and health [11]. It is paradoxical that small enterprises have lower levels of sickness absence than do larger enterprises. Small enterprises often have limited resources for any systematic interventions for environmental improvements, inadequate physical work environments, more work-injury hazards, problems with stress and time pressure and a higher sickness presence among employees [9]. On the other hand, positive features in small firms are a good team spirit, the variation of work tasks, work satisfaction due to high control and a holistic perception of work, short roads to decisions, and a personnel commitment to work [9]. These results are consistent with a Finnish study [28] which showed that

2

employees in small firms have more control over their work, better support from their managers and a better working climate than do those in larger enterprises. 2.2 Change strategies Change is simply defined as a wish to make things different and usually better. However, the notion of change is extremely broad [20]. According to Aronsson et al. [5], two factors have to be present to reach success in change projects: employee influence and participation in the change process, and top management’s active support. Results from research areas such as organizational change [14], leadership [12], quality development [7], stress management [26] and ergonomics [41] give support to this twofactor idea, although there are several other important factors for successful workplace interventions. Several researchers point out that change competence is a question of vital importance for coping with change [5,18]. Håkansson [18] also identifies four types of change strategy: the expert project, the problem-oriented strategy, the process-oriented strategy and the broad strategy. In her doctoral thesis, it was reported that for short running projects none of the strategies was superior to any other. For projects running over a longer period a broad strategy, which deals with several extensive problems concurrently while having broad leader/employee participation, gave significantly better results. This is in line with research by Ingelgård [20] where some of the findings were that the use of a learning strategy for change and high management involvement significantly correlated with outcomes concerning economic results and quality of working life. A surprising result was that the use of a learning strategy, where several problem areas are dealt with, a large number of employees participate in and have influence on change processes, and the processes are guided by a vision of the future was more strongly related to the outcomes than was top management involvement. According to Johansson [23], a learning-based change strategy has a particular relevance for small enterprises as planning and controlling are unusual in small firms. Too much planning can be an obstacle when change is rapid; formal planning is usually resisted. Employee engagement and influence is often improved when using a learningbased strategy. 2.3 Workplace change interventions Despite the extent and consequences of the major increases in sickness absence in Sweden and in several other Western countries, one recent review of research has stated that an understanding of the causes is limited [34]. In another review of workplace health promotion interventions [27], it was concluded that only a few of the examined articles focused on the organizational level. Concerning work stress prevention, Kompier et al. [26] argue that the lack of organization-level intervention studies is a barrier to progress in reducing work-related stress. In light of these factors it is desirable to carry on high quality studies with a longitudinal design. Particularly concerning larger organizations there are studies which have shown that workplace change processes have positive effects at an organizational level. For example, Huselid [17] found in a study of 968 US firms that when “high-performance work practices” increased there were significant reductions in employee turnover and

3

significant increases in productivity and profits. A link between quality and performance has been shown in studies where quality award winners were compared with a set of control companies [16]. When it comes to change processes in small enterprises, Hansson [15] found a link between successful TQM implementation and an advantageous financial performance development. In a study by Alpenberg & Karlsson [1], a finding was that investment in work environment improvements was related to increased economical efficiency. Another example is a study of 280 small enterprises that had taken part in competence development programs [8]. These enterprises significantly increased their productivity levels and financial performance when compared to a control group of 200 enterprises. One English study [35], identified some connections between competence development and economic outcomes in small firms - however, the results were weaker and contradictory relative to larger enterprises. In a Finnish study of change programs in small and medium-sized enterprises [28], one result was that more intensively planned interventions tend to have slightly more positive effects on sickness absenteeism and job satisfaction when compared to enterprises having minor or no interventions. In a Swedish case study of small manufacturing enterprises [23], results were that active and engaged employees, a dialogue between leaders and employees, and a positive learning climate characterized enterprises with successful work environment development activities. In summary, the literature described above demonstrates a need for more quantitative and qualitative longitudinally oriented studies concerning change processes and workplace health interventions in small enterprises. Also apparent in the literature is that research into effects of such processes and interventions is, to some extent, contradictory. 3. Methods The longitudinal panel data reported here is part of a larger research and development project concerning Swedish and Norwegian small enterprises [38]. In the present study, ten Swedish small enterprises with a population of 102 individuals (co-workers and leaders) participated in a before and after workplace health intervention study. The enterprises are not a representative sample. However, the sample used does represent key small business sectors such as information technology, manufacturing, retail trade and consulting. The research design consisted of a combination of quantitative and qualitative measures. The 102 individuals (co-workers and leaders) answered an individual questionnaire at two measurement points with a one and a half year interval (before and after performed workplace interventions). The leaders answered a questionnaire about aspects of content and process in performed workplace interventions, and were interviewed using a semistructured format. An examination of accomplished change programs was also carried out. From these latter measures it was possible to divide the enterprises in two change strategy groups: one using a broad change strategy and one using an expert/problembased strategy.

