Unwanted and wanted exposure to online pornography in a national sample of youth Internet users (2007)

Participants

We used telephone interviews conducted between March and June 2005 to gather information from a national sample of youth Internet users. The research was approved by the University of New Hampshire institutional review board.

Participants were 1500 youth aged 10 to 17 years (mean age: 14.24 years; SD: 2.09 years) who had used the Internet at least once per month for the past 6 months. Sample characteristics are shown in Table 1. Well-educated, prosperous families and white individuals were overrepresented in the sample but approximated the population of youth Internet users at the time of data collection.³⁰

Procedure

The sample was drawn from a national sample of households with telephones, developed through random-digit dialing. Details about the dispositions of the numbers dialed and a more-detailed description of the method can be found in other publications.^14,29 Short interviews were conducted with parents, and then youth were interviewed with parental consent. Youth interviews were scheduled at the youth’ convenience, when they could talk freely and confidentially. The average interview lasted ∼30 minutes.

The response rate, based on standard guidelines promulgated by the American Association for Public Opinion Research, was 45%.³¹ This rate, which is lower than rates typical of surveys in earlier decades, is in line with other recent scientific household surveys,³² which continue to obtain representative samples and to provide accurate data about the views and experiences of US populations, despite lower response rates.³³

Measures

Analyses

We used SPSS 14.0 (SPSS, Chicago, IL) for all analyses. First, we used descriptive statistics to examine rates of unwanted and wanted exposure to online pornography in the past year, on the basis of age and gender. Second, we used χ² cross-tabulations to determine which demographic, Internet use, prevention, and psychosocial characteristics were associated with unwanted and wanted exposure at the bivariate level. Third, we created a multinomial logistic regression model of the characteristics associated with unwanted or wanted exposure, with likelihood ratio tests for significant contribution to the overall statistical model at the .05 level. The reference category was youth with no exposure. Because we expected age and aspects of Internet use to exert strong influences on the results, we included all variables that were significant at the .25 level in bivariate analyses.³⁵

Unwanted and Wanted Exposure Among Youth Internet Users According to Age and Gender

Forty-two percent (n = 603) of youth Internet users had been exposed to online pornography in the past year.

Of the exposed youth, 66% (n = 400) reported only unwanted exposure and 34% (n = 203) reported either wanted exposure only (n = 91) or both wanted and unwanted exposure (n = 112).

Although only 1% of 10- to 11-year-old boys reported wanted exposure in the past year, the proportion increased to 11% of boys 12 to 13 years of age, 26% of those 14 to 15 years of age, and 38% of those 16 to 17 years of age (Fig 1).

Unwanted exposure also increased with age. Seventeen percent of boys 10 to 11 years of age had unwanted exposure in the past year, as did 22% of boys 12 to 13 years of age, 26% of those 14 to 15 years of age, and 30% of those 16 to 17 years of age. These were mutually exclusive categories and, for example, more than one half of male youth Internet users 14 to 15 years of age had been exposed to either unwanted or wanted online pornography in the past year, as had more than two thirds of those 16 to 17 years of age.

View larger version:

• In this page
• In a new window

• Download as PowerPoint Slide

FIGURE 1

Unwanted and wanted exposure to online pornography among boys (n = 727). Gender data were missing in 2 cases.

Little wanted exposure was reported by girls (Fig 2). Between 2% and 5% of girls 10 to 11 years of age, 12 to 13 years of age, and 14 to 15 years of age said they had gone to X-rated Web sites on purpose in the past year; 8% of girls 16 to 17 years of age had done so. Unwanted exposure in the past year increased with age among girls, from 16% of those 10 to 11 years of age to 38% of those 16 to 17 years of age.

View larger version:

• In this page
• In a new window

• Download as PowerPoint Slide

FIGURE 2

Unwanted and wanted exposure to online pornography among girls (n = 693). Gender data were missing in 2 cases.

Bivariate Associations of Unwanted and Wanted Exposure

The majority of youth who reported unwanted exposure were teens, 13 to 17 years of age, as were almost all of those who reported wanted exposure (Table 2). Otherwise, few demographic characteristics were related. However, most of the Internet use, prevention, and psychosocial characteristics measured were significant in bivariate analyses at ≤.01.

