The influence of social networking sites on high school students' social and academic development
iii Table of Contents
List of Tables
Chapter 1 :
Chapter 2: The Effect of Social Network Sites on Adolescents’ Social and Academic Development – Contemporary Issues and Research
Figure 2.1: Average Number of One’s Active Network
Chapter 3 :
Digital Divide and Social Network Sites
Students Participate in Social Media?
Chapter 4 :
The Effect of Social Networ k Sites on High School Students –
Cluster - Randomized Trial
Chapter 5 :
Future Directions for Research on Social Network Sites
iv List of Tables
Table 2.1: Digital Divides in Year 2000
Table 2.2: Overview of Empirical Studies on SNS
Table 3.1: Descriptive Statistics of Key Variables
Table 3.2: Cross Tabulations of Youth SNS Usage
Table 3.3 : Results of Binary Logistic Model on Probability of having a SNS Profile
Table 3.4 : Predicted Probability of Having a SNS Profile
Table 3.5: Internet Ownership and Internet Access for SNS Users
Table 4.1: Power Calculations when Classrooms are the Randomized Unit
Table 4.2: Required Classroom Clusters by ICC x MDE
Table 4.3: Characteristics of Classrooms in SNS and Control Groups
Table 4.4: Descriptive Statistics of Dependent Variables and Covariates
Table 4.5: Percentage of Missing Data on Dependent Variables
Table 4.6: Results of Random Effects Models for each Dependent Variable
This dissertation examines the effects of social network sites on youth social and academic development. First, I provide a critical analysis of the extant research literature surrounding social network sites and youth. I merge scholarly thought in the areas of Internet studies, digital divides, social capital theory, psychological well-being, identity development, academic engagement, and educational technology to understand how researchers might examine new social technologies and youth. Second, I examine the question of digital divide, or whether particular teenage populations do not have access to online social networks. Using a nationally representative dataset from the Pew Internet & American Life study, I explore whether there are disparities in teenage access to social network sites. Third, I report a cluster-randomized trial that was designed to explore whether social network sites have a beneficial impact when used in high school classrooms. A total of 50 classrooms, and nearly 1,400 students were randomly assigned to use an experimental social network site. The results highlight the challenges and potential of this technology when applied to school contexts.
1 Chapter 1: Introduction
As the world moves into the second decade of the 21 st century, one of the major markers of this era is the rise and use of online communities. In particular, a paradigm called Web 2.0 describes recent technologies that focus on networking mass numbers of individuals into distinct communities over the Internet (O’Reilly, 2007). Social networking sites (SNS) are online communities designed to connect individuals to wider networks of relationships, and are one major example of Web 2.0 applications. Sites such as Facebook have exploded in membership. In a short period of 2007 – 2010, Facebook estimates that its membership has grown from 50 million to over 400 million users (Facebook, n.d.). Online social networks are now an integrated part of daily life and compel questions of how these media platforms affect human development, relationships, and interaction. Teenagers are among the most avid users of technology in general and social network sites in particular (Lenhart, Madden, Macgill, & Smith, 2007b). Recent reports find that youth spend nearly 10 hours per day using some form of technology, with socially networked media playing a large role in their daily lives (Rideout, Foehr, & Roberts, 2010). New technologies are deeply intertwined with adult perceptions about teenage life. Mimi Ito and colleagues observe that, “Although today’s questions about ‘kids these days’ have a familiar ring to them, the contemporary version is somewhat unusual in how strongly it equates generational identity with technology identity” (Ito.
