Accepted by previous editor Maria Bakardjieva
Social Dynamics of Virtual Groups
Cultivating Social Resources on Social Network Sites: Facebook Relationship Maintenance Behaviors and Their Role in Social Capital Processes†
Version of Record online: 19 MAR 2014
© 2014 International Communication Association
Journal of Computer-Mediated Communication
Volume 19, Issue 4, pages 855–870, July 2014
How to Cite
Ellison, N. B., Vitak, J., Gray, R. and Lampe, C. (2014), Cultivating Social Resources on Social Network Sites: Facebook Relationship Maintenance Behaviors and Their Role in Social Capital Processes. Journal of Computer-Mediated Communication, 19: 855–870. doi: 10.1111/jcc4.12078
- Issue online: 18 JUL 2014
- Version of Record online: 19 MAR 2014
- Manuscript Accepted: 25 JAN 2013
- Manuscript Revised: 3 JAN 2013
- Manuscript Received: 15 AUG 2011
- social network sites;
- social capital;
- social grooming;
- relationship maintenance
- Top of page
- Social Capital
This study explores the relationship between perceived bridging social capital and specific Facebook-enabled communication behaviors using survey data from a sample of U.S. adults (N=614). We explore the role of a specific set of Facebook behaviors that support relationship maintenance and assess the extent to which demographic variables, time on site, total and “actual” Facebook Friends, and this new measure (Facebook Relationship Maintenance Behaviors) predict bridging social capital. Drawing upon scholarship on social capital and relationship maintenance, we discuss the role of social grooming and attention-signaling activities in shaping perceived access to resources in one's network as measured by bridging social capital.
Social network sites (SNSs) are web-based communication platforms that support socially relevant interactions among contacts (i.e., “Friends”) on the site (Ellison, Steinfield, & Lampe, 2011). Previous research has documented a relationship between use of the SNS Facebook and increased levels of social capital (Burke, Kraut, & Marlow, 2011; Burke, Marlow, & Lento, 2010; Ellison, Steinfield, & Lampe, 2007, 2011; Steinfield, Ellison, & Lampe, 2008; Valenzuela, Park, & Kee, 2009), a form of capital that describes resources embedded in social relationships and interactions within a network (Lin, 2001). Facebook may be especially well-suited for accruing bridging social capital, which speaks to the benefits associated with weaker, more heterogeneous social ties such as novel information and broadened worldviews (Burke et al., 2011; Ellison et al., 2007, 2011), in part because the site enables users to create “social supernets” of hundreds of social connections (Donath, 2007).
Previous research has examined the relationship between social capital and various Facebook activities, evolving from studies on overall measures of Facebook use (Ellison et al., 2007) to more granular metrics such as perceptions of “actual friends” on the site (Ellison et al., 2011), and engaging in directed communication with Friends (Burke et al., 2011). In this paper, we extend previous work to focus on the relationship between bridging social capital and users' engagement in a set of specific communication behaviors we call Facebook Relationship Maintenance Behaviors (FRMB). This measure assesses users' intention to respond to requests or needs articulated by members of their network and to engage in relationship maintenance activities such as wishing a Facebook Friend “Happy Birthday.” Although past work has identified a link between generic Facebook use and bridging social capital, here we unpack specific activities to better explicate the mechanism through which Facebook-enabled behaviors relate to social capital accrual. Specifically, we explore how individuals are using the site to maintain social relationships and how these activities relate to perceptions of social capital using survey data collected from a sample of nonacademic staff at a large Midwestern U.S. university (N=614). Our analysis draws upon concepts from social capital and relationship maintenance scholarship and considers the sociotechnical affordances of SNSs.
- Top of page
- Social Capital
Social capital (Bourdieu, 1986; Coleman, 1988) describes the ability of individuals or groups to access resources embedded in their social network. Social capital can be converted to other forms of capital such as favors (human capital) or new information (intellectual capital) (Resnick, 2001). Reciprocity is a key construct in many treatments of social capital, both at a generalized or individual level. Lin (2001) defines social capital as “investment in social relations with expected returns in the marketplace” (p. 19). In this framing, social capital is created through social interactions and the expectations of future social resources they engender.
Social capital is often separated into “bridging” and “bonding” (Putnam, 2000; Williams, 2006); these categories describe resources embedded in different types of relationships. Ties that connect different clusters within a network, often called “bridging” ties, help propagate novel information across those groups (Burt, 1992). As explicated below, weaker ties (such as a friend of a friend) are more likely to be bridging ties and thus provide access to novel information (Granovetter, 1973) and diverse perspectives (associated with bridging social capital). Stronger ties, on the other hand, are characterized by multiple iterative interactions and higher levels of trust, support, and intimacy; these ties typically provide access to the more substantive forms of capital conversion associated with bonding social capital (e.g., a financial loan). Facebook networks contain both strong and weak ties (Bakshy, Rosenn, Marlow & Adamic, 2012). Because individuals often use multiple channels to communicate with strong ties and fewer channels to interact with weaker ties (Haythornthwaite, 2005), the focus of this study is bridging social capital because we are interested in how Facebook enables greater access to resources held by weaker ties, who—unlike strong ties—may not be available through other channels.
Bridging Social Capital
One's ability to access useful informational resources, a key component of bridging social capital, is related to the composition of one's network, one's position in this network, and one's communicative and analytic skillset. Bridging ties—dyadic relationships spanning two clusters—allow novel information to spread across a network more effectively by creating information pathways that close “structural holes” between two otherwise unconnected groups (Burt, 1992). Empirical support for this notion is provided by Granovetter (1974), who found weak (bridging) ties were more likely than strong ties to provide useful employment information.
