JOURNAL OF COMPUTER-MEDIATED COMMUNICATION Journal of Computer-Mediated Communication Ethno-Racial Identity Displays on Facebook Sherri Grasmuck Jason Martin Shanyang Zhao Department of Sociology Temple University The present study investigates self-presentation in a nonymous setting and explores differences in self-presentation by distinct ethno-racial groups. Based on content analysis of 83 Facebook profiles of African Americans, Latino, Indian and Vietnamese ancestry students, supplemented by 63 in-person interviews, we found that ethno-racial identities are salient and highly elaborated. The intensive investments of minorities in presenting highly social, culturally explicit and elaborated narratives of self reflect a certain resistance to the racial silencing of minorities by dominant color-blind ideologies of broader society. In the nonymous environment of Facebook, various dimensions of identity claims appear to be grounded in offline realities as revealed in interviews and observations of campus social dynamics. doi: 10.111II). 1083-6101.2009.01498.X Introduction Most early studies of the impact of the Internet on identity production focused on online identity constructions in anonymous environments such as MUDs (Multiuser Dungeons), Chat Rooms, and Bulletin Boards (Rheingold, 1995; Surratt, 1998; Turkle, 1995). It was found that individuals tended to play-act at being someone else or act out their underlying negative impulses in the online world. More recently, researchers have shifted their attention to self-presentations in less anonymous online environments such as Internet dating sites (Ellison, Heino, & Gibbs, 2006; Gibbs, Ellison, & Heino, 2006; Yurchisin, Watchravesringkan, & McCabe, 2005). The nonymity (similar to "identifiability") of online environments seems to make people more "realistic and honest" (Ellison et al., 2006) in their self-presentation. Research on nonymous online sites uncovered the display of more affirmative identities reflecting efforts to project a "hoped for possible" self, one that is more socially desirable, better than their "real" offline identity (Yurchisin et al; Zhao, Grasmuck & Martin, 2008). This research suggested that self-presentations varied according to the nature of the settings: People tend to "play-act" at being someone else in anonymous settings and be more "realistic and honest" in nonymous environments. 158 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association The possibility of presenting online identities different from offline expressions creates novel social opportunities for cybernet participants. Yet, identity construction on the Internet is influenced by not only the characteristics of the online environment but also the characteristics of users' social positions including race and ethnicity. Competing identities - racial, gender, sexual, national - are often context specific, mobilized depending on the circumstances. It is thus likely that differing expressions of social identities may interact with distinct online environments. However, most scholarly attention related to the effect of the online environment on the articulation of race has concentrated on anonymous environments. Moreover, many studies that have explored issues of social difference, such as race/ethnicity/gender and the Internet, have focused on the impact of the Internet on offline social dynamics. A contested issue raised in such studies is whether growing online encounters across social differences ameliorate, worsen or have minimal consequences for offline social inequalities (Rheingold, 1995; Turkle, 1995; Kendall, 1998; Tynes, Reynolds, & Greenfield, 2004; Byrne 2007). Unlike most previous research in this area, the present study is focused on the interplay between race/ethnicity (social positions) and online nonymity (online environmental characteristics) in affecting identity construction on the Internet. Through an empirical analysis of 83 Facebook profiles, supplemented by 63 in-person interviews, we seek to understand the social dynamics that underlie different ethno-racial identity displays in a nonymous online environment. Literature Background Popular views of the impact of the Internet on race dynamics, taking their lead from technological visionaries, conceived of the online world as a disembodied, Utopian space where anything can happen (Negroponte, 1995). In a space such as this, social differences, especially those based on bodies and visual signs of differences, like those of race and gender, would melt away. The "identity play" possible on the Internet would foster views and expressions of the self as multifaceted (Turkle, 1995) and help expose the way racial and gender identities are socially constructed. Aspects of visual identities that lead to discrimination in the offline world would matter little in the fluid identities expressed in the democratic, disembodied world of cyberspace (Turkle, 1995). Following early Utopian visions of raceless and genderless cybernet subjectivity, by the mid 1990s researchers began to pronounce race and gender to be alive and well in Internet chat rooms. These studies have tended to focus on either a) the effect of offline social hierarchies such as race, class and gender, on online identity construction or b) the effect of online environments, such as anonymity, on the construction of online racial identities. Scholars began arguing that the "racelessness" of many online encounters happens because minorities are often silenced or marginalized in cyberspace in ways similar to their marginalization offline (Bailey, 1996). Others pointed to the dominance of white male participants in online communities, to the gender specific form of communication in bulletin board conversations, to the Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 159 depoliticized nature of conversations about race, and to unquestioned assumptions around an implicitly white cultural center that structures digital discourse (Herring, 1993; Gómez-Peňa, 2001; Miller, 2001; Morse, 2001; Poster, 2001). Kendall's research on chat-oriented MUDs, a type of online forum, found that minority marginalization was a function of a dominant online climate where race topics and expressions were generally avoided and participants rarely discussed race directly. Instead race issues emerged only indirectly around discussions of social events with racial implications. Moreover, in online encounters participants focused on their ethnic, rather than racial, heritage, similar to dominant racial ideology in the U.S. (Winant&Omi, 1994). In this way participants reproduced and intensified in their online chat encounters many features of the dominant offline discourse of "colorblindness" (Frankenberg, 1993; Bonilla Silva, 2003) where race continues to affect social inequality, but cannot be acknowledged in interactions. Kendall (1998, p. 149) concluded that online interactions merely sidestep, rather than call into question, essentialized views of race. In a related vein, Bailey (1996) observed in disembodied text-based communities, such as Internet newsgroups, online forms, and MUDs, a "quiet consensus" about the value of racial anonymity online. Over time the rapid spread of the Internet with its established decentralizing tendencies ushered in dramatic changes in the presence of marginalized groups. Some scholars began noting a growing presence of minorities on the Internet especially apparent in an emerging subculture of sites devoted to specific concerns of racial groups (Castells, 2001; Byrne 2007). Racial topics moved from the shadows to center stage within online communities such as Usenet, a global electronic bulletin board devoted to topics or interests. The speed, anonymity, and diffusion of newsgroup debate has given a semipublic airing to topics commonly confined to safe, private conversations among friends (Bailey, 1996). Although research on black SNSs is still noticeably sparse, several studies on black Internet usage underscore the importance of community-specific context to black participation online (Byrne, 2007). These developments raise new questions about the range of identity constructions we can expect to find as minority groups find new spaces for virtual expression. Some scholars who have explored the enhanced presence of minorities in cyberspace have challenged early claims that online racial identities are more fluid and flexible expressions of offline identities. Burkhalter (1999) explored how racial identities are achieved, questioned and re-established in Usenet newsgroups. He found that not only were racial identities highly salient in online encounters but that racial stereotyping may be even more influential and potent on the Internet than offline. Far from limiting the expression of racial difference, online environments, especially those limited to textual exchanges, established a racial world online similar to the offline world. This rinding is consistent with Castell's point that as the Internet has become more widely dispersed, it has been appropriated less by fantasy and more by social practice (Castells, 2001). Similarly, Tynes et al. (2004), who examined online conversations about race and ethnicity in teen chat rooms, found that groups previously underrepresented and silenced were now a strong online force. 160 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association Much of the research that has explored racial identity production in the Internet has treated the online environment mostly as a disembodied and anonymous environment (Turkle, 1995, 2001; Bailey, 1996; Kendall, 1998; Burkhalter, 1999). An exception is the research of Nakamura (2002) who explored racial identity construction in the semiembodied cybernet environment of NetNoir's chat space, Club Connect, an important example of minority presence on the Internet. This research illustrated the way a default graphic environment could erase considerations of race and politics from racial identity production. Although there are still relatively few culturally specific online social network studies, several new studies point in that direction. Two good examples are Banks' study of how members use black orality for text communication on BlackPlanet (Banks, 2005) and Byrne's exploration of the low impact of networking on BlackPlanet on participants' offline civic engagement (Byrne, 2007). Kim and Yun (2007) also creatively explored the way Cyworld, a Korean social network site, permitted users to transcend the high-context communication of Korean culture by offering an alternative means for emotional communication online and in turn a reframing of relational issues offline. We still know relatively little about what it means to construct identity in environments where visual cues about race, through photographs for example, are offered to the audience. It is reasonable to expect that race will become a more salient factor in such nonymous and semiembodied online settings than in anonymous and disembodied online settings, but the contours of such online identity expressions have not been carefully researched. In addition, much of the research on race and the Internet is based on adult encounters or performances (Tynes et al., 2004). Similarly, Herring (2008) points to a tendency for scholars of the Internet to exoticize youth practices by focusing too much on the technologies themselves rather than talking to young people about their communicative needs. Past research has established the importance of early and late adolescence for the exploration of ethnic and racial identities especially for minority youth (Phinney & Alipuria, 1990; Tatum, 1997; Kibria, 2001; Min, 2001; Willie, 2003). Yet, we know little about how these heightened ethno-racial sensibilities of this age group are expressed in embodied environments or social networking sites. Moreover, the