Digital Divides: What Types of Youth Use Social Network Sites?
Traditional digital divide scholarship asks whether particular populations have or do not have equal access to new technologies or platforms. Scholars assert that social media represent new skills and ways of participating in the world. If students are not allowed to use new technologies and contribute to online communities like SNS, they will not be able to develop the necessary skills and technical literacy that will be vital in the future (Jenkins, 2006). Stemming from this belief, researchers continue to wonder whether certain groups of students are systematically hindered from using new technologies. For example, Seiter (2008) observes that “Young people famously use digital communications—instant messaging, cell phone texting, and social networking Web sites—to maintain their social capital, at least with those peers who can afford to keep up with the costly requirements of these technologies” (p. 39). The statement succinctly outlines the concerns of digital divide scholars: (1) there is an understanding that many people are using technology, (2) the use has some positive outcome, i.e., developing social capital, and (3) questions remain as to the systemic and unequal access to the technology.
The emerging research literature suggests that SNS are becoming ubiquitous aspects of youth and young adult life. In a sample of college students, Hargittai (2007) finds few demographic differences between users and nonusers of SNS. Gender appears as a significant predictor, with females being 1.6 times more likely to use an SNS than males. In addition, having Internet access through friends or family also significantly predicted whether a college student used SNS. Other traditional indicators such as race and parent's education had no significant correlation with the use of SNS. Hargittai's study underscores the developing trend of mass adoption of SNS. Among the college students in her sample, there appeared to be few systemic inequalities in their access to SNS.
Hargittai (2007) also disaggregates her results based on different SNS—Facebook, MySpace, Xanga, and Friendster. She finds interesting and significant correlations between race and particular SNS communities. For example, Hispanic students were more likely to use MySpace, but less likely to use Facebook compared to Caucasian students. Asian students were significantly more likely to use Xanga and Friendster. Such patterns problematize some of the theoretical benefits of social networks. For example, Wellman et al. (1996) theorize that, “People can greatly extend the number and diversity of their social contacts when they become members of computerized conferences or broadcast information to other CSSN [computer supported social network] members” (p. 225). However, Hargittai notes that if particular groups of people gravitate to respective communities, offline inequalities may persist online.
Studies of digital divides and adolescent youth are less frequent. However, Ahn (2011) examines a nationally representative sample of 12–17-year-old teens from a 2007 Pew Internet and American Life survey. African-American youth were more likely to use SNS compared to their white peers, controlling for other factors. Furthermore, traditional divide indicators such as having Internet access at home were counterintuitive. A teenager whose primary Internet access was not at home or school was over twice as likely to use SNS as teens who had home access. Such results converge with ethnographic accounts and surveys of youth, which suggest that teenagers find different ways to connect to their online social networks despite socioeconomic status (Ito et al., 2009; Lenhart et al., 2010). Mobile devices and Internet access away from adult supervision may constitute new contexts where youth use social media. Such contexts are theoretically vital areas to explore because they may contribute to culture and behavior in SNS communities.
In addition to concerns about digital divide, understanding the characteristics of SNS users is necessary to properly assess any effects of participation. For example, perhaps one is concerned about whether using Facebook leads to higher levels of self-esteem among youth. The question cannot be adequately examined without taking into account the characteristics of youth who use Facebook or the network they interact with. Selection bias looms large in studies of SNS. Youth actively decide to use SNS, versus other tools, for particular communicative reasons such as keeping in touch with friends (Agosto & Abbas, 2010). Early studies in the field also imply that characteristics such as shyness, self-esteem, and narcissism are related to behavior in SNS (Barker, 2009; Buffardi & Campbell, 2008; Zywica & Danowski, 2008). Survey data find that female and male youth might use SNS in different ways (Lenhart, Madden, Macgill, & Smith, 2007). Several of the studies reviewed below also find that SNS use has differential effects for individuals with high/low levels of self-esteem or extraversion (i.e., Steinfield, Ellison, & Lampe, 2008; Zywica & Danowski, 2008). Youth enter these online communities with existing traits (gender, self-esteem, shyness, etc.). They also have varying motivations for using SNS. Such factors influence with whom youth interact, how they behave, and ultimately how they develop through their participation in SNS communities.