Multivariate Associations With Unwanted and Wanted Exposure

Compared with the no-exposure group, teenagers (13–17 years of age) were almost twice as likely to report unwanted exposure (odds ratio [OR]: 1.9; 95% confidence interval [CI]: 1.3–2.7), but no other demographic characteristic were related (Table 3). Only 1 characteristic of Internet use was associated with unwanted exposure. Youth who used file-sharing programs to download images from the Internet had almost twice the risk of encountering unwanted pornography (OR: 1.9; 95% CI: 1.3–2.9). However, youth who reported being harassed online (OR: 1.9; 95% CI: 1.1–3.2) or receiving unwanted sexual solicitations (OR: 2.7; 95% CI: 1.7–4.3) also had higher odds of unwanted exposure. Two types of prevention efforts seemed to afford some protection from unwanted exposure; having software (other than pop-up advertisement or spam e-mail blockers) to filter, to block, or to monitor Internet use on the computers youth used most often reduced the likelihood of exposure by 40%, and attending presentations about Internet safety led by law enforcement personnel reduced the likelihood by 30%. However, those who reported being talked to by parents or adults at school about online pornography had higher odds of exposure. Certain psychosocial characteristics were also related. Youth who reported offline interpersonal victimization (OR: 1.4; 95% CI: 1.1–1.8) and those who scored in the borderline or clinically significant range on the CBCL subscale for depression/withdrawal (OR: 2.3; 95% CI: 1.1–4.8) had higher risks of unwanted exposure.

Compared with unexposed youth, youth in the wanted exposure group were almost 9 times as likely to be 13 to 17 years of age (OR: 8.8; 95% CI: 3.8–20.6) and male (OR: 8.6; 95% CI: 5.2–14.3) (Table 3). Youth who used file-sharing programs to download images had higher risk (OR: 2.6; 95% CI: 1.6–4.4), as did those who were harassed online (OR: 2.6; 95% CI: 1.3–5.2), were solicited online (OR: 3.9; 95% CI: 2.1–7.1), talked online to unknown people about sex (OR: 2.6; 95% CI: 1.1–5.8), and used the Internet at friends’ homes (OR: 1.8; 95% CI: 1.1–3.0). Youth who had software (other than pop-up advertisement or spam e-mail blockers) to filter, to block, or to monitor Internet use on the computers they used most often had reduced risk of wanted exposure (OR: 0.6; 95% CI: 0.4–0.9). Offline interpersonal victimization (OR: 1.5; 95% CI: 1.013–2.2) and scoring in the borderline or clinically significant range on the CBCL subscale for rule-breaking (OR: 2.5; 95% CI: 1.2–5.4) were associated with higher risk of wanted exposure. Youth who scored in the borderline or clinically significant range on the CBCL subscale for depression were more than twice as likely to report wanted exposure, although this finding fell short of significance (OR: 2.3; 95% CI: 0.986–5.5; P = .054). In addition, bivariate analysis showed that, compared with other youth with wanted exposure, those with rule-breaking problems were more likely to view pornography when they were in groups with peers (63% of rule-breakers, compared with 39% of other youth; OR: 2.7; 95% CI: 1.3–5.6; P = .006; data not shown).

Unwanted Exposure

Forty-two percent of youth Internet users 10 to 17 years of age saw online pornography in the past year, and two thirds of those reported only unwanted exposure. Teens had higher risks, but preteen boys in particular had considerable unwanted exposure (17% of 10- and 11-year-old boys). No other demographic characteristics were related, however. The amount of Internet use was not related and, with 1 exception, what youth did online was not related. The exception was that youth who used file-sharing programs to download images were at risk for unwanted exposure; ∼1 of 5 youth with unwanted exposure had done this. This finding from a national survey confirms other reports that exposure to pornography is related to the use of file-sharing programs to download images.^6,10 Large volumes of pornography are transmitted through file-sharing, and some file-sharing software does not include filters for sexual material (or the filters are ineffective).