online racial performances of college students is especially fruitful terrain for enhancing our understandings of the evolving nature of late adolescence as it unfolds in a highly wired social milieu. Focus of the Present Study The present study seeks to extend the existing research on self-presentation in nonymous settings as well as to explore differences in self-presentation by distinct ethno-racial groups.1 We chose to examine identity construction on Facebook, a nonymous online environment that is less limiting than Internet dating sites. Created in 2004 by a Harvard University student for intracampus socializing, Facebook quickly spread to other university campuses and soon became the most popular social networking site among college students in the U.S. (Cassidy, 2006). Facebook Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 161 expanded to include high school campuses in 2005 and commercial organizations in 2006 (boyd & Ellison, 2007). Shortly thereafter, in September of 2006, Facebook expanded to include anyone aged 13 or older holding a valid e-mail address. Facebook is in many ways similar to another popular online social networking site, MySpace, but it also has some unique features that are crucial to the present study. Like MySpace, Facebook enables users to present themselves in a number of ways. Users can display pictures in their online albums, describe their personal interests and hobbies, and list their friends and social networks. There is also a communication function on Facebook that allows users to interact with one another through comments and messages. However, Facebook differs from MySpace in one important aspect: it is almost entirely a nonymous and (until 2006) "geographically bound community" (Ellison, Steinfield, & Lampe, 2007). If users wish to be a part of a major college or university network, which grants them access to other profiles on the same network (unless intentionally blocked), they are required to provide both an official campus e-mail account as well as their real name. This information is then automatically displayed in a user's profile. In other words, anyone is allowed to join Facebook, but it is the institutional connection that allows students from a given university to see other students' pages from that same university. The display and "searchability" (Lampe, Ellison, & Steinfield, 2006) of users' real names and their institutional affiliations make the Facebook environment for those who are part of a college or university network almost fully nonymous. The nonymity of Facebook is an ideal condition for examining identity construction in online environments where the relationships are anchored in offline communities (Zhao, 2006). Unlike Internet dating participants, who are primarily looking for romantic relationships among people previously unknown to them, Facebook users are looking for friendships as well as romantic relationships among mostly people they already know (Ellison et.al, 2007) but also people they do not yet know. This results in the logical potential for three types of relationship development: (1) relationship enhancement: deepening relationships with those whom the user already has an existing relationship; (2) relationship facilitation: bringing about a relationship with someone the user does not know but who shares a connection through a network of mutual friends;2 and (3) relationship creation: bringing about a relationship with someone the user does not know and with whom the user does not share any connection. Due to the mixing of these three different types of potential connections, Facebook allows users to engage in targeted performances by blocking certain viewers from viewing certain parts of their Facebook accounts. Just as people present themselves differently to different offline audiences, Facebook users tailor their online presentations to particular audiences. Building on the symbolic interaction tradition, we base our exploration of Facebook on a concept of identity that is fluid and context bound; identities develop and are sustained through real, and imagined, encounters with others. Goffman's dramaturgical perspective (1959) stresses the performance nature of our identities as we seek to enact positive images of ourselves to others, especially to others who matter. 162 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association One's sense of self is linked to membership in a range of social groups. Moreover, our composite identities often involve a hierarchical structure of who we are that can fluctuate according to context (Schlenker, 1985). Goffman's distinction between the "front stage" (where performances are enacted and organized, and impression management matters most because an audience is present) and "backstage" (where an audience is not present and subjects can "step out of character" without serious consequences) is particularly relevant to Facebook. Guerrero, Anderson and Affifi (2007) point out that Facebook pictures of self and others are often selected in private somewhat unguarded moments without an audience, or backstage, but are presented front-stage and viewed widely. The casual selection of photographs with close friends in mind may reveal more to a broad audience than a user would divulge in a face-to-face interaction where a sense of audience is more acute. Initial findings from our investigation of Facebook were consistent with research on Internet dating sites in the sense that Facebook users also tend to present their hoped-for possible selves rather than their "true" or hidden selves (Zhao, Grasmuck, & Martin, 2008). Facebook users emphasize or even exaggerate the parts of their possible selves that are socially desirable but not readily discernible in brief offline encounters, such as one's character, intelligence, and other important inner qualities. At the same time, they often seek to hide or de-emphasize the part of their selves they regard as socially undesirable, such as shyness, overweight, or stuttering. In so doing, Facebook users project a self that is typically highly socially desirable; "being popular among friends" was a claim that underlined many identity projects on Facebook. Additionally, there appears to be some unique features associated with the way in which Facebook users construct their "hoped-for possible selves." In making our claims we rely on self selected photographs as visual markers of identity. Autophotography, the use of self-generated or self-selected photographs, has been used by researchers as an analytical extension of "traditional verbal inquiries in the self concept" (Emmison and Smith, 2000). Self presentation in photographs has been reliably used to explore social shyness (Ziller and Rorer, 1985); gender differences in social connectedness (Clancy and Dollinger, 1993; Strano, 2008) and cross cultural differences in national identity (Ziller, 1990). Our detailed analysis of Facebook profiles revealed a continuum of strategies in identity constructions with varying frequency of use of different modes of expression (Zhao, Grasmuck, & Martin, 2008). First, on one end of the continuum there are the implicit visual claims demonstrated by photos and pictures uploaded by users as well as the "wall posts" of their friends. This is what we call the "self as social actor." Second, Facebook users frequently engage in an enumerative cultural self-description by listing a set of cultural preferences that they think define them, such as lists of consumption preferences and tastes such as movies, music and appreciated quotes. We call this the "self as consumer." Third, the "About Me" entry in the self-description section of Facebook permits users to offer the most explicit mode of identity claim by directly introducing themselves to users through a narrative self-description. We call this the "first-person self." On the continuum of identity claims, Facebook users apparently Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 163 prefer the most implicit method, with an almost universal selection of dense displays of profile photos and wall posts. Somewhat less preferable are the highly enumerated lists of cultural preferences associated with youth culture. Least popular are the more minimalist, first-person "about me" statements. In this way, we found that the visual possibilities of Facebook mean that users offer a mediated interaction to their audience, one that requires the audience to pay equal attention to the social milieu of the individual (my friends, my boyfriend, my music) as to the personal qualities of the Facebook users themselves. This mediated appeal, by "showing" rather than "telling," creates a triangular relationship of "mimetic desire" or interest between user, displayed friends/mates, and the audience (Zhao, Grasmuck, & Martin, 2008). This paper extends this line of research by raising several additional questions about modes of identity construction on Facebook favored by distinct ethno-racial groups of college students. The decades since the 1960s have seen a sharp rise in the numbers of racial minority students attending college, though the increases have been far from uniform across types of institutions. College campuses across America have been at the center of larger societal debates about racial equality, integration and racial representation in the curriculum. More recently, college campuses have witnessed greater racial heterogeneity related to dramatic increases in immigration post-1965 legal changes, especially in the numbers of Latino and Asian second generation immigrant students (Anderson, 2002). The growing presence of "in-between minorities" who do not neatly fit into the dominant black-white racial divide further complicates the negotiations around race and culture on many multicultural campuses. Many scholars have pointed to persistent racial segregation within university settings, sometimes referred to as "balkanization," as students strive to express their racial identities and find their place amongst their own racial groups (Takagi, 1992). Given the importance of the college years for the exploration of ethnic and racial identities and the salience of racial identities in the changing racial climate of many college campuses, Facebook, embedded in college settings, is a potentially rich site for enhancing our understandings of ethnic-racial identity construction and display. The interface of Facebook makes some choices for users unavoidable while others are unavailable. Facebook permits users to specify many dimensions of their identities: sex, sexuality ("interested in"), political and religious views, relationship status, etc. However, conspicuously absent among the dimensions of identity is race. This makes judgments about racial identities mostly a matter of visual display via pictures. Ginger (2007) has argued that by not making race an available category for users, Facebook reinforces the dominant colorblind mentality and perpetuates the racialized visual classification of others. Nonetheless, the question of how Facebook users respond to the environmental restrictions of Facebook remains. Recently researchers have also explored how online interactions reflect or interface with offline encounters (Lenhart & Madden, 2007). Donath and boyd (2007) concluded that most social networking sites primarily support pre-existing social relations. Specifically, one study reported that participants on one campus mainly 164 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association use Facebook to maintain or solidify existing offline ties rather than to create new ones or connect with strangers (Ellison et al., 2007). We are beginning to accumulate information on the differential access and use of Facebook among different constituencies of its users (Zywica & Danowski, 2008; Tong, Van Der Heide, Langwell & Walker, 