OBJECTIVE. There is increasing evidence that youth exposure to sexual content on television shapes sexual attitudes and behavior in a manner that may influence reproductive health outcomes. To our knowledge, no previous work has empirically examined associations between exposure to television sexual content and adolescent pregnancy.
METHODS. Data from a national longitudinal survey of teens (12–17 years of age, monitored to 15–20 years of age) were used to assess whether exposure to televised sexual content predicted subsequent pregnancy for girls or responsibility for pregnancy for boys. Multivariate logistic regression models controlled for other known correlates of exposure to sexual content and pregnancy. We measured experience of a teen pregnancy during a 3-year period.
RESULTS. Exposure to sexual content on television predicted teen pregnancy, with adjustment for all covariates. Teens who were exposed to high levels of television sexual content (90th percentile) were twice as likely to experience a pregnancy in the subsequent 3 years, compared with those with lower levels of exposure (10th percentile).
CONCLUSIONS. This is the first study to demonstrate a prospective link between exposure to sexual content on television and the experience of a pregnancy before the age of 20. Limiting adolescent exposure to the sexual content on television and balancing portrayals of sex in the media with information about possible negative consequences might reduce the risk of teen pregnancy. Parents may be able to mitigate the influence of this sexual content by viewing with their children and discussing these depictions of sex.
The teen pregnancy rate in the United States has decreased remarkably since 1991; however, this rate remains the highest among industrialized nations (76.4 per 1000 pregnancies were among 15-19-year-old individuals in 2002).1 Nearly 1 million young women 15 to 19 years of age, or 20% of all sexually active women in this age group, become pregnant each year; the majority of these pregnancies are unplanned.2,3 The effects of a teen pregnancy (or pregnancy before age 20) on young mothers and fathers, their children, and society as a whole can be profound. Young mothers are more likely than others to drop out of school, to require public assistance, and to live in poverty. In fact, teen mothers have earnings that average less than one half of the poverty level.4 Moreover, these mothers are less likely to have effective parenting skills or adequate social support as parents.5,6 Teen fathers face similar challenges. For example, teen fathers complete an average of 1.3 years less education than do men who delay fatherhood until age 21.4 Teen fathers often enter the labor market sooner than their counterparts because of fatherhood and ultimately earn less in their twenties than their peers.7 Compared with children of older mothers, children born to adolescent mothers are at higher risk of low birth weight, limited fine motor skills, and low math and reading abilities; these children are also more likely to drop out of high school themselves.6,8
The factors that contribute to teen pregnancy are complex and interrelated. Research has established that individual (eg, lack of school attachment), social (eg, peer norms regarding sexual behavior), and environmental (eg, availability of contraception) influences converge in predicting teen pregnancy.9–13 One factor that is likely to be related to teen pregnancy but has not received empirical attention is exposure to television sexual content. In 1995, the American Academy of Pediatrics issued a policy statement on the role of the media in shaping attitudes about sexuality and contraception.14 Youths watch an average of 3 hours of television per day,15–17 much of which deals with sexual themes.15 The notion that television may play the role of a “sexual super peer,”18 influencing the sexual decision-making of youths, is supported by a study that established a prospective link between exposure to television sexual content and earlier initiation of sex. This association held even with controlling for more than a dozen factors that might plausibly explain the relationship in noncausal terms.19 The current study extends that research by investigating the link between exposure to television sexual content and teen pregnancy.
Teens who initiate sex at a younger age are more likely to become pregnant or to be responsible for a pregnancy.20 Indeed, a delay in initiation of sexual intercourse and an increase in the correct and consistent use of contraceptives among teens are the leading reasons why the overall teen pregnancy rate has declined in the past decade.10,21,22 The earlier onset of sexual activity associated with viewing of television sexual content that was documented by Collins et al19 may indicate, by extension, that teens who see more sex on television are more likely to become pregnant or to be responsible for another person's pregnancy.