Facebook's technical and social affordances lower the cost of maintaining weaker ties (allowing users to maintain broader and more diverse social networks) and facilitate resource exchanges, thus enabling users to cultivate bridging social capital (Ellison, Lampe, Steinfield, & Vitak, 2010). Tong and Walther (2011) note that SNSs reduce the cost of relationship maintenance, presumably enabling more frequent message exchanges with a wider set of contacts. On SNSs , messages can either be directed to a specific individual, subset of individuals, or distributed to one's entire network. Facebook status updates allow commenters to interact with one another, and thus can facilitate interaction among “Friends of Friends” who are more likely to be sources of novel information (Granovetter, 1973).
The technical process of “Friending” another user on Facebook requires little effort and, once that user has been added, no action is required to keep the connection “live” unless it is purposely severed. Most Facebook networks are quite large: the average U.S. adult Facebook user reports 229 Friends (Hampton, Goulet, Rainie, & Purcell, 2011) and, on average, users actively engage with a very small percentage of their network while they “maintain relationships” with 2 to 2.5 times more Friends through passive strategies such as visiting profiles and clicking posted links (Facebook Data Team, 2009). Given these interaction patterns, recent scholarship has attempted to differentiate among different kinds of network ties and their relationship to social capital perceptions. Ellison et al. (2011) asked undergraduate Facebook users to report how many total Facebook Friends they had, as well as the number of these contacts they considered “actual” friends. On average, they reported about 25% were actual friends, suggesting that Facebook Friend networks include individuals of varying degrees of perceived relational closeness and that users are able to distinguish between them. Ellison and colleagues (2011) found that while the number of actual friends positively predicted bridging social capital, the total number of Facebook Friends did not, suggesting some relationships are more likely than others to provide access to the resources associated with bridging social capital.
Studies like Ellison et al.'s (2011), which employ college student samples, typically use social capital measures that assess perceptions of resources available in an offline, localized setting such as one's current university. As the site's population has expanded beyond college students, Friend networks now include individuals from many dimensions of a user's life such as family, friends, coworkers, and neighbors. Individuals use Facebook to fulfill a variety of needs (Joinson, 2008; Smock, Ellison, Lampe, & Wohn, 2011), but the most frequently cited reason for membership is maintaining offline relationships (Lenhart, 2009). In this study, we employ measures of social capital that reflect resources associated with the full composition of users' Facebook Friends network, which we call “Facebook-specific bridging social capital,” as well as their general social network (i.e., all connections, including online and offline connections), which we label “general bridging social capital.” These networks should overlap; Hampton et al. (2011) found Facebook users had, on average, “Friended” 48% of their total network. Taking this into consideration, along with Ellison et al.'s (2011) findings regarding “actual” friends, we propose hypotheses for both Facebook-specific social capital and general social capital:
H1: The greater the number of actual friends on Facebook, the greater users' reported (a) Facebook-specific and (b) general bridging social capital.
Relationship Maintenance via Social Network Sites
Lin's (2001) framing of social capital as an “investment in social relations” highlights the importance of actively contributing to and maintaining personal relationships. We consider relationship maintenance activities to include the processes through which individuals keep a relationship in existence, in a specified state, in a satisfactory condition, or in repair (Dindia & Canary, 1993). Duck (1991) describes the work involved in “relationshipping” (the work of relationship maintenance) and highlights the importance of having the skills to maintain and repair social bonds, pointing out that relationships “need overhaul, maintenance and servicing just as any other dynamic structure does” (p. 27). Many theories of relationship maintenance focus on dyadic, offline interactions, although recent research on relationship maintenance has extended study of these strategies into mediated environments (e.g., Rabby, 2007).
Facebook and other SNSs provide an ideal platform for relationship maintenance interactions to occur quickly, with multiple others, and with low transaction costs because they enable wide dissemination of messages and foster participation, feedback, and interaction through various communication channels (Tong & Walther, 2011). Relationship maintenance behaviors in the SNS context can be understood as a form of “social grooming” (Donath, 2007)—activities that signal attention, build trust, and create expectations of reciprocal attention. Grooming has been identified as a meaningful activity among nonhuman primates, functioning to enhance hygiene and general well-being of apes as well as to build social bonds between them; humans engage in equivalent trust-building and relationship-nurturing activities through language (Dunbar, 1996). On SNSs, social grooming occurs via interactions between connected members, with the content, frequency, and length of messages serving as signals of the strength and context of the relationship; for example, Donath (2007) argues that the “cost in time is a signal of the resources one is willing to commit to this relationship” (p. 238). Tufekci (2008) operationalized social grooming as the desire to stay in touch with friends and curiosity about others. She found that, when compared with nonusers, SNS users kept in touch with a higher number of people on a weekly basis. Thelwall and Wilkinson (2010) explored social grooming in a SNS context, asserting that public comments such as “happy birthday” posts generally served to initiate and maintain contact between Friends. Similarly, researchers report relationship maintenance behaviors, such as using SNSs to “keep in touch with an old friend” (Lampe, Ellison, & Steinfield, 2006, p. 169) or “maintaining relationships with people you may not get to see very often” (Joinson, 2008) are some of the most frequent SNS activities.
