Research on adolescent parenthood has found one of the most critical predictors of offspring’s well-being is social support; however, scholars have also found that expectant adolescents sometimes experience a lack of support from their face-to-face networks. Very little research has examined how adolescent parents might make up for this discrepancy by utilizing mediated social networks; therefore, this study aims to fill a gap in the literature by examining how adolescent parents and expectant adolescents utilize online networks for support. One hundred and fifty messages from two online forums of adolescent pregnancy Web sites were content analyzed to examine the type of social support solicited by individual members. Additionally, 150 pairs of messages were examined to determine whether other community members provided the type of support solicited in the original posts. Guided by the optimal matching model, findings revealed informational and emotional support were sought most frequently across the two forums, with few users soliciting esteem, network, and tangible support. A further examination of the responses provided to support seekers revealed these online communities’ members most frequently matched the type of support initially solicited, followed by situations where they provided more support than asked for. Out of all five support types, those original posts that solicited informational or tangible support were most likely to have responses that provided the support type that was requested and even though posters did not necessarily ask for it, community members often provided emotional and esteem support in their responses.
Researchers have examined online social support exchange in various environments, including discussion boards (Vayreda and Antaki, 2009) and social network sites such as Facebook (Schrag and Schmidt-Tieszen, 2014). One of the most commonly studied areas of online social support involves health-related forums. Researchers have examined the online social support exchange process in forums dedicated to topics such as individuals with disabilities (Braithwaite, et al., 1999), Amyotrophic Lateral Sclerosis (ALS; Loane and D’Alessandro, 2013), and depression (Keating, 2013).
Although the aforementioned studies have helped to dispel some myths concerning the inability of online networks to provide effective support compared to face-to-face networks, some contexts of online social support are still understudied. One of those contexts is adolescent pregnancy/parenting. Adolescent parents face a variety of stressors that make online social support a viable option for them, such as a potential lack of support from their face-to-face interpersonal networks, particularly friendship networks (Bunting and McAuley, 2004; Sherman and Greenfield, 2013). It is especially important for scholars to better understand this context because receiving social support is so integral to the success of adolescent parents and their children (Kim, et al., 2014). More knowledge of the support options available to adolescent mothers and which of those options are the most effective in particular situations could help mitigate negative outcomes associated with adolescent parenting.
In order to explore online social support exchange within the context of adolescent pregnancy/parenting, the current research utilized content analysis to investigate the following: (a) various types of support exchanged on adolescent pregnancy/parenting online forums; (b) the relationships between the type of support solicited and type of support provided by forum (or online community) members; and, (c) whether the replies given by members matched the social support solicited. This study contributes to previous research by addressing two gaps in the literature. First, the population of expectant adolescents and adolescent parents has been understudied in the social support literature, particularly in an online context. For example, Nolan, et al. (2017) did an extensive search of the literature to examine what previous scholars had found concerning the use of online social networks by adolescent mothers. The authors only found one study (i.e., Sherman and Greenfield, 2013) that examined the naturally occurring use of online networks by pregnant adolescents and/or adolescent parents. However, Sherman and Greenfield (2013) only examined posts made by forum users and did not consider how the responses users received compared to their original posts. Therefore, this study extends Sherman and Greenfield’s initial research by examining the interactive process of social support exchange communication (not just how individuals seek support). Second, this study answers a call by Vayreda and Antaki (2009) for more research that examines how the types of support provided in online forums compare to the support users solicit.
The information provided by this study is significant because it allows scholars to better understand the role online social networks play in the support exchange process of adolescent parents. Given the importance of social support to the success of adolescent parents and to the outcomes of children of adolescent parents (Furstenberg and Hughes, 1995; Pinzon and Jones, 2012), the more scholars understand about the support provided in online networks, the better practitioners will be able to help those adolescent parents who may lack support in their face-to-face networks.
Social support and adolescent pregnancy/motherhood
Social support, communicated via messages and actions that help people feel cared for, is particularly important in the context of adolescent pregnancy/motherhood (Furstenberg and Hughes, 1995). A social support network provides adolescent mothers with a system of individuals they can turn to in times of need, which has a variety of positive outcomes for their physical and mental health (Smith, et al., 1994). Although the number of children born to adolescent parents has declined in the past 30 years, adolescent parenting is still considered a social problem in many regions of the United States (SmithBattle and Leonard, 2012). This is likely because research has suggested children born to adolescent parents tend to have poorer life experiences than those with older parents. For example, previous research has linked adolescent parenting to poor behavioral, cognitive, and social outcomes for children (Clemmens, 2001; Kim, et al., 2014; Pinzon and Jones, 2012). Because of this research, scholars have continued to study what kind of life experiences might help to mitigate the negative outcomes children born to adolescent parents experience, and why some children have better outcomes than others. Many of the factors they have discovered were associated with social support the mother received.
For instance, Pinzon and Jones (2012) found early childhood care, provided by the infant’s family of origin, and support that helped the adolescent finish school were related to more positive outcomes for children of adolescent parents. Furthermore, Kim, et al. (2014) found adolescent mothers who reported receiving more support were less likely to experience postpartum depression. Furstenberg and Hughes (1995) suggested support from one’s family members was associated with a higher likelihood of completing school, avoiding trouble, and having more positive life outcomes for the children of adolescent parents. This research suggests familial social support plays a crucial role in the outcomes associated with adolescent parenthood.
Although support from one’s family should theoretically benefit the adolescent parent, empirical research has provided mixed results. For example, Bunting and McAuley (2004) suggested while support provided by family members could be beneficial for adolescent mothers, sometimes this support could also be a source of conflict. Adolescent parents are in a unique position in that they play the role of both child and parent. Because of this, there are times in which their own parents might provide social support that comes across as encroaching upon their independent-adult status (Bunting and McAuley, 2004). For example, longitudinal interview research revealed one adolescent mother complained about social support from her mother, suggesting that her mother’s involvement hindered her growth as a parent (SmithBattle and Leonard, 2012).
Overall, research suggests the way social support is enacted plays an important role in the outcomes of both children of adolescent parents and adolescent parents themselves. Since access to helpful social support within the family unit seems to vary, it is possible those adolescent parents that do not receive much support from their family seek support elsewhere. Some research has suggested friendship networks might be better equipped to support adolescent parents in a positive way (Bunting and McAuley, 2004; Nolan, et al., 2015). However, adolescent mothers tend to feel disconnected from their peer groups once they have a child (Sherman and Greenfield, 2013). Therefore, given the mixed effects of support provided by family members and potential disconnection from one’s peers, one of the resources adolescent parents might turn to in order to develop a support network could be the Internet.
Why online social support?