2008). One study found that students who live at home are considerably less likely to use Facebook than those who live with roommates or on their own (Hargittai, 2007). In addition, Hispanic students are more likely to use MySpace than are whites while Asians and Asian-Americans are more likely to use Xanga and Friendster than whites, possibly related to their greater popularity in Asia (Hargittai, 2007). Ellison et al. (2007) found significant differences in use of Facebook between whites and "nonwhites" but offered no speculation about the meaning of those differences owing to small numbers of nonwhites in the sample. Descriptive findings from a new dataset on Facebook found notable differences in the size, density and heterogeneity of Facebook networks of various ethno-racial groups of students. Specifically, after controlling for gender, SES, and online activity, the Facebook networks of black and Asian students far exceeded those of white students in size as well as heterogeneity. (Lewis, Kaufman, Gonzalez, Wimmer & Christakis, 2008). How distinct ethno-racial groups may use Facebook differently to build or maintain social capital remains largely relatively unexplored in the literature. Several questions guide our exploration. First, how invested in identity production are the different ethno-racial groups? Facebook is a social networking site grounded in educational institutions linked to upward social mobility. Historically stigmatized groups may have more to gain from the social connections made in filtered online environments. Therefore, we expect to find higher levels of investment in identity production among the minority students compared to the white students. Second, do students downplay or assert their racial identities on Facebook? Given our previous findings that Facebook users construct highly social and positive presentations and tend toward the implicit, visual, consumer, and less-narrated self, does this pattern hold consistently across distinct racial/ethnic groups? Alternatively, is the story more complicated such that distinct ethno-racial groups assert their identities differently? Given the heightened significance that ethnic and racial identity often takes on during the college years, we expect to see racial and ethnic dimensions as salient in the identity performances of minorities. Third, given the fact that Facebook is nonymous, where users can potentially have their claims challenged in offline encounters with fellow students, how realistic are Facebook claims? Is the range of identity claims made by distinct ethno-racial groups reflective in any way of their offline claims or of broader racial dynamics in the offline campus milieu? To return to CastelPs contention that the Internet increasingly merely appropriates social practice, do the identity performances of different ethno-racial groups reflect social practices and divisions on campus? Here we expect to find that the Facebook identities performed by the different ethno-racial groups will reflect "hoped for possible selves" or enhanced expressions of their offline selves as well as reflect the broader racial dynamics of the offline campus environment. Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 165 Methods This study is part of a broader research project on immigrant and racial/ethnic identity, friendship, courtship, and sexuality at a racially diverse urban public university in a large Northeastern city in the United States. The broader project focused on four racial minority groups, three of whom are second-generation immigrants: Vietnamese-Americans, Indian Americans, Latino-Caribbean (of Dominican, Colombian, or Puerto Rican backgrounds), and African Americans. Given the considerable socioeconomic stratification of immigrant groups in the broader American context, we included two Asian immigrant groups, highly dissimilar in terms of levels of education, income, and occupational profiles at the national level as well as cultural backgrounds. Indian immigrants in general have socioeconomic profiles above the native American population while Vietnamese immigrants' levels of education, income and poverty rates are near U.S. averages, but significantly below those of Indian immigrants (Leonard, 2007, p. 460; Rumbaut, 2007, p. 665). The research methodology included same-ethnic focus groups, interviews with administrators involved in student services regarding organizational and campus life, structured in-depth interviews with students from the four ethno-racial groups as well as analysis of these students' Facebook profiles. The data set is unique in Internet research in that, for the majority of the respondents, it permits a comparison of the results of quantitative coding of online profiles with the corresponding users' subjective interpretations of their own activities. The majority of the study subjects came from a list of students having responded to a comprehensive student survey conducted by the university administration (National Student Survey). In-depth interviews were completed with 63 students who met selected criteria.3 This included approximately 8 males and 8 females, or a total of 16 from each of the four groups with one honors student represented within each cluster of 8 same-sex groups. The interview guide contained questions related to ethnic identity, high school social life, perceptions of and involvement in campus social life, friendship networks, dating and partnering experiences, as well as Internet usage and Facebook activities. The present study of Facebook is based on an enlarged sample of 83 students. For comparative purposes, we decided to supplement our original sample of 63 nonwhite interviewees with 20 additional cases of white students randomly selected from the respondents of the aforementioned National Student Survey. Of the 83 cases, 11 students did not have Facebook accounts. The Facebook pages of the remaining 72 students were downloaded in March 2007. Of these 72 Facebook accounts, seven were entirely blocked except for the profile cover pictures.4 In addition to the blocked accounts, two other accounts were devoted exclusively to group activities, leaving us with a total of 63 analyzable Facebook accounts for our study. Table 1 shows the racial and gender compositions of both the initial sample (n = 83) and the resulting sample (n = 63). The social backgrounds of the sampled students reflect distinct socioeconomic origins. The two Asian groups provide the sharpest contrasts in class 166 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association Table 1 Ethnicity and Gender Composition of the Sample Initial sample Resulting sample Gender Gender Male Female Total Male Female Total African American 8 8 16 8 4 12 Latina/o 8 8 16 7 7 14 Indian 8 9 17 7 6 13 Vietnamese 6 8 14 5 4 9 White 10 10 20 7 8 15 Total 40 43 83 34 29 63 Source: Project Data, 2007. backgrounds, with almost 60 percent of Indian American students falling in the high social class category and 64 percent of the Vietnamese falling in the lower class group. The Caribbean-Latinos predominantly come from either lower or medium (social class backgrounds while the African Americans are the most evenly distributed among the three class categories.5 We developed a comprehensive coding scheme covering virtually all the major items that can be found in a Facebook account (see Appendix I). Using this coding scheme, the contents of the resulting 63 Facebook accounts were numerically recorded.6 For the purposes of the present study, we focused our analysis on the following sections of a Facebook account: user's profile, social networks, and self-description. Facebook automatically includes the user's real name and the name of his or her university in the profile section. The profile section also optionally includes a profile cover picture, additional photos of the user, and the information about the user's sex, birthday, relationship status (single, engaged, married, etc), and the relationship the user is looking for (friendship, dating, random play, etc.). Users can choose to disclose their e-mail addresses, IM screen names, mobile/land phone numbers, and residential locations in the contact information section. The social network section lists users' Facebook friends from both in and outside of their universities, as well as all the groups they have joined. In the self-description section, users can provide an "About Me" blurb, state their hobbies and interests (e.g., activities, books, music and movies), and share their favorite quotes. Thus, all users must make a range of formal decisions in launching their accounts. As it were, Facebook offers an excellent opportunity for studying identity construction in a nonymous online environment. Different social environments can trigger the expression of distinct dimensions of one's various social identities. It is important therefore to understand the campus environment where identity constructions on Facebook occur. Every university has its particular campus culture, traditions, and social environment. Victory State (a pseudonym), the site of this research, is no exception. Victory State is a public Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 167 university with more than 30,000 enrolled students located in a predominantly black inner-city neighborhood. The annual guide, The Best 361 Colleges, published by The Princeton Review (2006), ranked Victory State over a number of years among the most diverse universities in the U.S. giving it the moniker "Diversity University."7 The undergraduate racial/ethnic makeup of the university at the time of the study was 58 percent white, 19 percent African American, 9 percent Asian, 3 percent Hispanic, 9 percent "other," and 3 percent international students. Students at Victory State, like campuses across America, are highly conscious of the existence of racial boundaries among social groups and organizations on campus reflected in highly visible homogenous racial clusters of students in campus public spaces, the highly racialized nature of student organizational life, and the uneven distribution of racial groups across majors and colleges. This racially diverse campus environment, where significant numbers of students construct student organizations around their racial and ethnic identities, is the institutional context of the Facebook users of this investigation. It should be noted that while Victory State as an ethnically mixed large university is an appropriate site for the study of ethno-racial identity displays, some aspects of student life on this campus may not be representative of student lives on other U.S. campuses, and, as a result, caution must be exercised in extrapolating the broader implications of the findings of this study. We will come back to this important issue in the concluding section of this paper Results8 Degrees of Investment in Identity Construction of Distinct Ethno-Racial Groups Our findings indicate that racial/ethnic groups use different strategies in identity constructions on Facebook. Examining the continuum of strategies on Facebook, from visual and implicit ones, to the more enumerated and narrative techniques, we find consistent differences across the five comparison groups in frequency and intensity of use of each mode of expression. While three ethno-racial groups (the African Americans, the Latinos, and the Indian ancestry students) revealed more intensive investments in the production of all three modes of identity construction, the other two groups (the White students and Vietnamese ancestry students) showed less involvement. We will illustrate this by examining each mode of identity strategy separately. First, the most implicit, visual identity claims involve the display of photos and pictures uploaded by the users themselves or the pictures along with the "wall posts," comments