Adolescent Online Communication: Old Issues, New IntensitiesKaveri Subrahmanyam
California State University, Los Angeles
Children’s Digital Media Center, UCLA
There are many compelling reasons to study Internet use among young people. More than any other demographic group, today’s youth have embraced the Internet and other digital technologies. According to a 2005 Pew report, 87% of U.S. children between 12 and 17 are online and approximately 11 million of them access the Internet every day (Lenhart, Madden, & Hitlin, 2005). Marketing research suggests that youth spending on data services represents almost 50 percent of all mobile spending in most mature markets Not only are youth, the early adopters of most new technologies, they are also among the more sophisticated users of it as well. In fact, Livingstone and Bober (2005) have coined the term reverse socialization to characterize the shifting nature of socialization that occurs in many families as teenage children are often more knowledgeable than their parent about these technologies. Research has also suggested that adolescents’ interactions with these new technologies are often at the vanguard of trends to come (Greenfield & Subrahmanyam, 2003; Šmahel, 2006) and so studying their online behavior may provide a key to understanding virtual worlds of the future
Among adolescents, communication is the most important use of the Internet (Subrahmanyam, Greenfield, Kraut, & Grosss, 2001; Gross, 2004; Šmahel, & Machovcová, 2006) and the popular communication applications include, e-mail, instant messaging (private, one-on-one, text-based conversations), chat rooms (communication systems that allow text-based conversation among multiple users), and the newest craze social networking sites (connects people together) including blogs (www.livejournal.com), MySpace and Facebook. Although we know that adolescents are spending considerable amounts of time on these applications, many questions remain. Firstly, what are teens doing in communication forums such as chat rooms and blogs? What do they talk about? Secondly, are these online communication activities fundamentally changing adolescent behavior or are they simply providing new venues for “traditional” adolescent behavior? Lastly, what is the relation between teen virtual worlds and “real” worlds? Are teen participants leaving behind real lives when interacting online or do virtual worlds reflect themes that are traditionally adolescent ones? These questions are the focus of this presentation and they will help us better understand . the impact of adolescent Internet use on their development.
Prior Research on Chatrooms
I start by looking at one online forum, chat rooms. Chat rooms are disembodied in that information about bodies is not readily available; also missing are face-to-face cues such as gesture and eye gaze (Greenfield & Subrahmanyam, 2003). Participants can also be as anonymous as they want. In-depth analysis of a single conversation from a teen chat room revealed 3 important themes: identity (e.g., gendered nicknames – babygirl), sexual exploration (e.g., no sex until ur happily married. . .thatz muh rule), and partner selection (anybody here like 50 cent press 1234) (Subrahmanyam, Greenfield, & Tynes, 2004). Research also suggests that identity, sexuality, and partner selection are important tasks facing adolescents (e.g., Erikson, 1959 Hill, 1983) and that media play an important role in these developmental tasks (See Greenfield & Subrahmanyam, 2003). Teen virtual contexts, which combine peers and media appear to provide an online forum for enacting real life themes (Subrahmanyam et al., 2004)
Although, analysis of a single chat room suggests that real life themes may be enacted in virtual worlds, there are many unanswered questions. For instance, how often do these themes occur in chat rooms? What proportion of chatters utilize them? And importantly, do themes from real life occur in another online public forum such as blogs? Today I present two lines of studies that address these questions – the Chat studies (Subrahmanyam, Smahel, & Greenfield, 2006; Smahel & Subrahmanyam, In Press) and the Blog Project (Li, Lipana, & Subrahmanyam, 2006; Garcia, Harsano, & Subrahmanyam, 2006).
