Internet pornography exposure and risky sexual behavior among adult males in the United States
Sexually transmitted infections (STIs) continue to pose a threat to the public health in the United States. The Centers for Disease Control and Prevention (CDC) report that more than 1 million US adults and adolescents are currently infected with the human immunodeficiency virus (HIV) (Campsmith, Rhodes, Hall, & Green, 2008) and more than 50,000 new HIV transmissions occur each year (Hall et al., 2008). If current trends persist, approximately 50% of US Americans will contract an STI at some point in their lives (Planned Parenthood, 2011).
Many sexual behaviors increase an individual’s risk of STI contraction. Chief among these are having unprotected sex, having sex with multiple partners, and either paying for sex or having sex for pay (Campsmith et al., 2008; Workowski & Berman, 2010). Interventions designed to reduce the risk of STI transmission rely on behavioral research to identify factors that predict an increased likelihood of risky sexual behavior (Hall, Hughes, Dean, Mermin, & Fenton, 2011). The purpose of the present study is to explore the possibility that exposure to internet pornography is a sexual risk factor for adult males in the United States. Specifically, the present study uses General Social Survey (GSS) data from 2000, 2002, and 2004 to explore the association between exposure to internet pornography and having unprotected sex, engaging in paid sex, having extramarital sex, and having sex with multiple partners. Subsequent GSSs have not asked participants about exposure to internet pornography, and the GSS is the only ongoing, fullprobability, national survey assessing social beliefs and behaviors in the United States. Thus, the 2000, 2002, and 2004 GSSs provide a rare opportunity to explore at a national level how exposure to internet pornography associates with STI risk behavior among adult US males.
1.1. Pornography: theory
The term ‘‘pornography’’ is value laden and characterized by many meanings. Following Wright, Malamuth, and Donnerstein (2012), the present report associates no pejorative connotation with the term, using it only as shorthand for mediated content depicting nudity and explicit sexual behavior. Pornography is primarily targeted at males (Malamuth, 1996) and is a lucrative enterprise. The pornography industry in the United States grossed upwards of $7 billion in 2006 (Edelman, 2009). Although offline pornography continues to attract many consumers, economic indicators show that online pornography is the wave of the future (Edelman, 2009). The anonymity and accessibility of online pornography may encourage males who would not have gone to an adult book store or rented an adult film from their local video proprietor to explore pornography online (Carnes, Delmonico, &Griffin, 2001; Schneider & Weiss, 2001). The tremendous communication potential of the internet coupled with the allure of sexual media has led many to wonder about the social influence of online pornography (Brown & L’Engle, 2009; Peter & Valkenburg, 2006). Wright (2011) has developed a script acquisition, activation, application (3AM) model that explains how exposure to sexual media can impact sexual behavior. The model references numerous social learning and media theories (e.g., the heuristic processing model of cultivation effects, social cognitive theory, priming, uses and gratifications, media dependency), but draws primarily on the script theorizing of Huesmann (1986, 1988, 1998). A scripts perspective on sexual behavior has been recommended by several sex researchers (Gagnon & Simon, 1973; Laws & Schwartz, 1977). Huesmann (Huesmann, 1986) argues that mediated models may affect behavior by providing cognitive scripts which (a) suggest what events should and should not be happening, (b) how people should behave in response to what is or is not happening, and (c) what the outcomes of particular courses of action should be. Huesmann’s writings reveal three steps in the media effects on behavior process: the acquisition of behavioral scripts, the activation of behavioral scripts, and the application of behavioral scripts.
An acquisition effect occurs when an observer learns a novel behavioral script. Individuals can learn particular scripts from sexual media or ‘‘higher order scripts’’ (Huesmann, 1986, p.131). Higher order scripts are acquired by abstracting the behavioral philosophy guiding media models’ behavior. For example, a male might learn both how to solicit sex from exposure to pornography with pay-for-sex themes (specific script) and that sex for amusement, whether with a prostitute or a casual romantic partner, is acceptable and pleasing (higher order script). An activation effect occurs when media exposure cues a script that has already been learned. For instance, a sexual script encouraging group sex may be dormant in an individual’s mind until he views pornography with such a theme. An application effect occurs when a script that has been acquired and activated is applied at the level of behavior (e.g., actually exchanging sex for money or having group sex). Wright (Wright, 2011) proposes a number of individual difference, media content, and situational factors that may prevent the acquisition, activation, and application of sexual scripts provided by sexual media. Given the themes in such content (e.g., attractive models who are positively reinforced for their behavior), the attributes of many males (e.g., lacking in definitive familial sexual socialization), and the obstacles to controlled information processing in many sexual situations (e.g., sexual arousal, time pressure), however, it is probable that exposure to pornographic media results in script matching behavior more often than script mismatching behavior. Consequently, it is important to explore whether pornographic content provides scripts that encourage paid sex, having extramarital sex, having sex with multiple partners, and unprotected sex.