Social Network Sites and Youth Relationships: Safety Versus Social Capital
A major controversy surrounding SNS is youth safety and privacy. Approximately 70% of school districts block access to SNS, and the main reason for this trend centers on fears about student safety (Lemke et al., 2009). However, initial research on SNS suggests that these online communities help individuals build social capital. Social capital refers to the idea that one derives benefits—i.e., advice, information, or social support—through their network of relationships (Portes, 1998). A critical theoretical concern for youth is whether and how SNS facilitate detrimental behaviors such as bullying and interacting with strangers, versus positive outcomes such as developing wider networks of relationships.
The early picture concerning youth and online privacy is mainly positive. Nearly every major SNS offers privacy controls. In fact, “These privacy measures have given adolescent users a great deal of control over who views their profiles, who views the content that they upload, and with whom they interact on these online forums” (Subrahmanyam & Greenfield, 2008, p. 123). Current research finds that teenagers disclose a variety of personal information on their profiles, but they also proactively use privacy features to manage who can view their content (Hinduja & Patchin, 2008; Lenhart & Madden, 2007). Lenhart and Madden (2007) report from a nationally representative sample of youth that 66% of teenagers limit their profile to particular people in their network. A cross-sectional study of a college student sample also reports that privacy concerns did not hinder users' desire to share personal information on their profiles. Rather, students used privacy features to control and limit who could view their information (Tufecki, 2008).
Approximately 91% of youth who use SNS report that they utilize the sites to communicate with already known friends (Lenhart & Madden, 2007). Qualitative studies also converge with this finding that U.S. youth mostly use SNS to interact with friends and not to meet strangers (Agosto & Abbas, 2010; boyd, 2008). Studies also find that teenagers are less likely to experience unwanted sexual solicitations or harassment in SNS, while more likely to experience these dangers in instant messaging and chat room environments (Ybarra & Mitchell, 2008). This initial research suggests that the fears about SNS creating opportunities for predators to solicit children are overstated. Nevertheless, these and other detrimental behaviors such as cyberbullying are real concerns. Even if dangerous or negative experiences in SNS only account for a small percentage of online activity, each instance represents a significant concern for adults, parents, and educators.
A social informatics approach to understanding youth safety would compel researchers to consider two interrelated aspects of SNS: technical features and youth behavior. The features of a technology tool may influence the likelihood of contacting strangers on the Internet. Peter, Valkenburg, and Schouten (2006) find that youth who spend more time in chat rooms talk with more strangers. Ybarra and Mitchell (2008) also find that adolescents are less likely to be targeted for unwanted sexual solicitation in SNS compared to chat rooms. Chat rooms are often public and unmonitored spaces where multiple people talk synchronously. Perhaps such features are related to the higher frequency of risky behavior and unwanted interactions in these online forums.
Early research also notes variations within different SNS themselves. Dwyer, Hiltz, and Passerini (2007) find that MySpace users utilize the site to meet new people more often than Facebook members. Such patterns might be related to the norms of each site at that time. Facebook originally began as a college-campus based SNS, and thus established boundaries around one's social networks (boyd & Ellison, 2007). MySpace began as a broader and open network. As Facebook has slowly opened its network to high school students, then to any individual, these dynamics may have changed. The key point is that technical and social elements of a respective SNS community may facilitate or inhibit behavior, and this question requires further examination.
Beyond a keen eye towards the technical features of an SNS platform, additional studies are needed to identify those youth who might be prone to risky online behavior and why they participate in such activities. One theoretical question is to understand what characteristics—i.e., social, emotional, or behavioral—relate to adolescents seeking or experiencing negative behavior in online communities. Factors such as age, gender, experience level, and personality traits appear to influence youth risks in online setting. For example, in a survey of 412 Dutch teenagers, Peter et al. (2006) find that younger adolescents were more likely to talk with strangers. In addition, teens that used the Internet to explicitly meet new friends or to overcome their own shyness (social compensation) communicated with strangers more often.