The chat studies were conducted to study the online construction of themes from adolescent life such as identity, sexuality, and partner selection (Subrahmanyam, Smahel, & Greenfield, 2006; Smahel & Subrahmanyam, 2006). We conducted in-depth analyses of conversations in monitored and unmonitored teen chat rooms. Monitored chatrooms are provided by an Internet service that requires a subscription fee and provides an adult host; in the teen chat rooms an adult monitor or “host” moves from chat room to chat room reminding users about the rules (e.g., no swear words, no giving out identity information, etc), kicking out users who violate the rules, etc. Unmonitored chat rooms are free and have no adult monitor. We analyzed a total of 20 chat sessions, ten from the monitored and ten from the unmonitored service; each session was 30 minutes long and was recorded on weekday and weekends between 12-9 pm PST (for more details see Subrahmanyam, Smahel, & Greenfield, 2006).
All the utterances in our chat corpus were coded for the presence of the adolescent themes of identity, sexuality, and partner selection. The presence of identity was assessed by coding whether or not a user provided basic identity information (age, sex, location) (e.g., 15/ohio/f ) about himself/herself. Utterances were coded as sexual (ANY HOT CHICKS WANNA CHAT PRESS 69) or Non-sexual (Wassup everybody?); sexual utterances were further coded as either implicitly sexual (eminem is hot ) vs. explicitly sexual (a dork is a whale’s dick). Lastly, we coded whether an utterance contained a partner request (Ladies If Ya Sexy Press 11 ). Partner requests were further coded as either sexual (Ladies If Ya Sexy Press 11) or. non-sexual (skins im me).
All nicknames were coded for information about gender and sexualized identity. For information about gender identity, nicknames were coded as masculine if they contained male names and/or masculine stereotypes or personas (e.g., - RAYMONI8, Vikingdude123); nicknames were coded as feminine if they contained female names and/or feminine stereotypes or personas (e.g., - MandiCS12, Americangal). Nicknames that did not present a gendered identity were coded as gender neutral (e.g., soccerlover, Spoiledbrat). Nicknames were coded for sexual explicitness and were coded as either non-sexual (e.g., Bratiegurl2) or sexual. Sexualized nicknames were further coded as either sexually implicit (e.g., prettygirl) or sexually explicit (e.g., SexyDickHed).
I start by describing the chat environment. Our sample contained over 12,000 utterances, 6702 from the monitored chat rooms and 5556 from the Unmonitored chat rooms. There were about 1000 participants/nicknames - 583 from the Monitored and 567 from the unmonitored chat rooms.
How frequent were the different kinds of utterances? We found that over 12% of all utterances were identity declarations; this amounts to over 2 identity declarations/minute. Five percent of all utterances were sexual ones; this amounts to 1 sexual remark per minute. Lastly, 11% of all utterances were partner requests; this amounts to 2 partner requests per minute. With regard to identity expression, identity declarations were made by 632 nicks or 55% of participants. Interestingly, gender was the most frequently uttered identity information; gendered nicknames were also used by 46% of participants. There was almost no discordance between gendered nick and gendered identity statements (p = .00). Participants who described themselves as younger provided more identity declarations (p ‘s = .00 across both services).
In the area of sexual exploration, we found that participants who declared themselves as older were more likely to make explicit sexual utterances (p = .00); in fact, 13% of nicks who self-described as 16-17 years produced at least one explicit sexual utterance whereas 40% of nicks who self-described as 18-24 years produced at least one explicit sexual utterance. Gendered identity related to modes of sexual expression (p’s < .05): masculine nicks made 14% sexually implicit versus 18% sexually explicit utterances. Feminine nicks made 19% sexually implicit versus 12% sexually explicit utterances. Analysis of sexualized nicknames suggested revealed that there was no relation between declared age and sexualized nicks. Further, implicitly sexual nicknames used more often by participants who described themselves as female (26%) rather than male (10%) (p = .00). With regard to partner requests, at least one partner request was uttered by 53% of nicks. Participants who stated they were older more actively searched for a partner (p < .01); among 14-15 years, 11.1% were partner requests whereas among 16-17 years, 16.4% were partner requests.