Online support groups help people build weak-tie networks (Wright and Bell, 2003; Wright, et al., 2013). Granovetter (1973) claimed individuals have networks that consist of both strong and weak ties. Strong ties are individuals one has known for a long time, feels intimate with, and has intense emotional connection or exchanges reciprocal amount of resources (Wright and Miller, 2010). Weak ties constitute those with whom individuals are not necessarily close (Granovetter, 1973), but may provide novel information and perspectives.
Individuals who experience difficult situations may choose weak-tie networks over strong-tie networks because of the experiential similarity of community members, more objective feedback, and greater security and comfort (Wright and Miller, 2010; Wright, et al., 2010). For example, research suggests online support contexts make it possible for individuals to seek support from others who have gone through similar experiences (Tanis, 2008; Walther and Boyd, 2002). Therefore weak-tie networks establish a platform for individuals with similar experience to convey empathy, which facilitates the provision of emotional support (Wright and Bell, 2003).
Due to these features, online networks, such as online forums focused on adolescent pregnancy and parenting, might offer a useful alternative that allows adolescent parents to feel more connected to others who understand their situation (Dunham, et al., 1998). Research by Schotanus-Dijkstra, et al. (2014) suggests online support groups can be especially helpful for individuals with a lack of face-to-face support options. Additionally, Nolan, et al. (2015) found adolescent mothers felt more socially connected when they utilized online networks. Finally, adolescent pregnancy is oftentimes stigmatized (Wiemann, et al., 2005). Previous research suggested online support networks might be particularly useful to those who were stigmatized because it provided them with a safe space to connect with similar others who were less likely to judge them, as well as an opportunity to feel empowered (Rains, et al., 2015; Sherman and Greenfield, 2013). For example, according to a report by the Centers for Disease Control and Prevention, states with the highest rates of adolescent pregnancy are located in the South and Midwest of the United States (Matthews and Hamilton, 2019). The Pew Research Center (2014) suggests many of the states with the highest rates of adolescent pregnancy also tend to be politically conservative, and have many residents that practice the Christian faith (Pew Research Center, 2019). Given that unwed pregnancy is discouraged among Christians, it is possible those individuals who grow up in religious and politically conservative areas, but become pregnant and/or have children as adolescents, may struggle to find support systems in their immediate personal environment.
Types of support necessary for adolescent parents/expectant adolescents
One way scholars have examined social support is to classify social support into five large categories; informational support incorporates providing advice, referrals, and details about situations. Emotional support includes providing empathy and physical contact. Tangible support refers to providing practical help. Providing compliments is the main feature of esteem support, and offering access to other sources is conceptualized as network support (Cutrona and Suhr, 1992).
According to Cutrona and Russell’s (1990) optimal matching model, in order to determine the type of social support that is most beneficial to the individual in need of support, one needs to consider the controllability of the stressor. Stressors are controllable to the extent that the individual is able to do something to prevent and/or diminish the stress caused by the situation in question. Those stressors that are controllable are best managed with support that is action-facilitating, or supportive acts that seek to eliminate the stressor or decrease its severity (Cutrona and Suhr, 1992). In the context of Cutrona and Russell’s (1990) typology, both informational and tangible support qualify as action-facilitating support (Cutrona and Suhr, 1992). Stressors that are uncontrollable, however, are best managed with nurturing types of support, or supportive acts that diminish the severity of negative emotions such as grief or guilt. In the context of the Cutrona and Suhr (1992) typology, emotional and network support are considered to be nurturing types of support. The final type of support, esteem support, is said to be helpful for handling both controllable and uncontrollable stressors (Cutrona and Suhr, 1992).
Adolescent pregnancy and being an adolescent parent present an interesting scenario to apply Cutrona and Russell’s (1990) model because some aspects of the stress associated with being an adolescent parent/an expectant adolescent are controllable, but others are not. For example, research suggests expectant adolescents worry frequently about what to expect during pregnancy and delivery (de Anda, et al., 1992). Based on Cutrona and Suhr’s (1992) definition of controllability, this particular type of stress should be considered controllable because questions about pregnancy and delivery can be answered by reading relevant books, talking to their doctor or case worker, attending parenting classes, and/or by asking others who have gone through pregnancy and delivery about their experiences. In addition, Cutrona and Russell (1990) treated transition to parenthood as a controllable event.
However, de Anta, et al. (1992) also found that expectant adolescents frequently worried about being expected to act like an adult, but treated like a child by others; thus, their own identity and abilities. Additionally, Bierman and Streett (1982) believe adolescent pregnancy creates an identity crisis for girls because they are not yet fully developed adults, but are treated as such by society and are judged by adult standards. Relatedly, the stigma placed on adolescent parents and expectant adolescents by society and potentially even close others could be a source of stress and concern (SmithBattle and Leonard, 2012). These identity-related issues are all based on the judgments of others, which are largely out of the adolescent’s control and there might not be anything the adolescent can do to change the way others in society view them. Therefore, stress related to other people’s perceptions of an adolescent parent seems to be uncontrollable vis-à-vis Cutrona and Suhr’s (1992) definition.
With this in mind, it is possible that the stress associated with adolescent pregnancy and adolescent parenting is both controllable and uncontrollable, and that all five types of Cutrona and Russell’s (1990) support are potentially relevant to adolescent parents/expectant adolescents. Although all five types of support may be relevant in general, the nature of online support communities can make certain types of support more useful in an online context than others. This possibility will be described in more detail below.
Solicitation of types of social support online
A variety of research has utilized Cutrona and Russell’s (1990) typology to examine the most prevalent types of support individuals tend to elicit within certain online contexts. Rains, et al.’s (2015) meta-analysis revealed that among these five types of social support, informational and emotional support were most frequently exchanged in health-related online communities, whereas tangible support was least evident compared to other types of support, indicating that online communities primarily served the function of advising and comforting (Rains, et al., 2015).
Although Rains, et al.’s (2015) meta-analysis included numerous studies that examined the topic of health-related online social support, a search for literature concerning online social support for expectant adolescents and adolescent mothers yielded few results. Most of the articles were focused on online interventions for adolescent parents, not naturally occurring support networks (e.g., Dunham, et al., 1998; Kauppi and Garg, 2009). However, Sherman and Greenfield (2013) did examine online social support and adolescent parenting in a naturally occurring online context. They analyzed four online forums focused on adolescent pregnancy and/or parenting. Their results showed although most posts were community oriented, the bulk of them did reveal messages related to emotional, informational, and tangible support.
