PARENTING SUPPORT ON SOCIAL MEDIA Challenging the roles of ‘skilled’ professionals and ‘risky’ young mothers?

This paper is about how young parents in Sweden use social media to seek out, receive and give, parenting support to one another online. The internet, and especially social media, has opened up new venues that allow parents to interact in forum discussions and other online exchanges in ways that enable negotiating ideas about good and desired parenting practices largely outside the reach of conventionally appointed experts and governmental institutions. Young parents are a stigmatised group, and because of their age they have a tendency towards feeling a greater degree of affinity with new media. For such different reasons, peersupport online may be especially attractive to young parents. In the Swedish setting, young parents have a conspicuously low degree of participation in organised and public forms of parenting support (Petersson, Håkansson, and Petersson 1997; Fabian, Rådestad, and Waldenström 2006). This means that for some reason young parents are potentially out of reach of expert socialisation, even though they tend to be seen as ‘risky’, and in particular need of support (Beck & Beck-Gernsheim, 1995; Lupton, 1999; Goodwin & Hupaz, 2010; Lee et al., 2014). This, furthermore, begs the question whether these parents seek out support at all, and if so where? In this paper, we analyse three different Swedish Facebook groups that are administrated in three different ways; by professionals, by peers, and by a semi-professional. The overarching aim of the study is to assess to what extent, and how, social media may potentially function as alternative platforms for peer-support among young parents. What are the characteristics of the social formations that take shape, and the exchange of knowledge that takes place on such platforms? Furthermore, what are the features of the actual exchanges as regards topics and modes of expression? More specifically, we ask: - How do the social network relations, and the discussions, differ depending on the analysed Facebook group is administrated by professionals or peers? -In what ways may professional experts play a role in communication among peers online? 
 -How might parents use social media to ‘hide’ their discussions away from the controlling eye of governmental institutions? - What potential do social platforms, such as Facebook, have for being a tool for parents to take control over their own support needs?

Previous research The idea of parenting support has been imported to Sweden from countries such as the USA, Canada, and Great Britain (Sandin, 2011). Parenting support can either be referred to as universal support offered to all parents, with the aim of preventing any negative outcomes in the child’s future, or specific support offered to parents that are seen as risky, such as young parents which is the focus of this paper, but also adoptive parents, parents with a history of drug abuse, or families’ with disabled parents or children. The universal parenting support is offered both during the pregnancy and during the babies’ first year (Author a; SOU

2008:131). Both of these parenting groups have, in Sweden relatively high rates of participations, 70 percent (Hwang & Wickberg, 2001). The extensive investments in parenting support as well as the production of guidelines for parenting practice such as feeding routines, are based on the logic of parenting determinism according to which ‘good parenting’ is seen as the preventive cure for different kinds of social problems for example crime and poverty (Lee, Bristow, Fairlcloth and Macvarish, 2014). A foundational idea is that good parents raise ‘good and strong citizens’ which in turn results in a ‘good society’ (Moss, 2014). In one sense the idea of parenting determinism is based on a rather philanthropic and democratic idea in offering special interventions to ‘risky parents’, with the intention of giving all children as equal pre-conditions as possible (Sandin, 2011). But at the same time the idea of parenting determinism has been problematized by Lee et al (2014 see also Frank Furedi, 2001) since it firstly builds on the assumption that parents cannot raise their own children without governmental support and involvement. Secondly because it leads to a categorisation of some parents as being fit, while others are seen as unfit to parent. The dominant understanding of young parents, within research are that they are in lack of parental skills such as: low self-esteem regarding parenting (Letourneau, Stewart, & Barnfather, 2004), problematic attachment to their children (Flaherty & Sadler 2011) and inadequate knowledge about child rearing and child development compared with older parents (Emery, Paquette, & Bigras, 2008). The governmental idea of offering specific support for young parents probably leans on research pointing towards that young parents are in need of special support in order too increase their parenting skills. However there are research showing that young parents sees themselves as capable parents, and mostly are troubled by stigmatizations from society viewing them as less good parents (Duncan, 2010; Kirkman, Harrison and Priscilla 2001), as well as troubled by their specific poor economical situation, since they barley finishing school or entering labour market and sometimes relay on their own parents for housing and financial support (Author b; Phoenix, 1999). Perhaps the low participation in face-to-face support of young parents can be explained by a discrepancy between the professionals understanding of young parents as in need of support to increase their parental skills and the young parents own understanding of their support needs in other areas specific for young parenthood, that are available online by peer-support and within an space with less risk of being stigmatized. In addition to the expert guided face-to-face support offered by the government, we know that parents go online looking for information and to create and join online groups for social support both at dedicated discussion forums (Dworkin et al., 2013; Lind 2013; Plantin & Daneback 2009) and on social media platforms such as Facebook. Studies shown that parents prefers to search for parental information online since this are described as convenient and comfortable way of gathering information, and offers an immediate information of a more detailed and interactive form than provided by the face-to-face support groups. Through online support and information, parents are able to get information concerning their own child’s normative development and what them as a parent encounter in this where minute (Warren et al., 2010; Dworkin et al., 2013). Bernhard & Felters (2004) found that parents prefer online information from professionals, because it felt trustworthy, but prefers online parenting support from other parents on community based communication systems, messages or e-mail groups (Bernhard & Felters, 2004). When it comes to trustworthiness on community based communication systems, parents report an discrepancy, since they on one hand report that they prefer online support provided by professionals or peers, but on the other hand, when actually using online support, parents have very high rates of using commercial websites providing communication systems (Jang & Dworkin, 2012).