The Blog Project
Blogs are web logs and contain audio, video, textual, and iconic images. For this study, we identified a sample of 200 English language weblogs; blogs were drawn from 9 hosting sites. We only identified blogs that were maintained by authors who stated that they were between 14 and 17 years old. Finally, we only sampled blogs maintained by frequent bloggers (defined as a minimum of 5 entries within a one month period). >From each blog, 3 entries between March and April 2005 were analyzed yielding at total sample of 600 entries.
We coded both blogger demographics and identity as well as blog content. For blogger demographics, we coded the blog-authors’ self-presented information about their age, gender, location and the number of months their blogs had been in existence. For identity presentation, we coded individual bloggers’ username (gendered, activity- or interest-based, ambiguous) and the presence of user pictures; the latter was coded both for format (graphics, text, or graphics+text) and content (self-photograph, popular culture, and ambiguous). Finally, we coded the similarity between the blogger’s username and instant messaging screenname to assess the continuity in self-presented online identity.
The content of blog entries were coded for style (narrative, reflective, feedback, filter, quiz/survey and creative), and topic (peers, romance, structured life, future, identity, etc). Finally, we also coded for the number of comments that were left for each of the analyzed entries. All coding was done by two undergraduate coders, who were trained on a comparable sample before coding 15% of the study sample to calculate inter-rater reliability. The kappas were acceptable and ranged from 0.798 to 1.000.
Blog authors were 86% female and 4.3% male; for 9.7% of the sample, we could not clearly determine gender of the author. Although the majority of the bloggers in our sample were from the U.S. (55.9%), blog authors came from a number of countries including Australia, Canada, Indonesia, Malaysia, New Zealand, Singapore, Sweden, and the United Kingdom. The majority of usernames were neither gendered nor based on activities/interests. Furthermore, 49% of user pictures (N=110) consisted of self-photographs. We also found no continuity between different forms of bloggers’ online identity (e.g. blog username, e-mail address, instant messaging screennames, and forum username).
Content analysis of the entries suggested that the most entries were written in narrative and reflective styles. Entries written in narrative style recounted the events of a day in a story like fashion. Reflective entries were those in which the blog author seemed to ponder or reflect on some thought or event. Less common were entries asking for feedback, entries linking blog readers to other sites, and entries containing quizzes/surveys and more creative entries. The more frequent blog topics included peers (73%), structured life (65%), and families (48%); less common were entries about romance, identity, teen problems, and the future. We found an interaction of style and topic such as that many entries were narratives describing daily events, including interactions with peers, family, and at school. Less common were reflective entries that dealt with emotions and or opinions. Like the narratives, these entries also were often about the day’s events. Contrary to our expectations, majority of the entries did not have comments (51.8%), and 38% were found to have between 1 to 4 comments.
The goal of the chat and blog studies were to describe in detail the online environments with a view toward comparing and real and virtual worlds. With regard to the frequency of the different kind of utterances in the chat rooms, our results suggest that identity, sexual exploration, and partner selection were ubiquitous online. All of these themes are important in adolescent life. For instance, identity is considered the main developmental task of adolescence (Erikson, 1958; Johnson & Aries, 1983); it was rife in teen chat, and was actually provided by more than half the participants. Sexuality is similarly important during adolescence and research suggests greater sexual concerns and sexual involvement with increasing age (e.g., Cubbins, Santelli, Brindis, & Braveman, 2005). We found a similar trend online and users who self-declared as being older made explicit sexual comments and used obscene language at a much higher rate than those who described themselves to be younger. Finally pairing off and romantic partner selection are quintessential adolescent activities (Furman, & Shaffer, 2003) and these were similarly very frequent in the chat rooms. These trends suggest that identity, sexuality, and partner selection are part of adolescents’ lives in both physical and virtual settings.
Although we find similar themes in real and virtual worlds, their expression in virtual worlds are at times different and at other times similar to their expression in physical worlds. For instance identity information that we take for granted in the physical world, like information about gender, is missing in disembodied virtual contexts such as chat rooms. Such information becomes salient in the virtual world as illustrated by the frequency of gendered nicknames as well as utterances providing gender information.. However, just like in the physical world, gender identity appears to be stable within the time and space of a chat session. Finally, it has been suggested that younger teens may be more concerned with individual identity than older teens (Erikson, 1959) and this was the pattern that we found in our chat participants as well.