Television also may increase the risk of teen pregnancy by encouraging lax attitudes toward the need for contraceptive use. Mention of the risks or responsibilities of sex, including pregnancy, sexually transmitted infections, and condom or contraceptive use, occurs in only 10% to 15% of programs containing sexual content.16,17,23 Social learning theory posits that overemphasizing the potential benefits of sex without portraying the potential costs may lead teens to perceive little risk in engaging in unprotected sexual behavior.24,25 Some indirect evidence of a link between television sex exposure and contraceptive use is provided by a study that showed that teens with greater exposure to sexually oriented music videos were 1.5 times as likely as teens with low exposure to contract a sexually transmitted infection.26 This relationship suggests that teens with high levels of exposure were less likely than teens with low levels of exposure to have used condoms, a key form of contraception among youths. In summary, exposure to sex on television may increase the risk of a teen pregnancy either by accelerating the initiation of sexual intercourse or by creating the perception that there is little risk to sex without contraceptive use (or both).
In studying the association between television exposure and teen pregnancy, it is important to recognize that television is one of many factors that may contribute to girls' experiences of pregnancy or to boys' responsibility for pregnancy. There are notable racial/ethnic differences in rates of early pregnancy. Hispanic/Latino (137.9 pregnancies per 1000 teens 15–19 years of age in 2000) and black (64.8 pregnancies per 1000 teens 15-19 years of age) teens have significantly higher rates than white teens (27.5 pregnancies per 1000 teens 15–19 years of age).1 These differences have been attributed to more-positive views toward early parenthood, limited educational or economic opportunities in traditionally disadvantaged minority communities, and inadequate access to or use of contraception.27,28 Living in a single-parent household,29 low parent education,30 limited teen educational or career aspirations,31 and ambivalent or positive attitudes toward pregnancy32,33 have been identified as critical influences on the timing of sexual initiation and first pregnancy. There is also evidence that teen pregnancy risk is greater among teens with lower grades34 and those who engage in delinquent behaviors.35 Most of these variables also have been linked to more-frequent television viewing and to greater sexual content viewing among youths and thus could confound the relationship between television viewing and pregnancy if not accounted for in tests of their association.36,37
The current study uses data from a national longitudinal sample of youths to test for a prospective relationship between exposure to sex on television and pregnancy during adolescence. The study builds on previous work by Collins et al19 that established a relationship between exposure to sex on television and earlier initiation of intercourse among teens. By using an additional (third) wave of data collected from the same panel of participants, we tested whether baseline exposure to sex on television also predicted whether teens experienced a pregnancy between the baseline survey (2001) and the second follow-up survey (2004). We hypothesized that greater exposure to sex on television would predict teen pregnancy among girls and responsibility for pregnancy among boys. In our analyses, we controlled for the potential confounding influence of several variables known to correlate with both television sex exposure and teen pregnancy.
The data for this study were from a national longitudinal survey of youths 12 to 17 years old at baseline. Participants were interviewed via telephone in spring 2001and were reinterviewed 1 and 3 years later, in spring 2002 and spring 2004 (time 2 and time 3, respectively). The sample was drawn from a commercially purchased list of households (based on residential telephone listings, school enrollment records, car registrations, magazine subscriptions, and warranty registrations) with high estimated probabilities of containing a 12- to 17- year old adolescent. To address the underrepresentation of ethnic minorities in these published household lists, relative to national census statistics, we oversampled within the frame on the basis of both frame-reported race/ethnicity and the characteristics of the census tract of residence. Parental consent and adolescent assent were obtained before the interview. The study was approved by the Rand human subjects protection committee.
Without weights, the baseline sample of teens (N = 2003) had demographic characteristics similar to those of US adolescents but included fewer Hispanic youths and youths with highly educated parents. We created nonresponse weights with a regression equation predicting nonresponse at baseline. After applying these weights, we created poststratification weights to correct for a small departure from the demographic characteristics cited in the 1999 Current Population Survey. The nonresponse and poststratification weights were combined to form the final baseline weights.19,38
Attrition and Longitudinal Weights
A total of 2003 teen participants completed baseline interviews. Seventy-three percent of this sample (n = 1461) was retained at time 3. Multivariate logistic regression modeling of attrition from baseline to time 3 revealed some selective attrition. Overall, attrition rates were higher among all races for teens >14 years of age at baseline, boys, and teens whose parents had greater educational attainment. Among black teens, attrition was higher among those with the least sexual activity at baseline and was lower among those who, at baseline, had not engaged in intercourse but had engaged in genital noncoital sexual activity. Results from this modeling were used to generate inverse-probability attrition weights, which were combined with the final baseline weights to produce longitudinal weights. All analyses used these weights, appropriately accounting for their effects on SEs.