2 Baumer, Bittanti, boyd, & Cody et al., n.d.). The clear finding is that today’s youth are increasingly connected to the world through socially networked media. While teenagers are engaged with technology, they are ever more disengaged from another major component of their lives – school. National analyses find that nearly 30% of high school students do not obtain their diploma on time (Cataldi, Laird, KewalRamani, 2009). High school completion rates are difficult to measure, but various independent studies also suggest that nearly one-third of students ultimately drop out of school (Barton, 2005). When one compares these competing aspects of teenage life – technology versus education – a simple strategy clearly emerges. Perhaps if educators begin to integrate social technologies into learning, they will increase student engagement and achievement in school. Heeding the call of scholars (i.e. Jenkins, 2006; Ito et al. n.d.) recent policy and research efforts are now racing to develop new social media platforms and technologies for learning. For example, the Federal Department of Education and organizations such as the MacArthur Foundation have invested millions of dollars to build social media platforms, video games, and other digital tools for learning (Whitehouse, n.d.). Despite the optimism that social media tools might improve student engagement and learning, the stark reality is that these new technologies often conflict with the practices of K-12 schools. Surveys find that the vast majority of school district leaders believe social technology can improve student learning. However, these same district administrators typically block student access to online resources like social network sites (Lemke & Coughlin, 2009). The decision to ban students from accessing social network
3 sites underscores a major conundrum for educators. Online social networks widen a students’ access to resources and social support and may have beneficial effects on their development. Conversely, as student access to the world widens they are inevitably exposed to potentially negative material and interactions. The simplest strategy to limit liability and safeguard school districts is to ban access to these new digital tools. However, such policies neglect the potentially large benefits of using social media in the classroom. To alleviate this dilemma, educators and policymakers need a deeper understanding of social media and youth. Several questions are critical in the area of youth learning with social technologies, including: • Which youth are using particular social technologies? • How do they use these technologies to communicate, develop relationships, socialize, and learn? • What are the effects of these technologies on youth development? • What are the effects of these technologies when applied in educational contexts such as the classroom? In this dissertation, I explore these questions by examining a particular technology: the social network site. Communities such as Facebook and MySpace mediate teenage life, affecting how youth communicate and learn from one another. In addition, social networks are intertwined into just about every major online community today (Livingstone, 2008). These factors make SNS a particularly salient focus for evaluation. Throughout the following chapters I examine different questions surrounding the phenomena of social network sites and teenage youth. In Chapter 2, I review the extant
4 research literature that examines SNS. I consider several controversies around SNS and youth: (a) What kinds of youth are using social networking sites? (b) Does student participation in these online communities affect their privacy and social relationships? (c) Do student activities in SNS influence their personal development in terms of self-esteem and psychological well-being? (d) Does SNS use affect student grades and learning? The review highlights how research in this field is only just emerging. The few studies that examine social network sites are mainly exploratory. However, media researchers have a rich history of scholarship from which to draw new insights. I integrate previous thought on Digital Divides, Psychological Well-being, Social Capital Theory, and Cognitive and Social Learning theories to guide SNS researchers in future studies. In Chapter 3, I present an empirical analysis using a national dataset of teenagers from the Pew Internet & American Life Project (Lenhart et al., 2007b; Pew Internet & American Life Project, n.d.). In this study, I ask whether demographic variables such as education, socioeconomic status, and access to the Internet are significantly related to whether teenagers participate in social network sites. This line of analysis is typical of digital divide studies that examine whether particular populations have less access to new technologies. If new technologies do have positive benefits for individuals, but under- represented populations do not have access to such tools, there are tremendous issues of equity and access yet to be addressed (Jenkins, 2006). Most studies of digital divide and SNS examine adult and college-age populations. I present an analysis of teenage populations to examine their usage patterns. The results of this paper highlight how the association between demographic indicators and social media use are weaker in 2007
5 than seen in earlier studies. Teenage youth of all backgrounds increasingly find ways to connect with others using social network sites. In Chapter 4, I consider a question of particular importance to teachers and education leaders. Through a large-scale experiment, I examine whether using social network sites in urban classrooms has any causal effect on students’ social capital, engagement with school, or academic achievement. I build an experimental social network site that approximates the functionality seen in sites such as Facebook and MySpace. The key difference in this experimental condition is that the site is private to two urban, school districts and explicitly for use to exchange educational information. Working with 50 classrooms and nearly 1,400 students, I utilize a cluster-randomized trial, where class periods are randomly assigned to use the experimental site. Employing this randomized trial design, I find that an academic social network site does not necessarily improve student engagement with their peers, their classes, or increase student achievement. However, I find exploratory evidence that existing social network sites such as Facebook and MySpace improve students’ feelings of connection with their school community. The study offers evidence for one compelling idea: Perhaps schools should attempt to leverage students existing social networks, rather than block access to them or impose their own. In Chapter 5, I outline what is needed in future research about social network sites, and new technologies in general, to better inform the policies and practices of schools, educators, parents, and those interested in youth development. In particular, previous scholarly thought has focused on either a technologically deterministic or social
6 agency perspective. Technological determinism suggests that a media tool itself affects social outcomes such as learning, but a long history of research underscores the fallacy of this philosophy. Scholars who focus instead on social agency, explore how individuals use new technologies in cultural and social contexts. However, this stream of research neglects rigorous evaluation of how new media affect youth. Both perspectives in isolation offer incomplete analyses of how new media, such as SNS, impact youth. I argue that future researchers must develop and test finer hypotheses that simultaneously consider the technological affordances of social network sites, the social and cultural institutions within which SNS are used, and the actual interactions between individuals that occur in these online communities. The chapters in this dissertation examine the phenomena of social network sites and youth through different but complementary lenses: theoretical, descriptive, and experimental. The summative contribution of these analyses is a deeper picture of how teenage youth use SNS and its effects on their academic and social development. The studies show that youth of all backgrounds are increasingly connected via online social networks. The empirical analyses also show that social network sites are no silver bullet for improving learning in high school classrooms. The technology itself does not improve learning, but social media might help students become more connected and engaged with their school communities. The implications for educators and schools are numerous. Problems such as student disengagement with education are profoundly significant issues, and additional research is needed to better understand how online networks influence youth development and learning.