There do exist different governmental and expert guided online programs for parenting support, but no program or website that works nation wide or has gain at high usage and impact in peoples every day life of searching for information and parenting support as the bigger commercial or peer supported bases for communication system (Dworkin et al., 2013). For example, the ‘Mumsnet’ in Great Britain or the ‘Familjeliv’ forum in Sweden. One example of online peer support administrated by professions, is a program offering online parenting support for young lone mothers. The program had a preventive purpose, and reached 42 young mothers that where given the access of a specific network with peer support, under the frames of professionals. Dunhams (1998) evaluation of the program reported positive results, where the mothers reported a feeling of decreased parenting stress, through their participation in the program. Much like in other areas where like-minded individuals who share experiences of similar situations and challenges, there seems to be a need for coming together in groups of peers to exchange social support (ref). The processes of peer-support in general may be supervised or unsupervised to varying degrees, and just as the online space may be more or less open or closed as regards who can join, the exchanges can be based on differing levels of anonymity. Parents use online parenting peer supports as a way of make sense of their parental identity position and to improve their parental skills (Brady & Guerin, 2010). Through exchanging experiences and advices with others, for example about difficult infant sleep, breastfeeding, balancing work and family, parent are able to normalize their own experiences and find their own parental identity position and strategies to handle their every day parenting (Plantin & Daneback 2010; Hall & Irvine, 2008; Nicolas et al., 2009). The online shared experiences among peers works as a strategy to feel more safe and empowered in their everyday parenting (Madge & O´Connor, 2006). Lind (2013) found that parents communicating on online support becomes both their own experts and the experts of other parents. Indicating that the online parenting support can have possible implications for changing the understanding of the role as an parental expert - as in the parents themselves, other parents or the professionals or all of them. Also the communication system, functions as a base of knowledge in ‘good parenting’, and are used by parents as a frame or reference for their own parental practice (Dworkin et al., 2013; Lind, 2013). Although most studies writes about ‘parents’ use of online support, the use of online information and online parenting support are gendered, in where it often is the mother in the family that are the online consumer (Lind, 2013; Brady & Guerin, 2010).

When studying online social networks it is important to keep in mind that they may be quite different from offline face-to-face groups (Bambina, 2007:18). Therefore, it is important to explore online support networks on their own terms and to evaluate the extent to which they may cause a realisation of the potentials that are inherent in these platforms. A network approach, such as that employed in this study, focuses on the general characteristics of relationships rather than on the individual characteristics of unique interactions. As Bambina argues, social network analysis can make important contributions to research on online support groups since previous studies have tended to focus on considerably smaller groups, while backgrounding the larger social network context (Bambina, 2007:14-15). Internet forums and social media groups tend to be relatively large in numbers, with extremely varied levels of participation and loose commitments (Bambina, 2007:15.16). Various measures can be used to evaluate social networks. One group of measures aims to assess network structure in terms of size (number of individuals) or the degree to which relationship ties are intertwined (density). A second group is aimed at describing functional