posted to their accounts by others: "the self as social actor." Apart from the cover picture, users can show within their Facebook profiles as many photos of themselves as they wish. In our sample, while the number of additional profile pictures displayed by all users averaged 88 (median = 63) photos per user, three groups displayed notably higher averages: 93 for African Americans, 112 for Latinos and 110 for Indian ancestry students.9 In contrast, both Vietnamese ancestry and 168 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association Table 2 Self as Social Actor Mean "Popularity" Measures by Ethno-Racial Groups (Number of Cases in Parentheses) Photos Posts** On-campus Facebook "friends" Off-campus Facebook "friends"** African American 93 (12) 354 (12) 218 (11) 94 (11) Latina/o 112(14) 285 (12) 204 (14) 106 (14) Indian 110(12) 392 (13) 158(13) 124(13) Vietnamese 56 (8) 96 (7) 60(7) 52(8) White 61(14) 102 (13) 86 (15) 66 (15) Total 88 (60) 257(57) 150 (60) 90 (61) **ANOVA tests: p < .05. Source: Facebook Project Data, 2007. white students averaged only 56 and 61 photos respectively (Table 2). Similarly, in terms of the average number of "wall posts," the visible pictures or comments uploaded by users' friends, we see a notable variation among the groups with the same three groups among the seemingly "most social." Indians displayed an average of 392 posts, African Americans an average of 354 posts, and Latinos an average of 285 posts, compared to only 102 for whites and 96 for the Vietnamese. Scanning a profile online where more than 300 pictures of friends are displayed, typically smiling and often expressing affection, one is hard pressed not to be impressed by the depth of the social world of the Facebook user. However, in contrast to the pictures uploaded by users, wall posts are initiated by others and more directly reflect levels of sociability rather than identity performance per se. Nonetheless, users do have the option of removing such posts and passively leaving them signals something about oneself. The fact that most of these pictures and wall posts are not "blocked" by the user but rather visible to a general audience may illustrate merely a lack of concern about privacy or passivity but simultaneously a highly performative expression of identity is accomplished. Similar claims to dense social networks and popularity are evident in the section of Facebook "friends." The basic visibility rule of Facebook is that all the user's schoolmates can see the user's Facebook account unless blocked by the user. Those who are outside of the user's institution and not on the user's friends' list cannot view the user's account but they can find the user in the search results and are able to see the user's cover picture, view the user's friends, send the user a message or request to be a user's "friend." Users may choose to accept or reject requests to become a friend. Facebook prominently displays the total number of friends a user has both from the local institutional network as well as accepted friends from outside the network—a kind of public popularity proclamation. The first and third columns of Table 5 reveal the average number of on-campus and off-campus Facebook friends displayed by each ethnic/racial group. African Americans and Latinos surpass all other groups with an average of 218 and 204 on-campus friends, respectively, followed Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 169 by 158 for Indian American students, 60 for Vietnamese, and 86 for whites. This same pattern holds true for off-campus friends, with African Americans, Latinos, and Indians having approximately twice the average number of friends as do the Vietnamese and white students. Not surprisingly all groups display lower average levels of off-campus Facebook friends than local ones, though the ratio of off-campus to on-campus friends is higher for both the white and Vietnamese students, signaling that their lower local investments in networking may be partially offset with more off-site networking. In summary, the African Americans, Latinos, and Indian ancestry students project a visual self that is dramatically more social, in terms of the depth and extent of their ties, than do the Vietnamese or white students. This is reflected in photos, wall posts, and local and nonlocal Facebook friends. These online findings of ethno-racial differences may well reflect aspects of the campus environment, an issue we will return to later in the paper. The second type of identity claim on Facebook reflects a cultural self, the self of consumption preferences and tastes. Facebook users engage in enumerative cultural self-description when they list a set of cultural preferences they believe define them. There is a self-description section where users can tell viewers about their personal interests and hobbies including their favorite activities, quotes, movies, music, books, and TV shows. Table 3 shows that most Facebook users provide highly elaborated lists of such preferences, thereby signaling their precise cultural tastes. A large majority offered at least one appreciated quote (71.4%) and shared their favorite music and movie favorites (65.1% for both) with almost half listing favorite television programs (47.6%). However, ethno-racial groups invest differently in their displays of "the cultural self." African Americans and Latinos invest most intensively, Vietnamese and whites least intensively, and Indians fall somewhere in the middle of the continuum depending on the enumerated item. While all African American students (100%) and 78.6 % of Latinos opted to share their favorite music, fewer Indians (46.2%), Vietnamese (55.6%), and whites (46.7%) did so. African Americans and Latinos also enumerated a higher number of songs/artists on average (9 and 11 for African Americans and Latinos, respectively) than Indians (7), Vietnamese (5), and whites (6). Similarly, African Americans surpass all other groups in their average rates of sharing favorite movies (91.7%) and TV shows (75.0%). They enumerate an average of six favorite TV shows compared to an average of four for all groups. Latinos offer the most elaborated lists of movies of all the groups, offering an average of 12 favorite movies, compared to an average of eight for the other groups. With respect to favorite quotes, while all African American students (100%) and 78.6% of Latinos include a favorite quote, fewer Indians (53.8%), Vietnamese (66.7%) and whites (60.0%) did so. Also, among those students who did select this option, African Americans, Latinos, and Indians offered much more elaborated quotations, with more average lines of quoted text, or an average of 18,16, and 13 lines of text, respectively, compared to an average of only three to four lines offered by the Vietnamese and white users. The identity claims expressed in these consumption preferences represent a cultural self, somewhat more explicit than the claims of the "watch me" implied by the photos 170 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association Table 3 Self as Consumer Percent and Mean "Personal Interests" Measures by Ethno-Racial Groups (Number of Cases in Parentheses) Music Movies TV shows Quotes Percentage Mean items Percentage Mean items Percentage Mean items Percentage Mean lines Responded** listed responded listed responded listed* responded of type African American 100.0 (12) 9 91.7(11) 7 75.0 (9) 6 100.0 (12) 18 Latin a/o 78.6 (11) 11 64.3 (12) 12 35.7 (5) 3 78.6 (11) 16 Indian 46.2 (6) 7 61.5 (8) 9 46.2 (6) 4 53.8 (13) 13 Vietnamese 55.6 (5) 6 55.6 (5) 8 55.6 (5) 3 66.7 (6) 3 White 46.7 (7) 6 53.3 (8) 5 33.3 (5) 4 60.0 (9) 4 Total 65.1 (41) 8 65.1 (8) 8 47.6 (30) 4 71.4 (45) 13 *ANOVA tests: p < .1; **Chi Square Tests: p < .05. Source: Facebook Project Data, 2007. and posts, but still indirect: "see my music, my books, my shows, and know me." While there are relatively high levels of investment by all of these minority groups in displaying this cultural self, African Americans and Latinos invest more frequently and somewhat more intensively in expressing their taste identity than do Indians, Vietnamese, and whites. The third mode of identity claim on Facebook involves the most explicit verbal descriptions of the self. In the "about me" entry in the self-description section users can engage in a narrative self-description whereby they directly introduce themselves to their viewers. As shown in Table 4, African Americans (100%) and Latinos (85.7%) far exceeded the Indian (53.8%), Vietnamese 44.4%), and white (46.7%) students by opting to narrate something about themselves. Compared to the other types of claims, this category, the narrated self, tended to be the least elaborated of the three identity strategies; 54.8 percent (n = 23) of the 42 users who provided an "about me" blurb in the Facebook accounts wrote just one or two short sentences. Some typical examples of these brief "about me" statements are: "That is for me to know and you to find out" (African American male); "Busy with stuff' (Indian male); "What do you want to know??" (Vietnamese female); or "My right leg is 1.5 mm shorter than my left leg" (white male). Most of the "about me" statements are notably minimalist in their revelatory claims, reflecting a general preference for the more visual and cultural types of identity claims. Nonetheless, similar to the above findings, Latinos invested more intensively in making identity claims via first person narrations of themselves to the broader audience of Facebook users (with fully 83.3 percent writing longer responses compared to an average of 45.2 percent for the other groups). Typical among the more elaborated responses of Latinos were: [Accompanied by a design of a flag of the Dominican Republic produced with typed characters] "That's all you need to know! But! if you need more, here ya go: Outgoing personality, Like to have fun, down to earth, very open minded, Love Baseball (YANKEES ARE STILL THE BEST BY THE WAY!!!); Love Table 4 Narrative Self "About Me" Responses by Ethno-Racial Groups (Number of Cases in Parentheses) Users responded (%)** Short response (%)** Long response (%)** African American 100.0 (12) 75.0 (9) 25.0 (3) Latin a/o 85.7 (12) 16.7(2) 83.3 (10) Indian 53.8 (7) 71.4 (5) 28.6 (2) Vietnamese 44.4 (4) 50.0 (2) 50.0 (2) White 46.7 (7) 71.4 (5) 28.6 (2) Total 66.7 (42) 54.8 (23) 45.2 (19) **Chi Square Tests: p < .05. Source: Facebook Project Data, 2007. 172 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association architecture and making it my profession. Love watching movies at the theatre or at home and dancing all night. *R.I.P.* Arnold Boyd *R.I.P.