An important component of relationship maintenance and social grooming is signaling attention to others. In mediated environments, attention cues may need to be more explicitly signaled because nonverbal signals are unavailable (Walther, 1992). A delay or lack of response to an e-mail, for example, may represent intentional neglect of a message or may merely be due to a technical glitch; ultimately, communicators must interpret silence with the information they have available (Kalman & Rafaeli, 2011). Interpretation of nonresponses in a SNS environment is similarly ambiguous. A Facebook status update without comments or “likes” may signal lack of interest by one's network, or the update may not have been widely displayed or noticed—potentially due to the sheer amount of information being pushed through the News Feed or the mechanics of the site's algorithm for displaying content. Given the lack of visible signals of attention, SNS users must respond in a manner that leaves an observable marker of attention as a way of cultivating, or grooming, their connections and thus increasing access to the resources they represent. On a SNS, explicitly responding to another user via activities that leave visible traces, such as commenting or clicking the “Like” button, is the most reliable way to indicate one has seen and attended to any individual piece of content on the site.
These activities are presented in the News Feed, an aggregated collection of contributions from one's network which uses the “EdgeRank” algorithm to select News Feed content based on factors such as how recently the post was created and past interaction patterns with the poster (Newman, 2011). Although the News Feed can be helpful for managing large, heterogeneous networks of “actual” and dormant Friends, most users are unlikely to see all their Friends' content under the default setting. Users can manually filter their feed by “hiding” specific Friends or adjusting their News Feed preferences, thus minimizing the cognitive effort needed to monitor all the updates, links, and photos produced by their network. Regardless of whether they are aware of EdgeRank and how it works, users train Facebook's filters by interacting with specific content they find valuable or relevant because these activities are used to determine which content is presented in the Feed. Of course, activities such as commenting on a Friend's post, which Tong and Walther (2011) identify as relationship maintenance signals, represent an investment in a given relationship (Lin, 2001) and thus help develop social capital.
Previous research has shown that publicly viewable forms of interaction on the site, such as commenting on a Friend's status update, are significantly related to perceptions of social capital. For example, Burke et al. (2011) collected server-level data to examine the role of three behaviors in predicting bridging social capital: passive consumption of information through the News Feed, broadcasting information through public posts, and engaging in directed communication with another user. Only inbound directed communication significantly predicted bridging social capital. Burke et al.'s (2011) measure of directed communication consisted of a frequency count of comments, Wall posts, “likes,” messages, and tags received by a participant. In contrast, our focus is on the content of these exchanges and the degree to which they constitute signals of attention, investment in one's network, and social capital contributions such as offering information or conveying social support. Our measure also includes an item about publicly posting birthday wishes to a Friend's Wall, which we believe signals attention to the recipient and performs a social grooming function. In summary, engaging in these relationship maintenance and attention-signaling behaviors may be linked to expectations of access to social resources, both from one's Facebook network and from one's social network more generally.
H2: The more users engage in Facebook Relationship Maintenance Behaviors, the greater their reported (a) Facebook-specific and (b) general bridging social capital.
Next, we consider the relationship between these relationship maintenance behaviors and users' network composition. Research on network structure (Burt, 1992; Granovetter, 1973) found that weaker, bridging ties are more likely to be sources of nonredundant information, while Ellison et al. (2011) found “actual” friends were more predictive of social capital than total network size. By exploring the interaction between actual friends and relationship maintenance behaviors, we may be able to gain insight into the nature of actual friends and the kinds of ties that are the recipients of these behaviors. Understanding how these variables interact and their effect on bridging social capital would complement insights provided by Burke et al. (2011), who highlight the role of directed communication on the site. Given the lack of specific literature on this topic, we pose the following research question:
RQ: How do number of actual friends and Facebook Relationship Maintenance Behaviors interact in predicting bridging social capital?
- Top of page
- Social Capital
Participants and Procedure
Data were collected during February and March of 2011. A random sample of 2149 nonfaculty staff at a large Midwestern U.S. university were invited, via e-mail, to participate in a study about their use of online communication tools; 614 usable responses were received for a response rate of 28.9%. The average participant was female (66%), 45 years old (SD=11.0), and a college graduate (40.1% had a bachelor's degree, 32.1% did postgraduate work). Twenty-two percent of participants (N=134) did not use Facebook. Comparing this dataset to a U.S. dataset from the same time period (Madden & Zickhur, 2011), the national sample was less educated (18.8% had a bachelor's degree, 13.9% did postgraduate work), older (M=53, SD=20.2), less likely to be female (57%), and less likely to be a SNS user (50%) than our sample.
Unless otherwise noted, all scale-based variables were measured using a Likert-type response scale ranging from 1=Strongly Disagree to 5=Strongly Agree.
Bridging social capital.
In this study, we adapted Williams' (2006) 10-item bridging scale—which captures aspects such as contact with diverse others, feeling part of a broader group, and engaging in reciprocal behaviors with one's community—to measure individuals' perceptions of their resources within two types of networks. Research from the Pew Internet Project (Hampton et al., 2011) points to benefits of Facebook use related to access to various resources, such as the diversity of their network, level of political engagement, and the amount of support received from their network. These effects may be associated in part with the ease and convenience of maintaining relationships through the site. Therefore, we chose to capture how participants' perceptions may vary when thinking of their entire social network as well as a separate measure of bridging social capital which focuses specifically on the resources associated with one's Facebook Friends. First, General Bridging Social Capital (Cronbach's α=.88; M=3.74, SD=0.58) reflects broad-based perceptions of social capital experienced through interactions with one's entire social network. These items used the phrase “social network” (replacing Williams' “online/offline”) and included the following directions: “For the next series of questions, think about your entire social network, including relatives, close and distant friends, coworkers and acquaintances.” Second, Facebook Bridging Social Capital (Cronbach's α=.93, M=3.33, SD=0.76) captures participants' perceptions of bridging benefits associated with interactions with their Facebook Friends, and used “in my Facebook network” in all items rather than “online/offline.” Participants were instructed to only think about Facebook Friends when responding to the 10 items. It is important to note that the items did not specify interactions on the site, but rather used phrasing such as, “Interacting with people in my Facebook network makes me want to try new things.” Item wording, means, and standard deviations for both General Bridging Social Capital and Facebook Bridging Social Capital are presented in Appendix A, online at: http://www-personal.umich.edu/~enicole/socialcapitalscales.html.