aspects, which in the context of support networks has to do with how well the network provides emotional and instrumental resources as well as affirmation of value and a sense of connection (Balaji et al., 2007). It is of interest to study in which ways the support forum is something more than ‘just an aggregate of people’ (Wenger, 1998:74). Online support forums are based on forms of communication that not necessarily rely on mutual engagement. For a sense of community to develop, there has to be a component of people doing things together, establishing and maintaining relationships. There has to be a social complexity and community maintenance. Furthermore, there needs to be some sort of joint enterprise which is collectively negotiated and which is marked by mutual accountability.

Data and methods In order to respond to the research questions, we strategically selected three Facebook groups to represent different degrees of openness and kinds of ‘leadership’ (see case descriptions below). A Facebook group is a function within Facebook’s platform in where people are free to create a page (similar to a homepage) with the purpose of gathering people with a joint interest or on a common grounds, such as in this case, being young when entering parenting and in need of interaction and discussion similar topics. In the process of selecting these three groups for closer study, we did Facebook searches for groups aimed towards young parents, using a variation of relevant search criteria. We identified a total of X groups. In line with previous research, we noted a gender difference, where all identified groups, but one, where administrated by young mothers (Lind, 2013; Brady & Guerin, 2010). The one administrated by young fathers had few members (X) and groups administrated by young mothers had few male members and even more rare that the male members where active in posting or commenting. Therefore we will from here on talk about young mothers online parenting support. It was a common pattern that groups administrated by professional were open, while groups administrated by young mothers themselves demanded applications for membership, and were thus out of reach for experts. The three selected cases were analysed using three different types of methods: network analysis, online ethnography, and interviews with the administrators. The network analyses served as starting points offering broad information and a mapping of each group in rather quantitative terms. Through this analyse a number of questions concerning the function and aim of the administrators, what the communication consisted of and if it where any differences of the conversational tone between the three groups. To answer these questions we furthering our exploration with an online ethnographical analysis. The online ethnography entailed following the three groups daily during a period of… The interaction within each groups, in terms of common pattern, interactive behaviour that stood out and so on where noted in field notes on an every day basis. We also tried to be sensitive to the roles of, and power relations between, the administrator and members, as well as among members. After concluding the network analysis and the gathering of the online ethnographical data, we also carried out interviews with the main administrators of two of the three Facebook groups. The group administrated by peers that, after internal dialogue, decided not to agree on an interview, with the motivation that they wanted to keep their group strictly closed to outsiders. This is of course an interesting finding in itself. The interviews, where conducted by one of the authors, and served to get an direct interaction with the group, asking for any information about the foundation of the groups and functions that where not visible on there Facebook group. Through the interview we where also able to ask about the conversational tone, power

and tools of regulating the conversation, relationship between the administrators and the members as well as the administrators thoughts on peer support versus expert guided support, and what supportive function they thought that their group had for young mothers. The data gathered through online ethnography and through interviews (that been transcribed), where analysed with an inspiration of Braun and Clarke’s (2006) thematic content analysis. Where we firstly read the data repeatedly, thereafter made initial coding and started to formulating tentative themes, and continued the procedure of coding until our central themes where identified: the content of the communication (advice about maternal practices, legislations, stigma and relations to the father, partner, grandparents or others), the administrators functions in terms of power and taking on an position as the expert, and lastly the tone and level of intimacy in the communication. While the network analysis gives us the broader picture of the groups communication, the online ethnographies and the interviews gave an deeper insight regarding the content of the communication and to learn more about the power relation between the young mothers and the semi-professional and who have the control of the communication and the flow of interaction. The limitation of these three methods, is that it measure parents interaction, e. i. the communicative bound between too or more persons. However, the comments in the groups are accessible for any member in the group or others if the group have an open access. Within Social media, people can ‘lurk’ around and taking part of the conversation without being visual participating (Plantin & Daneback 2010; Lind, 2013). This means that the information shared between two persons can also been read and communicated through more than these two persons.