Although Sherman and Greenfield’s (2013) study provided a glimpse into the world of naturally occurring online support groups for adolescent parents and/or adolescents who are expecting, there are still a variety of questions to be answered. For example, their study examined only three types of support: emotional, informational, and instrumental, but other types of support, particularly esteem support, are likely relevant to adolescent parents. Given that adolescent parenthood is stigmatized and adolescent parents report feeling like others treat them differently and look down upon them (SmithBattle and Leonard, 2012), it is possible adolescent parents reach out to their online support networks for a boost in self-esteem and to help them recover from the identity-related damage caused by stigmatization. More importantly, Cutrona and Suhr (1992) suggest esteem support is helpful for individuals experiencing both controllable and uncontrollable stressors, and adolescent pregnancy/adolescent parenthood is indeed associated with both controllable and uncontrollable stressors; thus, esteem support is likely important in the context of adolescent parenthood. Since Sherman and Greenfield (2013) limited the scope of their study to only three types of support, we do not yet know if adolescent parents/expectant adolescents utilize their online networks for seeking esteem support as well.
Additionally, Cutrona and Russell’s (1990) optimal matching model suggests tangible support is useful for those who experience controllable stressors, while network support is important for those experiencing uncontrollable stressors. Although adolescent parents and expectant adolescents likely experience stressors that are both controllable and uncontrollable, individuals may not turn to online support networks for tangible and network support because online networks tend to be geographically varied and may not easily provide the type of resources necessary for these two types of support. In an effort to have a more holistic understanding of the online social support exchange process for adolescent parents/expectant adolescents, the present study proposed the following research question:
RQ1: How frequently do members of adolescent pregnancy/adolescent parenting online support groups solicit a) informational; b) emotional; c) tangible; d) esteem; and, e) network support in their forum posts?
In addition to studying support solicited by individuals, it is also important to understand the type of support that is actually offered in online support communities. Importantly, previous research found that users tended to utilize online forums specifically so the type of support they sought could be matched by the support they received (Robinson and Turner, 2003; Walther and Boyd, 2002). However, just because posters solicited a particular type of support that matches the type of support they need does not mean a match is always provided. Vayreda and Antaki (2009) examined whether individuals’ posts in an online forum for bipolar disorder received responses that matched their desired support type. Their research showed many times when a member made a post announcing their bipolar disorder it was often met with unsolicited advice, indicating a mismatch and ineffective communication between what the posters wanted and what the responders provided.
Therefore, in order to know whether online support forums offer a viable alternative for adolescent parents/expectant adolescent mothers and interactive communication environment, researchers need to examine whether the support offered in these forums is actually helpful (i.e., whether the type of support offered matches the type of support solicited). To the authors’ knowledge, no previous research has examined the match between support solicited and support offered on online teen pregnancy/parenting Web sites; therefore, the second research question was proposed:
RQ2: How frequently are the types of support solicited by users of adolescent pregnancy/adolescent parenting forums matched by others’ responses to their posts?
The population of interest for this study was posts made within online forums dedicated to pregnant adolescents and/or adolescent parents. In order to sample the forums for this study, the first author conducted a Google search using the search terms “teen pregnancy forum” and “adolescent pregnancy forum.” This method is similar to that used by Sherman and Greenfield (2013). Additionally, since Google searches produce results that are listed based on popularity and relevance, the authors took into consideration only those boards in the top ten Web sites produced by the search. These two decisions were made because forums provided on the first page of a Google search would be most likely to be utilized by adolescents when searching for a forum to use. Out of those sites identified using the above criteria, the first author chose two for analysis. These two sites were chosen based on multiple factors, including: a) they were accessible by the general public, which meant one did not have to be a member of the forum in order to view posts; b) the posts were organized chronologically, which allowed random selection of posts for analysis, and assured the researchers they chose posts off of each site that occurred during a specific time period; and, c) the Web sites had a relatively equal amount of posts during the time period in question, which ensured differences in terms of frequency of user posts on each site would not cause a problem in any analyses. The importance of chronological order of the posts is discussed in more detail in the procedures section.
Forum one was a sub-community of a more general pregnancy Web site. This forum had 22,697 posts as of mid-October 2016. The Web site itself contained only forum posts and responses, with no other community features. The second forum was a community strictly dedicated to adolescent parenting, with a focus on adolescent motherhood and a smaller community for adolescent fatherhood. The site had 22,154 posts as of mid-October 2016. This second Web site contained many sections, including a dedicated forum for Q&A, as well as a space to post personal stories, and links to local pregnancy support services.
Prior to engaging in any sort of data collection, the authors obtained the Institutional Review Board’s exemption for research because we were studying publicly accessible documents that do not require registration. After choosing the Web sites for the study, the authors randomly sampled threads in order to answer the two research questions. The goal of the first research question was to examine what types of support members of online forums associated with adolescent pregnancy solicited in their posts; in order to answer this research question, the authors randomly sampled 150 posts made between 2015 and 2016 from each of the two sites. This timeline was chosen because the authors were interested in the most recent trends in social support messages at the time of data collection. Additionally, to determine the number of posts that should be analyzed, the authors conducted an a priori power analysis using G*Power. According to the analysis, in order to detect medium effects with alpha at .05 and 95 percent power, 145 threads should have been analyzed. Since 145 posts were the minimum, the authors decided to add five additional posts per site, resulting in the total of 150 per site. These posts were then coded based on the typology provided by Cutrona and Suhr (1992) as seeking a) informational; b) tangible; c) emotional; d) esteem; and/or, e) network support. The authors recognized that one individual post might contain requests for multiple types of support; thus, each message was coded as having each type of support either present or not present. For example, the authors examined every post requesting informational support (present or not present), then re-examined each post for requests of emotional support (present or not present) and so on and so forth for each of the five types of support.
Before actually engaging in the coding process, it was important to establish inter-coder reliability; therefore, once the individual posts were collected, 20 percent of the posts were randomly selected and the coding was performed as described above. This first 20 percent served as coder training. After the training was complete and all authors felt comfortable with the codebook, they then randomly selected another 20 percent of data that each author coded separately and calculated inter-coder reliability by utilizing Krippendorf’s (2011) alpha. All reliabilities for different types of support were 1.00, so the authors split up the remaining data and coded separately.
Procedures for the second research question were similar, except the authors randomly selected 150 posts from the previous three years. This change was made because it was necessary to obtain a more diverse sample of posts for this particular question. After engaging in analysis for the first research question and briefly examining the results, we noticed the forums seemed to contain more posts in previous years, with only a few posts per month within the last year. Therefore, we realized a longer time period might offer more insight into how the forums were utilized during a high traffic time. Coding procedures were similar to those used for RQ1, but instead of coding only the type(s) of support solicited in the initial posts, the authors also coded the type of support offered in first response to each post. The first response was chosen for analysis because in some instances only one response was given to support-seeking posts. Additionally, in many situations the first response would be the first message the original poster sees, making it the first possible instance of receiving support. These first messages could be particularly influential to those posters who were seeking time-sensitive advice.