Facebook Relationship Maintenance Behaviors
(FRMB; Cronbach's α=.90, M=3.55, SD=0.83). FRMB captures individuals' likelihood to engage in directed communication behaviors that represent relationship maintenance activities and signal attention and investment in one's contacts on the system through small but meaningful actions. To construct this measure, we included a series of items in our instrument assessing Facebook users' engagement in interactive communication with their Facebook Friends, including measures of behaviors (e.g., “When I see someone asking for advice on Facebook, I try to respond”), frequency (e.g., “How often do you respond to questions from your Facebook friends?”), and motivations (e.g., “I answer questions on Facebook because I like helping other people”). We then conducted exploratory factor analysis on the 10 items using principal components analysis with Varimax rotation. After removing cross-loading items, the remaining items loaded cleanly onto one, six-item factor; however, one item (“I get respect from others when I answer questions”) was removed from the final scale because it slightly lowered the alpha and assessed motivation as opposed to behavior. Confirmatory factor analysis confirmed the model including the five items was a good fit to the data (χ2=7.91, p >.05; RMSEA=.05; CFI=1.00; GFI=.99). Items, means, and standard deviations for the measure are presented in Table 1. A correlation matrix is presented in Appendix B, online at: http://www-personal.umich.edu/~enicole/correlations.html.
|When I see a friend or acquaintance sharing good news on Facebook, I try to respond.||3.70||.94|
|When I see a friend or acquaintance sharing bad news on Facebook, I try to respond.||3.46||.99|
|When I see someone asking for advice on Facebook, I try to respond.||3.27||.99|
|When a Facebook friend has a birthday, I try to post something on their wall.||3.71||.1.06|
|When I see someone asking a question on Facebook that I know the answer to, I try to respond.||3.64||.93|
|Full Scale (α=.901):||3.55||.83|
Following creation of the scale, we conducted a series of tests to look for differences in engagement in Facebook Relationship Maintenance Behaviors across gender, age, and site use. An independent samples t-test revealed that women (M=3.69, SD=0.75) were more likely to perform these behaviors than men (M=3.28, SD=0.93), t(456)=−5.156, p<.001. FRMB also varied by age, as indicated by an analysis of variance, F(453)=4.211, p<.05. According to Tukey's HSD post hoc test, those aged 36 to 50 years of age (M=3.66, SD=0.77) were significantly more likely to engage in FRMB than those aged 51 and older (M=3.39, SD=0.95); however, neither of these age groups differed significantly in their performance of FRMB than younger participants aged 20 to 35 (M=3.59, SD=0.78). A second ANOVA revealed that FRMB varied significantly based on the amount of time spent on Facebook, F(459)=50.276, p<.001. Tukey's post hoc HSD test demonstrated those who spent 0 to 5 minutes on Facebook per day (M=2.93, SD =0.94) and who spent 5 to 15 minutes per day (M=3.58, SD=0.61)—approximately half of our sample—were significantly less likely to perform Facebook Relationship Maintenance Behaviors than those who spent approximately 15 to 45 minutes per day on Facebook (M=3.82, SD=0.62) and over 45 minutes per day (M=4.00, SD=0.62).
Facebook use variables.
In order to test the relationship between number of actual friends on Facebook and our two measures of bridging social capital, we included the item employed by Ellison et al. (2011): “Approximately how many of your TOTAL Facebook friends do you consider actual friends?” (open-ended). In addition, we included a measure of total Facebook Friends as a control variable, using the same wording as Ellison et al. (“Approximately how many TOTAL Facebook Friends do you have?”). Participants reported a mean of 207 total Friends (median=120, SD=288.16) and 76 “actual” friends (median=40, SD=101.07) on the site. On average, participants reported 37% of their total Facebook Friends were “actual” friends. Finally, time spent on Facebook (M=33.89 minutes, median=15, SD=47.90) was measured through the open-ended item, “In the past week, on average, approximately how many minutes PER DAY have you spent actively using Facebook?” We specified “actively” to avoid responses that included time spent with the site open but inactive.
In addition to gender (dichotomous), age (continuous), and education (ordinal), we included self-esteem as a control variable in our regressions because it has been a significant predictor of social capital (Ellison et al., 2007, 2011; Steinfield et al., 2008). Self-esteem (Cronbach's α=.86) was measured through seven items from the Rosenberg Self-Esteem Scale (Rosenberg, 1989) and had a mean score of 4.33 (SD=.56) . Also in line with previous research, we included Weekly Internet Use as a control variable. This measure uses a weighted mean of two items (time spent using the Internet on a typical weekday and a typical weekend day). Mean Internet time per day was 3 hours, 10 minutes (SD=2 hours, 46 minutes).
- Top of page
- Social Capital
Our analysis focuses on the relationship between two main predictor variables—actual friends on Facebook and engagement in Facebook Relationship Maintenance Behaviors — and both general and Facebook-specific bridging social capital. To test the relationship between these two forms of bridging social capital, control variables, and the predictors we describe above, we ran a series of nested OLS regressions.
