pediatrics
January 2012, VOLUME129 /ISSUE 1

Prevalence and Characteristics of Youth Sexting: A National Study

  1. Kimberly J. Mitchell, PhD,
  2. David Finkelhor, PhD,
  3. Lisa M. Jones, PhD, and
  4. Janis Wolak, JD
  1. Crimes against Children Research Center, University of New Hampshire, Durham, New Hampshire

Abstract

Objectives: To obtain national estimates of youth involved in sexting in the past year (the transmission via cell phone, the Internet, and other electronic media of sexual images), as well as provide details of the youth involved and the nature of the sexual images.

Methods: The study was based on a cross-sectional national telephone survey of 1560 youth Internet users, ages 10 through 17.

Results: Estimates varied considerably depending on the nature of the images or videos and the role of the youth involved. Two and one-half percent of youth had appeared in or created nude or nearly nude pictures or videos. However, this percentage is reduced to 1.0% when the definition is restricted to only include images that were sexually explicit (ie, showed naked breasts, genitals, or bottoms). Of the youth who participated in the survey, 7.1% said they had received nude or nearly nude images of others; 5.9% of youth reported receiving sexually explicit images. Few youth distributed these images.

Conclusions: Because policy debates on youth sexting behavior focus on concerns about the production and possession of illegal child pornography, it is important to have research that collects details about the nature of the sexual images rather than using ambiguous screening questions without follow-ups. The rate of youth exposure to sexting highlights a need to provide them with information about legal consequences of sexting and advice about what to do if they receive a sexting image. However, the data suggest that appearing in, creating, or receiving sexual images is far from being a normative behavior for youth.

KEY WORDS
  • sexting
  • child pornography
  • Internet
  • naked images
  • Abbreviations:
    AAPOR
    American Association for Public Opinion Research
    YISS
    Youth Internet Safety Survey
  • What’s Known On This Subject:

    Educators, public health authorities, and law enforcement are confronting an increasing number of cases in which youth made sexual images of themselves and other minors and transmitted them via cell phones and the Internet.

    What This Study Adds:

    This study provides the first detailed and comprehensive national estimate of the percentage of youth who create and distribute various kinds of sexual images.

    Several concerns have fueled the considerable attention to the problem of “youth sexting” among the media, parents, professionals, educators, and law enforcement.13 (Sexting generally refers to sending sexual images and sometimes sexual texts via cell phone and other electronic devices.) One concern is that youth may be creating illegal child pornography, exposing them to possibly serious legal sanctions.4,5 Another is that youth may be jeopardizing futures by putting compromising, ineradicable images online that could be available to potential employers, academic institutions, and family members.

    These concerns have been abetted by frequently cited statistics about the supposed widespread teen involvement in sexting. The most common reference has been to a National Campaign to Prevent Teen and Unplanned Pregnancy study6 revealing that 20% of teeangers had sent or posted nude or semi-nude pictures of themselves. However, this research, as well as other often cited studies,7,8 has flaws that compromise the findings.9 For example, the authors of the National Campaign study used an Internet panel rather than a true population sample and included 18 and 19 year olds, and not just minors.

    Moreover, none of these studies has made distinctions that allow a careful assessment of the problem from a policy perspective. The authors of studies have asked respondents about “nude or semi-nude,” “nearly nude,” or “sexually suggestive” images that might, in fact, be no more revealing than what someone might see at a beach. In some studies, sexting was defined to include text messages that could contain no images. And many studies did not distinguish between taking and sending an image of oneself as opposed to receiving or disseminating an image of another youth. For policy purposes, it is important to look at whether images are created or simply received and whether images might qualify as child pornography, but such information is not currently available.

    Our research is the first to assess in detail the range of youth sexting behaviors, including the content of images that youth receive, distribute, and create. It is intended to give parents, policy makers, and professionals a more accurate assessment of the scope of sexting.