Once both the original post and the first response were coded for the type(s) of support they contained (i.e., informational, emotional, tangible, esteem, and/or network), the authors then examined whether the response a) contained no support message; b) contained some support but did not fully match the type(s) solicited; c) fully matched the type(s) of support solicited; or, d) fully matched and provided extra support that was not solicited. This was determined based on how the type(s) of support offered in the response to the initial message compared to the type(s) of support solicited in the initial message. For example, if the original message asked for informational support, but the response contained informational support and emotional support, it was considered a full match that provided extra support that was not solicited. If a message had asked for informational support and received just informational support, this was considered a full match. If a post asked for informational support, but instead received tangible support, it was coded as containing some support, but not the type that was solicited. Finally, if the original post asked for informational support, but the response did not provide any type of support at all, this was considered a no support message.
It was important to differentiate between these four categories of support match because previous research has suggested examining social support messages for matches should be more nuanced than simply looking for a complete match or no match (Vayreda and Antaki, 2009). The authors then went through the same coder training process mentioned above, with all but one category having 1.0 Krippendorf’s alpha reliability (network support for the response was α = .88). The authors discussed their discrepancy until they agreed upon the proper way to code for network support and then coded the remaining data separately.
The typology provided by Cutrona and Suhr (1992) was utilized to code for types of support solicited and provided in posts. The types of support examined were: informational, tangible, emotional, network and esteem support. Informational support was defined as supportive messages that provide factual information, guidance and advice. Tangible aid referred to messages providing practical help (Cutrona and Suhr, 1992). Emotional support was defined as messages that provide comfort, empathy, warmth, and encouragement. Network support was defined as messages providing a sense of membership and belonging. Finally, esteem support was defined as messages providing compliments and reassurance that speak to the support seekers’ identity and/or self-esteem.
RQ1 was concerned with the types of support users solicited in their forum posts. Frequency counts showed informational support (n = 108, 67.33 percent) was the most common type of solicited support, followed by emotional support (n = 61, 40.67 percent), esteem support (n = 13, 8.67 percent), network support (n = 3, 2 percent) and no posts contained tangible support. A chi-square test of independence was conducted to see whether types of support solicited differed based on the forum in which the post was made. All tests were non-significant, indicating no significant differences in the types of support solicited based on the forum: χ2information (1, N = 150) = 2.42, p = .12; χ2tangible (1, N = 150) = 1.03, p = .31; χ2emotional (1, N = 150) = .40, p = .53; χ2network (1, N = 150) = 3.14, p = .08 χ2esteem (1, N = 150) = .67, p = .41. Table 1 shows examples of messages related to RQ1.
Table 1: Various types of social support solicited by members’ posts in adolescent parenting forums.
Note: All messages are displayed as written by the community members. Any grammatical/spelling mistakes errors are presented verbatim to preserve the authenticity of the post.
Support Example Informational Hi there So occasionally I would be 2 days late but i would start my period. But this is the longest time I havent started. I took two tests (on separate days) and both say negative. I have been working out excessively, I am on the track team. My start date was supposed to be the 17th but my track meet was on the 16th. Could all the training and running postpone that? Im usually on time with my period. Not to mention. I have been eating a lot more and there is white discharge. My back hurts but it always does. Im bloated. I havent been eating healthy in two weeks. The last time I had protected sex was last sunday. What could this be? Why is it like this? Is it the exercise and unhealthy eating habits? Tangible N/A Emotional Hey. Maybe somebody still know me? I am a Girl from Germany and got my little son 4 years ago. Today I tested positive again — I was really shocked — cause I don’t know how! I am scared ... Scared of my parents Scared of my boyfriend
Scared of the future :-/ Why?
Network Hey girls, Im 4 weeks pregnant, thought it would be nice having some pregnancy buddies! Im russian and my baby daddy is ethiopean, so thats a rainbow baby Im excited Anyone just found out shes pregnant too? Nice to meet you ladies Esteem I feel like I’m now starting to be more depressed, and it’s been a a year since I’ve had my son. I feel like I’m vanishing and this new..boring and sad girl is taking over. I just feel so blah and not myself. I feel SOOO sad and depressed and I just cry. I cry because I feel like I’m not good at what I’m supposed to be doing. I’m not the perfect mother and I just want my son to enjoy life like i didnt. I want to be the mother i didnt have. I feel bad now as a mother, as I look back at how I wish i was treated as a child. i feel selfish for wanting to be the best. I feel like I’m being lazy by not accomplishing everything. I’m not graduating on time. I don’t have imagination, I feel like a fat lazy bum when my son is happy playing and i just dont have the energy or want, to sit down there and play and laugh. Don’t get me wrong, i talk and sing and read and play with him all the time. But i feel like its my obligation to do that ALL THE TIME. Because I didn’t have any of that. I just want to have a healthy and happy baby boy. I feel like I’m going crazy and that no one feels the same. this is the first time I’ve gotten this off my chest.. :/
RQ2 concerned the types of support community members offered in response to a support solicitation message, and whether the type of support given matched the type of support sought. For each type of support, we examined whether the solicited support type was matched, unmatched, or overmatched (support was provided when not solicited). For informational support, about 84 percent of original messages (n = 89)  were matched with the solicited type of support while 16 percent (n = 17) were unmatched. There were a total of seven cases (15.9 percent, out of 44) of overmatching. See Table 2 for the cross-tabulation between the original post and its first response in terms of presence and absence of informational support.
A chi-square test of independence was also conducted to examine whether there was a significant difference between an original post and the reply post in terms of the presence or absence of informational support. The results were significant, χ2 (1, N = 150) = 62.50, p < .001, suggesting if the original post requested informational support, the reply post also provided informational support more frequently than expected by chance. The same goes for the absent case of original post; if the post did not ask for informational support, the reply most likely did not provide it. All in all, this analysis confirmed that in these two teenage pregnancy sites studied, informational support was matched between support seekers and providers.
Table 2: Cross-tabulation of informational support, original post with informational support, reply.
Note: Each subscript letter denotes a subset of informational support, Reply categories whose column proportions do not differ significantly from each other at the .05 level.
Informational support, reply Absent Present Total Informational support, original post Absent Count 37a 7b 44 Expected count 15.8 28.2 44.0 % within informational support, original post 84.1% 15.9% 100.0% % within informational support, reply 68.5% 7.3% 29.3% Present Count 17a 89b 106 Expected count 38.2 67.8 106.0 % within informational support, original post 16.0% 84.0% 100.0% % within informational support, reply 31.5% 92.7% 70.7% Total Count 54 96 150 Expected count 54.0 96.0 150.0 % within informational support, original post 36.0% 64.0% 100.0% % within informational support, reply 100.0% 100.0% 100.0%
For emotional support, only 32.8 percent (n = 20) of original messages were matched with the solicited type of support while 67.2 percent (n = 41) were unmatched. There were a total of 20 (22.5%, out of 89) cases of overmatching (see Table 3 for the cross-tabulation). Compared to the informational support case, emotional support was much less frequently matched, but more frequently overmatched as well. A chi-square test of independence was conducted to examine whether there was a statistically significant difference between an original post and the reply post in terms of the presence or absence of emotional support. The results were not significant, χ2 (1, N = 150) = 1.97, p = .16. Therefore, when an original post sought emotional support, many times, the first response did not provide it; also, when an original post did not ask for an emotional support, the reply post often provided it anyway, which indicates a poor matching of this particular support.