To be eligible for inclusion in the current analyses, participants were required to have known sexual initiation status at time 3 (1315 of 1461 participants), to have engaged in sexual intercourse by time 3 (744 of 1315 participants), and to have complete information on pregnancy history at time 3 (718 of 744 participants). We did not exclude participants who had experienced a pregnancy before the start of the time 1 television viewing season (September 2000; n = 5), but those pregnancies were not counted toward the pregnancy outcome because they occurred before measured television exposure. The final analysis sample was 57% male, 66% white, 12% Hispanic, and 17% black. Sixty-nine percent of this sample resided in 2-parent families. The average age of these respondents at time 3 was 18.1 years (SD: 1.5 years; range: 13–21 years). Approximately 59% had ≥1 parent who had completed at least some college, and nearly 70% reported achieving mostly A's or B's in school. At baseline, 31% had already had sexual intercourse, and 13% expected to have their first child before they were 22 years of age. At time 3, the average time elapsed since respondents first had sexual intercourse was 2.3 years (SD: 1.7 years; range: 0–8 years).
Approximately 14% (n = 91; 58 girls and 33 boys) of the analysis sample had experienced or reported being responsible for a pregnancy after baseline. Among those who had experienced such a pregnancy, the average time since the most recent pregnancy was 1.2 years.
The survey measured television viewing, sexual knowledge, attitudes, and behavior, and a large set of demographic and psychosocial variables known to predict television viewing habits or sexual behavior. Teens were encouraged to respond to the survey in private. Questions were asked in a way that acknowledged the sensitive content and were designed so that no one listening to the responses (eg, yes, no, or most of the time) could infer the question content.
Pregnancy since baseline television exposure was the main outcome variable. For girls, the survey asked, “Have you ever been pregnant.” For boys, the survey asked, “Have you ever gotten a girl pregnant?” For both girls and boys, we asked, “In what month and year [were you most recently/did you most recently get a girl] pregnant?” Those reporting a pregnancy in any month after their baseline interview were coded as experiencing a post–television exposure pregnancy. (Although for girls the phrasing of the question could tap the end date for pregnancy, rather than the date of conception, availability of baseline responses to the question and an instruction to include current pregnancies allowed us determine that all pregnancies captured by our derived variable began after baseline television exposure.)
Exposure to Sexual Content on Television
As part of the baseline survey, teens indicated how frequently they watched each of 23 programs during the previous television season (from 1 = never to 4 = every time it's on). We selected programs that were popular with teens at the time of the survey and that contained high levels of sexual content. The selections included programs appearing on broadcast networks and basic and premium cable channels and encompassed animated and live-action shows, reality shows, sitcoms, and dramas.39 Kunkel et al16 developed methods to measure the sexual content of these programs as part of a much larger study, and we used extensively trained and experienced coders from that study to analyze our data. These coders categorized at least 3 and up to 14 episodes of each program into distinct scenes and then coded each scene for the presence of sexual behavior (with 5 subcategories, ie, physical flirting, passionate kissing, intimate touch, intercourse implied, and intercourse depicted) and sexual talk (including talk about own/others' sexual plans or desires, talk about sex that occurred in the past, talk between characters leading up to sex, expert advice, and talk about sex crimes). A given scene in a television program could contain talk and behavior, one or the other, or neither. Raters also indicated the degree of focus on sex in the scene as major or minor. In the larger study by Kunkel et al,16 interrater reliability ranged from 90% to 100% for the content and focus variables.
For each television series examined in the present study, the amount of sexual content was calculated as the average number of scenes per episode containing a major focus on sexual behavior plus the average number of scenes containing a major focus on talk about sex. The exposure measure was derived by multiplying indicators of the sexual content amount in an average episode of each program by how much the respondent watched the program and then adding these values across programs. We combined talk and behavior because we were interested in sexual content exposure more generally, and the 2 subtypes of exposure are correlated (when separate measures of exposure to talk and exposure to behavior were calculated, they were correlated at 0.93 among baseline participants). The combined measure has been shown to predict reliably intercourse initiation and other sexual activity19 and has been subjected to a wide range of sensitivity analyses to confirm its validity.39
Total Television Viewing
At baseline, we also measured how much time respondents spent watching television in general, with a set of items that assessed hours of viewing during different times of the day and days of the weeks. These items were averaged to derive a continuous measure of average viewing time (α = .70). Because youths who watch more sex on television also watch more television overall, it is critical to control for overall television viewing when investigating the influence of sexual content on television.