In the regression predicting Facebook-specific bridging social capital, the control variables accounted for 5.4% of the variance, with gender (β=.207, p<.001) and self-esteem (β=.098, p<.05) emerging as significant, such that women and those with higher self-esteem reported higher perceived Facebook bridging social capital. The addition of the Facebook usage variables—minutes on Facebook, total Facebook Friends, and actual friends—increased the R2 to .131, with minutes (β=.144, p<.01) and actual friends (β=.218, p<.001) positively predicting Facebook bridging social capital, supporting H1a. Finally, the addition of Facebook Relationship Maintenance Behaviors (β=.569, p<.001) increased the model's R2 to .394 and provided support for H2a. Full results are presented in Table 2.
|Model 1: Controls||Model 2: Time & Friend Variables||Model 3: FRMB||Model 4:Interaction|
|Standardized Betas (t-scores)|
|Gender: Women||.207 (4.29)c||.172 (3.67)c||.052 (1.28)||.045 (1.14)|
|Ethnicity: White||.008 (0.17)||.022 (.483)||.041 (1.07)||.043 (1.14)|
|Age||−.130 (−2.70)b||−.029 (−.586)||−.006 (−.141)||.013 (.314)|
|Education||−.029 (−.602)||−.015 (−.324)||.018 (.460)||.026 (.667)|
|Self-Esteem||.098 (2.07)a||.061 (1.33)||.047 (1.22)||.038 (.995)|
|FB Mins Per Day||.144 (2.93)b||.000 (.012)||.007 (.172)|
|Total FB Friends||.023 (.410)||.001 (.018)||−.009 (−.205)|
|Actual Friends||.218 (4.04)c||.137 (3.02)b||.967 (3.93)c|
|FRMB||.569 (13.52)c||.657 (13.46)c|
|Interaction: Actual Friends by FRMB||−.857 (−3.43)c|
|Constant||n/a (7.96)c||n/a (7.16)c||n/a (2.68)b||n/a (1.69)|
In the regression predicting general bridging social capital, the control variables accounted for 9.3% of the variance; as with Facebook-specific bridging social capital, gender (β=.182, p<.001) and self-esteem (β=.255, p<.001) were significant predictors. The addition of the time and Friends variables increased the R2 to .168. Total Facebook Friends (β=.118, p<.05) emerged as a significant predictor in this model; however, the impact of actual friends (β=.190, p<.001) on general bridging social capital was stronger, supporting H1b. Adding Facebook Relationship Maintenance Behaviors to the model increased the R2 to .229, and the significant positive relationship between FRMB and bridging social capital (β=.278, p<.001) provided support for H2b. See Table 3 for the full regression model.
|Model 1: Controls||Model 2: Time & Friend Variables||Model 3: FRMB||Model 4:Interaction|
|Standardized Betas (t-scores)|
|Gender: Women||.182 (3.92)c||.162 (3.59)c||.101 (2.27)a||.099 (2.23)a|
|Ethnicity: White||−.006 (−.138)||.011 (.261)||.025 (.597)||.029 (.688)|
|Age||−.056 (−1.21)||.056 (1.17)||.069 (1.49)||.083 (1.79)|
|Education||−.024 (−.519)||−.021 (−.465)||−.006 (−.131)||.001 (.019)|
|Self-Esteem||.255 (5.59)c||.219 (4.94)c||.212 (4.97)c||.204 (4.78)c|
|FB Mins Per Day||.077 (1.63)||.009 (.181)||.009 (.187)|
|Total FB Friends||.118 (2.17)a||.106 (2.03)a||.107 (2.05)a|
|Actual Friends||.190 (3.62)c||.150 (2.94)b||.701 (2.65)b|
|FRMB||.278 (5.94)c||.337 (6.21)c|
|Interaction: Actual Friends by FRMB||−.571 (−2.13)a|
|Constant||n/a (10.28)c||n/a (9.44)c||n/a (6.75)c||n/a (5.96)c|
Our research question probed the relationship between the number of “actual” friends and users' engagement in relationship maintenance activities (FRMB), as they relate to bridging social capital. First, we conducted an OLS regression (see Model 4 in Tables 2 and 3) using the same variables as used in previous analyses as well as an interaction term (actual friends by FRMB). The interaction term was significant in both the general (β=−.571, p<.05) and Facebook-specific (β=−.857, p<.001) bridging models, suggesting that the impact of FRMB on bridging social capital may be moderated by users' reported number of actual friends in their Facebook network. The significant negative term indicates that for Facebook users who report fewer actual friends on Facebook, greater engagement in FRMB was correlated with a larger increase in bridging social capital than for users who reported more actual friends on the site.
To further examine this finding, we conducted a simple slopes analysis (Aiken & West, 1991). We observed effects of FRMB on bridging social capital at three levels of actual friends: (1) reported value, (2) reported value minus one standard deviation, and (3) reported value plus one standard deviation. Results indicated that FRMB's influence on both Facebook-specific and general bridging social capital significantly differed across the three measurement points for actual friends. For all three levels, greater engagement in FRMB was correlated with higher perceived bridging social capital (see Figure 1). However, those with the fewest actual friends saw the greatest gains in bridging social capital as their engagement in FRMB increased. An analysis of variance shows this difference across the three levels of actual friends is significant for Facebook-specific bridging social capital, F(82)=33.67 p<.001, and trends towards significance with general bridging social capital, F(86)=3.204 p=.077.