Table 3: Cross-tabulation of emotional support, original post with emotional support, reply.
Note: Each subscript letter denotes a subset of emotional support, Reply categories whose column proportions do not differ significantly from each other at the .05 level.
Emotional support, reply Absent Present Total Emotional support, original post Absent Count 69a 20a 89 Expected count 65.3 23.7 89.0 % within emotional support, original post 77.5% 22.5% 100.0% % within emotional support, reply 62.7% 50.0% 59.3% Present Count 41a 20a 61 Expected count 44.7 16.3 61.0 % within emotional support, original post 67.2% 32.8% 100.0% % within emotional support, reply 37.3% 50.0% 40.7% Total Count 110 40 150 Expected count 110.0 40.0 150.0 % within emotional support, original post 73.3% 26.7% 100.0% % within emotional support, reply 100.0% 100.0% 100.0%
Unfortunately, there was no original post that specifically asked for tangible support; thus, how many were matched by its solicited type could not be calculated. However, in reply messages, there were 24 cases (16 percent, out of 150) of providing tangible support when none was solicited, which belonged to the case of overmatching. Due to the absence of soliciting tangible support in the original post, a chi-square test could not be conducted.
For network support, nine out of 11 cases (81.8 percent) were matched with their solicitation. There were 28 cases (20.1 percent, out of 139) of overmatching that provided unsolicited network support (see Table 4 for the cross-tabulation). A chi-square test of independence was conducted to examine whether there was a significant difference between an original post and the reply in terms of the presence or absence of network support. The results were significant, χ2 (1, N = 150) = 20.86, p < .001, indicating that, if the original post requested network support, the reply post also provided network support more frequently than expected by chance (eight out of ten times). The same goes for the absent case of original post; if the post did not ask network support, the reply most likely did not provide it, either (also, eight out of ten times). In summary, this analysis confirmed that in these two teenage pregnancy Web sites, network support was matched between support seekers and providers.
Table 4: Cross-tabulation of network support, original post with network support, reply.
Note: Each subscript letter denotes a subset of network support, Reply categories whose column proportions do not differ significantly from each other at the .05 level.
Network support, reply Absent Present Total Network support, original post Absent Count 111a 28b 139 Expected count 104.7 34.3 139.0 % within network support, original post 79.9% 20.1% 100.0% % within network support, reply 98.2% 75.7% 92.7% Present Count 2a 9b 11 Expected count 8.3 2.7 11.0 % within network support, original post 18.2% 81.8% 100.0% % within network support, reply 1.8% 24.3% 7.3% Total Count 113 37 150 Expected count 113.0 37.0 150.0 % within network support, original post 75.3% 24.7% 100.0% % within network support, reply 100.0% 100.0% 100.0%
For esteem support, 60 percent (6 out of 10) were matched with their solicitation in the original post, and the rest (40 percent) were unmatched. There were eight cases (5.7 percent, out of 140) of overmatching (see Table 5 for the cross-tabulation results). Again, a chi-square test of independence was conducted to examine whether there was a statistically significant difference between an original post and the reply post in terms of the presence or absence of esteem support. The results were significant, χ2 (1, N = 150) = 32.50, p < .001, suggesting if the original post requested esteem support, the reply post also provided esteem support more frequently than expected by chance (six out of ten times). The same goes for the absent case of original post; if the post did not ask for esteem support, the reply most likely did not provide it, either (nine out of ten times). In summary, this analysis confirmed that in these two teenage pregnancy forums, esteem support exchange was close to being matched between support seekers and providers.
Table 5: Cross-tabulation of esteem support, original post with esteem support, reply.
Note: Each subscript letter denotes a subset of esteem support, Reply categories whose column proportions do not differ significantly from each other at the .05 level.
Esteem support, reply Absent Present Total Esteem support, original post Absent Count 132a 8b 140 Expected count 126.9 13.1 140.0 % within esteem support, original post 94.3% 5.7% 100.0% % within esteem support, reply 97.1% 57.1% 93.3% Present Count 4a 6b 10 Expected count 9.1 0.9 10.0 % within esteem support, original post 40.0% 60.0% 100.0% % within esteem support, reply 2.9% 42.9% 6.7% Total Count 136 14 150 Expected count 136.0 14.0 150.0 % within esteem support, original post 90.7% 9.3% 100.0% % within esteem support, reply 100.0% 100.0% 100.0%
The current study analyzed social support messages both solicited and provided in online adolescent pregnancy/parenting forums in order to bring a new context of online social support into light. The results of this study support Sherman and Greenfield’s (2013) findings, but also build upon them by examining the role of both support solicited and support provided in online communities for adolescent parents and adolescent parenting. Becoming an adolescent parent is associated with a variety of life changes, some of which can cause an identity crisis (e.g., stigmatization), sadness (e.g., loss of a friendship network), or confusion (e.g., not knowing what to expect after becoming a parent) (de Anda, et al., 1992; Sherman and Greenfield, 2013). Because support is so integral to the success of adolescent parents (Furstenberg and Hughes, 1995), but many adolescent parents and expectant adolescents may see a decrease in their face-to-face support networks after becoming pregnant (Sherman and Greenfield, 2013), it is important for scholars to understand the role online communities may play in their support exchange processs.
This research was based on the optimal matching model provided by Cutrona and Russell (1990), which infers the types of support that should be most relevant to individuals on how controllable stress might be, especially stress associated with a specific event. According to research about the stress associated with adolescent pregnancy, stressors expectant adolescents and adolescent parents experience are likely both controllable and uncontrollable (de Anda, et al., 1992); therefore, the authors expected to see all five types of support solicited in the online social support forums examined. However, we were unsure how often each type of support would be solicited, particularly in an online context. It is important to understand what type of support users seek online most often because this information gives researchers insight into the types of stress expectant adolescents and adolescent parents are dealing with in their face-to-face networks.
A content analysis completed for research question one revealed members of adolescent pregnancy support groups most frequently solicited informational support, followed by emotional, esteem, and network support, which mimics the results found by Rains, et al. (2015). Specifically, solicitation of informational support was present in 67.3 percent of posts, while solicitation of emotional support was present in 40.7 percent, esteem support in 8.7 percent, network support in only two percent, and no posts contained solicitation of tangible support. These results highlight the fact that expectant adolescents and adolescent parents seem to turn to online networks to manage the controllable aspect of their stress most often, although the frequency of emotional support also suggests they turn to online networks for help managing uncontrollable stressors as well.