* Daria Chentsova." (Dominican male) The Indian students, however, do not here cluster with the African American and Latinos in that they are more reluctant (44.4%) to narrate something about themselves and, when they do, they offer minimalist responses more typical of the Vietnamese10 and white students. In summary, the African Americans, Latinos, and Indian ancestry students project a visual self that is dramatically more social than do Vietnamese or white students. African Americans to some extent and Latinos in particular also stand out in their greater willingness to invest in the more direct "about me" narrations than do the other three groups. The Salience of Racial Identities Among Ethno-Racial Groups Beyond the intensive involvement in identity production and the high levels of "social traffic" projected by some minority groups on Facebook, we explored the content of identity claims of each group. There were great variations in the kinds of self-images produced on Facebook. Some self-images were carefully choreographed and well polished, others were simple and rough. Regardless of level of sophistication, Facebook users in our sample all attempted to project a self that is socially desirable (popular, well connected and well rounded) (authors 2008). However, the groups differed in the degree to which they expressed clear ethno-racial identities. Quoting someone else was a frequently used strategy for describing oneself on Facebook. A qualitative analysis of the favorite quotes section revealed group differences in the types of identity claims. While all ethno-racial groups tended to use these indirect expressions of identity often taking the form of uplifting reflections on life, each group tended to employ a distinct style. African Americans, Latinos and the Indian ancestry students stood out in the types of identity claims they made, revealing more group-oriented, and racially conscious identities often tinged with religious or spiritual themes. They also invested much more heavily in elaborating their quotations with longer and more numerous entries than white or Vietnamese students. The quotations selected by African Americans were highly infused with inspirational quotations about racial injustice from literature and popular culture as well as religiously themed quotes. One young man included fully 64 lines of text with quotes from Tupac, Kanye West, Michael Jordan, and Jay-Z among others, including the following one by Muhammad Ali listed here among a sample of typical inclusions by African American students: 1. "Champions aren't made in gyms. Champions are made from something they have deep inside them - a desire, a dream, a vision. They have to have last-minute stamina; they have to be a little faster; they have to have the skill and the will. But the will must be stronger than the skill." Muhammad Ali (African American male) 2. "Blood of a slave, heart of a king" (African American male) Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 173 3. "Black Love is Black Wealth" Nikki Giovanni (African American female) 4. "Our deepest fear is not that we are inadequate. Our deepest fear is that we are powerful beyond measure. It is our light, not our darkness that most frightens us. We ask ourselves, Who am I to be brilliant, gorgeous, talented, fabulous? Actually, who are you not to be? You are a child of God. Your playing small does not serve the world. There is nothing enlightened about shrinking so that other people won't feel insecure around you. We are all meant to shine, as children do. We were born to make manifest the glory of God that is within us. It's not just in some of us; it's in everyone. And as we let our own light shine, we unconsciously give other people permission to do the same" Marianne Williamson (African American female) Like African Americans, the Latino students mixed religious and inspirational quotes, often referencing ethnic themes or alluding to their Latino origin by using untranslated Spanish quotes. As illustrated in the first quotation below, one young man incorporated extensive quotes from Fidel Castro, Herbert Marcuse, Foucault, Sartre, and Nietzsche extending over 99 lines of type focusing on themes of political resistance, social change, and revolution. The other three quotes give a flavor of some of the ethnic and religious references of many others. 1. "By virtue of the way it has organized its technological base, contemporary industry tends to be totalitarian. For totalitarianism is a terroristic political coordination of society in a nonterroristic economic technical country which operates through the manipulation by vested interests." Marcuse (Colombian-American male). 2. "Cooooono! [Dominican expletive] Dejame dormir!!" (Dominican male) ["Let me sleep!" our translation] 3. "No hay que llorar, que la vida es mas bello vivir cantando." (Colombian female) ["No need to cry. Life is more beautiful if lived singing," our translation of unattributed quote ofCelia Cruz, Cuban singer/song writer.] 4. "God's delays are not his denials." (Puerto Rican female) Indian students also tended to incorporate long entries or multiple quotations involving thoughtful and sometimes humorous themes. Though somewhat less inclined to include ethnic-specific quotations than African America or Latino students, ethnic pride was often signaled through quotations from Mahatma Gandhi. 1. "Live as if you were to die tomorrow. Learn as if you were to live forever. Gandhi"(Indian male) 2. "To laugh often and much; to win the respect of intelligent people and the affection of children; to earn the appreciation of honest critics and endure the betrayal of false friends; to appreciate beauty; to find the best in others; to leave the world a little better; whether by a healthy child, a garden patch or a redeemed social condition; to know even one life has breathed easier because you have lived. This is the meaning of life." (Indian female) 174 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 3. "Unity to be real must stand the severest strain without breaking. Mahatma Gandhi" (Indian female) 4. "C is for cookie... and cookie is for me. Cookie Monster" (Indian female) In contrast to the salient expressions of racial consciousness and pride revealed above, white and Vietnamese students almost never selected quotes that signaled racial or even ethnic identification. Vietnamese students favored a few short, one-liner quotes. They often echoed the theme that it is both important to accept your life while making the most of it. However, there were sometimes undercurrents of resentment about social exclusion in the selected quotes especially those of males: 1. "Time may change me but I can't time" (Vietnamese female) 2. "I'd rather make a life, than make a living" (Vietnamese male) 3. "To the bastard, talking down to me; Your whipping boy, calamity; Cross your fingers, I'm gonna knock it all down." Third Eye Blind-Graduate (Vietnamese male) 4. "If they think they're the shit, treat them like it."(Vietnamese male) When white students included a quotation, it tended to be a "raceless one-liner," often drawn from popular culture with little signaling of identification with a particular white ethnic group, such as Italian or Irish heritage, as shown below. 1. "Too much of a good thing can be wonderful" Mae West (white female) 2. "Nobody realizes that some people expend tremendous energy merely to be normal"(white male) 3. "Did anyone ever tell you that you look like a penis with a little hat on?" Tom Hanks (white female) 4. "I do not intend to tip-toe through life only to arrive safely at death" Christina Aguilera (white female) The uplifting and often inspirational quotes related to racial themes of injustice frequently included by the African American, Latino and Indian students convey a sense of group belonging, color consciousness, and identification with groups historically stigmatized by dominant society. Such modes of identity production reflect a certain resistance to the racial silencing of minorities by dominant colorblind ideologies of broader society where direct referencing of race remains taboo. They often seem to be talking to other members of their own ethno-racial group, signaling solidarity, rather than out-group observers. Group consciousness or ethnic pride was less often expressed by the Vietnamese students, though themes of social exclusion did appear. The selected quotes by white students rarely signaled group identification or ethno-racial themes, reflecting "strategies of racelessness" (Willie, 2003) more typical of research on white students in offline contexts (Bonilla-Silva, 2006). Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 175 Correspondence Between Online Presentations And Offline Accounts of Friendship How close are the Facebook identities constructed for their audiences in this nony-mous environment to the identity claims users make in the shared offline world of the campus? For example, do the high rates of social traffic for three minority groups, evidenced by high numbers of Facebook friends, reflect extensive offline social networks? Are the implicit Facebook claims of popularity real? Because our project included interviews with all the Facebook users, with the exception of the white students, we may compare some of the information reported during the interviews with Facebook claims. On Facebook, our sampled students displayed an average number of 150 on-campus friends and 92 off-campus friends (Table 5). In our interviews, we asked a similar question about the total number of "chums" students had, defined as all the friends they might hang out but with whom they were not necessarily close.11 In the offline interview, students claimed an average number of 29 on-campus friends, and 19 off-campus friends. These are restrained numbers compared to those displayed on Facebook and reflect what we have described elsewhere as the "hoped-for possible selves." This is a clear example of how Face-book empowers users to claim the more socially desirable versions of their offline realities. However, the question remains as to why there are marked ethno-racial differences in online claims about friendship networks? For example, why are African American, Latino, and Indian students likely to claim an online network of friendships that is far more extensive than that of white and Vietnamese students? Table 5 compares the on-campus and off-campus Facebook friends with those reported in interviews for the four ethno-racial groups. For on-campus offline friends, we find a similar pattern of higher average by African American (32), Latino (37) and Indian (26) students compared to those of the Vietnamese students (19). A similar descending order obtains for off-campus offline friends, from a high of 37 for African Americans, 20 for Latinos, 15 for Indians and 6 for the Vietnamese students. In all cases the friendship Table 5 Online-Offline Sociability Comparison Mean Number of "Friends" by Ethno-Racial Groups (Number of Cases in parentheses) On-campus friends Off-campus friends Facebook Offline Facebook** Offline African American 218 (11) 32 (15) 94 (11) 37 (14) Latin a/o 204 (14) 37 (15) 105 (14) 20 (15) Indian 158 (13) 26 (15) 124 (13) 15 (15) Vietnamese 60(7) 19 (13) 59(8) 6(13) White 86 (15) n.a. 67 (15) n.a. Total 150 (60) 29 (57) 92 (61) 19 (57) **ANOVA Tests: p < .05. Source: Interview Project Data, 2006; Facebook Project Data, 2007. 