Sociodemographic variables at baseline included participants' age (in years, to 1 decimal place), gender (0 = male and 1 = female), race/ethnicity (Hispanic, black, and white/other, because there were insufficient numbers in the other group to analyze individually), family structure (0 = other type of family and 1 = 2-parent family), highest level of education completed by either parent (1 = high school to 6 = graduate or professional degree), and self-reported grades (1 = mostly A's to 5 = mostly F's). Participants' intention to have children before age 22 was assessed at baseline with the following items: “Do you think you'll have children?” and, for those responding yes, “Do you think you'll have your first child at age 17 or younger, age 18 to 21, or 22 or older?” We derived a dichotomous variable to discriminate those who intended to have children before age 22 from all others. We chose age 22 as the cutoff point to provide an approximate indicator of intent to have children before the completion of college. Educational aspirations were assessed with an item at baseline that asked, “What is the highest level of school you plan to finish?” (1 = high school to 4 = graduate or professional degree). Deviant behavior was measured with the sum of 6 baseline items drawn from previous studies of adolescent risk behavior (α = .65).40 Participants indicated how many times in the past 12 months they had been sent out of class/from work; had broke into a house, school, or place of business; skipped school; had cheated on a test; had damaged something on purpose that did not belong to them; or had stolen something (1 = not at all to 4 = ≥10 times).
Missing Data Imputation
Our predictor variables were missing at very low levels (<1%), with 2 exceptions, described below. Participants who were missing data on one variable were typically not the same people who were missing data on another and, if we had removed respondents with any missing values from our analysis, the remaining sample would have been significantly smaller. Also, those remaining might have been different from the original group in some way (eg, more educated), which would have introduced bias in our multivariate regression analyses.41 Therefore, we imputed missing predictor values by using regression-based imputation plus random residuals to reduce this bias. Regression imputation involves using participants' scores on other variables to predict responses to a missing item. Seventy-three respondents (10%) were missing data on perceived pregnancy risk from unprotected sex, and 140 respondents (19%) who admitted having had sex at ≥1 of the 3 survey waves were missing data on the age at first sex. A sensitivity test added an imputation flag for age at first pregnancy as a covariate to a secondary model. The flag was not significant (P > .05).
In preliminary analyses, we examined bivariate associations between exposure to sexual content on television and pregnancy and between pregnancy and each covariate. We also used logistic regression analyses to test the partial association between baseline exposure to sexual content on television and pregnancy after baseline, with adjustment for total television exposure. Our previous work18,34 indicated that youths who see more sexual content on television are also youths who watch more television overall; failure to account for this association results in sexual content exposure acting in part as a “stand-in” for overall viewing. This partial correlation better captures the “true” relationship between sexual content and pregnancy (before adjustment for demographic and other individual characteristics).
Our key analysis was a multivariate logistic regression analysis that examined the association between exposure to sexual content on television at baseline and pregnancy after baseline, controlling for all covariates. Given that reporting of pregnancy may differ according to gender (ie, the factors involved in knowing that you are pregnant versus knowing that you impregnated someone else are possibly different), we used multiple-group models to test whether parameters for each of our models differed according to gender. Our analyses found no evidence that any of the candidate predictors functioned differently in predicting pregnancy for female versus male teens. Therefore, we report the results of single-group analyses that were conducted with the sample as a whole.
The bivariate associations between the baseline predictor variables and pregnancy after baseline are shown in Table 1. The simple association between sexual content exposure and pregnancy was positive but not statistically significant. However, overall viewing was nonsignificantly negatively related to pregnancy, which suggests that there may be a suppression effect. Indeed, sexual content on television was strongly associated with overall frequency of television viewing (β = .29; P < .001; data not shown); when we controlled for overall viewing, the association between sexual content and pregnancy was statistically significant (β = .30; P < .05; data not shown).