4

3.1 Sample and procedure At Time 1 (November-December 1999), before performed workplace health interventions, 261 individuals in the enterprises received the questionnaire. The response rate was 88 percent or 229 individuals. At Time 2 (April-May 2001), after performed interventions, there were 211 individuals in the companies and 69 percent answered the questionnaire. The final panel consisted of 102 individuals (48 percent of the participants from 2001) who responded to the questionnaire on both occasions. This selection made it possible to compare results for the same individual before and after accomplished workplace interventions. The characteristics of the individuals in the final panel were: 54% women, 58% under 40 years of age, 16% in the 50+ age group, 12% with a basic education and 23% with a university education. The characteristics of the individuals in the two studied enterprise groups (broad strategy and expert/problem-based strategy) are similar concerning gender, age, education and sickness absence levels. 3.1.1 Individual level questionnaire The questionnaire used contained a total of 162 items covering perception the work environment, job and organizational characteristics as well as aspects that concerned organizational health and performance. Corresponding to an earlier study [38], eleven indicators were developed on the basis of substantive considerations and factor analyses (principal factor analyses with the varimax method). Internal consistency data was calculated using Cronbach’s alpha (α) coefficient. According to relationships covered in the literature presented above, some of the indicators are seen as organizational determinants and some as outcome indicators. The content of the eight determinant indicators is as follows. The indicator customer oriented quality practices (seven items) covered feedback on quality to employees, programs for improvements, quality goals and cooperation with customers. The respectful leadership indicator (seven items) covered leadership consideration, information about changes, ethical issues and dialogue between leaders and co-workers. The creative work indicator (six items) measured developing and learning at work, meaningful and stimulating work tasks and engagement at work. The team spirit indicator (two items) measured team functioning and team climate at work. Work intensity demands had three items about hard and demanding work and stress at work. The data stress indicator consisted of two items about stress related to computer problems. The decision latitude indicator (three items) covered worker autonomy/control and possibilities for influencing work. The indicator restructuring worries had two items about concerns related to possible restructuring and its potential influence upon continued employment. Three indicators were seen as outcome indicators of health and performance. The indicator self-assessed health consisted of two items that sought to measure physical health and psychological well-being. The indicator workplace adaptability (three items) covered effectiveness and adaptability to external demands and changes. The indicator customer satisfaction (four items) sought to measure internal judgments about degree of