    Methods

    The Third Youth Internet Safety Survey (YISS-3) was conducted to quantify and detail unwanted or problematic technology-facilitated experiences among youth, including sexting. Data collection occurred between August 2010 and January 2011. YISS-3 was conducted via telephone surveys with a national sample of 1560 youth Internet users, ages 10 to 17, and their parents. A sample size of 1500 was predetermined based upon a maximum expected sampling error of ±2.5% at the 5% significance level. Human subject participation was reviewed and approved by the University of New Hampshire Institutional Review Board and conformed to the rules mandated for research projects funded by the US Department of Justice.

    Abt Schulman, Ronca, and Bucuvalas, Inc, a national survey research firm, conducted the sampling, screening, and telephone interviews for YISS-3. The main sample was drawn from a national sample of households with telephones developed by random digit dialing. Using standard dispositions as defined by the American Association for Public Opinion Research (AAPOR),10 the cooperation rate was 65% (AAPOR Cooperation Rate 4-interviews/estimated eligible) and the refusal rate was 24% (AAPOR Refusal Rate 2-refusals/estimated eligible). Only a minority (1.3%) of cooperating households were not eligible due to no or limited Internet access. Due to increasing reliance of the US population on cell phones only,11,12 a cell-phone random digit dial sample was included in addition to the landline sample in the YISS-3 study. The original intention was to include a sample of 300 respondents from the cell phone sample in the final target sample of 1500. However, due to problems with cell phone sample response rates, and given the required timeframe for the study, a decision was made to complete the survey once a total of approximately 1500 landline completions had been reached. At the end of data collection, 45 interviews had been completed by cell phone in addition to 1515 landline interviews, resulting in a total sample size of 1560. Analysis of youth demographic and Internet use characteristics between the cell phone and landline samples indicated the cell phone sample was accessing a harder to reach population of youth. Specifically, youth in the cell phone sample were more likely to be of Hispanic ethnicity and come from families with a single, never married parent.

    Sample

    Eligible respondents were youth, ages 10 to 17, who had used the Internet at least once a month for the past 6 months from any location and a caregiver in each household. Eligibility criteria were consistent with 2 previous YISSs.13,14Table 1 provides details of the sample characteristics.

    TABLE 1

    Demographic Characteristics Based on Reports of Appearing in, Creating, or Receiving Nude or Nearly Nude Images or Videos

    Procedure

    In households with eligible children, interviewers asked to speak with the adult who was most familiar with that child’s Internet use, and after receiving informed consent, interviewers asked a series of questions about Internet use. Then the interviewer requested permission to interview the child. Interviewers told parents that the youth interview would be confidential and include questions about “sexual material your child may have seen on the Internet,” and that youth would receive $10 for participating. In households with more than 1 eligible youth, the one who used the Internet the most often was chosen as the respondent.

    After receiving parental permission, interviewers spoke with the youth and requested permission to conduct an interview. Interviewers assured youth that answers would be confidential and they could skip any question and end the interview at any time. Steps were taken to help ensure confidentiality and safety for youth participants, including asking mostly yes or no questions, checking at regular intervals that youth were in a private location, and providing Internet safety resources at the end of the interview. The average youth interview lasted 30 minutes and the average adult interview lasted 10 minutes.

    Measurement

    We created a series of 5 screener questions that asked about 3 types of sexting involvement: (1) receiving “nude or nearly nude” images, (2) forwarding or posting such images, and (3) appearing in or creating such images. When youth answered yes to screeners, follow-up questions gathered details about their responses, including the content of the nude or nearly nude images. The screeners asked:

    1. Has anyone ever sent you nude or nearly nude pictures or videos of kids who were under the age of 18 that someone else took?

    2. Have you ever forwarded or posted any nude or nearly nude pictures or videos of other kids who were under the age of 18 that someone else took?