Based on the likelihood that adolescent pregnancy/parenthood is associated with both controllable and uncontrollable stressors, network (uncontrollable) and tangible (controllable) support should have also been popular in users’ posts; however, solicitation of network support was only present in only two percent of posts in this study and solicitations for tangible support was not found in any of the posts. Perhaps this was because many users on the two sites were from all around the world. Soliciting and receiving network and tangible support becomes difficult as they may require support group members to meet in person.
The lack of esteem support solicitation was less intuitive, given the stigma historically associated with adolescent pregnancy (Wiemann, et al., 2005) and esteem support being associated with identity and self-esteem reinforcement. Additionally, esteem support is said to be helpful for any stressor regardless of its controllability, so why was esteem support not solicited frequently in this study? Perhaps pregnant adolescents did not think online support networks were an appropriate place to go to receive esteem support and instead turned to those close others they had face-to-face relationships with for this type of very personal support. Granovetter’s (1973) distinction between strong and weak ties help explain why users may have felt like the connections they made within their online networks have not been able to provide adequate esteem support. Granovetter (1973) says strong-tie networks are made up of those that individuals have known for a long time and that they have a close emotional connection with, whereas weak-tie networks (associated with online support sites) are most useful for receiving different perspectives and new information. If someone needs their esteem to be boosted, they may not feel like individuals in online networks are capable of providing it because they are part of their weak tie network and do not necessarily know them well. This is not to deny the possibility of online support networks providing esteem support; in fact, many members of the adolescent pregnancy Web sites we studied provided esteem support in their responses for the support seekers who may have felt uncertainty and low self-esteem due to the unexpected pregnancy. Our interpretation is from the support seekers’ perspective and why they may not have actively sought out esteem support from these Web sites.
According to Cutrona and Suhr (1992), stressors that are expected to last a long time tend to be associated with seeking informational support. Pregnancy certainly is a life-changing and long-lasting stressor especially for adolescents who did not expect this to happen; therefore, this might account for why informational support message exchanges were most common. Additionally, research in other health-related contexts has found situations that are complicated, controversial, and difficult to manage require more informational support in order to relay technical information and professional knowledge (Coulson, 2005). In fact, many of the posts examined for the first research question involved posters asking for advice from others and help recognizing pregnancy symptoms. For example, one user posted:
Anyone familiar with pregnancy symptoms? I haven’t gotten my period in almost 2 months and I've been experiencing a few things like nausea, bloating, minor headaches, discharge, stomach makes noises ... I took 2 home pregnancy test and both came back negative...? I took them later in the day because I hadn't known it’s most effective in the morning. I haven’t gone to doctors because I figured the home test were correct but I still haven’t gotten my period ... Pretty concerned.
This example is indicative of the types of questions posters asked, and highlights the use of online networks for health-related information. Pregnancy tends to be complicated and difficult to manage on one’s own, especially for a young parent, which might also explain why informational support was frequently sought. Finally, Rains, et al.’s (2015) meta-analysis found that informational support was the most sought out type of online social support in health-related forums. Since adolescent parenting/pregnancy could be considered a health-related issue (and many posts were concerned with health during pregnancy/finding out if one was pregnant), it makes sense that these results were consistent with research findings in other health-related contexts.
The second research question of our study was to understand the types of responses community members provided to individuals who solicited support online, and whether the type of provided support matched the type of solicited support. Results of the content analysis revealed members in online communities for adolescent pregnancy/parenting provided replies that fully matched the support solicited in the original message most frequently, followed by replies that contained extra support, partially matched the solicited support, and no support. It was apparent that a large portion of replies in adolescent pregnancy/parenting forums met or even exceeded the needs of community members who sought social support, suggesting a relatively effective communication environment, positive atmosphere, and utility of these online communities. These results also resonated with Sherman and Greenfield’s (2013) finding that the majority of comments on adolescent pregnancy/parenting forums were friendly and supportive.
However, there were noticeable differences in how each type of support was matched in these two forums. While informational and network support was matched with the solicited type of support (80 percent or above), emotional support was much less frequently matched (around 33 percent). Esteem support’s matching rate was somewhere between those two (i.e., 60 percent). Interestingly though, emotional support that was less frequently matched than other types of support, was often overmatched; when the original post did not ask for emotional support, the reply post often provided it anyway. There were many cases of overmatching for esteem support as well. For example, in response to a post asking for advice about managing her failing relationship with her unborn child’s father, one community member wrote:
Hi! Being pregnant at any age is a lot to handle, but especially being young, it can be difficult to know what to do. But there are people to help! First I would suggest telling your parents. I know that’s not easy, trust me I’ve been there. Secondly, I would strongly suggest getting to a doctor or pregnancy clinic because 4 months is pretty far along. You can find a pregnancy center near you on our sites home page. About the boyfriend, if he isn’t there for you now, give him time for the shock to wear off, if he still isn’t supportive then let him go. You only want supportive and positive people around you right now. Let us know if you need anything. Good luck with everything. Let us know how you’re doing.
In this response you can see that the community member addressed both emotional and esteem concerns for the original poster (e.g., “Being pregnant at any age is a lot to handle, but especially being young ...”, “I know that’s not easy, trust me I’ve been there”, etc.); however, the original post only asked for advice about communicating with the father. This situation exemplified how community members seemed to recognize the needs support seekers might have and respond based on those needs instead of focusing only on providing the exact support type solicited (for the above case, it would have been informational support).
These results highlight the idea that perhaps community members realized how important emotional and esteem support can be to adolescent parents and expectant adolescents and provided these types of support even when individuals did not solicit them in their original posts. In this context, these results provide evidence for how helpful online support networks could be to those dealing with adolescent pregnancies and/or parenting issues. Members of these communities seem to provide support that is geared towards helping people manage the negative emotions and stress associated with becoming a young parent, such as guilt due to stigmatization or sadness from the loss of a friend network, even if the young parents themselves do not realize they need that type of support, or are too afraid to ask for it.
Limitations and future directions
While the results of this study offer useful insights about online adolescent pregnancy forums and supportive communication exchanged there, the results are not generalizable beyond the two sites that were studied. Future research should therefore examine the social support process in a larger sample of online adolescent pregnancy forums. Additionally, the sample size of posts for this study was relatively small, 150 for research question one and 150 pairs (300 total messages) for research question two. Sherman and Greenfield (2013) examined 200 posts; however, they also examined four websites in total (50 posts per site). Therefore, the number of posts examined per forum website actually exceeded previous research on this topic. However, future researchers should seek to examine larger samples of support messages across a longer time frame to see whether there is any noticeable difference across distinctive sites over time.
Finally, future researchers should continue to examine how the type of support solicited might influence the match in support that is provided, and consider what aspects of the forums’ topics may contribute to this relationship. This study found significant differences between the types of support and how they were matched by responses, which seemed to reflect the experiences and needs of adolescent parents. More research unraveling this difference is necessary so scholars can understand these relationships more fully.