Early studies of college students and Facebook find that peer influence is related to safety behaviors in SNS. Students appear more likely to have a private profile if their friends or roommates also used privacy settings (Lewis, Kaufman, & Christakis, 2008). Such results offer a hypothesis that peer effects influence the safety behaviors of youth online. Finally, a recent report by the Berkman Center for Internet & Society observes that while the Internet may potentially provide access to negative experiences for children, technology alone is not the causal mechanism (Berkman Center for Internet & Society, 2008). Factors such as age influence when and how youth experience unwanted sexual solicitation or cyberbullying. In addition, youth characteristics are related to both those who are victims of online harassment, and those who bully their peers. The underlying social, psychological, and emotional characteristics of youth influence whether they engage in negative activity, and technology provides another avenue (but is not a cause) for these behaviors.
Studies about adolescent privacy and safety focus on the potential negative relationships that can be formed online. However, scholars also posit that the Internet widens our social networks and provides positive benefits in the form of social capital (Wellman et al., 1996). Various theorists focus on disparate elements of social capital theory, which often leads to confusion on the part of research studies that use the framework (Portes, 1998). For example, Pierre Bourdieu (1986) focuses his definition on people's membership of social groups that have cultural and financial wealth. If one is a member of a group with many resources, he or she can accrue benefits—financial, cultural, or social—from having that access.
James Coleman (1990) defines social capital in terms of relationship and group norms. Groups that exhibit a high level of trust have more social capital because they are more likely to help each other. Putnam (2000) also popularizes the term in his book Bowling Alone and SNS researchers have utilized his ideas of bridging and bonding capital in recent studies. Putnam observes that diverse social groups provide bridges to new information and ideas, while homogenous groups most often offer bonding relationships based on social support. The diverse perspectives on social capital are worth noting because SNS scholars often evoke one or more of these definitions under the banner of social capital theory. Portes (1998) offers a more general definition that highlights the explicit conceptual link between SNS and the theory: “Despite these differences [in definitions], the consensus is growing in the literature that social capital stands for the ability of actors to secure benefits by virtue of membership in social networks” (p. 6).
Hypothetically, SNS have the potential to widen a person's social networks and provide access to valuable resources, information, and social support (Wellman et al., 1996). A series of studies with college students and Facebook test these particular social capital hypotheses. For example, Ellison, Steinfield, and Lampe (2007) find that higher Facebook use is positively correlated with bridging and bonding social capital in a sample of college students. The researchers also find interactions between Facebook use, measures of self-esteem (SE) and life satisfaction (LS). Participants who were low in self-esteem, but frequently used Facebook, had higher bridging social capital than their peers who were already high in self-esteem. The results suggest that college students who have low self-esteem or life satisfaction might benefit more from Facebook usage. Subsequent studies also find a positive relationship between Facebook use and social capital (Steinfield, Ellison, & Lampe, 2008; Valenzuela, Park, & Kee, 2009). For example, Valenzuela et al. (2009) examine a sample of college students in Texas and find that Facebook usage is positively correlated to life satisfaction, social trust, and civic engagement. However, the authors note that the relationships were small and conclude that SNS might not be the most effective means to develop social capital.
Almost no studies of SNS and social capital have considered adolescent youth, with most considering college-age users. However, emerging studies suggest that SNS may also connect younger teenagers to the broader community. For example, Ahn (2010) finds that high-school students who are members of Facebook and MySpace report substantially larger levels of social capital than their peers who are not members. However, beyond this exploratory evidence, further research is needed to understand whether and how youth participation in SNS connects them to their broader community. Questions of whether youth also develop different types of social capital (i.e., bridging and bonding) are fruitful avenues for study. Such research is particularly helpful because the benefits of social capital are numerous for youth. Children with more social capital appear to achieve higher academically, attend college at greater rates, and are less likely to drop out of school (Dika & Singh, 2002). If SNS increasingly mediate adolescent interactions, youth relationships with others in these online communities may prove to be a vital mediating variable for a variety of life outcomes.
Does Participation in SNS Affect Psychological Well-Being and Self Esteem?
Self-esteem and psychological well-being are the two most common outcomes of interest in prior Internet and SNS studies. Researchers typically measure self-esteem using established scales such as the Rosenberg Self-Esteem Scale (used in Ellison et al., 2007). Psychological well-being often refers to various measures that capture an individual's satisfaction with life. Scholars use a variety of scales that include measures of loneliness, depression, and overall life satisfaction (i.e., Kraut et al., 1998). A key debate among researchers considers whether higher use of the Internet affects one's self-esteem and psychological well-being (Kraut et al., 1998; Valkenburg & Peter, 2009a). Such Internet research informs how SNS researchers examine psychological well-being.