Two types of prevention efforts were associated with lower risks of unwanted exposure. The first was filtering, blocking, or monitoring software. This is consistent with other findings that filtering and blocking software has a modest protective effect on unwanted exposure.¹⁹ The software that seemed to have a preventive effect was distinguished from pop-up advertisement blockers and spam e-mail filters, which suggests that more-comprehensive software is required for effectiveness. However, it is also important to emphasize that the high rate of unwanted exposure to online pornography occurred despite the use of filtering and blocking software by more than one half of families with home Internet access.¹⁴ This suggests that filtering and blocking software alone cannot be relied on for a high level of protection against unwanted exposure and other approaches are needed.

Attending a law enforcement presentation about Internet safety was also associated with reduced odds of unwanted exposure. Since the late 1990s, there has been a concerted effort among certain law enforcement agencies to provide Internet safety information to youth, and specific programs have been developed for this purpose.^36,37 Some law enforcement programs provide specific information about how pornography is marketed online, how it can get on a person’s computer, and how to avoid or to remove it.³⁷ Youth may pay more attention or give more weight to information provided by law enforcement personnel. Also, simple presentations may be particularly effective when aimed at a problem such as unwanted exposure, which may not be an outgrowth of difficult-to-change youth characteristics or behaviors. However, youth who said they were talked to by parents or adults at school about online pornography had higher odds of exposure. One explanation for this finding is that many conversations between parents and youth happen after incidents of unwanted exposure.

We also found that certain youth seemed to be more vulnerable to unwanted exposure. There were associations between unwanted exposure and offline interpersonal victimization and borderline or clinically significant depression. These findings are similar to previous results showing associations between online harassment or sexual solicitation and offline interpersonal victimization and psychosocial challenge.³⁸ Some common underlying features, such as impulsiveness or compromised judgment, may explain these associations. For example, impulsive youth may have poor judgment or less ability to avoid unwanted online pornography or to make use of prevention information. Depression may put some youth Internet users at risk for similar reasons.

It is important not to overstate the relationship between unwanted exposure and characteristics such as offline interpersonal victimization or depression, however. These associations were not strong. Ours was a general population sample, and most youth with unwanted exposure were not victimized or depressed. Overall, the findings suggest that much unwanted exposure arises from normal Internet use and, except for downloading images with file-sharing programs, is not related strongly to specific behaviors or characteristics that increase risk.

It is also important to note that not all unwanted exposure incidents were unintentional. In 21% of incidents, youth said they knew sites were X-rated before they entered the sites.¹⁴ These episodes were not otherwise distinguishable from other instances of unwanted exposure. Some youth might have been motivated by curiosity and, even in incidents that were entirely unintentional, some degree of curiosity might have been involved. Also, most youth were not upset by the images they saw.¹⁴ Many youth may be somewhat inured to sexual images because of exposure from other sources, such as television, magazines, and R-rated films.

Wanted Exposure to Online Pornography

The great majority of youth with wanted exposure were teenage boys, and rates of wanted exposure increased with age. More than one third (38%) of male Internet users 16 to 17 years of age had visited X-rated sites on purpose in the past year. Interest in sexuality is high in this age group, and it was not surprising that wanted exposure was associated with talking online with unknown people about sex, which could be seen as another form of sexual curiosity.

Like unwanted exposure, wanted exposure was associated with using file-sharing programs to download images. Youth who used the Internet at friends’ homes also had higher risk of wanted exposure. If using the Internet at friends’ homes meant using it in pairs or groups, then this might reflect a group dynamic at play among some youth, because 44% of youth with wanted exposure said they had gone to X-rated sites on purpose when they were “with friends or other kids.”¹⁴ We also found that filtering and blocking software, other than pop-up advertisement and spam e-mail blockers, reduced the odds of wanted exposure.

Having delinquent tendencies seemed to be a factor in wanted exposure. Youth who scored at the borderline or clinically significant level on the CBCL rule-breaking subscale were more than twice as likely to report wanted exposure. One possible explanation is a link between rule-breaking behavior and an underlying tendency for sensation seeking.^15,39–41 A possible association between wanted exposure and depression could have a similar explanation, in that some depressed youth might seek the arousal of online pornography as a means of relieving dysphoria.^42–44 Although the association between wanted exposure and depression fell short of significance, the OR indicated a possible relationship.