    3. Have you ever taken nude or nearly nude pictures or videos of yourself?

    4. Has someone else ever taken nude or nearly nude pictures or videos of you?

    5. Have you ever taken nude or nearly nude pictures or videos of other kids who were under the age of 18?

    When youth responded positively to a screener question, interviewers asked if the incident occurred in the past year. Interviewers then asked extensive follow-up questions about up to 2 unique past year sexting episodes. Our prevalence estimates were created based on youth-level data, some of whom reported more than 1 sexting type incident. An algorithm was used to choose incidents for follow-up with a hierarchy that selected first for incidents in which pictures were taken and second for incidents in which pictures were distributed. No youth were left uncounted based on this algorithm.

    Demographic information was gathered from both the parent or caregiver and the youth respondent. Parents provided information about youth age and gender; parental marital status; and household composition, education, and income. Youth reported on their race and ethnicity.

    Results

    Types of Sexting Involvement

    A total of 149 youth (9.6%) reported appearing in or creating nude or nearly nude images or receiving such images in the past year. (Only 1 youth reported forwarding or posting, and that was part of a first priority incident, so it was not counted separately.) See Table 1 for the personal characteristics of youth involved in sexting.

    We differentiated 3 dimensions of the incidents that youth reported (Fig 1). The first dimension was whether youth appeared in or created images versus receiving images. Of youth reporting involvement in sexting, 39 (2.5%) appeared in or created images; 110 youth (7.1%) received images but did not appear in or create them. The second dimension, among those who appeared in or created images, was whether a youth was pictured in an image. The third dimension was whether images were sexually explicit (ie, might qualify as child pornography; more discussion below).

    FIGURE 1

    Types of sexting involvement (percentages based on N = 1560).

    Respondents Who Appeared in or Created Images

    Of the 39 youth who appeared in or created images, 61% were girls, 72% were ages 16 or 17, and 6% were 10 to 12 (Table 2). Most youth created images of themselves (1.8% of sample, n = 28); some were photographed by someone else (0.3%, n = 5); and some photographed other youth (0.4%, n = 6). Youth stated:

    TABLE 2

    Youth and Incident Characteristics (n = 149)

    “I was just dating a boy and he wanted a picture and I just sent him my picture.”

    “Well, I did not have a boyfriend at this time, and I was curious as to what my body would look like to other people …, so I took some pictures.”

    “I was getting dressed at my boyfriend’s house and this girl was there and she took a pic of me and sent it to someone and it got around the school, and after a month it went away.”

    “We were just messing around and being guys. It wasn't anything sexual.”

    Respondents Who Received Images

    Of the 110 youth who received images but did not appear in or create them, 56% were girls; 55% were ages 16 or 17, and none were younger than 12. Youth stated:

    “Someone sent me a picture of my boyfriend and another girl, and he's no longer my boyfriend.”

    “[I was] sitting in room and playing guitar. Got text message. Opened it. It showed pictures of breasts, vagina. I immediately erased it.”

    “This boy had 4 pictures of a naked girl—he was showing everybody in the classroom.”

    “In girls’ locker room and some girl asked if anyone wanted to see a pic of her and her boyfriend and we thought it would be them hanging out but they were in bed together.”

    Were Images Sexually Explicit?

    One of the goals of this study was to determine how youth define nude or nearly nude because this phraseology has been used in previous studies and been the basis of reported statistics on sexting. We asked youth whether images “showed breasts, genitals, or someone’s bottom.” Only 54% of the 39 youth who appeared in or created images reported pictures that met these criteria, as did 84% of the 110 youth who received images (Table 3). For 46% of youth appearing in or creating images and 16% of those receiving images, nude or nearly nude included youth wearing underwear or bathing suits, sexy poses with clothes on, and pictures focused on clothed genitals.

    TABLE 3

    Characteristics of Nude or Nearly Nude Images or Videos of Minors (N = 149)

    Other Key Features of Sexting

    Emotional Impact

    Twenty-one percent of respondents appearing in or creating images reported feeling very or extremely upset, embarrassed, or afraid as a result, as did 25% of youth receiving images (Table 2).

    Disclosure

    Twenty-eight percent of youth who appeared in or created images and 28% of those who received images either reported incidents to an authority (eg, parent, teacher, or police) or an authority found out in some other way.