7 Chapter 2: The Effect of Social Network Sites on Adolescents’ Social and Academic Development – Contemporary Issues and Research
The current tools of teenage communication go by a peculiar set of names. Wall Posts, Status Updates, Activity Feeds, Thumbs Ups, Facebook Quizzes, and Profiles are some of the ways that youth today communicate with one another. These tools are features of social network sites (SNS), such as Facebook and Myspace. SNS are part of a suite of recent web applications, also called social media, which utilize Web 2.0 principles. The term Web 2.0 defines websites that are designed to: (a) rely on the participation of mass groups of users rather than centrally controlled content providers, (b) aggregate and remix content from multiple sources, and (c) more intensely network users and content together (O’Reilly, 2007). People use these web applications to interact in hyper-aware ways and the scale of this mass communication phenomena is significant. As of May 2009, Facebook ranked as the 4 th most trafficked website in the world and Myspace ranked 11 th highest (Alexa, n.d.). That high school youth are connected to these global online communities is both a frightening prospect for parents and educators and an intriguing area for social science research. Educators and parents in the United States face difficult quandaries concerning students and SNS. No one denies that youth use these technologies to communicate with the world, and they do so with high frequency and intensity (Lenhart et al., 2007b). Many scholars suggest that students learn in new ways using social media and that educators should embrace these new platforms (Ito et al., n.d.; Jenkins, 2006). In a recent national
8 survey, the vast majority of school district leaders report that they view social media as a positive development for education (Lemke & Coughlin, 2009). Nevertheless, 70% of districts also report that they banned all access to SNS in their schools. Despite the clear understanding that social media can be vital to student learning and digital literacy, educators currently struggle with how to comply with regulations like the Children’s Internet Protection Act (CIPA), as well as overcome general fears about student interactions in social network sites. To inform both the policy concerns of district leaders and the local practices of teachers and parents, research is needed to understand how youth use SNS and what effects it has on their social and academic development. In this chapter, I consider several key controversies around youth usage of SNS, and review relevant research that begins to inform these debates. I first define the media effects framework and outline how this research tradition attempts to understand the effects of new technologies on social outcomes. Second, I define social network sites and describe studies that capture how youth use these technologies to develop relationships, hang out with friends, and learn new skills. Third, the chapter reviews relevant research that informs several controversies concerning SNS and adolescents. I also connect these contemporary debates with previous scholarly thought about students’ out-of-school time (OST) and traditional concerns about the effect of technology on learning. The specific controversies reviewed are: • What kinds of youth are using social networking sites? • Does student participation in these online communities affect their privacy and social relationships?
9 • Do student activities in SNS influence their personal development in terms of self-esteem and psychological well-being? • Does SNS use affect student grades and learning? Finally, I outline the overall condition of research on SNS and youth. The current state of the literature is suggestive of the effects on adolescent social and academic development, and primarily consists of ethnographic and cross-sectional data. I outline the future questions that will be critical for the field and suggest relevant methodological directions to move this emergent research stream forward. What Can We Learn from a Media Effects Framework? Many of the controversial questions concerning social network sites ask what kinds of effects these technologies have on youth development. Given this focus, I work primarily from a media effects tradition of research. Media effects scholars examine the outcomes that arise when people use new technologies. Talking about effects engenders important theoretical discussions that must be laid clear when examining studies. Most significantly, the term implies a focus on causality. Studies in this framework imply that a media form, or the features of the technology, causally influences some outcome (Eveland, 2003). The structure of questions from this perspective is usually in the form of: Does media affect learning? Does television influence student achievement? Or do social network sites affect the psychological well-being of adolescents? Media effects scholars in a variety of fields have quickly come to realize that the answers to these
10 questions are more complex. Very rarely, if ever, is there a direct causal relationship between a technology and a social outcome such as learning (Clark, 1983; Clark, 1991; Schmidt & Vandewater, 2008). Early media questions often used a technological framework or object-centered approach (Fulk & DeSanctis, 1999; Nass & Mason, 1990). Such a perspective assumes and tests whether a technology itself causally affects a social outcome. For example, in Education a major question of technology research is whether media affects learning. Education researchers now firmly conclude that media does not affect student learning (Clark, Yates, Early, & Moulton, In Press). Numerous studies show that the media tool neither improves nor negatively impacts learning when compared to the same teaching strategy in the classroom (Bernard, Abrami, Lou, Borokhovski, Wade, Wozney et al., 2004; Clark, 1983; Clark, 1991). What matters is not the computer, but the learning behaviors that occur within the software or educational program. The findings of non-significant media effects on student learning do not mean that technology has no influence. For example, Richard Mayer (2001) shows through a series of experiments that the design of a multimedia presentation affects student learning of a topic. Putting words and pictures closer together on the screen, when they are relevant to each other, helps students retain more knowledge than when the elements are placed further apart on the screen. These results do not validate a technological orientation, where one expects that the computers themselves improve learning. Rather, the pedagogical strategy of placing relevant words and images together in a presentation affects cognition. Media researchers understand that the features of a technology afford