5

customer satisfaction, quality in products and services, and potential to respond to customer desires. All indicators were based on questions on a Likert scale that ranges from 1 (strongly agree) to 5 (strongly disagree). One exception was the indicator decision latitude that had a range of 1-4. Note that items were reversed for some questions, which means that a high value on an indicator always indicate a more positive value. 3.1.2 Leader questionnaire, interviews and grouping of enterprises After performed workplace interventions, the ten managers at studied enterprises answered a questionnaire used in other studies [21] about content and process in the interventions. The used 18 items formed three indices: learning strategy index (ten items about visions of the change process where several problem areas are addressed, whether both employees and leaders were taking part in and influencing change processes and how both planning and improvisation has characterized change processes); a top management involvement index (four questions about managers pursuing the change process which considered opinions and different views, about managerial beliefs in positive outcomes and strong identification with visions); and quality of working life index (four questions about to what degree the actual change process has contributed to improved working climate, co-operation, involvement and more varied work tasks). All indicators were based on questions on a Likert scale that ranges from 4 (to a great extent) to 1 (Not at all). The internal consistency data in earlier studies [21] using Cronbach’s alpha (α) coefficient for the three indices were: learning strategy index (α=0.76-0.81), top management involvement index (α=0.84-0.88) and quality of working life index (α=0.88-0.92). Semi-structured interviews were performed with all managers, after the enterprises had completed their workplace change programs. The interviews were at respondent workplaces and lasted approximately two hours. In accordance with an earlier study [37], content focused on change methodologies used, changes about job/organizational factors, health and performance, change strategies, and obstacles in the processes. Notes were made during interviewed. They were reviewed soon after each interview. Clarifications and follow-up questions were then made by telephone. Documents about workplace interventions were also incorporated into qualitative method used during the study. As a result of the leader questionnaire and interview, the enterprises were classified into two different change strategy types, a broad strategy and an expert/problem-based strategy - this was similar to what was used by Håkansson [18]. The broad strategy deals with several extensive problems concurrently, the degree of participation is high and several arenas are used during development processes. The expert/problem-based strategy has a lower degree of participation, is more expert oriented and deals with only a few problem areas.

6

3.2 Statistical analyses Statistical analyses consisted of comparisons of first and second questionnaire responses of individuals using the previously described determinant and outcome indicators. For all used indicators, Mean (M), Standard Deviation (SD) and Cronbach’s alpha (α) – were calculated for both measuring points (Time 1 and 2). As measures were performed for the same individual at Times 1 and 2, the t-test to measure differences between groups was based on paired samples test. Pearson’s product correlation coefficient (r) was used for measures of associations between determinant and outcome indicators. P-values less than 5% were regarded as statistically significant. Comparisons were also made between groups of enterprises (broad strategy versus expert/problem-based strategy). First, mean changes for all determinant and outcome indicators were calculated within both enterprise groups. Due to that it was possible to follow the same individual before and after performed interventions; a paired samples ttest were used here. Second, an independent samples test with equal variances assumed was used for comparing differences in indicator mean changes for the two enterprise groups. 4. Analysis results 4.1 Descriptive analyses Table 1 shows the reliability coefficients (Cronbach’s alpha) and the means and standard deviations for Time 1 and Time 2 variables (before and after workplace interventions) for the whole study group. High values indicate more positive values, and t-values and significance indicated for paired sampled test. With the exception of self-assessed health, the reliability coefficients were sufficient, ranging from .68 to .90. The test-retest correlations, ranging from .30 to .58 showed that the scales for job characteristics, health and performance are fairly stable over time. For the indicators customer oriented quality practices, respectful leadership, creative work, team spirit, restructuring worries, self-assessed health, workplace adaptability and customer satisfaction, there are significantly lower values at Time 2.

7

Table 1 Basic statistics of the study indicators (n=102) Indicator Customer oriented quality practices Respectful leadership Creative work Team spirit Work intensity demands (low) Data stress (low) Decision latitude Restructuring worries (low) Self-assessed health Workplace adaptability Customer satisfaction

