Introduction
The following article will be discussed in this paper:
Ellison, N. B., Vitak, J., Gray, R., & 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(4), 855-870.Web.
The Purpose of the Study
The purpose of the study by Ellison, Vitak, Gray, and Lampe (2014) was to scrutinize the relationship between certain types of Facebook-enabled communication, and the perceived bridging social capital. This study is important because exploring such a relationship may allow for finding out whether certain behaviors on Facebook is associated with a wider, more profound network of social connections that an individual has; when more information on this topic is obtained, this knowledge may be used to e.g. expand one’s social network and obtain additional social capital. It might be hypothesized that the authors of the article conducted the study because of the increasing importance and prevalence of the use of social networks, which warrants research pertaining to the changes these networks bring to human communication.
The Theory on Which the Study Is Based
The study by Ellison et al. (2014) is based on a theory of social capital. The notion of social capital refers to the ability of people or groups of people to gain access to various resources that exist within the network of their social relationships. For instance, it is possible to convert social capital into different forms of resources such as favors or information. Social capital is accumulated via the process of maintaining contacts with other individuals. People generally communicate via a greater number of channels with those with whom they have stronger ties, and these channels of communication are not always available for interaction with those who are only weakly tied to an individual (e.g., friends are strongly-tied, but acquaintances of friends are weakly-tied) (Ellison et al., 2014). Importantly, social networks allow for communication between weakly-tied people who would otherwise be unable to communicate, and also permit contributing to and maintaining numerous relationships simultaneously (Ellison et al., 2014). The theory that includes this knowledge provided a basis for the study by Ellison et al. (2014).
It should be noted that two types of social capital are commonly considered: bridging social capital (connections which relate different groups inside a network, therefore supplying access to new information) and bonding social capital (reinforcing already existing ties) (Ellison et al., 2014). The study in question concentrates on the phenomenon of bridging social capital, investigating how Facebook allows for better access to resources that are available via weaker ties and may be unavailable via other channels of communication. For this study, the authors also propose using two additional notions: Facebook-specific bridging social capital (the capital which is composed of resources available via the Facebook connections of an individual) and general bridging social capital (all connections, including both offline and online connections) (Ellison et al., 2014, p. 857).
The Method and Sample
For this study, authors propose two hypotheses, each one composed of two parts: H1: “The greater the number of actual friends on Facebook, the greater users’ reported (a) Facebook-specific and (b) general bridging social capital,” and H2: “The more users engage in Facebook Relationship Maintenance Behaviors, the greater their reported (a) Facebook-specific and (b) general bridging social capital,” where the term Facebook Relationship Maintenance Behaviors (FRMB) refers to actions such as liking posts or saying things such as “happy birthday,” to preserve contacts (Ellison et al., 2014, pp. 857-859).
The authors used a survey that was emailed to a random sample of 2149 non-faculty personnel of a large university in the U.S., obtaining 614 usable responses (28.9% response rate) (Ellison et al., 2014). The sample was mainly female (66%); aged 45 on the average, with a standard deviation (SD) of 11.0; well-educated (40.1% possessed a bachelor’s degree, 32.1% did postgraduate studies); also, 134 respondents (22%) did not utilize Facebook. The sample was more educated, had a larger percentage of females, and was more likely to use Facebook than a national sample from another study from the same time period (as cited in Ellison et al., 2014, p. 860). Thus, it seems that the sample was not representative of the general population when it came to demographic variables, which limits its external validity. Nevertheless, it is apparent that this method was appropriate, at least because it allowed for gaining a random sample of individuals, and, although the sample was not representative of the national population, it permitted making conclusions at least about university non-faculty staff.
The survey included several parts comprised of multiple questions, mainly using 5-point Likert scales. The authors assessed the respondents bridging social capital (general and Facebook-specific; the same scale was used twice, asking about all contacts at first, and only about Facebook friends the second time); their Facebook relationship maintenance behaviors (FRMB); the number of their Facebook friends, and the number of Facebook friends they considered actual friends; and the following control variables: gender, age, education, self-esteem (as per Rosenberg Self-Esteem Scale), and weekly Facebook use. The Cronbach’s alphas for all the constructs were high: no Cronbach’s alphas were lower than.80. For FRMB, exploratory and confirmatory factor analyses were employed to gain the final scores (Ellison et al., 2014). All the scales can be found in studies referenced by the authors. Therefore, the researchers adequately defined their variables.
To carry out the study, the authors conducted several descriptive calculations and inferential statistical tests, such as means, SDs, t-tests, ANOVAs with posthoc tests, regressions, and so on. These procedures allowed for finding the characteristics of the sample and making inferences about the differences in the population.
Main Findings
Ellison et al. (2014) list numerous findings in this study, but the main ones pertain to the two predictor variables (actual friends on Facebook, and FRMB) and two outcome variables (general and Facebook-specific bridging social capital). After running a regression with control variables as predictors, it was found that they accounted for 5.4% of the variance (gender and self-esteem were significant predictors). Next, adding the variables of weekly Facebook use, total friends on Facebook, and actual friends on Facebook resulted in an increase of the variance in Facebook bridging social capital explained by the model. This supported the H1a hypothesis. Finally, adding FRMB as a predictor also increased the variance in Facebook bridging social capital explained by the model. This supported the H2a hypothesis (Ellison et al., 2014).
As for the general social capital, the control variables explained 9.3% of the variance in the data; gender and self-esteem were also significant predictors in this model. Adding the variables of weekly Facebook use, total friends on Facebook and actual friends on Facebook resulted in an increase of the variance in general bridging social capital explained by the model. Thus, the hypotheses H1b and H2b were supported (Ellison et al., 2014).
Therefore, all the hypotheses made in this study were supported.
Limitations
There are several limitations to the research by Ellison et al. (2014). For instance, social capital measures may not have sufficient construct validity, and further studies are required to validate these measures or create new ones. Also, because the study was cross-sectional, it is unclear whether the predictor variables caused changes in the outcome variables or vice versa; it is stated that a longitudinal study might shed more light on the direction of this relationship. There are some more limitations pertaining to the constructs used in the study; for instance, FRMB measures should be more detailed to provide better insights into the interpretation of high scores on this scale. It is also difficult to explain the exact nature of the term “actual friends” employed in the study. Also, the fact that the sample was not representative of the adult U.S. population limits the generalizability of the results of the study (Ellison et al., 2014).
Implications of the Study
The study implies that individuals who have more actual friends on Facebook may have greater bridging social capital, both Facebook-specific and general; that is, they have access to a larger number of groups and possess more informational resources at their disposal. Also, the more people engage in FRMB, the more bridging social capital (both Facebook-specific and general) they accumulate. It should be noted, however, that the nature of this association is unclear; that is, it is unknown whether having more friends on Facebook and engaging in FRMB increases the amount of social capital, or vice versa, or both.
The “real world” implications of the study are that having more Facebook friends is associated with greater social capital, and engaging in FRMB and similar behaviors might increase one’s social capital. This may be beneficial for people who wish to expand their network of social contacts and may help resolve some personal matters as well, which may, e.g., be used by counseling specialist as advice to their clients. For instance, a lonely, depressed person who feels shy to restore their old contacts might be recommended to engage in FRMB or communicate via social networks, which may give them additional social capital, which then can be turned into e.g. emotional support. Future studies should further explore the nature and direction of the association between behaviors in social networks and social capital.
References
Ellison, N. B., Vitak, J., Gray, R., & 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(4), 855-870. Web.