176 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association "activity" represented by Facebook numbers reflects an enhancement of the social network of friends in the offline world of students. In general, the on-campus and off-campus Facebook friends are approximately five times the number of offline friends that students claim in in-person interviews (an enhancement ratio of 5.2 for on-campus and 4.8 for off-campus friends). The Vietnamese not only have notably lower numbers of offline friends, but their friends apparently engage with them much less frequently in Facebook activity which translates into an enhancement of only about three times as many more Facebook friends as offline friends. Nonetheless, we see a direct correspondence between the highly social identities and popularity claims implicitly made by African Americans, Latinos, and Indians on Facebook and their more extensive offline social connections on and off campus. Similarly, the more modest identity claims displayed by the Vietnamese students online apparently correspond to their smaller network of campus and off campus friends. In general these findings illustrate that in a nonymous environment like Facebook, identity claims regarding the extensiveness of social networks appear to be grounded in offline realities. Degrees of "friendship traffic" on Facebook seem to correspond proportionately to existing social networks in the offline world. Discussion Several features of the campus institutional context help explain the differences in projected Facebook identities found above. First, the formal organizational culture of student life at Victory State is dominated by organizations that express ethno-racial identities. In addition to student interviews, we also gathered information on student organizations from campus administrators involved with student services. Based on this information, we analyzed all registered student organizations on campus in terms of their organizational title, their stated purpose, and estimates of the size and ethnic composition of the organization by the administrative staff who worked with them. Of a total of 152 student organizations registered in 2006, the year of the interviews, 21.5 percent were organized explicitly in racial/ethnic terms and 44.7 percent were implicitly ethno-racial or de-facto ethnic/racial, in terms of membership composition, and 28.3 percent signaled no ethnic or racial theme or consisted of a mixed-race group of students. Thus 66.2 percent of all registered student organizations on campus could be considered ethno-racial social spaces devoted to the identities of nonwhite students. This means that the organizational context on this campus stresses the mobilization of ethno-racial identities. Second, ethno-racial minorities are in general more highly mobilized into student organizations at Victory State than are white students. That is, we found different degrees of mobilization of different ethnic groups into these somewhat racialized student organizations. While white students constitute 58% of the student body, they represent only 38.1 percent of membership in student organizations. All of the racial minorities are mobilized into student organizations in disproportionately higher numbers compared to their relative size on campus. This is especially the case for African Americans who constitute approximately one-third of all student Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 177 membership while only 19.0 percent of the student body and Asians with 19.3 percent of all membership compared to their 9.0 percent in the student body. This means that minorities are disproportionately mobilized into the formal social spaces available to students on campus. That is, the actual student organizational structure at Victory State, with relatively higher levels of formal expressions of sociability by ethno-racial minorities than by white students, for example, parallels the more highly social and group-oriented identities expressed on the Facebook pages of Victory State students. It makes sense that these cultural differences in face-to-face sociability, or organizational participation, might translate into more "Facebook traffic," expressed in friends, photos, and wall posts. Third, in terms of the hierarchy of sociability and group orientation expressed in the Facebook identities, with African Americans projecting the most social, most culturally expressive and most narrated identities, we find other correspondences in campus life. For one, African American leadership is highly salient in student government as well as in the student organization on campus with dominant influence in selecting most campus-wide social programs and public events. Moreover, the racial composition of the membership of this hegemonic campus organization has been about 90 percent black for the past decade12. Fourth, several factors relevant to the Vietnamese students on campus may help explain why they are distinct from the other ethno-racial groups. We considered the possibility that our sample of Vietnamese ancestry students included more 1.5 immigrants (those who were born abroad but came to the U.S. before the age of 12) than was the case among the Indian sample. This might have explained lower levels of social integration leading to the less socially engaged portraits expressed on Facebook. However, this did not turn out to be the case, as both groups contained about equal numbers of those born here, though a few more of those born in Vietnam came at later ages than the students born in India. Another possibly relevant factor lies in the differences in majors of the different groups and the very high concentration of the Vietnamese in time-demanding majors. Fully 64 percent (9 of 14) of the Vietnamese were in the natural sciences or health tracks, almost exclusively in the two majors of biology and pharmacy. In contrast, only 41.2 percent of the Indians, 12.5 percent of the Latinos, and none of the African Americans were in the natural sciences or health tracks.13 Equally important, Vietnamese students had different residential experiences than that of the other groups, a factor related to intensity of Facebook usage in other research (Hargittai, 2007). When taking a look at freshman year housing (the year in which the very high proportions of students tend to live on campus) we find that only 11.1 percent of the Vietnamese students were living on campus. For this same year, Latinos and African Americans had the highest numbers living on campus (50.0 percent and 33.3 percent respectively). On the other hand, fully 44.4 percent of the Vietnamese group lived at home during their freshman year. For freshman and sophomore years, the Vietnamese students commuted from home in substantially higher portions than any other group. 178 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association These campus contextual factors - the high salience of ethno-racial identities in the organizational culture of students on campus, the disproportionately high mobilization of minorities into these relatively ethnically segregated clubs, the salience of black leadership and membership in one of the most prominent and publicly visible of all student organizations, and the apparently lower levels of social integration into campus life of the Vietnamese - help explain the variation we uncovered among the ethno-racial groups in terms of their investments in racial identity display on Facebook. On campus, minorities such as the African Americans, Latinos and Indian students are highly networked into public social and organizational spaces and, as a result, quite feasibly have larger extended networks of (albeit weak) ties than do the more marginalized Vietnamese or the organizationally less mobilized white students. Conclusion While the presence of minority groups may have been marginalized or silenced in the earlier days of the Internet or in anonymous online communities, their presence is alive and extremely well articulated in the Facebook community of this study. As postulated, we found that ethno-racial identities are salient and highly elaborated. African Americans, Latinos, and Indian ancestry students project a visual self that is dramatically more social, they invest more frequently and intensively in displaying a cultural self marked by specific consumer and popular cultural preferences, and they invest more in the direct "about me" narrations than do Vietnamese or white students. The uplifting and often inspirational quotes related to racial themes of injustice frequently included by the African American, Latino, and Indian students convey a sense of group belonging, color consciousness, and identification with groups historically stigmatized by dominant society. The profiles of white students and Vietnamese students rarely signaled group identification or ethno-racial themes, reflecting "strategies of racelessness" (Willie, 2003) more typically discussed in research focused on white students in offline college contexts (Gallagher, 1994; Winant, 2004; Bonilla-Silva, 2006). The intensive investments of minorities in presenting highly social, culturally explicit, and elaborated narratives of self in the Facebook profiles are consistent with preoccupations about and heightened awareness of racial identities during this stage of life. Such modes of identity production reflect a certain resistance to the racial silencing of minorities by dominant color-blind ideologies of broader society where direct referencing of race remains taboo. Research on nonymous heritage and identity forums devoted to ethno-racial concerns has confirmed the importance of same-race encounters for strengthening the cultural and positive racial identities of participants (Byrne, 2008). Yet, in the more "race-neutral" context of Facebook, we found that many of the minorities and immigrant students are also strongly "putting themselves out there" and doing so with well articulated performances tinged with racial consciousness. Mostly their performances express positive racial identities, or at least strive in that direction, which psychologists underscore as essential to dealing Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 179 with the negative toll of racism. Indeed, Marcia (1989) has argued that positive adjustment to diverse environments depends on the development of healthy cultural identities among adolescents and young adults. Facebook appears to permit such affirming, elaborated expressions of self. This finding adds to the literature on online empowerment of minority and marginalized users (Hu & Leung, 2003; Mehra, Merkel & Bishop, 2004). As Amichai-Hamburger, McKenna, & Tal (2008) point out, empowerment is not a personal trait, but a concept that links individual strengths and competences to the wider social and political environment in which they live. The Internet has served as a social technology that reshapes the terrain of human interaction and enables the marginalized and the powerless to find their voice in the online world. Even though the emergent online interpersonal sphere does not fundamentally alter the power structure of society, it provides virtual "identity workshops" (Amichai-Humburger et al., 2008) that allow users to reframe themselves. Our findings of relatively high investments by minority users in the presentation of identities that are highly social, as well as culturally and narratively elaborated, may well reflect their deeper concerns with building networks, with like-minded or same-group audiences at this critical time of their lives (Tatum 1992; Willie 2003). Facebook is one of those online venues that help many minority students express and project a positive ethno-racial identity that enhances their self-conception. We also found that in a nonymous environment like Facebook, identity claims regarding the extensiveness of social networks appear to be grounded in offline realities; high rates of "friendship traffic" on Facebook do seem to correspond proportionately to extensive social networks in the offline world. Similarly, online social dynamics in the nonymous environment of Facebook are reflective of offline social practices. In this case the intensive nature of racial identity performances by several of the minority groups corresponds to a highly racialized campus culture and the salience and visibility of black leadership in prominent student organizations and campus events. A limitation of this study is that it is, in effect, an institutional case study of one undergraduate community. The Facebook community we studied is closely coupled with the bounded campus of Victory State, possibly an outlier when it comes to minority representation. It may well be that on such a campus minorities have reached a critical mass of visibility and empowerment that reinforces the enthusiastic elaboration of their racial identities. The highly salient ethno-racial identities we found may relate both to the unique aspects of this campus (including high rates of minority representation) as well as the unique nature of the undergraduate moment, where racial identity concerns are heightened. In contexts where minorities represent a small proportion of students or are culturally marginalized on campus, they may be less inclined to highly invest in performances of their racial identities and restrict that activity to heritage and identity forums where they are more assured of a sympathetic audience. Yet even then such a distinct outcome - less salient racial identity performances in contexts of minority 180 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association marginalization - would reinforce the broader implications of our findings, namely that nonymous cybernet environments are highly reflective of the social practices of their bounded offline communities where participants can expect that their claims, even if merely their "hoped for selves," can be challenged, or gossiped about, next week at the local campus food court. Everyone in the audience may not be known, but everyone in the audience is knowable and with relatively low transaction costs. Indeed, because the imagined audience of Facebook is very close to home, it restrains the identity performances of its participants. Acknowledgements The authors are grateful to Elisa Bernd and Jennifer Kim for early research assistance and to the Department of Sociology for research support. Both Sherri Grasmuck and Shanyang Zhao would like also to acknowledge the support provided by a study leave awarded by Temple University during the data collection phase of this project. Finally, we thank the three anonymous reviewers for their helpful comments. Notes 1 The identities of racialized minorities must be understood as simultaneously ethnic and racial, or "ethno-racial," because these identities are formed in the context of societies with externally imposed definitions of race and color but are negotiated and constructed in terms of their unique histories and cultural practices (Candelario, 2007). 