α .85

Time 1 M 3.19

SD 0.92

α .87

Time 2 M 2.83

SD 0.93

t 3.96

Sign. .000

.90

3.29

0.95

.88

2.95

0.96

3.99

.000

.80 .76 .82

3.93 4.15 3.26

0.75 0.77 1.04

.81 .81 .85

3.63 3.96 3.25

0.80 0.86 1.11

4.11 2.05 0.07

.000 .043 .946

.79

3.96

1.18

.68

4.07

1.07

-0.94

.348

.80

2.88

0.81

.78

2.84

0.78

0.55

.580

.80

4.52

0.85

.89

3.61

1.56

5.97

.000

.61

4.29

0.71

.78

3.95

0.98

3.64

.000

.76

3.67

0.75

.83

3.42

0.91

3.03

.003

.78

3.93

0.62

.88

3.75

0.85

2.43

.017

4.2 Interrelations of health and effectiveness outcomes and job characteristics For the entire study group, the relations between job and organizational characteristics, and outcomes that were assumed to measure for health and performance were analysed (Table 2). In this table, bold figures indicate a significant correlation at the 0.05-level and bold and underlined figures a significant correlation at the 0.01-level. The determinant indicators respectful leadership, creative work and team spirit related significantly in the assumed direction with the outcome indicators self-assed health, workplace adaptability and customer satisfaction. The outcome indicator, self-assessed health, also related significantly in the assumed direction with the indicators work intensity demands and data stress at both measuring points. For customer oriented quality practices there were no significant relationships, and for decision latitude and restructuring worries there was a relationship only at Time 2 with self-assessed health.

8

Table 2 Correlation coefficients (Pearson’s r) between job and organizational characteristics, and health and performance outcomes (n=102) Indicator

Customer oriented quality practices Respectful leadership Creative work Team spirit Work intensity demands (low) Data stress (low) Decision latitude Restructuring worries (low)

Self-assessed health

Workplace adaptability

Customer satisfaction

Time 1

Time 2

Time 1

Time 2

0.05

0.07

0.55

0.51

Time 1 0.51

Time 2 0.58

0.23

0.34

0.63

0.59

0.55

0.55

0.27

0.30

0.51

0.52

0.52

0.51

0.33 0.23

0.22 0.32

0.34 -0.03

0.51 -0.10

0.36 -0.19

0.51 -0.12

0.28

0.30

0.15

0.25

0.14

0.04

0.14

0.28

0.03

0.18

0.04

0.25

0.05

0.29

0.23

0.34

0.28

0.17

For the performance outcome indicators of workplace adaptability and customer satisfaction there were also strong relationships with the assumed determinants for quality practices at both Time 1 and Time 2. For indicators concerning work demands and data stress, the relationships were not significant except for data stress and workplace adaptability at Time 2. The indicator decision latitude correlated significantly with customer satisfaction at Time 2. For restructuring worries the relationships with the outcome variables were significant except for the relationship to customer satisfaction at Time 2. 4.3 Effects of the workplace interventions for two change strategy types In order to get a picture of the relationships between change strategy types and changes in the study’s indicators, the reported questionnaire data from both leaders and coworkers were used. From the leader questionnaire and the interviews, it was possible to group the workplace interventions into two types: broad strategy and expert/problembased strategy. In Table 3, the leaders’ views on indicators for learning strategy, top management involvement and quality of working life are presented for the two change strategy types. High values indicate more positive values.

9

Table 3 Leaders view of learning strategy, top management involvement and quality of working life for two change strategy types Indicator

Learning strategy index Top management involvement index Quality of working life index

Broad strategy

Expert/problem-based strategy

3.17 3.72

2.55 2.61

3.81

2.32

The leaders in enterprises using a broad strategy in the workplace change processes had higher values on both the learning and top management involvement index, and they also judged that quality of working life aspects have improved to a greater extent when compared to responses from leaders of enterprises using a expert/problem-based strategy (all differences are significant using t-test for independent samples). In Table 4 below, the individual’s judgement of changes concerning job and organizational characteristics and outcomes for health and effectiveness are presented; they are grouped into the two change strategy types. When comparing mean changes for indicators used by enterprises that adopted a broad strategy with enterprises having an expert/problem based strategy, there are significant differences concerning customer oriented quality practices, respectful leadership, creative work, restructuring worries, workplace adaptability and customer satisfaction (see Table 4).