Several of the covariates were related to pregnancy. Adolescents living in 2-parent families had a lower probability of pregnancy; female participants, black participants, and youths with more deviant behavior had higher probabilities, as did youths who intended to have children early.
The positive association between exposure to sexual content and pregnancy became even stronger when all covariates were included in the multivariate regression model (β = .44; P = .03) (Table 2). To illustrate the magnitude of this effect, we used the full sample to translate the coefficients from our multivariate regression model into predicted probabilities of pregnancy at the 10th (low), 50th (medium), and 90th (high) percentiles of baseline exposure to sexual content on television, holding all other covariates at their naturally occurring values. First, the sample was assigned the exposure value that was observed at the 10th percentile and predicted values were obtained, then the sample was reassigned the value at the 50th percentile, and so on. This produced estimates of the probability of pregnancy at each exposure level that were based on the estimated associations between exposure and pregnancy in the original regression model and that accounted for standing on covariates.42 We did this for each of 5 age groups. As Fig 1 shows, higher levels of exposure were associated with a higher predicted probability of pregnancy, with ∼2 to 3 times the risk of pregnancy for youths exposed to high versus low levels of televised sexual content. For example, if the entire sample had been exposed to low levels of sexual content on television at baseline, then 5% of 16-year-old youths would have been expected to experience a pregnancy between baseline and wave 3, compared with 12% if everyone had been exposed to high levels of sexual content on television.
Covariates that remained significant predictors of an increased likelihood of pregnancy in this multivariate model include female gender, deviance or problem behavior, black race, and not living in a 2-parent family. Intention to have a child early in life was not significantly associated with pregnancy in the multivariate model, possibly because of its association with ≥1 of the other risk factors. Two predictors that were not significantly associated with pregnancy at the bivariate level, that is, total television viewing and age, emerged as significant predictors in the multivariate model. Watching more television in general was negatively associated with the likelihood of teen pregnancy (perhaps because sexual content was controlled). Older members of our sample were more likely to have experienced or caused a pregnancy.
This research represents an important step in understanding the relationship between exposure to sexual content in the media and adolescent sexual and reproductive health. Our results indicate that frequent exposure to sexual content on television predicts early pregnancy, even after accounting for the influence of a variety of other known correlates of each. Previous analyses of this longitudinal data set identified a prospective association between exposure to televised sexual content and initiation of sex.19,43 Although earlier sexual initiation is associated with negative health outcomes, our previous work did not establish a direct link between exposure to television sexual content and this reproductive health outcome. The results of the present study suggest that adolescents who view substantial televised sexual content have an increased risk of experiencing a pregnancy before age 20, compared with youths who view less sexual content on television. Given the negative consequences of early pregnancy for young people and for society, it is critical to recognize and to understand this association.
To our knowledge, this study is the first to demonstrate an association between exposure to sexual content on television and a reproductive health outcome. A previous study showed a relationship between music video viewing and sexually transmitted diseases,26 but that investigation did not specifically link the sexual content in videos to that negative health outcome. The 2 studies are clearly complementary, however, and together suggest that television viewing can influence multiple aspects of reproductive health among youths. Young people receive a considerable amount of information about sex through television, and this information does not typically highlight the risks and responsibilities of sex. Our results suggest that television may have a substantial role in the high rates of teenage pregnancy in the United States. High rates of exposure corresponded to twice the rate of observed pregnancies seen with low rates of exposure. Although the present analysis cannot establish definitively how much, if any, of the observed association between exposure to sexual content and pregnancy is causal, the magnitude of the association is strong enough that, even if only a fraction of the observed association is causal, reducing the exposure of US teens to sexual content might substantially reduce teen pregnancy rates.