Research focused on support messages exchanged in online forums has flourished over the past decade in communication and technology literature; however, very few have examined the online support process of adolescent parents and/or adolescents who are expecting. This study supported previous research by suggesting adolescents seeking online support were most likely to solicit informational and emotional support. Additionally, the results of this study added to previous research by showing that solicitations of support were either fully matched by other users’ responses, or matched and given additional types of support that were not originally solicited. This indicates messages adolescents posted on pregnancy forums were generally met with sufficient support, or replied with more support than posters solicited. The findings beget the popular notion of online space being dangerous for adolescents full of inappropriate contents and predators. At least the two forums we studied showed mostly positive and supportive communication environments that is helpful to adolescent parents or those who are expecting. Finally, the amount of match that a response provided seemed to differ based on the type of support solicited in the original post. Informational and network support were more matched than esteem or emotional support. Exchange of social support is integral to the success of families with adolescent parents; thus, we recommend that future research continues to examine the support exchange process adolescents engage in various online environments.
About the authors
Eryn Bostwick, Ph.D., is an Assistant College Lecturer in the School of Communication at Cleveland State University. Her research specializes in understanding interpersonal and family communication processes, and she has frequently examined the role that technology plays in managing relationships and relational issues. Her research has been published in the Journal of Family Communication, Journal of Social and Personal Relationships, Studies in Media and Communication, and Communication Reports.
Direct comments to: e [dot] n [dot] bostwick [at] csuohio [dot] edu
Danni Liao is a doctoral student in the Department of Communication at University of Illinois at Urbana-Champaign. Danni’s research interests lie at the intersection of health, interpersonal, and intercultural communication. Her research focuses on cross-cultural care, examining international medical graduates’ influence on patient evaluation and their communication competence.
E-mail: dannil3 [at] illinois [dot] edu
Sun Kyong (Sunny) Lee, Ph.D., is an Associate Professor in the Department of Communication at the University of Oklahoma. Her research examines socio-cultural aspects of communication technology uses and organizational communication networks of ethnic and religious immigrant communities. Her research on social support networks of a Korean immigrant community can be found in Communication Research, Journal of Computer-Mediated Communication, and International Journal of Intercultural Relations. Other studies have been published in First Monday, Computers in Human Behavior, and Management Communication Quarterly.
E-mail: sunklee [at] ou [dot] edu
The authors would like to thank the anonymous reviews for their feedback on previous drafts of this manuscript. Their comments and guidance have helped to clarify issues and improve the quality of this paper.
1. The percentage was calculated based on the cross tabulation between the original post and the reply post in their presence and absence of a particular type of support.
B.R. Bierman and R. Streett, 1982. “Adolescent girls as mothers: Problems in parenting,” In: I.R. Stuart and C.F. Wells (editors). Pregnancy in adolescents: Needs, problems, and management. New York: Van Nostrand Reinhold, pp. 407–426.
D.O. Braithwaite, V.R. Waldron, and J. Finn, 1999. “Communication of social support in computer-mediated groups for people with disabilities,” Health Communication, volume 11, number 2, pp. 123–151.
doi: https://doi.org/10.1207/s15327027hc1102_2, accessed 12 August 2019.
L. Bunting and C. McAuley, 2004. “Research review: Teenage pregnancy and motherhood: The contribution of support,” Child & Family Social Work, volume 9, number 2, pp. 207–215.
doi: https://doi.org/10.1111/j.1365-2206.2004.00328.x, accessed 12 August 2019.
D. Clemmens, 2001. “The relationship between social support and adolescent mothers interactions with their infants: A meta-analysis,” Journal of Obstetric, Gynecologic & Neonatal Nursing, volume 30, number 4, pp. 410–420.
doi: https://doi.org/10.1111/j.1552-6909.2001.tb01560.x, accessed 12 August 2019.
N.S. Coulson, 2005. “Receiving social support online: An analysis of a computer-mediated support group for individuals living with irritable bowel syndrome,” CyberPsychology & Behavior, volume 8, number 6, pp. 580–584.
doi: https://doi.org/10.1089/cpb.2005.8.580, accessed 12 August 2019.
C.E. Cutrona and J.A. Suhr, 1992. “Controllability of stressful events and satisfaction with spouse support behaviors,” Communication Research, volume 19, number 2, pp. 154–174.
doi: https://doi.org/10.1177/009365092019002002, accessed 12 August 2019.
C.E. Cutrona and D.W. Russell, 1990. “Type of social support and specific stress: Toward a theory of optimal matching,” In: B.R. Sarason, I.G. Sarason, and G.R. Pierce (editors). Social support: An interactional view. New York: Wiley, pp. 319–366.
D. de Anda, P. Darroch, M. Davidson, J. Gilly, M. Javidi, S. Jefford, R. Komorowski, and A. Morejon-Schrobsdorf, 1992. “Stress and coping among pregnant adolescents,” Journal of Adolescent Research, volume 7, number 1, pp. 94–109.
doi: https://doi.org/10.1177/074355489271007, accessed 12 August 2019.
P.J. Dunham, A. Hurshman, E. Litwin, J. Gusella, C. Ellsworth, and P.W.D. Dodd, 1998. “Computer-mediated social support: Single young mothers as a model system,” American Journal of Community Psychology, volume 26, number 2, pp. 281–306.
doi: https://doi.org/10.1023/A:1022132720104, accessed 12 August 2019.
F.F. Furstenberg and M.E. Hughes, 1995. “Social capital and successful development among at-risk youth,” Journal of Marriage and the Family, volume 57, number 3, pp. 580–592.
doi: https://doi.org/10.2307/353914, accessed 12 August 2019.
M.S. Granovetter, 1973. “The strength of weak ties,” American Journal of Sociology, volume 78, number 6, pp. 1,360–1,380.
doi: https://doi.org/10.1086/225469, accessed 12 August 2019.
C. Kauppi and R. Garg, 2008. “Development of cybermoms: A computer-mediated peer support group to address the needs of young mothers,” Currents: Scholarship in the Human Services, volume 7, number 2, at https://journalhosting.ucalgary.ca/index.php/currents/article/view/15894, accessed 12 August 2019.
D.M. Keating, 2013. “Spirituality and support: A descriptive analysis of online social support for depression,” Journal of Religion and Health, volume 52, number 3, pp. 1,014–1,028.
doi: https://doi.org/10.1007/s10943-012-9577-x, accessed 12 August 2019.
T.H.M. Kim, J.A. Connolly, and H. Tamim, 2014. “The effect of social support around pregnancy on postpartum depression among Canadian teen mothers and adult mothers in the maternity experiences survey,” BMC Pregnancy and Childbirth, volume 14, article number 162, at https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/1471-2393-14-162, accessed 12 August 2019.
doi: https://doi.org/10.1186/1471-2393-14-162, accessed 12 August 2019.