Similar to the physical worlds, our results suggests that in online chat rooms, female identity tended to be associated with implicit sexual communication whereas male identity was associated with explicit sexual communication. Similarly we found that just as sexual concerns and involvement increases with age in the physical world (Cubbins, et al., 2005), self-declared older participants were more likely to utter sexually explicit utterances. Traditional roles in sexualized interaction were also seen the chatrooms, where nicks with a masculine identity were more active and females were more passive in sexual exploration. Finally self-declared older participants more actively searched for a partner – this finding is consistent with older adolescents’ increased sexual concerns (Erikson, 1959; Cubbin, et al., 2005) and consequently their greater need for intimacy and romantic involvement (Pombeni, Kierchler, & Palmonari, 1990). Overall it appears that for adolescent chat users, identity and sexual exploration and the search for partners are present in both offline as well as online contexts. However, it does appear that frequency of sexual utterances and partner requests are intensified in the virtual world.
Next I turn to the results of the blog project. First of all, it is important to note that adolescent blogging is clearly an international phenomena, with bloggers coming from a variety of countries. Equally important, it appears to be a primarily female pastime with the majority of bloggers being female. However, since we only sampled blogs that were updated at least 5 times in a one-month period, our sample might have ended up with more females, who may be more likely consistently and frequently update their blogs. At the same time, keeping dairies has traditionally been a female habit and online dairies may be no different.
Interestingly, the presence of self-photographs in user pictures suggests that there is identity stability between blogs and the physical world. Furthermore, blog usernames do not appear to be used for sharing identity information about age and gender unlike nicknames in other online contexts such as chat rooms. This difference between chat rooms and blogs reveals that online users find ways to reveal identity information if such information is not readily available; at the same time, the presence of such identity information is a source of worry for those concerned for the well-being of adolescents.
The presence of various adolescent themes such as family, peers, romance, and structured life in the blog entries point again to the continuity between real and physical worlds. Since many blogs are used in the narrative style to discuss peers, family, structured life and romance, teens may be using blogs to “talk” about every day life. A the same time, the low frequency of comments suggest that audience feedback may not be crucial for blog authors. These results point to the emergence of online monologues in addition to other forms of dialogue such as phone conversations and instant messages to keep up with friends and family. Interestingly, the relatively low number of blogs that mention teenage problem behavior such as drinking, cutting, or drug usage gives a different view than the media-created stereotype of teenage blogging. These results suggest that adolescents are using new communication forms, such as blogs, to cope in new ways, with traditional developmental issues from their everyday lives.
Summary, Future Directions and Challenges
Across two different online forums, chat rooms and blogs, we found that new behaviors were not being created; instead it appeared that the virtual world may actually serve as a screen for playing out issues from the physical world. But because of different affordances (e.g., disembodied participants), these behaviors are enacted sometimes with different intensities compared to the real world. Thus it appears that physical and virtual worlds are psychologically connected and not separate as many scholars have claimed. Future research should assess factors that influence the relation between real and virtual worlds. Some mediating factors are age, gender, personality variables (e.g., extraversion/introversion, social competence), and context (MMROPG vs. Social networking sites). An important issue concerns situations where real and virtual worlds are disconnected – such disconnection should be a red flag for parents, counselors, and others working with adolescents. Finally a fertile area of research is examining the influence of the virtual on the real; some areas include the impact of multitasking on cognitive skills and the impact of online language use on written discourse.
The challenges facing researchers in the field include keeping up with the fad-like nature of online communication forms, while at the same carrying out good quality research. Because of the constantly changing nature of the Internet, researchers have to also take into account the changing environment of the Internet when interpreting research. Finally, the biggest challenge of all is connecting the virtual to the real at the individual level.