The results of this study have several practical applications. Although advocating for self-regulation of the media is challenging in many ways, it is important to continue to encourage industry leaders of the media to include in television programs messages about the consequences young people may face as a result of engaging in sexual activity, such as pregnancy. Efforts to improve literacy regarding the media among adolescents should highlight the relative absence of portrayals of negative consequences of sex on television and should encourage youths to consider alternative outcomes of the sexual behavior portrayed on television. The study findings also reinforce the importance of educating pediatricians about television's influence on child health, as part of residency training programs. In 1997, the American Academy of Pediatrics implemented Media Matters, an initiative to educate pediatricians about the role of the media and to offer clinical tools such as a media history form to use in medical histories.44 However, Rich and Bar-on45 found that, despite more awareness about the influence of the media on child health, very few pediatric training programs actually teach about exposure to the media. Pediatricians should consider incorporating conversations about the use of media (eg, unrealistic portrayals of sex on television) as part of anticipatory guidance for parents and/or as part of conversations and/or screening tools with adolescents (eg, Guidelines for Adolescent Preventive Services).46,47 In addition, pediatricians should advocate parental coviewing and should encourage parents to discuss television portrayals of sex and the consequences of this behavior with their children. Our analyses also are consistent with previous research indicating that girls and black teens are at greater risk for pregnancy during adolescence.1,48 Our findings suggest that these groups may be important to include in interventions; however, our data do not provide evidence on the differential impact of television exposure among these groups, to argue definitively for targeting education regarding media or related interventions. As additional research clarifies the mechanisms through which television sex exposure and teen pregnancy are linked, more-focused intervention strategies should become apparent.
This study also highlights avenues for future research. Given previous research that asserts the protective role of consistent and correct contraceptive use against early pregnancy,9,20 future studies on this topic should include contraceptive use profiles that allow a test of whether inconsistent contraceptive use may help to explain the relationship between exposure to sex on television and pregnancy. Future analyses also might examine whether watching sex on television predicts particular birth outcomes (eg, abortion versus carrying a pregnancy to term).
A few limitations of our study should be considered. First, although our model included a wide range of potentially confounding factors as covariates, there is the possibility that we did not account for all factors that may alternatively explain the relationship we uncovered. The magnitude of the observed association is so great, however, that, if even 20% of the prospective association that remained after controlling for covariates was causal, then sex exposure in the media would be a policy and public health arm of tremendous leverage. It should be noted that, for these analyses, we focused on the overall exposure to sexual content on television and not the types of messages. Therefore, we cannot describe the differential influence of content, although we know that very little television programming contains messages related to sexual risks and responsibilities.16,17,23 Second, the small sample sizes for black and Hispanic teens precluded examination of the relationship within racial subgroups. Future research should attempt to determine whether the relationship we observed is invariant across racial and ethnic subgroups. Given that there are varying norms and values related to pregnancy according to race/ethnicity and that viewing patterns differ according to race/ethnicity,49 the social learning provided by television may have a different effect among these groups. Brown et al50 recently found that the relationship between exposure to sexual content across multiple media and sexual initiation was strong among white adolescents but was not statistically significant among black adolescents, even after adjustment for factors such as parental disapproval of sex. Third, as mentioned earlier, we were unable to explore the birth outcomes for all of the early pregnancies. For example, we do not know whether these pregnancies were unwanted or unplanned or whether they resulted in abortions or live births. We do know that, in our data set overall, one quarter of pregnancies (girls' reported pregnancies and boys' reported pregnancies of female partners) resulted in an abortion. We did not have sufficient statistical power to assess whether sexual content exposure predicted abortion in our data set, but future research with larger sample sizes or greater proportions of high-risk youths may be able to address this.
Despite these limitations, our study clearly suggests that television plays a role in shaping adolescent reproductive health outcomes. Given the high rates of teen pregnancy in the United States, our findings argue for continued evaluation of television's role in teen pregnancy and for prevention strategies that factor in the role of portrayals of sexuality on television and in other media. Moreover, this work supports the American Academy of Pediatrics recommendations that pediatricians encourage family conversations about the influence of media on sexual behavior.
- Accepted February 26, 2008.
- Address correspondence to Anita Chandra, DrPH, Rand Corp, 1200 South Hayes St, Arlington, VA 22202. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Early findings suggest that exposure to greater amounts of sexual content on television may be linked to early initiation of sexual intercourse among adolescents. How this exposure is linked to sexual and reproductive behaviors has not been explored thoroughly.
What This Study Adds
We identified a link between exposure to sexual content on television and the experience of a teen pregnancy.