Case descriptions Group 1: Administrated by professionals This Facebook group ) was open for anyone to join, and was administrated by professionals who worked parallel with face-to-face parenting support groups for young mothers in a Swedish municipality. It was started in 2010 by a group of 2-3 professionals who had already been working for many years with face-to-face parenting support. Their face-to-face group mainly consisted of a core of 10-15 regular participants, mostly mothers, during each halfyear term. The support workers took the group online, to the early Swedish social media site Lunarstorm, already in 2003. As the popularity of that site was declining, they switched over to a Facebook group in 2010. At the time of our study, in August 2016, the group had 46 members. In the group description, the administrators have written at length about how their parenting support, for young mothers and young pregnant women, mainly takes place offline. They say that the purpose of the Facebook group mainly is to inform young mothers, and pregnant young women, about the possibility to get support offline, and explained what type of support they were offering face-to-face.

Group 2: Self-administrated by young mothers This Facebook group was administrated by young mothers themselves and demanded that users applied for membership. In order to get access to data, one of the authors of this paper applied for membership and informed about the purpose of our study. The group was started in November 2014 by a young mother. At the time of our study, the group had 950 members, while the number of administrators had grown into 6 young mothers all with different functions and “titles”. The site description states that the group is for young mothers to share their thoughts about pregnancy and motherhood. It also says that the maximum age for joining is 30 years old. Other Facebook groups that are administrated by young mothers, and that had closed access, had similar or more expressive descriptions, such as ‘There is

nothing wrong with being a young parent!’, or ’Once a young parent, always a young parent :)’. Group 3: Administrated by a ‘semi-professional’ This Facebook group was administrated by a semi-professional, namely a social worker who was also the mother of a young mother. This group was also closed and accessed through applying for a membership for one of the authors. In this group, the administrator wanted the author to present herself as a new member joining this group as a researcher. The group was started in November 2011 by the administrator. At the time of our study, the group had 181 members. The group description states that the group is for ‘proud’ young parents, to share good advice and to take part in each other’s everyday lives. In the interview conversation, the administrator said that the aim of starting the group where to create a safe meeting space where all mothers are young, and thus age does not matter. The group, she said, was to be the one place where they can be ‘just mothers’ not ‘young mothers’.

Already through the work in sampling out three cases, we can see an important results measured in terms of numbers of members, where the second group administrated by young mothers has far more members than the groups administrated by professionals and a semiprofessional. Most groups administrated by professionals within the frames of a municipality had a number of members around 40 or often fewer. While groups administrated by young mothers themselves often had around 200 members or more depending on if the group mainly where limited to a geographical border such as for all young parents within one county. These numbers indicates that most young mothers prefers parenting support online by peers rather than by professionals either online or at face-to-face groups. Now we will look deeper on how these groups work in terms of parenting support. This is first done through a network analysis, studying the connectivity and ‘communicative paths’ among the members, in the different groups.

Network analysis We have analysed the social connectivity in the Facebook groups. Connections between participants have been operationalised in terms written comments to posts. If a participant A has posted in a discussion thread that was initiated by a participant B, then there is a connection between the two. Analysing these connections, we present metrics for network density. Network density can vary between 0 and 1 and measures the connectivity between participants in a group. If all participants are connected to everyone else, the connectivity is 1, and if no one is connected to no one else, the connectivity is 0. Most networks of the type studied here tend to be very sparse, which means that they have low density scores. Still, some differences can be seen. As densities can be misleading, especially when values are small, we also present data for average degree (Wasserman & Faust, 1994:182). The average degree of a network represents the average number of connections of a participant. This means, in the context of this study, that a participant in a group with an average degree of 10 would on average have connections to five other participants. Figure 1 shows the results of the network analysis of the three Facebook groups. Figure 1. Social network analysis of the three Facebook groups

Group 1 Number of members: 46 Density: 0,009 Average degree: 0,76

Group 2 Number of members: 950 Density: 0,027 Average degree: 0,86

Group 3 Number of members: 181 Density 0,186 Average degree 5,2

Node size equals normalized degree. Edge width equals number of co-occurrences in threads. Bottom 24% of edges, and nodes with degree