10

Table 4 Mean changes (M) in job and organizational characteristics and outcomes for health and performance for two change strategy types (n=102) Indicator

Broad strategy (N=31)

Expert/ problembased strategy

M

(N=71) M

t

Sig.

Customer oriented quality practices

-0.09

-0.49

-1.98

0.050

Respectful leadership

-0.02

-0.48

-2.52

0.014

Creative work

0.01

-0.44

-2.86

0.005

Team spirit

0.00

-0.27

-1.36

0.178

Work intensity demands (low)

-0.09

0.03

0.54

0.591

Data stress (low)

0.03

0.15

0.44

0.659

Decision latitude

-0.04

-0.04

0.01

0.991

Restructuring worries (low)

0.08

-1.34

-4.73

0.000

Self-assessed health

-0.16

-0.43

-1.29

0.199

Workplace adaptability

0.08

-0.40

-2.66

0.009

Customer satisfaction

0.05

-0.28

-2.06

0.042

4.4 Leader interview results Some comprehensive patterns for enterprises using a broad change strategy are that several areas such as work organization, leadership development and quality improvements are combined with individual measures concerning health promotion. Managers in these enterprises can be characterized as those who were involved, proactive and not afraid to take a stand. They also have a strong belief in connections between working environment, employee health, effectiveness and quality aspects. Other characteristics are that employees to a relatively high degree are participating in the change processes, and that several arenas for the developing the workplace exist. These leaders also tend look upon change processes as continuous improvements and that human resource issues must be integrated with business issues. Although this picture described above points towards several positive aspects, the leaders also identify obstacles in the way of long-term improvements. Lack of resources for development work, insufficient competence concerning successful measures and external factors are commonly mentioned obstacles.

11

Studied enterprises using a more expert/problem-based strategy work with more narrow change processes, and with only a couple measure areas. Even if the managers were participating in the processes, they delegated more to external experts and they didn’t have the same clear values about connections between human resource development and business outcomes. In some of these enterprises only a few employees were participating and in some the strategy was more informative without any real opportunity for employees to influence change processes. It also appeared that the processes were more fragile and easily interruptible in these enterprises. 5. Discussion and conclusions Small business research often focuses on managers and leaders. The study here is innovative in the sense of its longitudinal design and that it contains responses from both leaders and co-workers. The findings of rather strong relationships in the assumed direction between the determinant indicators respectful leadership, creative work and team spirit and outcome indicators of health and performance gives some support for models about healthy work organizations as applied to small companies. This is in line with a study of larger workplaces where a comprehensive model was tested [42], a study of Finnish small and medium-sized enterprises [28] and research about the specific importance of these factors in small enterprises [9]. The findings that there were strong relations between the indicator of customer oriented quality practices and assumed performance indicators (workplace adaptability and customer satisfaction) was in line with other studies about quality development in small enterprises [2,15]. The findings that there were no strong links between indicators for quality practices and self-assessed health and weaker relationships between work intensity demands, decision latitude and performance outcomes opens possibilities for further research in this area. The clear relationship between the indicators restructuring worries and self-assessed health differ between Time 1 and Time 2 is possibly connected to general trends in working life regarding increased insecurity and organizational change as reported by Härenstam et al. [19]. When looking at the effects of the interventions for the whole group through a comparison of indicator mean values at both measuring points there was a significantly change for the worse concerning almost all indicators. The exceptions were the indicators for work intensity, data stress and decision latitude. There are several possible explanations behind this result. The general trends in Swedish work life during the follow-up period were characterized by increased sickness absence, deteriorating psychosocial conditions, insecurity and worry about organizational change [19]. As described in the earlier review of theoretical aspects, small enterprises have limited resources for the carrying out of change processes which affects the quality of change processes. With relatively few exceptions this was the case for the group of small enterprises studied. However, when enterprises using a broad strategy are compared with enterprises using a more expert/problem-based strategy there are significant differences concerning mean