- ↵Ventura SJ, Abma J, Mosher W, Henshaw S. Recent Trends in Teenage Pregnancy in the United States, 1990–2002. Hyattsville, MD: National Center for Health Statistics; 2006
- ↵Maynard RA. Kids Having Kids: Economic Costs and Social Consequences of Teen Pregnancy. Washington, DC: Urban Institute; 1997
- ↵Terry-Human E, Manlove J, Moore KA. Playing Catch-up: How the Children of Teen Mothers Fare. Hyattsville, MD: National Center for Health Statistics; 2005
- ↵Pirog-Good MA. The education and labor market outcomes of adolescent fathers. Youth Soc.1996;28 (2):236– 262
- ↵American Academy of Pediatrics, Committee on Communications. Sexuality, contraception, and the media. Pediatrics.1995;95 (2):298– 300
- ↵Rideout VJ, Foehr UG, Roberts DF, Brodie M. Kids and the Media @ The New Millennium: A Kaiser Family Foundation Report: A Comprehensive National Analysis of Children's Media Use. Menlo Park, CA: Henry J. Kaiser Family Foundation; 1999
- ↵Kunkel D, et al. Sex on TV3: A Biennial Report to the Kaiser Family Foundation. Menlo Park, CA: Henry J. Kaiser Foundation; 2003. Available at: www.kff.org/entmedia/upload/Sex-on-TV-3.pdf. Accessed November 2, 2006
- ↵Lowry DT, Shindler JA. Prime time TV portrayals of sex, “safe sex,” and AIDS: a longitudinal analysis. Journal Q.2003;70 (3):628– 637
- ↵Collins RL, Elliott MN, Berry SH, et al. Watching sex on television predicts adolescent initiation of sexual behavior. Pediatrics.2004;114 (3). Available at:www.pediatrics.org/cgi/content/full/114/3/e280
- ↵Kunkel D, Cope-Farrar K, Biely E, Farinola WJM, Donnerstein E. Sex on TV2: A Biennial Report to the Kaiser Family Foundation. Menlo Park, CA: Henry J. Kaiser Foundation; 2001. Available at: www.kff.org/entmedia/upload/A-Biennial-Report-to-the-Kaiser-Family-Foundation-Report.pdf. Accessed September 17, 2007
- ↵Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall; 1986
- ↵Bushman BJ, Huesmann LR. Effects of televised violence on aggression. In: Singer D, Singer J, eds. Handbook of Children and the Media. Thousand Oaks, CA: Sage; 2001:223– 254
- ↵South SJ, Baumer EP. Deciphering community and race effects on adolescent premarital childbearing. Soc Forces.2000;78 (4):1379– 1407
- ↵Roberts DF, Foehr UG, Rideout V. Generation M: Media in the Lives of 8–18 Year-Olds. Menlo Park, CA: Henry J. Kaiser Foundation; 2003. Available at: www.kff.org/entmedia/entmedia030905pkg.cfm. Accessed June 3, 2006
- ↵Martino SC, Collins RL, Elliott MN, Strachman A, Kanouse DE, Berry SH. Exposure to degrading versus nondegrading music lyrics and sexual behavior among youth. Pediatrics.2006;118 (2). Available at:www.pediatrics.org/cgi/content/full/118/2/e430
- ↵Collins RL, Elliott MN, Miu A. Linking media content to media effects: the RAND Television and Adolescent Sexuality (TAS) study. In: Kunkel D, Jordan A, Manganello J, Fishbein M, eds. Media Messages and Public Health: A Decisions Approach to Content Analysis. Mahwah, NJ: Lawrence Erlbaum Associates. In press
- ↵Rich M, Bar-on M. Child health in the information age: media education for pediatricians. Pediatrics.2001;107 (1):156– 162
- ↵Schmidt ME, Rich M. Media and child health: pediatric care and anticipatory guidance in the information age. Pediatr Rev.2006;27 (8):289– 298
- ↵American Medical Association. Guidelines for Adolescent Preventive Services (GAPS). Available at: www.ama-assn.org/ama/pub/category/1980.html. Accessed September 15, 2007
- ↵Brown JD, L'Engle KL, Pardun C, Guo G, Kenneavy K, Jackson C. Sexy media matter: exposure to sexual content in music, movies, television, and magazines predicts black and white adolescents' sexual behavior. Pediatrics.2006;117 (4):1018– 1027
- Copyright © 2008 by the American Academy of Pediatrics