11 certain possibilities for activity. A multimedia video on the computer allows one to design words and images on the screen, while a computer simulation might guide a learner using models of real-world cases. A media tool allows for different possible learning behaviors (Kozma, 1991). This subtle difference in theoretical orientation is what scholars call an emergent perspective (Fulk & DeSanctis, 1999) or a variable-based approach (Nass & Mason, 1990). Scholars using an emergent or variable-based approach view technology as a structuring factor. Features of a technology, not the technology itself, enable and constrain how one uses that tool. Conversely, social forces such as cultural norms and behavioral practices influence how one ultimately uses a technology. William Eveland (2003) offers five characteristics of media effects research that help define how studies take into account both technological and social variables. Media effects studies have: (1) A focus on an audience, (2) Some expectation of influence, (3) A belief that the influence is due to the form or content of the media or technology, (4) An understanding of the variables that may explain the causality, and (5) The creation of empirically testable hypotheses. A focus on audience compels researchers to understand the characteristics of the youth who use SNS. Knowing who uses, or does not use, social network sites is an important sociological question for scholars of digital divide. In addition, Hornik (1981) notes the possible differential effects for disparate populations, “If communication researchers have learned anything during the previous three decades, it is that communication effects vary with members of the audience” (p. 197). Current media
12 studies also focus on the form or content of a technology, and move away from making black-box comparisons between technologies. Questions that ask whether Facebook is related to lower grades, or if MySpace is unsafe for children, are broad and uninformative directions for future media effects studies. Instead, the pivotal questions explore how the features of SNS enable or constrain behavior. Future media studies about SNS and youth should not frame questions using a technologically deterministic perspective where one expects the technology to cause an outcome. Instead, media scholars identify how youth interaction, communication, and information sharing are the critical variables in understanding SNS effects on social and academic outcomes. This understanding of media effects research helps define finer-grained hypotheses of why a tool like SNS might affect student development, under what uses, for whom, and when. What are Social Network Sites and How Do Youth Use Them? When a teenager joins a site like Facebook they first create a personal profile. These profiles display information such as your name, relationship status, occupation, photos, videos, religion, ethnicity, and personal interests. What differentiates SNS from previous media like a personal homepage is the display of one’s friends (boyd & Ellison, 2007). In addition to exhibiting your network of friends, other users can then click on their profiles and traverse ever widening social networks. These three features – profiles, friends, traversing friend lists – represent the core, defining characteristics of social networking sites. One will notice that SNS also include other media tools such as video and photo uploading and many websites now employ social networking features. For example,
13 YouTube is primarily a video sharing service, but users can add others as their friends or subscribe to a member’s collection of videos. Using boyd & Ellison’s (2007) definition, YouTube can be included as a type of social network site. As researchers examine the effects of SNS on social behaviors, they will undoubtedly come across these blurring of technologies. Sonia Livingstone (2008) notes that SNS invite “convergence among the hitherto separate activities of email, messaging, website creation, diaries, photo albums and music or video uploading and downloading” (p. 394). This convergence of technologies may complicate what one means by the term social network site. Amidst the sea of what websites can be termed SNS, the technical definition of social network sites still provides a shared conceptual foundation. Comparing across common features – i.e. profiles and friend networks – researchers can begin to understand how various communities co-opt these characteristics to create entirely new cultural and social uses of the technology. Patricia Lange’s (2007) ethnographic study of YouTube shows that users deal with issues concerning public and private sharing of video. Some YouTube users post videos intended for wide audiences, but share very little about their own identities. Their motivations might be to achieve Internet fame and gather viewers. Other members upload videos intended for a small network of friends and may restrict the privacy settings to only allow access to those individuals. The concepts of friend and social network for these users are entirely distinct. Dodgeball, an early and now defunct mobile-SNS, is another social network site that has been studied. In Dodgeball, a user broadcasts their location via cell-phone to their network of friends:
14 For example, when users get to a bar or cafe, they can "check in" by sending a text message to Dodgeball such as "@ Irish Pub." Dodgeball then broadcasts their location via text message to people in their Dodgeball network. Users can also be alerted when friends of friends who have checked in to Dodgeball are within a 10- block radius (Humphries, 2007, para. 5).