S.S. Loane and S. D’Alessandro, 2013. “Communication that changes lives: Social support within an online health community for ALS,” Communication Quarterly, volume 61, number 2, pp. 236–251.
doi: https://doi.org/10.1080/01463373.2012.752397, accessed 12 August 2019.
T.J. Matthews and B.E. Hamilton, 2019. “Total fertility rates by state and race and Hispanic origin: United States, 2017,” National Vital Statistics Report, volume 68, number 1 (10 January), pp. 1–10, and at https://www.cdc.gov/nchs/data/nvsr/nvsr68/nvsr68_01-508.pdf, accessed 12 August 2019.
S. Nolan, J. Hendricks, and A. Towell, 2015. “Social networking sites (SNS): Exploring their uses and associated value for adolescent mothers in Western Australia in terms of social support provision and building social capital,” Midwifery, volume 31, number 9, pp. 912–919.
doi: https://doi.org/10.1016/j.midw.2015.05.002, accessed 12 August 2019.
S. Nolan, J. Hendricks, S. Ferguson, and A. Towell, 2017. “Social networking site (SNS) use by adolescent mothers: Can social support and social capital be enhanced by online networks? — A structured review of the literature,” Midwifery, volume 48, pp. 24–31.
doi: https://doi.org/10.1016/j.midw.2017.03.002, accessed 12 August 2019.
Pew Research Center, 2019. “Conservatives: Religious composition of conservatives,&redquo; at http://www.pewforum.org/religious-landscape-study/political-ideology/conservative/, accessed 12 August 2019.
Pew Research Center, 2014. “Party affiliation by state,” at http://www.pewforum.org/religious-landscape-study/compare/party-affiliation/by/state/, accessed 12 August 2019.
J.L. Pinzon and V.F. Jones, 2012. “Care of adolescent parents and their children,” Pediatrics, volume 130, number 6, pp. 1,743–1,756.
doi: https://doi.org/10.1542/peds.2012-2879, accessed 12 August 2019.
S.A. Rains, E.B. Peterson, and K.B. Wright, 2015. “Communicating social support in computer-mediated contexts: A meta-analytic review of content analyses examining support messages shared among individuals coping with illness,” Communication Monographs, volume 82, number 4, pp. 403–430.
doi: https://doi.org/10.1080/03637751.2015.1019530, accessed 12 August 2019.
J.D. Robinson and J. Turner, 2003. “Impersonal, interpersonal, and hyperpersonal social support: Cancer and older adults,” Health Communication, volume 15, number 2, pp. 227–234.
doi: https://doi.org/10.1207/S15327027HC1502_10, accessed 12 August 2019.
M. Schotanus-Dijkstra, P. Havinga, W. van Ballegooijen, L. Delfosse, J. Mokkenstorm, and B. Boon, 2014. “What do the bereaved by suicide communicate in online support groups?” Crisis, volume 35, number 1, pp. 27–35.
doi: https://doi.org/10.1027/0227-5910/a000225, accessed 12 August 2019.
A. Schrag and A. Schmidt-Tieszen, 2014. “Social support networks of single young mothers,” Child & Adolescent Social Work Journal, volume 31, number 4, pp. 315–327.
doi: http://dx.doi.org/10.1007/s10560-013-0324-2, accessed 12 August 2019.
L.E. Sherman and P.M. Greenfield, 2013. “Forging friendship, soliciting support: A mixed-method examination of message boards for pregnant teens and teen mothers,” Computers in Human Behavior, volume 29, number 1, pp. 75–85.
doi: https://doi.org/10.1016/j.chb.2012.07.018, accessed 12 August 2019.
C.E. Smith, K. Fernengel, C. Holcroft, K. Gerald, and L. Marien,1994. “Meta-analysis of the associations between social support and health outcomes,” Annals of Behavioral Medicine, volume 16, number 4, pp. 352–362.
L. SmithBattle and V. Leonard, 2012. “Inequities compounded: Explaining variations in the transition to adulthood for teen mothers’ offspring,” Journal of Family Nursing, volume 18, number 3, pp. 409–431.
doi: https://doi.org/10.1177/1074840712443871, accessed 12 August 2019.
M. Tanis, 2008. “What makes the Internet a place to seek social support?” In: E.A. Konijin, S. Utz, M. Tanis, and S.B. Barnes (editors). Mediated interpersonal communication. New York: Routledge, pp. 290–308.
A. Vayreda and C. Antaki, 2009. “Social support and unsolicited advice in a bipolar disorder online forum,” Qualitative Health Research, volume 19, number 7, pp. 931–942.
doi: https://doi.org/10.1177/1049732309338952, accessed 12 August 2019.
J.B. Walther and S. Boyd, 2002. “Attraction to computer-mediated social support,” In: C.A. Lin and D. Atkin (editors). Communication technology and society: Audience adoption and uses. Cresskill, N.J.: Hampton Press, pp. 153–188.
C.M. Wiemann, V.I. Rickert, A.B. Berenson, and R.J. Volk, 2005. “Are pregnant adolescents stigmatized by pregnancy?” Journal of Adolescent Health, volume 36, number 4, pp. 352.e1–352.e7.
doi: https://doi.org/10.1016/j.jadohealth.2004.06.006, accessed 12 August 2019.
K.B. Wright and S. Bell, 2003. “Health-related support groups on the Internet: Linking empirical findings to social support and computer-mediated communication theory,” Journal of Health Psychology, volume 8, number 1, pp. 39–54.
doi: https://doi.org/10.1177/1359105303008001429, accessed 12 August 2019.
K.B. Wright and C.H. Miller, 2010. “A measure of weak-tie/strong-tie support network preference,” Communication Monographs, volume 77, number 4, pp. 500–517.
doi: https://doi.org/10.1080/03637751.2010.502538, accessed 12 August 2019.
K.B. Wright, S. Rains, and J. Banas, 2010. “Weak-tie support network preference and perceived life stress among participants in health-related, computer-mediated support groups,” Journal of Computer-Mediated Communication, volume 15, number 4, pp. 606–624.
doi: https://doi.org/10.1080/03637751.2010.502538, accessed 12 August 2019.
K.B. Wright, L. Sparks, and H.D. O’Hair, 2013. Health communication in the 21st century. Second edition. Malden, Mass.: Wiley-Blackwell.
Received 26 March 2019; revised 5 June 2019; revised 30 July 2019; accepted 30 July 2019.
“Could I be pregnant? A study of online adolescent pregnancy forums for social support” by Eryn N. Bostwick, Danni Liao, and Sun Kyong Lee is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Could I be pregnant? A study of online adolescent pregnancy forums for social support
by Eryn N. Bostwick, Danni Liao, and Sun Kyong Lee.
First Monday, Volume 24, Number 9 - 2 September 2019