12

changes for the determinant indicators of quality practices, respectful leadership, creative work and the outcome indicators judging workplace adaptability and customer satisfaction. This result could be interpreted to mean that the broad strategy companies have “balanced” an otherwise negative development caused by the negative working life trend. If correct, this interpretation then suggests that a broader change strategy, to some extent, is more likely to be the more effective strategy for small enterprises. Several researchers from different areas give support to the idea that more holistic change programs with a high degree of participation by both leaders and co-workers are more effective (e.g Håkansson [18], Ingelgård [20], Lindström et al. [28]. An interesting result is that there are no significant differences for self-assessed health, when comparing the two enterprise groups. Apart from the above discussion of general trends, another explanation could be that changes in health status first begin to show a couple of years after the measures have been started [32]. This line of reasoning gives some support to the concept that sustainable workplace change processes and workplace health interventions are important. The findings that the leaders in the broad strategy enterprise group judged significantly higher values concerning learning strategy, top management involvement and outcomes related to quality of working life correspond to other research results [20]; this indicates that a combination of a high learning strategy and high top management involvement gives the most favorable results for small enterprise performance. The leader interview results also show that it is possible to combine different organizational and individual oriented measures in small firms along with a high commitment from both leaders and employees. However, the interview results also point at obstacles concerning workplace change processes as lack of resources, insufficient competence and influence of external factors. This result point at the need for more integrated models for workplace health intervention; particularly when considering the circumstances of small enterprises. In summary, this study provides evidence concerning links between such small enterprise organizational factors as leadership, creative work and team spirit, and outcomes related to self-assessed health, judged workplace adaptability and customer satisfaction. The study also indicates that high quality change processes with a broad learning strategy and high top management involvement can apply to small enterprises and help their performance. However, it is important to find out more about how such change strategies should be adopted to particular circumstances in small enterprises. 5.1. Limitations It is important to note that the findings in this study come from a limited sample of Swedish small enterprises and that the results should therefore be interpreted with caution. The empirical base needs to be broadened to further investigate effects of different change strategies with the ultimate goal being to find out more about successful factors and obstacles for such strategies. Although the longitudinal design is a strength (it was possible to follow the same individual before and after workplace interventions) it is also a weakness as the study is dependent on this assessed questionnaire data. For future research, it will be important to combine quantitative and qualitative subjective data with more objective data concerning for instance, sickness

13

absence figures, customer satisfaction judged by the customers, and financial and quality performance outcomes. References [1]

J. Alpenberg and F. Karlsson, Arbetslivsfondens satsningar på småföretag – en effektutvärdering (The Working Life Fund’s work with small businesses – a result assessment), Centrum för Småföretagsutveckling (Centre for Small Business Development), Högskolan i Växjö (Växjö University), 1994.

[2]

M. Anderson and A.S. Sohal, A study of the relationship between quality management practices and performance in small business, International Journal of Quality & Reliability Management 16(3) (1999), 859-877.

[3]

A.B. Antonsson, L. Birgersdotter and S. Bornberger-Dankvardt, Small enterprises in Sweden: Health and safety and the significance of intermediaries in preventive health and safety, Work and Health 2002:1, National Institute for Working Life, Stockholm, 2002.

[4]

G. Aronsson and T. Lindh, Långtidsfriskas arbetsvillkor: En populationsstudie (Work conditions among workers with good long term health: A population study), Work and Health 2004:10, National Institute for Working Life, Stockholm, 2004.

[5]

G. Aronsson, L. Svensson, K. Leksell and A. Sjögren, Förändringskompetens: Projektledares erfarenheter från 300 arbetslivsfondsprojekt (Change and innovation capability: project manager experiences from 300 Working Life Fund projects), National Institute for Working Life, Stockholm, 1995.

[6]

C. Axelsson, Formalisering som hinder och möjlighet i små företag: en studie av arbetsmiljöarbete ur ett lärandeperspektiv (Hinderences and opportunities in small business due to formalism: A study of work environment research from a theoretical perspective), Doctorial Dissertation 2002:14, Luleå University of Technology, 2002.