- Top of page
- Social Capital
In this study, we explore the relationship between bridging social capital and two components of Facebook use through survey data collected from a sample of U.S. adults. We employed two separate measures of bridging social capital—one assessing the benefits associated with interactions with one's network of Facebook Friends and the other focusing on one's interactions with his or her entire social network. Confirming and expanding on previous studies of Facebook use (e.g., Burke et al., 2011; Ellison et al., 2011), we find significant positive relationships between these measures of bridging social capital and both the number of “actual” Facebook friends participants report that they have and their engagement in Facebook Relationship Maintenance Behaviors. FRMB assesses the extent to which subjects report they engage in activities that signal attention to and purposefully cultivate relationships on the site; these activities include responding to Friends' implicit or explicit resource requests and writing “happy birthday” on a Friend's Wall. This study contributes to our understanding of the interpersonal dynamics of SNS use by (1) identifying how users' engagement in the social grooming behaviors identified above relate to perceptions of bridging social capital—both within their Facebook Friend network and their social network more broadly—and explicating the interpersonal, structural, and attention-filtering dynamics which contribute to this relationship; (2) exploring the differences between perceived total and “actual” Friends on the site in regards to bridging social capital outcomes; and (3) examining differences in perceptions of general bridging social capital between users and nonusers of the site.
Considered in conjunction with previous research on this topic, these findings provide insight into the mechanisms through which bridging social capital is cultivated and the role that Facebook-enabled communication practices may play in this process. Our bridging social capital measure, adapted from Williams (2006), captures the extent to which participants feel that interacting with their social network promotes “outward thinking” and allows them to encounter new ideas, experiences, and people. Scholarship on the information diffusion properties of network structures (Granovetter, 1973; Burt, 1992) suggests one explanation for the relationship between Facebook use and perceptions of bridging social capital: Because weaker ties are more likely to be bridging ties, the bridging social capital benefits of Facebook use may lie in technical features of the site that lower the cost of maintaining and communicating with a larger network of weak ties. SNS scholars point to other relevant affordances, such as the fact that SNSs enable users to request informational or other resources from these connections (Gray, Ellison, Vitak, & Lampe, 2013). Similarly, the identity information available about users in the profile and elsewhere may make these exchanges more visible, persistent, and meaningful (Ellison et al., 2011). However, our findings suggest that understanding social capital dynamics in the context of SNS use may entail a more complex calculus that considers not just the number and strength of one's ties, but also issues of visibility and attention, subjective—and perhaps idiosyncratic—distinctions between “actual” and nonactual Friends, and specific communication behaviors that reshape information flow across networks by enabling participants to gain access to Friends of Friends. In other words, it is not enough to just add connections on the site; rather, individuals must engage in intentional behaviors that signal attention to components of their network and contribute to expectations of reciprocity. In the following section, we describe the implications of FRMB on social capital in three domains—interpersonal, structural, and technical.
Our focus on relationship maintenance behaviors highlights their potentially important role in binding networks in meaningful ways and creating an environment in which social capital-enhancing exchanges can flourish. First, engaging in FRMB enables users to explicitly signal they are attending to specific individuals in their network and are available for reciprocal interactions. We argue these activities may signal attention within a system in which attention cues must be explicitly constructed, and that these signals support relationship maintenance and serve to activate expectations of reciprocity.
Second, Facebook enables interactions with users outside of one's articulated Friend network through communication affordances such as comments on a mutual Friend's status update, a network-spanning communicative act that enables users to interact with Friends of their Friends. Recent empirical work on voting behavior finds that users who learned via the site that their Facebook Friends voted were more likely to vote themselves, and that this message was influential even for Friends of Friends (Bond et al., 2012). Interactions with Friends of Friends may increase perception that users have access to diverse individuals, are interested in the greater world, and see themselves as part of a larger community—all elements of bridging social capital articulated by Williams (2006) and captured by the scales adapted for this study.
Third, although paying attention to individuals in one's network is important in offline contexts as well, the specific sociotechnical context of SNSs means the attention-signaling behaviors captured by our FRMB measure help shape users' experiences on the site and specifically the content that is filtered into and out of their News Feeds. Given that many users have hundreds of Friends on Facebook, engaging in these behaviors serves to signal attention to specific ties within one's network on an interpersonal level, but it also serves a technical purpose within the specific context of Facebook. Responding to a Friend's request for advice not only signals willingness to help, but also may indirectly increase the visibility of one's content in others' News Feeds because interactions between individuals are treated as a signal of relevance by the News Feed algorithm (Newman, 2011). The content that appears in one's News Feed is a function of a proprietary algorithm created by Facebook to increase the visibility of relevant posts and to decrease the visibility of those more likely to be uninteresting. Thus, behaviors captured by the FRMB scale may elevate the visibility of users who engage in them.
As an example of how these interpersonal, structural, and attention-filtering (technical) domains may interact, consider the act of commenting on a Friend's status update. Comments are fairly common: Hampton et al. (2011) found that more than half of adult American Facebook users comment on Friends' updates at least once a week. Although our FRMB items do not specify the channel users employed to respond to broadcasted requests (with the exception of the “happy birthday” item), a common response to the kinds of status update requests described in the FRMB items would be in the form of a comment. As we argue above, commenting on a Friend's post signals an investment in the relationship, and reciprocated interactions are a primary component in the generation of social capital. However, the specific features of SNSs, namely the public nature of interactions, may also play a role in these processes. Comments on Friends' posts may be more likely to generate bridging social capital than status updates or other broadcasting behaviors because these comments have the potential to be seen by all of the poster's Friends and thus constitute an effective strategy for accessing “Friends of Friends.” In this way, the behaviors included in our scale may amplify the “bridging” function of the site by providing an opportunity for Friends of Friends to interact. Importantly, commenters on Friends' posts are presented to this extended network not as total strangers, but rather as individuals with at least one shared social connection, which, as Donath and boyd (2004) suggest, may signal trust and serve as a tool for verifying legitimacy. Furthermore, these Friends of Friends can often obtain personal information about one another via the user profile, helping to establish common ground (Lampe, Ellison, & Steinfield, 2007) and potentially encouraging them to activate these latent ties and convert them to weak and bridging ties (Ellison et al., 2007, 2011; Haythornthwaite, 2005). Perhaps some of the “benefits of Facebook Friends” (as discussed by Ellison et al., 2007) are actually embedded in interactions with Friends of Friends and the relationships these interactions may cultivate.