 Cubbin, C., Santelli, J., Brindis, C.D., & Braveman, P. (2005). Neighborhood context and sexual behaviors among adolescents: findings from the National Longitudinal Study of Adolescent Health. Perspectives on Sexual and Reproductive Health, 37, 125-134.
 Erikson, E. (1959). Identity and the life cycle. New York: W. W. Norton.
 Furman, W., & Shaffer, L. (2003). The role of romantic relationships in adolescent development. In: Florsheim, P., ed. Adolescent romantic relations and sexual behavior: Theory, research, and practical implications. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 3-22.
 Garcia, E. C., Harsano, L. S., Subrahmanyam, K., & Greenfield, P.M. (2006). Blogs as a window into teenage lives. Poster presented at the Psychology Undergraduate Research Conference, UCLA, May 12, 2006.
 Greenfield, P. M. & Subrahmanyam, K. (2003). Online Discourse in a Teen Chatroom: New Codes and New Modes of Coherence in a Visual Medium. Journal of Applied Developmental Psychology, 24, 713-738.
 Gross, E. F. (2004). Adolescent Internet use: What we expect, what teens report. Journal of Applied Developmental Psychology, 24, 633-649.
 Hill, J. (1983). Early adolescence: A research agenda. Journal of Early Adolescence, 3, 1-21
 Johnson, F. L. & Aries, E. (1983). Conversational patterns among same-sex pairs of late adolescent close friends. The Journal of Genetic Psychology, 142, 225-238.
 Lenhart, A., Madden, M., & Hitlin, P. (2005). Teens and technology: Youth are leading the nation to a fully wired and mobile nation. Pew Internet and American Life Project, Washington, D.C. http://www.pewinternet.org/pdfs/PIP_Teens_Tech_July2005web.pdf
 Li, J. S., Lipana, L., Subrahmanyam, K., & Greenfield, P.M. (2006). Quantitative analysis of blogger demographics and online self-presentation. Poster presented at the Psychology Undergraduate Research Conference, UCLA, May 12, 2006.
 Livingstone, S & Bober, M.. (2005). UK Children Go Online: Final Report of Key Project Findings. Economic & Social Research Council. Retrieved July 15, 2006, from http://www.children-go-online.net.
 Pombeni, J.L., Kierchler, E., & Palmonari, A. (1990). Identification with peers as a strategy to muddle through the troubles of the adolescent years. Journal of Adolescence, 13: 351-370.
 Šmahel, D. (2006). Czech Internet report: The rise of the Internet generation? In: Cyberspace 2005. Polcak, R., Skop, M. & Šmahel, D. (Eds.) Brno: Masaryk University. p. 261-268. WWW: http://www.terapie.cz/materials/wip-smahel-new-generation2006.pdf
 Šmahel, D. & Machovcová, K. (2006). Internet use in the Czech Republic: Gender and Age differences. In Cultural Attitudes Towards Technology and Communication 2006, Sudweeks, F., Hrachovec, H. & Ess, Ch. 2006. Murdoch, Australia : School of Information Technology, Murdoch University, 2006, p. 521-533. WWW: http://www.terapie.cz/materials/catac-tartu-2006.pdf
 Šmahel, D. & Subrahmanyam, K. (In Press). Any girls want to chat press 911: Partner Selection in Monitored and Unmonitored Teen Chat Rooms. CyberPsychology and Behavior.
 Subrahmanyam, K., Greenfield, P. M. Kraut, & Gross, E. (2001). The Impact of Computer Use on Children’s Development. Journal of Applied Developmental Psychology, 22, 7-30.
 Subrahmanyam, K., Greenfield, P. M, & Tynes, B. (2004). Constructing Sexuality and Identity in an Online Teen Chatroom. Journal of Applied Developmental Psychology, 25, 651-666.
 Subrahmanyam, K., Šmahel, D., & Greenfield, P. M. (2006). Connecting developmental processes to the Internet: Identity presentation and sexual exploration in online teen chatrooms. Developmental Psychology, 42, 1-12.