The case of Dodgeball highlights how the social network is utilized differently with this technology. In YouTube, individuals interact with their social network in ways that are influenced by the video-sharing focus of the site. In Dodgeball, the SNS technology is used to mobilize and meet up with friends in real-world spaces. The cases of YouTube and Dodgeball show how friends, friend networks, and profiles are used in very diverse ways depending on the individual’s motivations and the cultural norms of a particular online community. The early research on what is commonly understood as SNS (i.e. Facebook and MySpace) also explores how teenagers utilize profiles, friends, and friend-networks. The process of creating profiles has been a major focus of theoretical and empirical discussion. The common features of profiles include personal information such as one’s name, location, school affiliation, occupation, and personal interests such as favorite movies or music. Other vital components of the profile are pictures, videos, and the comments one’s peers leave on the page. Profiles can be updated at any time and some sites like MySpace allow individuals control as to how their profile looks. Using programming techniques, youth frequently apply “skins” to their MySpace profiles that completely alter the visual design or interface of their pages (boyd, 2008). Researchers have understood profiles as both an internal and external process of identity development. Some studies focus on how individuals display and experiment
15 with their identities through their profiles (boyd, 2008; Liu, 2007; Livingstone, 2008; Manago, Graham, Greenfield & Salimkhan, 2008; Schmitt, Dayanim, & Matthias, 2008). Other studies focus on how one’s network of contacts interpret and assess the profile (boyd, 2008; Walther, Van Der Heide, Kim, Westerman, & Tong, 2008; Walther, Van Der Heide, Hamel, & Shulman, 2009). The emerging picture is that youth make explicit decisions to disclose information about themselves on their profiles, and their networks provide social feedback to those profile displays. Most interactions in social network sites follow this cycle of self-disclosure and feedback. The process of displaying elements of one’s identity to the public in SNS is similar to how humans interact off-line. Donath (2007) observes that, “Whether face-to- face or online, much of what people want to know is not directly observable” (para. 10). She contends that much of human interaction consists of signals that communicate the status and characteristics of an individual. Donath utilizes signaling theory to examine how one’s self-presentation in SNS develops identity and trust with others. For example, when a user displays a contact as a “friend” he or she is – in an indirect way – vetting that that person is in fact who they claim to be. Thus, members who indiscriminately add any and all friend requests (including fake profiles or people they do not know) in an effort to seem popular may instead damage their credibility and trustworthiness to others. Among teenagers, boyd (2008) finds that “it is cool to have Friends on MySpace but if you have too many Friends, you are seen as a MySpace whore” (p. 129). In face-to-face settings, identity signals come in forms such as speech, body movements, facial expressions, and taste statements like fashion. On SNS, these signals
16 are written textually and visually on the personal profiles of members (boyd, 2008). Qualitative studies of adolescents and their media use suggest that (1) identity development is a major element of teenage life, and (2) youth use media such as profiles and personal pages to express themselves (Livingstone, 2008; Schmitt et al., 2008). Participation in social network sites also allows youth to participate in different social groups, and experiment with their identity displays (Manago et al., 2008). In their study of college students, Manago and her colleagues also note the possible negative aspects of identity experimentation. They find that the males and females in their sample often portray themselves in stereotypical, gendered ways. In addition, social comparison plays a large role in SNS. When profile and friend information is readily available, students frequently check in on their peers to see how they stack up. In addition to deciding what to present on a profile, friendship practices play a particularly large role in adolescent life. Teenagers need to navigate a variety of decisions including whom to accept as their SNS friend, and setting privacy controls to allow certain members to view their profile (boyd, n.d.). Studies suggest that the technical design and cultural norms of a respective SNS influences friendship behaviors. In YouTube, the motivations for sharing video influence the privacy controls a member sets on their media (Lange, 2007). Tools such as Dodgeball, that explicitly intends to facilitate off-line meeting up, have members whose real world networks more closely match their online contacts (Humphries, 2007). When MySpace introduced its Top 8 function, where users designated their top friends on their profile, it set off a firestorm of social drama amongst teens. boyd (2006) noted, “There are tremendous politics behind the Top 8, not