|Home | About | Journals | Submit | Contact Us | Français|
To evaluate whether online friends of adolescents who display sexual references on a social networking site also display references. The method used was content analysis. The result of this study was that adolescents who displayed explicit sexual references were more likely to have online friends who displayed references. Thus, social networking sites present new opportunities to investigate adolescent sexual behavior.
Adolescents frequently display sexual references on social networking websites (SNSs); these websites may present new opportunities to identify adolescents at risk for or engaging in sexual behavior . SNSs allow users to create profiles and interact with friends by linking profiles through “friending.” Therefore, references to sexual behaviors can be assessed for a profile owner's online friend network. The purpose of this study was to test the hypothesis that teens who display sexual references on MySpace.com are more likely to have online friends who also display sexual references.
This observational study applied content analysis to publicly available MySpace profiles between June and August 2008, and received Institutional Review Board exemptions from the University of Washington and University of Wisconsin.
Our first goal was to identify a group of 20 MySpace profiles: 10 profiles without sexual references (nondisplayers) and 10 profiles with sexual references (displayers) (n = 20). Inclusion criteria for these 20 profiles consisted of the following: publicly available profile, self-reported age of 18 years, profile content in English, profile activity within the last 30 days, and United States resident. Finally, the profile owner was required to have eight friends listed in their “Top 8 friends” profile section. At the time of this study, the “Top 8” friends on MySpace were displayed in a separate section from remaining friends.
A total of 94 profiles were assessed for eligibility: eight were excluded due to “private” security settings, nine were excluded due to having less than eight friends in the “Top 8” friends section. After identifying10 nondisplayers, an additional 57 profiles were excluded in the process of identifying 10 displayer profiles (due to absence of sexual references). After identifying 10 nondisplayers and 10 displayers, all “Top 8” friends from each profile were examined. Therefore, a total of 160 friends' profiles were examined.
From each of our subjects' profiles we recorded demographic data, including age, gender, and race/ethnicity. We evaluated each profile for sexual references using criteria developed by the Kaiser Family Foundation, which define a sexual reference as any depiction of sexual activity or sexually suggestive behavior . Examples of sexual references could include text describing sexual experiences or images suggesting sexual acts.
Two investigators conducted content analysis (L.B. and C.R.) using a codebook created in our previous study . A third investigator (M.M.) evaluated 20% random profile sample to assess inter-rater reliability. Cohen's Kappa statistic was used to evaluate agreement in the coding of the Web profiles. The Kappa value for sexual references was .70.
After initial data collection, all sexual references were re-examined independently by three authors (M.M., L.B., C.B.). Each was evaluated for whether it represented an explicit sexual reference. Explicit sexual references were defined as those in which the profile owner referenced personal sexual behaviors or displayed personal revealing images. In order for a reference to be further categorized as explicit, the consensus of all three authors was required.
Statistical analyses used STATA version 9.0 (Statacorp, College Station, TX). Our sample size was determined by our previous findings regarding proportion of profiles displaying sexual references . Analysis 1 used the χ2 test to compare the proportion of nondisplayers' friends' profiles that displayed sexual references with the proportion of displayers' friends' profiles that displayed sexual references. Analysis 2 conducted an identical comparison for explicit sexual references.
Of our 20 subjects' profiles, 11 profile owners were female (55%). Among displayers, nine were female (90%), and among nondisplayers, two were female (20%) (χ2 = 9.8, p = .001). The majority of profile owners were White/Caucasian (59%); there were no significant differences in race/ethnicity between nondisplayers and displayers (Table 1).
Of our 160 “Top 8” friend profiles, only 63 profile owners reported gender (35%); of these, 24 were female (38%). Of the profiles that included data, there were no significant gender, race/ethnicity, age, or profile security differences between nondisplayers' friends versus displayers' friends.
Of the 160 “Top 8” friends' profiles, sexual references were present on 38 profiles (24%). The proportion of profiles that displayed sexual references was not significantly different between nondisplayers' friends (n = 17) and displayers' friends (n = 21).
Of the 38 profiles with sexual references, 18 were further categorized as explicit. Of these 18 references, five were present on nondisplayers' friends' profiles compared with 13 on displayers' friends' profiles (χ2 = 4.1, p = .04). Table 2 includes examples of sexual and explicit sexual references.
Sexual references were evenly distributed among friends' profiles linked to both displayers and nondisplayers, perhaps representative of adolescents' normal developmental curiosity about sex expressed on SNSs. This finding is supported by previous studies illustrating that sexual references are common on adolescents' SNS profiles [1, 3]. However, when references were limited to explicit sexual references, adolescents displayed these references in patterns similar to their online peers. This suggests that adolescents within “Top 8” friend groups may have similar sexual intentions, sexual experiences, or place similar levels of importance on displaying sexual references. These references have potential to influence younger teens who wish to emulate adolescents in a particular friend group, or even a potential romantic partner of a profile owner within a friend group . Our previous study found that adolescents interpreted peers' SNS alcohol references as representative of actual alcohol use .
Limitations to our study include the uncertain validity of self-report on SNSs; however, previous work supports the validity of most online disclosures [6, 7]. Furthermore, we evaluated profiles from only one SNS. For consistency, we chose a single website that remains popular among teens . Finally, we only included public profiles; private profiles may feature different content. We were interested in profiles that are accessible to any viewer, such as a younger teen.
In spite of these limitations, our findings have valuable implications. Given the influence of peer networks and media on adolescent sexual behavior, SNSs may represent a novel way to investigate adolescent sexual behaviors within peer networks [4, 9]. A previous SNS intervention showed promise in reducing online sexual references, future studies involving online peer networks may be possible .
This project was supported by Award Number K12HD055894 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health & Human Development or the National Institutes of Health. We would like to acknowledge Marlon Mundt, PhD and Katie Egan for their assistance with this manuscript.