It is also important not to overstate associations between wanted exposure and delinquency or depression. Sexual curiosity among teenage boys is normal, and many might say that visiting X-rated Web sites is developmentally appropriate behavior. However, some researchers have expressed concern that exposure to online pornography during adolescence may lead to a variety of negative consequences, including undermining of accepted social values and attitudes about sexual behavior, earlier and promiscuous sexual activity, sexual deviancy, sexual offending, and sexually compulsive behavior.^{2–4,6,8,9,44}

It is by no means established that online pornography acts as a trigger for any of these problems in youth or adult viewers. However, if it can promote deviant sexual interests or offending among some youth viewers, then the subgroup of youth Internet users with delinquent tendencies could include the youth most vulnerable to such effects, given the association between juvenile sexual offending and antisocial behavior.⁴⁵ Also, some researchers have found relationships between depression and online sexually compulsive behavior.^42–44 This suggests that the group of depressed youth Internet users could contain some who might be at risk for developing online sexual compulsions, which could interfere with normal sexual development or impair their ability to meet daily obligations and to develop healthy relationships with peers.

Implications

The high rate of exposure to online pornography among youth Internet users merits more attention, as does the fact that most such exposure is unwanted. Surveys have found high rates of unwanted exposure since the late 1990s, when Internet use became widespread among youth.^{6,14,17–19,21} Exposure to online pornography might have reached a point where it can be characterized as normative among youth Internet users, especially teenage boys. Medical practitioners, educators, other youth workers, and parents should assume that most boys of high school age who use the Internet have some degree of exposure to online pornography, as do many girls. One clear implication is that professionals should not shy away from this topic. Frank direct conversations with youth that address the possible influences of pornography on sexual behavior, attitudes about sex, and relationships are needed.

A focus on the unwanted aspect of much exposure to online pornography is also needed. Despite varying views on restricting adult voluntary access to legal pornography, we think that there is a consensus that youth, using a modicum of care, should be able to use the Internet without coming across pornography they do not want to see. This requires finding ways to restrict the use of aggressive and deceptive tactics to market pornography online. We also need to urge technology companies to make Internet filtering and blocking easier, more built into systems, and less reliant on individual initiative, technologic skill, and financial resources and to promote the use of filtering and blocking software in households with children. In addition, we need to educate youth about the technical details of how unwanted pornography is distributed online and to help them to protect themselves against it.

Methodologically sound empirical research about whether and how exposure to online pornography may be influencing youth is also in order. There is some evidence that youth reactions to sexual material are diverse and complex, especially among older youth,⁷ and many teens may respond thoughtfully and critically to the content of the images they see. However, there has been very little research about the impact on youth of viewing pornography, either wanted or, more relevantly, unwanted. There is no research that sheds light on whether, how, and under what circumstances unwanted exposure to pornography may trigger adverse responses in youth. Clearly, the extent of exposure is great enough that, even if adverse effects occur for only a small fraction of youth, the numbers in absolute terms could be fairly large. Researchers in the field of sexual development do not know whether there are important “primacy effects” related to early exposure of youth to pornography or what the effects of such exposures might be on anxieties, normative standards, or patterns of arousal in some youth.^1,2

As this study shows, it is possible to collect data on sensitive topics from youth informants. In addition to research about whether and under what circumstances viewing online pornography affects the sexual behavior and psychological health of youth, we need information about factors that could influence youth reactions to online pornography, such as family attitudes, psychological attributes, formats and content of pornography, effects of group dynamics among youth, and whether and under what circumstances unwanted exposure may lead to wanted exposure (or vice versa).

Limitations

Research about youth and the Internet is a relatively new undertaking. Procedures for inquiry have not been standardized, and measures have not been validated. The topic of exposure to pornography is a charged one, and there is room for a substantial amount of subjectivity in responses, as well as possibilities of nonresponse and evasive responses. For example, some youth might have characterized exposure incidents as unwanted because they were embarrassed to admit that they sought out such material. The study is hampered also by the limited information it gathered about wanted exposure incidents. In addition, some youth declined or were barred from participating, and their inclusion could have changed the results.

Finally, our numbers are only estimates, and samples may be unusual. For most of our major findings, statistical techniques suggested that estimates were within ≤2.5% of the true population percentage for 95 of 100 samples like this one, but there is a small chance that our estimates are farther off than 2.5%.