This work contributes to our understanding of how mediated communication practices support relationship maintenance, and how these interactions contribute to social capital accrual and development. Other than the “happy birthday” item, our FRMB items intentionally did not specify the channel through which responses were communicated, in order to capture responses that occurred offline as well as within the Facebook environment. While future research should confirm these speculations, based on media multiplexity research (Haythornthwaite, 2005), we assume that weaker ties are more likely to respond via the site, whereas stronger ties may use other channels. For instance, prompted by a Friend's complaint about a bad day, it is more likely that weaker ties who respond would do so through a comment or Facebook message, whereas closer ties may respond through the site or via a telephone call, private email, or in-person visit. In order to more fully understand the mechanism by which FRMB is related to social capital, future work should employ more nuanced measures that probe which channels—both on Facebook and elsewhere—are being used to engage in these relationship maintenance behaviors.
Actual Friends. The connection between these relationship maintenance behaviors and the number of “actual friends” in one's Facebook network was another focus of this study. Through a series of analyses, we found that the relationship between FRMB and bridging social capital is moderated by the number of actual friends a user reports in their network. In other words, as engagement in FRMB increased, users with fewer actual friends reported higher perceived bridging social capital—both Facebook-specific and general—than users with more actual friends. This may be because those with fewer “actual” friends are performing FRMB with Facebook Friends they do not consider actual friends. These weaker connections in their network might represent more diverse values, informational resources stores, and networks (Granovetter, 1973). Additionally, by interacting with these “nonactual” friends through more public, visible channels on the site, users might gain access to the Friend networks of these nonactual friends and the richly diverse set of perspectives and information they likely represent.
Limitations. Limitations to this study include the fact that our measure of social capital may not have adequate construct validity; future research in this area should develop new measures of social capital and validate existing measures. As this research study was cross-sectional, we are unable to demonstrate that social capital is an outcome of the activities we measure or the cause of these behaviors; thus, future research should employ a longitudinal approach, which might provide more insight into the directionality of these relationships. More nuanced FRMB measures, especially if combined with behavioral measures accessible from server-level data, would give us a better sense of how to interpret high scores and could add support to our interpretation of these data. Additionally, we cannot make definitive claims about the exact nature of “actual” friends due to our measure of this construct, which intentionally did not provide participants with a definition of what constituted an actual friend. Future work should probe user conceptions of actual friends and whether this is determined by visibility, closeness, recency of last interaction, or other related concepts. Due to space constraints, we do not consider bonding social capital in this piece, but future research may wish to explore the relationship between our maintenance behaviors and this other important form of social capital. Finally, our sample limits the generalizability of our findings, as these individuals differed from the population of U.S. adults in several ways; this work should be replicated with other samples.
In summary, this research shows that actively engaging with one's Facebook network—both through responses to Friends' requests and posting content directly on a Friend's wall—is positively linked to higher levels of bridging social capital. Importantly, our findings regarding general bridging social capital extend work in this area because they imply that engaging in specific practices related to Facebook use have the potential to shift perceptions regarding access to social resources, even within a more general (not just Facebook-specific) context. We argue that behaviors captured by the Facebook Relationship Maintenance Behaviors scale serve a relationship maintenance function by signaling attention and contributing to expectations of reciprocal assistance, enabling individuals to access networks outside of their own, and reshaping attention filters within the system. Although future research should empirically assess our claims regarding the specific dynamics of these exchanges, the significant statistical relationship between FRMB and perceptions of bridging social capital suggests that responding to implicit or explicit requests from Friends plays a key role in social capital processes, on and off the site.
- Top of page
- Social Capital
In this paper, we extend our knowledge of the social capital implications of social network site use through both the identification of specific SNS-enabled relationship maintenance behaviors that predict general and Facebook-specific bridging social capital gains. We believe that the social and technical affordances of Facebook make it easier for individuals to invest in and extract social resources from their network and to access Friends of Friends, who might provide novel informational resources. Most importantly, these findings highlight the importance of actively managing, grooming, and maintaining one's network, suggesting that social capital is not generated simply by the existence of connections on a SNS, but rather is developed through small but meaningful effort on the part of users as they engage in relationship maintenance behaviors such as responding to questions, congratulating or sympathizing with others, and noting the passing of a meaningful day. This work contributes to our understanding of relationship maintenance activities in social networks and suggests that the true benefit of social network sites may not just be the technical connections they make possible, but by creating an environment in which meaningful communicative exchanges, and the potential social capital benefits they embody, can flow.
- Top of page
- Social Capital
This work was supported by the National Science Foundation (HCC 0916019). We thank Brandon Brooks, Joe Walther, Toni Botsford, and our anonymous reviewers for valuable feedback on and assistance with this study.