The often-cited HomeNet study by Kraut et al. (1998) recorded the number of hours individuals spent on the Internet (using tracking software on the participant's computers) and its relationship to future measures of social involvement and psychological well-being. The researchers found that longer use of the Internet was related to increased depression, loneliness, and smaller social circles. The results suggest that Internet use isolates individuals from their friends and family, and has a negative impact on one's psychological well-being. This effect is known as the reduction hypothesis (Valkenburg & Peter, 2009a).
After the HomeNet project, Internet studies exhibited a wide variety of findings concerning psychological well-being. For example, a longitudinal follow-up to the original HomeNet study found no long-term effects of Internet use on loneliness or depression (Kraut et al., 2002). Valkenburg, Peter, and Schouten (2006) note a major shortcoming of previous Internet research. Many of the studies treat Internet use as a one-dimensional activity. In reality, individuals use the Internet for many goals such as information gathering versus social interaction. In addition, prior studies often do not specify what activities might affect self-esteem and well-being, and why those specific activities might plausibly affect these outcomes. Binary specifications of whether a teenager uses a particular technology or not will likely prove to be an inconclusive predictor of self-esteem and well-being. Instead, media scholars are now moving towards finer definitions of the technological environment, activities within that environment, and theoretical specifications about why those interactions would affect social and psychological outcomes.
Current media studies that examine online interactions instead of broad Internet use generally find positive outcomes for youth. For example, Valkenburg et al. (2006) find that within a sample of over 800 Dutch adolescents, SNS use is related to self-esteem and psychological well-being. Adolescents who frequently use an SNS have more friends on the site and also more reactions on their profile (i.e., friends posted more comments and wall posts). In addition, the researchers report that having more positive reactions on one's SNS profile is correlated with higher self-esteem, and higher self-esteem is significantly correlated with satisfaction with life. The results highlight the emerging sense that use of SNS itself does not cause feelings of well-being. Rather, the positive or negative reactions that youth experience within the site are a key mechanism for their social development.
Why might earlier Internet studies report negative psychological outcomes, while recent studies find positive personal development? Valkenburg and Peter (2009a) observe two changes in Internet behavior that help explain recent, positive results of SNS. First, the authors contend that when prior studies occurred, “…it was hard to maintain one's existing social network on the Internet because the great part of this network was not yet online” (p. 1). In the late 1990s, one had fewer family members and friends online with whom to communicate. Past Internet applications such as chat rooms and forums were designed to facilitate conversation between strangers. The situation now is starkly different as teenagers and parents, youth and adults, all find themselves connected in SNS. Adolescents typically do not join Facebook to meet strangers. Instead, they join because their friends are already members and have invited them to participate. The Internet is no longer isolating, but connecting people.
The fact that youth frequently encounter known friends and family online underscores a second change in the Internet (Valkenburg & Peter, 2009a). Web 2.0 or social media applications are designed to facilitate interaction and communication through networks. Prior uses of the Internet primarily focused on an individualistic process of presenting or finding information. Information exchange still plays a prominent role in online communication. However, current tools make one's social network an explicit and visible resource from which to get that information. SNS, through the use of profiles and friend networks, enhance the ways in which people share information about themselves, their friends, and their lives. Again, the focus of Web 2.0 applications has been to connect persons rather than information.
Self-disclosure also plays a large role in SNS effects on well-being. Specifically, researchers posit that when youth disclose and express more information about themselves the quality of their relationships improves. These positive interactions lead to improved self-esteem and psychological well-being (Valkenburg & Peter, 2009a,2009b). This theoretical direction is directly related to scholarly thought in other frameworks, including signaling theory (Donath, 2007) and warranting theory (Walther et al., 2009). Future studies of SNS and youth must consider more detailed measurement of behaviors within the online community. These interactions—positive, negative, informative, or social—may then better predict outcomes of youth well-being.