    Chronicity

    Youth stated that over half of the incidents in both categories occurred more than once in the past year.

    Context and Aggravating Features

    In most of the episodes, the person responsible (when it was not the respondents themselves) was someone the youth knew in person. The most commonly reported reason for incidents was “romance as part of an existing relationship”; pranks and jokes or trying to start a relationship were also noted. A notable minority of incidents where youth appeared in or created images (31%) included an aggravating component—usually alcohol or drug use (Table 3). Adults were involved in a minority of sexting incidents; they were all young adults, ages 18 to 21.

    Distribution

    One of the concerns about sexting is that youth will forward and distribute images they create or receive. However, in follow-up questions, only a small proportion of youth reported forwarding or posting images. Photographs were distributed in 10% of incidents when youth appeared in or created images and in 3% when youth received images.

    Prevalence Rates and 95% Confidence Intervals

    Table 4 provides national prevalence estimates and 95% confidence intervals using both more and less restrictive definitions of sexting. If sexting is defined as youth creating images of themselves that include their naked breasts, genitals, or bottom, the rate of involvement is 1.0%. If sexting is defined as receiving images that depict the breasts, genitals, or bottom of a minor, the rate is 5.9%. If a wider category of appearing in, creating, or receiving nude or nearly nude images is used, the rates rise to 2.5% and 7.1%, respectively.

    TABLE 4

    Prevalence Rates (%) and 95% Confidence Intervals for Sexting Involvement

    Discussion

    This study reveals that estimates of youth involved in sexting vary considerably depending on what activities are included in the concept of sexting. The percentage of youth who have, in the past year, appeared in or created sexually explicit sexual images that potentially violate child pornography laws is low (1%). But if sexting is defined as appearing in, creating, or receiving sexually suggestive rather than explicit images, the survey reveals 9.6% of youth who used the Internet in the past year involved in this way. The authors of many previous surveys on sexting have used the more expansive definition that captures sexually suggestive images and includes receiving such images, with percentages similar to our 9.6%.9 However, the much more detailed information obtained by the current survey suggests that the percentages of youth who appear in or create sexually explicit photographs that could meet the definition of child pornography are much lower.

    In the face of some widely cited, but flawed, studies claiming to show as many as 1 in 5 youth “involved in sexting”6 these results are to some extent reassuring. Only a low percentage of young people are appearing in or creating sexting images that could be considered illegal child pornography. Moreover, few of these images were being forwarded or posted, situations that could put youth at risk for having their images circulated online. Our lower and more accurate measurements may be particularly important for those interested in applying a “norms-based” approach to prevention, one that tries to dissuade youth from the perception that risky behaviors are the norm within their peer group. But receiving and thus possession of potentially illegal images among young people is widespread enough that education about this and its consequences is strongly warranted. Young people need to be instructed that the possession of sexually explicit images of minors is currently a criminal offense and that such images should be deleted and never retransmitted.

    Sexting of explicit images involves a low percentage but still a considerable number of youth. This raises the question of how the law should treat such cases. Subjecting youth to severe penalties for activities that would be legal for an 18 year old as long as no exploitation was involved is increasingly being recognized as draconian. States such as Vermont have taken steps to decriminalize some forms of this behavior, whereas others have reduced the severity to misdemeanor status.15,16

    Our findings also raise the question of how sexting should be defined. As is often the case with popularly inspired neologisms, the term sexting may be fatally compromised by its multiple and expansive colloquial use. We recommend that authors of all future research on sexting report results in such a manner that researchers and policy makers can identify policy relevant subsets, for example, those involving sexually explicit images. The findings also reveal that it is misleading to try to assess sexting behavior with single screening questions and ambiguous terms. Clearly, for many youth nude or nearly nude encompasses pictures that do not show naked breasts or genitals. Researchers and clinicians need to directly ask about the content of images. The findings also reveal that it is important to distinguish whether youth simply received images or appeared in or created them. Many fewer youth are involved in the latter than the former, and once again, follow-up questions are essential to establish how central a role youth played.

