The publisher's final edited version of this article is available at J Adolesc Health See other articles in PMC that cite the published article. Methods Data were collected online between and with 3, randomly selected to year-old youth across the United States. Results Seven percent of youth reported sending or showing someone sexual pictures of themselves, where they were nude or nearly nude, online, via text messaging, or in-person, during the past year.
Although females and older youth were more likely to share sexual photos than males and younger youth, the profile of psychosocial challenge and sexual behavior was similar for all youth. After adjusting for demographic characteristics, sharing sexual photos was associated with all types of sexual behaviors assessed e.
Adolescents who shared sexual photos also were more likely to use substances and less likely to have high self-esteem than their demographically similar peers. Whether there are adolescent health implications, however, is less well understood. In a study of high school students across 7 schools in Texas state, youth who reported sharing sexual photos of themselves were more likely to be dating and to have had sex [ 3 ]. This would suggest that sharing or posting sexual pictures is reflective of typical sexual expression in romantic relationships among adolescents.
Rates may differ for a variety of reasons, including: Sexually curious behavior is reflective of typical sexual development during adolescence [ 13 — 15 ]. Sharing or posting sexual pictures may therefore be reflective of usual sexual expression in romantic relationships in adolescence. We also examine how this behavior relates to psychosocial functioning, as this is less well understood.
To examine whether potential differences in previous findings are perhaps related to age differences, correlates are examined for younger and older youth separately. Participants were recruited from: Because the focus in this paper is on the general population of adolescents, the current analyses are restricted to the HPOL sample. Members were recruited through a variety of methods, including targeted mailings, word of mouth, and online advertising.
Panelists enrolled in the opt-in panel at the HPOL website, http: The median survey length was 23 minutes. Recent survey response rates are noticeably lower than in the past [ 17 , 18 ]. It was calculated as the number of individuals who started the survey divided by the number of email invitations sent, less any email invitations that were returned as undeliverable.
Measures Sexting was defined as sexual photo-sharing through any mode. The behavior was queried based on a question developed by Lenhart and colleagues [ 10 ]: We are talking about times when you wanted to do these things. Please keep in mind that these things can happen anywhere including in-person, on the Internet, and on cell phones or text messaging.
Youth who had shared pictures online were asked follow-up questions about the most recent incident, including whether they knew the recipient offline and the age difference between the respondent and the recipient. A range of sexual activities ever engaged in were also queried.
Items 1, 2, 5, and 6 were modified from the Protecting the Next Generation project [ 19 ] and items 3 and 4 were created for this survey: Depressive symptomatology was measured with the Center for Epidemiologic Studies Depression Scale-revised item version for adolescents [ 20 ], social support with the Multidimensional Scale of Perceived Social Support [ 21 ], and past year substance use using measures from the Youth Risk Behavioral Survey [ 22 ].
Further detail is available upon request. Next, a validity check was applied i. Finally, missing responses i. The final analytical sample was 3, youth. Statistical significance was determined using F statistics, which are chi-square statistics that take the weighting scheme into account, for categorical data and linear regression for continuous data. Differences were again tested for statistical significance, using F statistics for categorical data and linear regression for continuous data.
Differences were quantified using logistic regression. Odds ratios were adjusted for demographic characteristics: The percentages by mode were similar to the overall sample. As shown in Table 1 , male and female youth who sent or showed sexual pictures were significantly older and more likely to be Hispanic as well as lesbian, gay, bisexual, or other non-heterosexual identity LGB. Males who sent or showed sexual photos were more likely to be living in a small town compared to males who did not send or show sexual pictures.
Females who sent or showed sexual photos were less likely to be born again Christian. Aside from age and sexual identity, differences were not noted between youth who did and did not send or show sexy photos among younger youth Table 2.
Among older youth, those who sent or showed sexual photos were more likely to be female, Hispanic ethnicity, or LGB and were less likely to be born again Christian.