[7]

J. Axelsson, Quality and Ergonomics - Towards successful integration, Doctorial Dissertation, Linköping University, 2000.

[8]

L. Bager-Sjögren, O. Gustafsson and L. Pettersson, Växtkraft Mål 4 – en effektstudie, kompetensutveckling i små företag (Growth Policy Goal 4 – a study of the effects of competence in small businesses), Rapport A2003:001, Institutet för tillväxtpolitiska studier (Swedish Institute for Growth Policy Studies), Östersund, 2003.

[9]

S. Bornberger-Dankvardt, C.G. Ohlson and P. Westerholm, Arbetsmiljö- och hälsoarbete i småföretag - försök till helhetsbild (Work environment and occupational health in small business – seeking the full picture), Work Life in Transition 2003:1, National Institute for Working Life, Stockholm, 2003.

[10] J. Curran and R.A. Blackburn, Researching the small enterprise, SAGE Publications, London, 2001. [11] J. Eklund, Development work for quality and ergonomics, Applied Ergonomics 31 (2000), 641-648.

14

[12] G. Ekwall and J. Arvonen, Leadership Profiles, Situation and Effectiveness, Creativity and Innovation Management 3(3) (1994), 139-161. [13] N. Goldschmidt and B.G. Chung, Size does matter: the effect of organisational size on customer satisfaction, Journal of Quality Management 6 (2001), 47-60. [14] B. Gustavsen, A. Wikman, M. Ekman-Philips and B. Hofmaier, Concept-Driven Change: The Core Element in a National Workplace Development Program, Concepts and Transformation 1(2/3), 193-211. [15] J. Hansson, Total Quality Management – Aspects of Implementation and Performance: Investigations with a Focus on Small Organisations, Ph.D. Dissertation, Luleå University of Technology, 2003. [16] K.B. Hendricks and V.R. Singhal, Does implementing an effective TQM program actually improve operating performance? Empirical evidence from firms that won quality awards, Management Science 43(9) (1997), 1258-1274. [17] M.A. Huselid, The impact of human resource management practices on turnover, productivity and corporate financial performance, Academy of Management Journal 38 (1995), 635-672. [18] K. Håkansson, Förändringsstrategier i arbetslivet (Change strategies in working life), Ph.D. Dissertation, University of Göteborg, 1995. [19] A. Härenstam and E. Bejerot, Multilevel analyses of organizational change and working conditions in public and private sector, European Journal of Work and Organizational Psychology 13(3) (2004), 305-343. [20] A. Ingelgård, On Macroergonomics and Learning Strategies in Improving Working Conditions, Ph.D. Dissertation, Göteborg University, 1998. [21] A. Ingelgård and F. Norrgren, Effects of change strategy and top-management involvement on quality of working life and economic results, International Journal of Industrial Ergonomics 27 (2001), 93-105. [22] B. Johannisson and H. Landström, Images of Entrepreneurship and Small BusinessEmergent Swedish Contributions to Academic Research, Studentlitteratur, Lund, 1999. [23] B. Johansson, En motivationsbaserad analys av arbetsmiljöarbetet i en grupp av mindre tillverkande företag (A motivation based analysis of the working environment activities in a group of small manufacturing enterprises), Ph.D. Dissertation, Luleå University of Technology, 1998. [24] J.V. Johnsson and E.M. Hall, Job strain, workplace social support and cardiovascular disease: A cross-sectional study of a random sample of the Swedish working population, American Journal of Public Health 78(7) (1988), 1336-1342. [25] R. Karasek and T. Theorell, Healthy work: Job Stress, Productivity and the Reconstruction of Working Life, Basic Books, New York, 1990.