    Our findings should be interpreted within the confines of the limitations. First, as with all self-report measures, some youth respondents may not have disclosed their sexting involvement. Second, limiting participants to those that speak English is a drawback to the study. Finally, YISS-3 consists of a sample of young Internet users; sexting involvement does not necessarily have to involve the Internet. This could result in an undercount of youth involved in sexting, although we feel the number of youth missed is low given the high percentage of youth (93% of teens, aged 12 to 17) who used the Internet from any location in 2009.17

    Conclusions

    There is a tendency in our rapidly evolving society to be easily alarmed about changing youth mores, a tendency we have referred to elsewhere as “juvenoia.”18 Sexting has been greeted in many media portrayals as yet another sign of the hypersexualization of youth and extreme risk-taking. In fact, however, many indicators of youth sexual behavior such as teenage pregnancy and the number of youth with multiple sexual partners have been improving in recent years,19 in spite of such concerns. It is incumbent on youth-serving professionals not to respond or abet media portrayals that promote alarm. Sexting may not indicate a dramatic change in youth risk-taking or youth sexual behavior. It may just make some of that behavior more visible to adults and other authorities. Good research and sympathetic clinical assessment is necessary to understand the nature and extent of activities such as sexting before strong recommendations about how to counsel and educate youth and their families are developed and disseminated.

    Acknowledgment

    For the purposes of compliance with Section 507 of PL 104-208 (the “Stevens Amendment”), readers are advised that 100% of the funds for this program are derived from federal sources. This project was supported by grant 2009-SN-B9-0002 awarded by the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, US Department of Justice. The total amount of federal funding involved is $734 900.

    Footnotes

      • Accepted September 19, 2011.
    • Address correspondence to Kimberly J. Mitchell, PhD, Crimes against Children Research Center, University of New Hampshire, 10 West Edge Dr, Suite 106, Durham, NH 03824. E-mail: kimberly.mitchell{at}unh.edu
    • All authors have made substantial intellectual contributions to this study and article in each of the following categories: (1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be published.

    • Points of view or opinions in this article are those of the authors and do not necessarily represent the official position or policies of the US Department of Justice.

    • FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

    • COMPANION PAPER: A companion to this article can be found on page 4, and online at www.pediatrics.org/cgi/doi/10.1542/peds.2011-2242.

    References

    SMILEY FACES: I could hardly believe my eyes. There, at the end of a short text from my wife, was an emoticon. This was surprising for many reasons. First, my wife is not a natural when it comes to texting. She uses an older phone in which each button represents three letters. So for her, texting takes a long time as she deliberately taps each key and waits for the correct letter to appear on the screen. Secondly, we have known each other a very long time and can usually ascertain each other's mood or intent easily through written or spoken words. So why the ☺? Did she think I would not understand? As reported in The New York Times (Fashion: October 21, 2011), emoticons, which have been a mainstay of emails and texts between teens, can now be found in conversations between adults and even among professionals in the business community. Some use them to make sure the receiver understands the intent and to avoid any miscommunication. This may be particularly important in an age where much communication is devoid of tone. Others use them to provoke a smile particularly if not a demonstrative person. While teens may use emoticons all the time, in the professional world they tend to be reserved for use in congenial relationships. As the use of emoticons has exploded so has the number of symbols. There are symbols for happiness and sadness of course, but also action (e.g. a hug), or an activity (e.g. music). This can lead to some problems. While a little yellow smiley face (or frown) can be helpful in conveying a particular emotion, not all symbols transfer across platforms well. For example, a face or hug on one platform may appear as a series of punctuation marks in another, some of which may be confusing. Others find the use of emoticons abhorrent. Language should be specific enough to convey emotions and the need for pictorial representation is yet another example of the degradation of writing skills. While I never use emoticons in my professional correspondence, I happily returned my wife's text simply with a ☺.

    Noted by WVR, MD