1.2. Pornography: content
1.2.1. Paid sex, extramarital sex, and sex with multiple partners It has been argued that there are three overarching sexual scripts in US culture: the procreational script, the relational script, and the recreational script (Ward, 1995). The procreational script views sex as a means to generate offspring within the milieu of the traditional family. The relational script views sex as a mechanism through which a devoted couple can create intimacy and communicate liking. The recreational script views sex as a means of receiving physical gratification and sensual ecstasy; such gratifications can be obtained regardless of whether one cares for or is committed to one’s partner or partners. Both the procreational and relational scripts would discourage paid sex, extramarital sex, and having sex with multiple partners. The recreational script would be more likely to accept such pursuits
Male-targeted pornography tends to embrace a recreational approach to sex (Malamuth, 1996). According to Kline (2001) ‘‘the chief proclamation of pornography is great sexual joy without any attachment, commitment or responsibility’’ (p. 8). Zillmann and Bryant (1988) agree, observing that ‘‘pornographic scripts dwell on sexual engagements of parties who have just met, who are in no way attached or committed to one another, and who will part shortly, never to meet again’’ (p. 521). To conclude, the sexual scenarios presented in pornography may encourage many of the behaviors under investigation in the present analysis due to their encouragement of a higher order recreational sex script. Some evidence suggests that males may not need to engage in any higher order abstraction to have their beliefs about paid sex, extramarital sex, and sex with multiple partners affected by pornography exposure, however, as these themes appear to be overt in pornographic fare. Content analyses from the 1970s, 80s, and 90s suggest that approximately 5% to 20% of pornography features a prostitution theme (Brosius, Weaver, & Staab, 1993; Palmer, 1979; Palys, 1986; Smith, 1976). Reports from Smith (1976) and Winick (1985) suggest that extramarital sex is a common occurrence in printed pornography. Group sex occurred in 35% of pornographic scenes in videos studied by Brosius et al. (1993) and in 47% of scenes studied by Palys (1986). Internet pornography with these themes can be readily found. Examples of prostitution pornography websites are ‘‘mommyneedsmoney.com,’’ ‘‘crackwhoreconfessions.com,’’ and ‘‘hookerporn.net.’’ Examples of extramarital sex websites are ‘‘cheatwife.com,’’ ‘‘best-cheating-wives.com,’’ and ‘‘cheatingwife.tv.’’ Examples of group sex websites are ‘‘myxxxgroupsex.com’’ and ‘‘drunksexorgy.com.’’
1.2.2. Unprotected sex Very little research exists on the use of condoms in pornography. Wingood et al. (2001) state that sexually explicit films ‘‘rarely show contraceptive or STD preventive behaviors’’ (p. 1118), but do not cite any content analyses to substantiate this assertion. The present study could find only one report on the use of contraception in pornography: (Brosius et al., 1993). This study of video pornography found that only 1% of scenes mentioned contraception. Whether this trend has persisted in our more STI conscious era remains a question for future research.
1.3. Pornography: potential effects Research into the effects of pornography has primarily focused on its potential to catalyze males’ aggression towards females. Comparatively fewer studies have focused on pornography’s effect on sexual behavior (Zillmann & Bryant, 1988). Studies that have addressed behavioral associations relevant to the outcome variables of interest in the present study are reviewed below. Two recent US studies have asked whether individuals who view pornography are more likely to have multiple sexual partners. Braun-Courville and Rojas (2009) interviewed adolescents and young adults at a health center in New York. Youth who had visited sexually explicit websites were 1.8 times more likely to have had multiple sexual partners over the course of their lifetime and to have had more than one sexual partner in the last few months. Using a comparable methodology, Wingood et al. (2001) surveyed black adolescent females recruited from health related settings (e.g., community clinics, school clinics). Girls who viewed pornographic films were twice as likely to have had multiple sexual partners. A study of Swedish high school students aligns with these findings (Haggstrom-Nordin, Hanson, & Tyden, 2005). This study found that male students in the upper echelon of pornography consumption were approximately twice as likely to have engaged in group sex. Thus, the present study’s first hypothesis is: H1. Exposure to internet pornography is positively associated with the number of past sexual partners reported by adult US males.
Results from Wingood et al. (2001) and Braun-Courville and Rojas (2009) concerning contraception use were varied, on the other hand (Haggstrom-Nordin et al., 2005, did not report whether contraception use was associated with pornography consumption). Wingood et al. (2001) found that pornography consumers were 1.5 times less likely to have used contraception the last time they had intercourse. No relationship was found between use of contraception at last sexual encounter and exposure to pornography in Braun-Courville and Rojas (2009). As alluded to earlier, this inconsistency could be because condom use in pornographic fare has increased over time. For example, Wicked Pictures, a dominant force in the pornography industry, requires that its actors wear condoms (Strauss, 2008). Accordingly, the present study asks:
R1: Is there an association between adult US males’ condom use and exposure to internet pornography?