- Top of page
- Social Capital
- 1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage Publications, Inc. & (
- 2012, April). The role of social networks in information diffusion. Paper presented at the World Wide Web conference, Lyon, France. , , , & (
- 1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). New York: Greenwood. (
- 2012). A 61-million-person experiment in social influence and political mobilization. Nature, 489(7415), 295–298. , , , , , , et al. (
- 2010). Social network activity and social well-being. Proceedings of the 2010 Annual Conference on Human Factors in Computing Systems (pp. 1909–1912). New York: ACM. doi:10.1145/1753326.1753613 ., ., & . (
- 2011). Social capital on Facebook: Differentiating uses and users. Proceedings of the 2011 Annual Conference on Human Factors in Computing Systems (pp. 571–580). New York: ACM. doi:10.1145/1978942.1979023 , , & . (
- 1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. (
- 1988). Social capital in the creation of human capital. The American Journal of Sociology, 94, 95–120. (
- 1993). Definitions and theoretical perspectives on maintaining relationships. Journal of Social and Personal Relationships, 10, 163–173. , & (
- 2007). Signals in social supernets. Journal of Computer-Mediated Communication, 13, 231–251. doi: 10.1111/j.1083-6101.2007.00394.x (
- 2004). Public displays of connection. BT Technology Journal, 22(4), 71–82. doi: 10.1023/B:BTTJ.0000047585.06264.cc , & . (
- 1991). Understanding relationships. New York: Guilford Press. (
- 1996). Grooming, gossip, and the evolution of language. Cambridge, MA: First Harvard University Press. (
- 2007). The benefits of Facebook "friends:" Social capital and college students' use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168. doi: 10.1111/j.1083-6101.2007.00367.x , , & (
- 2011). Connection strategies: Social capital implications of Facebook-enabled communication practices. New Media & Society. doi: 10.1177/1461444810385389 , , & (
- 2010). With a little help from my Friends: Social network sites and social capital. In Z. Papacharissi (Ed.), A networked self: Identity, community and culture on social network sites (pp. 124–145). New York: Routledge. , , , & (
- Facebook Data Team. (2009, March 9). Maintained relationships on Facebook (Memo). Retrieved from http://www.facebook.com/notes/facebook-data-team/maintained-relationships-on-facebook/55257228858
- 1973). The strength of weak ties. American Journal of Sociology, 78, 1360–1480. doi: 10.1086/225469 (
- 1974). Getting a job: A study of contacts and careers. Cambridge, MA: Harvard University Press. (
- 2013). Who wants to know? Question-asking and answering practices among Facebook users. In Proceedings of the 16th Annual Conference on Computer Supported Cooperative Work and Social Computing. New York: ACM. doi: 10.1145/2441776.2441913 , , , & (
- 2011). Social networking sites and our lives. Washington, DC: Pew Internet & American Life Project. , , , & (
- 2005). Social networks and Internet connectivity effects. Information, Communication & Society, 8, 125–147. doi: 10.1080/13691180500146185 (
- 2008, April). “Looking at”, “Looking up” or “Keeping up with” people? Motives and uses of Facebook. Proceedings of CHI 2008 (pp. 1027–1036). New York: ACM. doi:10.1145/1357054.1357213 (
- 2011). Online pauses and silence: Chronemic expectancy violations in written computer-mediated communication. Communication Research, 38, 54–69. doi: 10.1177/0093650210378229 & (
- 2006). A face(book) in the crowd: Social searching vs. social browsing. Proceedings of CSCW 2006, Banff, Alberta, Canada. , , (
- 2007). A familiar face(book): Profile elements as signals in an online social network. In Proceedings of the SIGCHI Conference on human factors in computing systems (pp. 435–444). New York: ACM. doi: 10.1145/1240624.1240695 , , & (
- 2009). Adults and social network websites. Washington, DC: Pew Internet & American Life Project. (
- 2001). Social capital: A theory of social structure and action. London: Cambridge University Press. (
- 2011) 65% of online adults use social networking sites. Washington, DC: Pew Internet Project. & (
- 2011, August 16). The ultimate guide to the Facebook Edgerank algorithm [blog]. Retrieved from http://econsultancy.com/us/blog/7885-the-ultimate-guide-to-the-facebook-edgerank-algorithm. . (
- 2000). Bowling alone: The collapse and revival of American community. New York: Simon & Schuster. doi: 10.1145/358916.361990 (
- 2007). Relational maintenance and the influence of commitment in online and offline relationships. Communication Studies, 58, 315–337. (
- 2001). Beyond bowling together: Socio-technical capital. In J. Carroll (Ed.), HCI in the new millennium (pp. 647–672). New York: Addison-Wesley. doi: 10.1145/358916.362079 (
- 1989). Society and the adolescent self-image (Rev. ed.). Middletown, CT: Wesleyan University Press. (
- 2011). Facebook as a Toolkit: A uses and gratification approach to unbundling feature use. Computers in Human Behavior. doi: 10.1016/j.chb.2011.07.011 , , & (
- 2008). Social capital, self-esteem, and use of online social network sites: A longitudinal analysis. Journal of Applied Developmental Psychology, 29, 434–445. doi: 10.1016/j.appdev.2008.07.002 , , & (
- 2010). Public dialogs in social network sites: What is their purpose? Journal of the American Society for Information Science and Technology, 61, 392–404. doi: 10.1002/asi.21241 & (
- 2011). Relational maintenance and CMC. In K. B. Wright and L. M. Webb (Eds.), Computer-mediated communication in personal relationships (pp. 98–118). New York: Peter Lang Publishing. , & (
- 2008). Grooming, gossip, Facebook, and MySpace. Information, Communication & Society, 11, 544–564. doi: 10.1080/13691180801999050 (
- 2009). Is there social capital in a social network site?: Facebook use and college students' life satisfaction, trust, and participation. Journal of Computer-Mediated Communication, 14, 875–901. doi: 10.1111/j.1083-6101.2009.01474.x , , & (
- 1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19, 52–90. doi: 10.1177/009365092019001003 (
- 2006). On and off the ‘net: Scales for social capital in an online era. Journal of Computer-Mediated Communication, 11, 593–628. doi: 10.1111/j.1083-6101.2006.00029.x (