Does Social Network Site Use Affect Student Grades and Learning?
Research on social networking sites and learning achievement is particularly slight when compared to studies of privacy, safety, social capital, and psychological well-being. To date, two studies exemplify the debate surrounding SNS, youth, and educational achievement. A conference paper by Karpinski (2009) received much media attention with findings that college Facebook users have lower GPAs than students who are not users of the site. Karpinski offers several hypotheses for these findings. For example, perhaps Facebook users spend too much time online and less time studying. However, the study did not rigorously examine counter hypotheses and remains a rather exploratory, basic attempt to understand the effect of SNS on learning.
Pasek, more, and Hargittai (2009) note several clear limitations of the Karpinski study. First, the sample of students is clearly limited. Second, the study utilizes few control variables in the analysis. And finally, Pasek et al. take issue with the liberal conclusions of Karpinski, namely, that the original study offers strong evidence for a negative relationship between Facebook use and grades. Pasek et al. offer three additional analyses that use a larger sample of undergraduate students, a nationally representative sample of 14–22 year olds, and a longitudinal dataset. The authors utilize more control variables including race, socioeconomic status, and previous academic achievement variables. From this analysis, the researchers find that Facebook usage has no significant relationship to GPA in any of their datasets.
The researchers in this debate suggest that the Facebook/GPA relationship is an interesting avenue for future studies. However, aside from the fact that many youth use Facebook, there appear to be no substantive theoretical reasons why Facebook use might influence GPA. As noted earlier, adolescents use the Internet for diverse communication and social goals. If perhaps a large percentage of youth interactions on Facebook were school-or academic-related, one might find a relationship to measures such as GPA. However, measurement of these communication patterns is lacking in the current literature and is a critical area for additional studies.
The work of new media literacy researchers provides one avenue to better specify behaviors that might lead to learning. Most studies of social media and youth education define learning from a literacy perspective (Greenhow & Robelia, 2009; Ito et al., 2009; Jenkins, 2006). The literacy perspective focuses on learning practices, such as creating media, rather than traditional measures of learning such as grades or standardized assessments. Hull and Schultz (2001) note that one major contribution of literacy scholars is to understand the concept of practices. Children's activities in school—i.e., listening to a teacher's lecture, practicing problems on worksheets, taking tests to assess their learning—can be seen as specialized literacy practices. Formal schooling is designed to teach students to perform well in those behaviors. However, literacy practices outside of school may serve very disparate functions than expected in the classroom. In the context of new technologies, youth today communicate and learn very different practices outside of school. Engaging in social networking interactions is a different literacy practice than successfully completing a multiple-choice test.
This direction is particularly fruitful to consider how youth's everyday practices with technology constitute learning in and of itself, and how these activities are in stark contrast to practices within school. Jenkins (2006) observes that youth today must be literate in several practices within social media environments. For example, he defines performance as the ability to adopt different identities for the purpose of discovery. Perhaps SNS, which are ideal identity building tools, can be used to aid students in exploring different characters, voices, and perspectives during the learning process. Jenkins characterizes appropriation as a skill to remix content from disparate sources to communicate ideas. SNS are environments that integrate numerous media tools, and could theoretically be applied to help students collect, synthesize, and remix content. He defines networking as the capacity to search for, integrate, and disseminate information. Similarly, SNS offer a natural environment to examine youth information practices.
The early studies of youth literacy with social media suggest that adolescents do in fact practice these skills. Ethnographic studies find that teens use social technologies to delve deeper into interest-driven communities and activities (Ito et al., 2009). Perhaps SNS provide a platform for youth to participate in communities that help them learn, and practice skills, within particular knowledge areas. Greenhow and Robelia (2009) examine the SNS use of 11 low-income youth and find numerous social behaviors that provide a theoretical link to learning outcomes. For example, students in their study use MySpace profiles to display creative work and receive feedback from their network. Youth report experiencing social support for school-related tasks, daily stresses, and problems. SNS help blend school and outside life for the teenagers in this study.