15

[26] A.J. Kompier, C.L. Cooper and A.E. Geurts, A multiple case study approach to work stress prevention in Europe, European Journal of Work and Organizational Psychology 9(3) (2000), 371-400. [27] C. Källestål, eds, Hälsofrämjande arbete på arbetsplatser – Effekter av interventioner refererade i systematiska kunskapsöversikter och i svenska rapporter (Workplace health promomotion – Effects reported in systematic literature overviews and in Swedish reports), Statens folkhälsoinstitut (Swedish National Institute of Public Health), Stockholm, 2004. [28] K. Lindström, K. Schrey, G. Ahonen and S. Kaleva, The effects of promoting organizational health on worker well-being and organizational effectiveness in small and medium-sized enterprises, in: L.R. Murphy and C.L. Cooper, eds, Healthy and productive work – an international perspective, Taylor & Francis, London, 2000, pp. 83-104. [29] J.P. Machenbach, J. van der Bos, M.A. Joung, H. van der Mheen and K. Stronks, The determinats of excellent health: Different from the determinants of ill-health?, International Journal of Epidemiology 23(6) (1994), 1273-1281. [30] E. Menckel and L. Österblom, Ledningssystem för hälsosamma arbetsplatser (Management systems for healthy workplaces), in: Framtidens arbetsmiljö- och tillsynsarbete (Working environment and supervision in the future), B. Johansson and J. Johansson, eds, Studentlitteratur, Lund, 2004. [31] L.R. Murphy and C.L. Cooper, Healthy and productive work – an international perspective, Taylor & Francis, London, 2000. [32] D. Nutbeam, Building evidence of effectiveness in health promotion: measures, methods and madness, Abstract presented at 3rd Nordic Health Research Conference, Tampere, Finland, 6-9 September, 2000. [33] P.D. Reynolds and S.B. White, The entrepreneurial process: Economic growth, men, women and minorities, Quorum Books, Westport, 1997. [34] Statens Beredning för medicinsk Utvärdering - SBU (The Swedish Council on Technology Assessment in Health Care), Sjukskrivning – orsaker, konsekvenser och praxis: En systematisk litteraturöversikt (Sickness Absence - causes, consequences and practices), Stockholm, 2003. [35] D. Storey, Human resource management policies and practices in SMEs in the UK: Does it really influence their performance?, The Centre for Small & Medium Sized Enterprises, Warwick, 1999. [36] W.A. Taylor, Leadership challenges for smaller organizations: Self-perceptions of TQM Implementations, Omega International Journal of Management Science 25(5) (1997), 567-579. [37] S. Vinberg, Change processes and health outcomes in micro-enterprises and small public workplaces in rural areas, Finnish Journal of Rural Research and Policy 12(4) (2004), 151-162.

16

[38] S. Vinberg, Organizational and health performance in small enterprises in Norway and Sweden, Work: A Journal of Prevention, Assesment and Rehabilitation 24(3) (2005), 305-316. [39] D. Walters, Health and Safety in Small Enterprises, European Strategies for Managing Improvement, P I A-Peter Lang S A, Brussels, 2001. [40] P. Westerholm and S. Marklund, Strategies for occupational health research in an changing Europe, Work and Health 2002:12, National Institute for Working Life, Stockholm, 2002. [41] R.H. Westgaard and J. Winkel, Review article, Ergonomic intervention research for improved musculoskeletal health: A critical review, International Journal of Industrial Ergonomics 20 (1997), 463-500. [42] M.G. Wilson, D.M. DeJoy, R.J. Vandenberg, H.A. Richardson and A.L. McGrath, Work characteristics and employee health and well-being: Test of a model of healthy work organization, Journal of Occupational and Organizational Psychology 77 (2004), 565588. [43] J. Wiklund, Small Firm Growth and Performance: Entrepreneurship and Beyond, Ph.D. Dissertation, Jönköping International Business School, 1998. [44] J. Wiklund, P. Davidsson and F. Delmar, What Do They Think and Feel about Growth? – An Expectancy-Value Approach to Small Business Managers’ Attitudes Toward Growth, Entrepreneurship Theory and Practice, Spring 2003.

17