2 Adamic, Buyukkokten, and Adar's study of a specialized online social networking site revealed that users tended to most often create online network ties with those who share similar interests and characteristics. Furthermore, they found that these people tended to be fewer than four steps removed from the user in the online network. Therefore, friends of friends were the most likely candidates in the creation of new online connections (Adamic, Buyukkokten, & Adar, 2003). 3 For academic diversity, we also solicited interviews from a universal listing of honors students organized by ethnic ancestry. In order to diversify the representation of distinct social clusters on campus, our within group sampling goals were the following: a) not more than two cases from a sorority/fraternity within each group; b) not more than two male or two females from each group from the Honors Program; c) a mixture of majors/colleges within the university: at least some from nontypical majors for each ethnic group; d) not more than four from each group highly active in student organizations. 4 Females were much more likely to block their page from the view of other university students (six out of seven of the blocked pages belonged to women). A careful analysis of gender differences revealed that, while males tended to be slightly more active on Facebook, there were few, if any, significant variations by gender. 5 Our class index variable combined the educational attainment of both parents with total family income. The distribution of each group in terms of high, medium, and low social class background (respectively) was: 58.8%, 35.3 % and 5.9% for Indians; 7.1%, 28.6% and 64.3% for Vietnamese; 6.3%, 50.0% and 43.8 % for Latinos; and 33.3%, 40.0% and 26.7% for African Americans. Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 181 6 Both the Facebook pages as well as the interview data were coded by a trained graduate student. Because the coding was enumerative in nature, it did not require subjective interpretation and created only a few incidents of uncertainty during the coding process and were resolved by the authors on a case by case basis. For each ethnic group, means were calculated for profile photos posted, wall posts, friends in the campus network, friends in off campus networks, songs listed, movies listed, television shows listed, and lines of type in the quotes section. We also calculated the percentage of each group listing at least something in each of these categories. For the "about me" section, we coded one to two sentences as a short response and anything longer than that as a long response. We arrived at this break point primarily because there was a natural break there. That is, respondents tended to write either very short responses or very elaborate ones. The coding for romantic connections through Facebook and other social networking sites was coded from responses to very specific interview questions. As in the case of the Facebook coding, in the few cases of ambiguity that required a judgment call, the author who had personally carried out the interview was consulted and a mutual decision was made. 7 Despite Victory State's high ranking for its demographic diversity by the Princeton Review, it did not make even the top 20 list of colleges with the "most race/class interaction" ["Do different types of students 'black/white, rich/poor' interact frequently and easily?"]. The two separate indicators, one for diversity and one for race/class interaction, apparently reflect the fact that they do not necessarily correlate highly. 8 We carried out significance tests for each element of tables 2 through 5. In cases where we present the percentage that responded to an item, we used the chi-square test. In cases where we present a comparison of means, we used an ANOVA (analysis of variance) test. We have noted instances of significance at the .1, .05, and .01 levels. Due to a small sample size, there are relatively few elements that produced significant results. Consequently, we have reported the number of cases in parentheses next to the reported percentage or mean. The purpose of this is to allow the reader to compare the raw number values as well as the percentages. 9 There are no available national averages for the displayed pictures or other features of Facebook. One case study of Carnegie Mellon Facebook users reported an average of 78.2 campus friends and 54.0 non-campus friends (Gross and Acquisti 2005). Ellison, Steinfield and Lampe (2007) found that Facebook users at Michigan State displayed an average of 150-200 campus friends (no national means were provided). Andon (2007) uses the number of Facebook friends in another campus study as an index measuring intensity of use but does not report those averages beyond the index scores. In our study the mean number of campus friends (150) and noncampus friends (90) falls somewhat in the middle of the above cited averages. Over time the average number of posts (friends, photographs displayed) may increase as the reliance on Facebook among college students expands. 10 It should be noted that there were only four Vietnamese respondents who actually included an "About Me" entry. 11 In the interviews, respondents were asked about both numbers of "chums" and "close friends." Chums are articulated as those with whom a respondent has some friendship connection, but not necessarily a very strong one. In recognizing that Facebook friends tend to be generally reflective of what Granovetter refers to as "weak tie" networks, we 182 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association compare the number of Facebook friends to the number of "chums" reported in the interviews (Granovetter, 1983). 12 This estimate of 90% African American membership was confirmed by more than five student service administrations, numerous students interviewed including the organizational leadership and observed by Grasmuck in campus observations. 13 At Victory State, 81.3% of all African Americans and 75% of Latinos are in either business or liberal arts majors; 64% of all Indians are in either the natural sciences/health tracks or in the business school. Although not presented here, we analyzed the association between race, major and Facebook usage. While students' majors confound the relationship between race and Internet usage to some degree, race remains as an effect independent of student major, similar to findings of Korgen, Odell and Shumacher (2001). References Adamic, L. A., Buyukkokten, O., & Adar, E. (2003). A social network caught in the web. First Monday, 8(6). Amichai-Hamburger, Y., McKenna, K. Y. A., and Tal, S. (2008). E-empowerment: Empowerment by the Internet. Computers in Human Behavior, 24(5), 1776-1789. Anderson, J. D. (2002). Race in American higher education: Historical perspectives on current conditions. In W. A Smith, P. G. Altback, and K. Lomotey (Eds.) The racial crisis in American higher education: Continuing challenges for the twenty-first century (pp. 2-3). Albany: State University of New York Press. Andon, S. P., (2007). Evaluating computer-mediated communication on the university campus: The Impact ofFacebook.com on the development of romantic relationships. Master's Thesis, Florida State University. Retrieved 12-17-08 from: http://etd.lib.fsu.edu/ theses/available/etd-01052007-170005/ Bailey, C. (1996). Virtual skin: Articulating race in cyberspace. In D. Trend (Ed.) Reading digital culture (pp. 334-346). Maiden, MA: Blackwell Publishers. Banks, A (2005). Race, rhetoric, and technology: Searching for higher ground. Mahwah, NJ.: Lawrence Erlbaum Associates. Bonilla-Silva, E. (2006). Racism without racists: Color-blind racism and the persistence of racial inequality in the United States. 2nd ed. Rowman & Littlefield Publishers, boyd, d. & Ellison, N. (2007). Social network sites: Definition, history and scholarship. Journal of Computer Mediated Communication. 13 (1) article 1. Retrieved August 30, 2008 from: http://jcmc.indiana.edu/voll3/issuel/boyd.ellison.html Burkhalter, B. (1999). Reading race online: Discovering racial identity in Usenet discussions. In Smith, Marc, and Peter Kollock (Eds.), Communities in Cyberspace. London: Routledge. Byrne, D. N. (2007). Public discourse, community concerns, and civic engagement: Exploring Black social networking traditions on BlackPlanet.com. Journal of Computer-Mediated Communication, 13(1), article 16. Retrieved April 8, 2009 from: http://jcmc.indiana.edu/voll3/issuel/byrne.html Byrne, D. (2008). The future of (the) 'race': Identity, discourse and the rise of computer-mediated public spheres. In A Everett (Ed.), Learning race and ethnicity: Youth and digital media. The MacArthur Foundation Book Series on Digital Learning. Volume (pp. 15-38). Cambridge, MA: MIT Press. Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 183 Candelario, G. (2007). Black behind the ears: Dominican racial identity from museums to beauty shops. Durham: Duke University Press. Castells, M. (2001). The Internet galaxy: Reflections on the Internet, business, and society. Oxford: Oxford University Press. Cassidy, J. (2006). Me media. The New Yorker (May 15), 50-59. Clancy, S. M. and Dollinger, S. J. (1993). Photographic depictions of the self: Gender and age differences in social connectedness. Sex Roles, 29(7-8), 477-495. Donath, J. &boyd, d.. (2004). Public displays of connection. BT Technology Journal, 22 (4), 71-82. Retrieved September 1, 2008 from: http://www.danah.org/papers/Public Displays.pdf. Ellison, N., Heino, R. & Gibbs, J. (2006). Managing impressions online: Self-presentation processes in the online dating environment. Journal of Computer-Mediated Communication, 11 (2), article 2. Retrieved August 30, 2008 from: http://jcmc.indiana. edu/voll l/issue2/ellison.html. Ellison, N. B., Steinfield, C, & Lampe, C. (2007). The benefits of Facebook "friends:" Social capital and college students' use of online social network sites. Journal of Computer-Mediated Communication, 12(4), article 1. Retrieved August 30, 2008 from: http://jcmc.indiana.edu/voll2/issue4/ellison.html Emmison, M. and Smith, P. (2000). Researching the visual: Images, objects, contexts and interactions in social and cultural inquiry. London: Sage. Gallagher, C. 1994. White reconstruction in the university. Socialist Review, 94,1-2. Gibbs, J. L., Ellison, N. B. & Heino, R. D. (2006). Self-presentation in online personals: The role of anticipated future interaction, self-disclosure, and perceived success in Internet dating. Communication Research, 33(2), 152-177. Ginger, J. (2007). The implications of race and identity on Facebook. Retrieved July 4 2007 from http://facebookproject.blogspot.com. Goffrnan, E. (1959). The presentation of self in everyday life. New York: Doubleday. Gomez-Pena, G. (2001) The Virtual barrio @ the other frontier (or the Chicago Interneta). In D. Trend (Ed.) Reading digital culture (pp. 281-286). Maiden, MA: Blackwell Publishers. Granovetter, M (1983). The Strength of weak ties: A network theory revisited, Sociological Theory, 1,201-233. Gross, R & Acquisit, A. (2005). Information revelation and privacy in online social networks (the Facebook case). Pre-proceedings version. MCM Workshop on Privacy in the Electronic Society (WPES). Retrieved December 17, 2008 from http://www.heinz.cmu. edu/~acquisti/papers/privacy- facebook-gross-acquisti.pdf Guerrero, Laura K., Peter A Andersen & Walid A Afifi (2007) Close encounters: Communication in relationships. London: SAGE. Frankenberg, R. (1993J. White women, race matters. Minneapolis. University of Minnesota Press. Hargittai, E. (2007). Whose space? Differences among users and non-users of social network sites. Journal of Computer-Mediated Communication, 13(1), article 14. Retrieved April 8, 2009 from: http://jcmc.indiana.edu/voll3/issuel/hargittai.html Herring, S. (1993). Gender and democracy in computer-mediated communication. Electronic Journal of Communications, 3(2). -. (2008). Questioning the generational divide: Technological exoticism and adult construction of online youth identity. In: D. Buckingham (Ed.), Youth, identity, and 184 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association digital media (pp. 71-94). Cambridge, MA: MIT Press. Retrieved September 1, 2008 from: http://ella.slis.indiana.edu/~herring/macarthur.pdf Hu, S. L. Y. & Leung, L. (2003). Effects of expectancy-value, attitudes, and use of the Internet on psychological empowerment experienced by Chinese women at the workplace. Telematics and Informatics, 20(4), 365-382. Kendall, L. (1998). Meaning and identity in 'Cyberspace': the performance of gender, class, and race online. Symbolic Interaction 21(2), 129-153. Kibria, N. (2001). College and notions of "Asian Americans": Second-generation Chinese Americans and Korean Americans in Pyong Gap Min (Ed.) The second generation: Ethnic identity among Asian American (pp. 183-206). Walnut Creek, Cal. Alta Mira Press. Kim, K. H., & Yun, H. (2007). Cying for me, Cying for us: Relational dialectics in a Korean social network site. Journal of Computer-Mediated Communication, 13(1), article 15. Retrieved April 8, 2009 from: http://jcmcindiana.edu/voll3/issuel/kim.yun.html Korgen, K., Odell, P. and Schumacher, P. (2001) "Internet use among college students: are there differences by race/ethnicity?" Electronic Journal of Sociology (ISSN: 11983655). Retrieved September 2, 2008 from: http://www.sociology.org/content/vol005.003/korgen.html Lampe, C, Ellison, N., & Steinfield, C, (2006). A Face(book) in the crowd: Social searching vs. social browsing. Proceedings of the 2006 20th Anniversary Conference on Computer Supported Cooperative Work (pp. 167-170). New York: ACM Press. Lenhart, A., & Madden, M. (2007, April 18). Teens, privacy, & online social networks. Pew Internet and American Life Project Report. Retrieved September 2, 2008 from: http://www.pewInternet.org/pdfs/PIP_Teens_Privacy_SNS_Report_Final.pdf Leonard, K. I. (2007). India. In M. Waters and R. Ueda (Eds.), The new Americans: A guide to immigration since 1965 (pp. 460-468). Cambridge, Mass: Harvard University Press. Lewis, K., Kaufman, J., Gonzalez, M., Wimmer, A., Christakis, N. (2008). Tastes, ties, and tie: A new social network dataset using Facebook.com. Social Networks, 30, 330-342. Mehra, B., Merkel, C, and Bishop, A. P. (2004). The Internet for empowerment of minority and marginalized users. New Media & Society, 6(6), 781-802. Miller, L. (2001). Women and children first: gender and the settling of the electronic frontier. In D. Trend (Ed.), Reading digital culture (pp. 214-220). Maiden, MA: Blackwell Publishers. Min, P. G. (2001). Introduction in Pyong Gap Min (Ed.) The second generation: Ethnic identity among Asian Americans, (pp. 1-17) Walnut Creek, Cal. Alta Mira Press. Marcia, J. E. (1989). Identity and intervention. Journal of Adolescence 12, 401-410. Morse, M. (2001). Virtually female: body and code. In D. Trend (Ed.) Reading digital culture (pp. 87-97). Maiden, MA: Blackwell Publishers. Nakamura, L. (2002). Cybertypes: Race, ethnicity and identify on the Internet. New York: Routledge. Negroponte, N. (1995) Being digital. New York: Alfred Knopf. Phinney, J, and Alipuria, L. L. (1990). Ethnic identity in college students from four ethnic groups. Journal of Adolescence 13, 171-183. Poster, M. (2001). Cyberdemocracy: the Internet and the public sphere. In D. Trend (Ed.), Reading digital culture, (pp. 259-271). Maiden, MA: Blackwell Publishers. Princeton Review. (2006) The best 361 colleges. New York: Random House. Rheingold, H. (1995). The virtual community: Finding connection in a computerized world. London: Seeker & Wargurg. Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 185 Rumbaut, R. (2007). Vietnam. In M. Waters and R Ueda (Eds.), The new Americans: A guide to immigration since 1965 (pp. 665-673). Cambridge, Mass: Harvard University Press. Schlenker, B. R (1985). The self and social life: McGraw Hill series in social psychology and personality. New York: McGraw Hill. Strano, M. M. (2008). User descriptions and interpretations of self-presentation through Facebook Profile images. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 2(2), article 1. Retrieved from: http://cyberpsychology.eu/view.php?cisloclanku=2008110402&article=l. Surratt, C. G. (1998). Netlife: Internet citizens and their communities. New York: Nova Science. Takagi, D. Y (1992). The retreat from race: Asian American admissions and racial Politics (New Brunswick, J.J.: Rutgers University Press). Tatum, B. D. (1997). "Why are all the black kids sitting together in the cafeteria?": And other conversations about race. New York: Basic Books. Tong, S.T., Van Der Heide B., Langwell, L., Walther, J.B. (2008). Too much of a good thing? The Relationship between number of friends and interpersonal impressions on Facebook. Journal of Computer-Mediated Communication, 13(3), 531-549. Trend, D. (Ed.). (2001). Reading digital culture. Maiden, MA: Blackwell Publishers. Turkle, S. (1995). Life on the screen: Identity in the age of the Internet. New York: Simon & Schuster. -. (2001). Who am we? In D. Trend (Ed.) Reading digital culture (pp. 236-250). Maiden, MA: Blackwell Publishers. Tynes, B., Reynolds, L., & Greenfield, P. (2004). "Adolescence, race and ethnicity on the Internet: A comparison of discourse in monitored vs. unmonitored chat rooms" Applied Developmental Psychology (25), 667-684. Walker, K. (2000). "It's difficult to hide it": The presentation of self on Internet home pages. Qualitative Sociology, 23(1), 99-120. Willie, S. S. (2003). Acting black: Collage, identity, and the performance of race. New York: Routledge. Winant, H. and Omi, M. (1994). Racial formation in the United States. New York: Routledge, Second Edition. Winant, H. (2004). The new politics of race: Globalism, difference, justice. Minneapolis: University of Minnisota Press. Yurchisin, J., Watchravesringkan, K., & McCabe, D. B. (2005). An exploration of identity re-creation in the context of Internet dating. Social Behavior and Personality, 33(8), 735-750. Zhao, S. (2006). Cyber-gathering places and online-embedded relationships. Paper presented at the Annual Meetings of the Eastern Sociological Society in Boston. Zhao, S. Grasmuck, S & Martin, J. (2008). Identity construction on Facebook: Digital empowerment in anchored relationships. Computers in Human Behavior, 24, 1816-1836. Ziller, R.C. (1990). Photographing the self: Methods for observing personal orientations. London: Sage. Ziller, R.C. and Rorer, B.A. (1985). Shyness-environment interaction: A View from the shy side through auto-photography. Journal of Personality, 53(4), 626-639. Zywica, J. and Danowski, J. (2008). The Faces of Facebookers: Investigating social enhancement and social compensation hypotheses, predicting Facebook™ and offline popularity from sociability and self-esteem, and mapping the meanings of popularity with semantic networks. Journal of Computer-Mediated Communication, 14(1), 1-34. 186 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association About the Authors Sherri Grasmuck is Professor of Sociology at Temple University. Her research interests include International Migration and Integration/ Kinship/ Gender and Masculinities/ Urbanism and Place. Address: 1115 W. Berks St, 7th floor, Gladfelter Hall, Philadelphia, PA. 19122 Email: sgrasmuc@temple.edu Jason Martin is a Ph.D. candidate in the Sociology Department at Temple University. His research interests include Urban and Organizational Sociology / Space and Geography / Race, Religion and Social Capital. Address: 1115 W. Berks St, 7th floor, Gladfelter Hall, Philadelphia, PA. 19122 Email: jasonmar@temple.edu Shanyang Zhao is Professor of Sociology at Temple University. His research interests include Internet and Human Interaction/ Metatheory/ Self and Identity. Address: 1115 W. Berks St, 7th floor, Gladfelter Hall, Philadelphia, PA. 19122. Email: bzhao001@temple.edu Appendix I: Rules for Coding Facebook Pages (Abbreviated) Variable Value & Value Label Variable Value & Value Label Case # Friend_C R's Campus Friends Picture R's Profile Cover Picture 0 - Missing 0 - Blank -# of campus friends 1 - Self Friend_0 R's Other Friends 2 - With others 0 - Missing 3 - Avatar -# of all other friends Gender R's Gender Group R's Groups loined 0 - Missing 0 - Missing 1 - Male -# of groups joined 2 - Female Activity Activities R Likes Sex Sex R Interested In 0 - Missing 0 - Missing -# of activities listed 1 - Men 98 - Unspecified 2 - Women Interest R's Personal Interests Relation R's Current Relationship 0 - Missing Status 0 - Missing -# of interests listed 1 - Single 98 - Unspecified 2 - In a relationship Music R's Favorite Music 3 - In an open 0 - Missing relationship Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association 187 Appendix I: continued Variable Value & Value Label Variable Value & Value Label Look_For Birthday Hometown Residence Contact High School Class Year 4 - Engaged/ married 5 - It's complicated Relationship R TV Looking for 0 - Missing 1 - Friendship 2 - A relationship 3 - Dating Movie 4 - Random play 5 - Whatever I can get R's Birthday 0 - Missing Book 1 - Month & day & year 2 - Month & day R's Hometown Name 0 - Missing Quote 1 - Not missing R's Residence Information 0 - Missing 1 - Complete About Me 2 - Partial R's Online Contact Information 0 - Missing 1 - Email address 2 - IM screen name Album 3 - Mobile phone/ land phone 4 - Website R's High School Name Post 0 - Missing 1 - Not missing R's Year of Class 0 - Missing - Year specified -# of music listed 98 - Unspecified R's Favorite TV Shows 0 - Missing -# of shows listed 98 - Unspecified R's Favorite Movies 0 - Missing -# of movies listed 98 - Unspecified R's Favorite Books 0 - Missing -# of books listed 98 - Unspecified R's Favorite Quotes 0 - Missing -# of quotes listed 98 - Unspecified R's Narrative Self-Description 0 - Missing 1 - One or two short sentences 2 - One or two short paragraphs 3 - Long paragraphs R's Albums 0 - Missing -# of albums Wall Posts in R's Account 0 - Missing -# of wall posts 188 Journal of Computer-Mediated Communication 15 (2009) 158-188 © 2009 International Communication Association