These ethnographic studies offer rich accounts of new and vital literacy practices among youth. Similarly, research on college-age youth find that they produce a tremendous volume of writing via tools like SNS, blogs, emails, and other social media environments (Fishman, Lunsford, McGregor, & Otuteye, 2005; Stanford Study of Writing, n.d.). For researchers of social media effects, these exploratory accounts of media practices provide a vital link to learning outcomes. Perhaps SNS that: (a) are used for particular educational means, (b) have strong academic cultures that are built within the online community, and (c) encourage particular information and social learning behaviors will lead to better learning outcomes. These are open hypotheses for social media scholars. This area is ripe for interdisciplinary studies that combine insights from literacy, media effects, and information perspectives. Ultimately, researchers interested in traditional academic outcomes such as high-school completion, academic engagement, grades, and test scores must specify what practices would theoretically improve these outcomes.
The research on SNS, social capital, and psychological well-being offers an additional link to student learning through the mechanism of academic engagement. The concept of engagement can be defined in behavioral, emotional, and cognitive terms (Fredericks, Blumenfeld, & Paris, 2004). Behavioral engagement refers to participation in academic, social, or extracurricular activities. Emotional engagement describes the positive and negative feelings students may have towards teachers, peers, and the broader school community. Cognitive engagement depicts the idea that a student is willing to expend the energy to comprehend difficult concepts and learn new skills. As noted in this review, much of the research on SNS suggests that as students more frequently interact with their network, they develop higher quality relationships with others. Education researchers who examine the social context of learning in areas such as out-of-school time, extracurricular activity, and classroom climate also find a link between high-quality relationships, students' academic engagement, and achievement (Eccles & Templeton, 2002; Feldman & Matjasko, 2005; Martin & Dowson, 2009).
A major hypothesis among education researchers is that youth participation in extracurricular and school activities increases their social connectedness with teachers and peers (Eccles & Templeton, 2002; Feldman & Matjasko, 2005). This connectedness is related to increased engagement with school and academics. Engagement has also been related to a lesser likelihood to drop out of school (Fredericks et al., 2004). These hypotheses are still major questions for education research. SNS offer a new context within which to observe how relationships influence school engagement, grades, and student achievement.
Researchers of SNS also have the ability to directly observe how online relationship networks may facilitate this social learning process. What interactions in SNS might a researcher expect to affect student engagement? Martin and Dowson (2009) offer some hypotheses culled from a variety of social learning theories such as expectancy theory, goal theory, self-determination theory, and self-efficacy. Expectancy theory and goal theory suggests that one's peers communicate which behaviors and goals are of value. For example, a student will value achieving good grades and set this as a goal, if his or her friends also strive for high achievement. Similarly, Eccles and Templeton (2002) also suggest that peer groups transmit a social identity that affects student behaviors. Self-determination theory proposes that if a student's psychological need to belong is met, he or she is much more likely to take academic risks, explore more ideas, and persist when presented with difficult work. Self-efficacy, a major part of Bandura's (2002) social cognitive theory, describes how capable one feels about accomplishing a task. When teachers, parents, and friends model the kinds of behavior that lead to academic success (i.e., study habits or information seeking), a student subsequently feels more capable about achieving success.
Martin and Dowson (2009) observe that high-quality relationships with adults, teachers, and peers impact these social learning mechanisms. These theories also highlight the educational impact of SNS. Quality relationships might allow students to feel more connected to school and thus take academic risks. Other peers might communicate what goals and behaviors are valued, through their status messages and wall posts. Finally, students might model positive academic behaviors by posting their behaviors or sharing information in SNS. These types of interactions begin to specify how relationship development in SNS may contribute to increased engagement and learning. Perhaps teachers can utilize SNS to engage their students, develop closer relationships, and model positive learning behaviors over time. Such educational hypotheses have yet to be tested in formal studies.
Finally, SNS researchers can learn much from past studies in television and adolescent learning. For example, Karpinski (2009) offers a possible hypothesis that Facebook users might spend less time studying, thus explaining their lower GPA. This idea is called the displacement hypothesis, and has been examined by early television researchers who posited that television took away students' study time (Hornik, 1981). Studies of students' extracurricular activities instead suggest that new media, such as Facebook, replace or enhance other leisure activities, but do not take away time from youth (Roberts & Foehr, 2008). The critical question for future studies is not whether youth use one technology or another, but what kinds of interactions and content they experience in these virtual settings.