No research was uncovered assessing associations between pornography consumption and extramarital sex. A few studies, however, have found that exposure to pornography is associated with beliefs and attitudes that may support extramarital sex scripts. Zillmann and Bryant (1988) conducted an experiment wherein student and nonstudent adults viewed or did not view 6 h of video pornography over the course of 6 weeks. Pornography viewing led to higher estimates of extrarelational sex among adults in committed relationships, more favorable attitudes towards extramarital sex, more tolerance for affairs, more acceptance of individuals having multiple sexual partners even if they are in a relationship, and an increased intention of having an affair if given the opportunity. Similarly, Lo and Wei (2005) found that the more Taiwanese adolescents viewed internet pornography the more they agreed with statements such as ‘‘It is allowable for married men to have extramarital sex’’ and ‘‘It is allowable for married men to have more than one sex partner’’ (p. 228). As a result, the present study hypothesizes that: H2. Exposure to internet pornography is positively associated with having engaged in extramarital sex among adult US males. Studies correlating paid sex with pornography exposure are rare. Braun-Courville and Rojas (2009) intended to investigate the association between these behaviors, but were unable to do so because so few of their respondents had engaged in paid sex. Given that pornography with prostitution themes seems to be fairly common and the recreational approach to sex generally portrayed in pornography, however, the present study predicts a positive association between these two variables: H3. Exposure to internet pornography is positively associated with having engaged in paid sex among adult US males. 2. Method 2.1. Participants The data for the present study were provided by the General Social Survey (GSS) Davis & Smith, 2009. The GSS employs a fullprobability sample of US adults aged 18 and above. The GSS has been conducted since 1972. Participants for the present report were 1079 US males who reported having used the internet from GSS 2000, 2002, and 2004 (2000n = 331; 2002n = 462; 2004n = 286). The 2000, 2002, and 2004 GSS were selected because exposure to internet pornography was measured in these years. Subsequent GSSs have not asked participants about exposure to internet pornography. Participants ranged in age from 18 to 87 (M = 41.19; SD = 14.05). 85% of participants were White, 8% were Black, and 7% fell into an alternative ethnic category.
2.2. Measures
2.2.1. Independent variable Exposure to internet pornography was assessed with a single item: ‘‘In the past 30 days, how often have you visited a web site for sexually explicit material?’’ Response options were 1 = never, 2 = 1–2 times, 3 = 3–5 times, and 4 = more than five times (M = 1.33; SD = 0.71).
2.2.2. Dependent variables Five dependent measures were tested: number of sex partners in the last year, number of sex partners in the last 5 years, use of a condom during last sexual encounter, ever having engaged in extramarital sex, and ever having engaged in paid sex. Number of sex partners in the last year was assessed with a single item: ‘‘How many sex partners have you had in the last 12 months?’’ Response options were 0 = No partners, 1 = 1 partner, 2 = 2 partners, 3 = 3 partners, 4 = 4 partners, 5 = 5–10 partners, 6 = 11–20 partners, 7 = 21–100 partners, 8 = more than 100 partners (M = 1.36; SD = 1.22). Number of sex partners in the last 5 years was assessed with a single item: ‘‘How many sex partners have you had in the last 5 years?’’ Response options were 0 = No partners, 1 = 1 partner, 2 = 2 partners, 3 = 3 partners, 4 = 4 partners, 5 = 5–10 partners, 6 = 11–20 partners, 7 = 21–100 partners, 8 = more than 100 partners (M = 2.21; SD = 1.93). Use of a condom during last sexual encounter was assessed with a single item: ‘‘The last time you had sex, was a condom used?’’ Response options were 0 = Used a condom during last sexual encounter; 1 = Did not use a condom during last sexual encounter. Of the 921 participants who said they either did or did not use a condom, 26.9% used a condom during their last sexual encounter; 73.1% did not use a condom during their last sexual encounter. Extramarital sex was assessed for participants who had ever been married with a single item: ‘‘Have you ever had sex with someone other than your husband or wife while you were married?’’ Response options were 0 = No and 1 = Yes. Of the 615 ever been married participants who gave a ‘‘no’’ or ‘‘yes’’ response to this question, 75.6% had not had extramarital sex; 24.4% had engaged in extramarital sex. Paid sex was assessed with a single item: ‘‘Thinking about the time since your 18th birthday, have you ever had sex with a person you paid or who paid you for sex?’’ Response options were 0 = No and 1 = Yes (85.1% indicated they had not engaged in paid sex; 14.9% had engaged in paid sex).
2.2.3. Controls
A variety of potential confounds were included in analyses. Age was controlled because younger, testosterone driven males may be both more likely to consume pornography and to take sexual risks (Baumeister, 2000). Ethnicity (operationalized as White/Nonwhite) was controlled because there are differences in both the sexual behavior and sexual media consumption of Whites and Nonwhites (Brown & L’Engle, 2009). Religiosity (operationalized as frequency of attendance at church services) was controlled because less religious individuals may be more likely to engage in recreational
sex and consume sexual media (Collins et al., 2004). Having ever used crack cocaine was controlled because experimenting with recreational drugs may be a sign of a compulsive or sensation seeking personality, both of which potentially predict risky sexual behavior and sexual media consumption (Brown & L’Engle, 2009; Collins et al., 2004; Kalichman & Rompa, 1995).
3. Results
Logistic and multiple regression analyses were used to test the study’s hypotheses and research question. Analyses for each dependent variable were conducted for each GSS year and for the composite 2000, 2002, 2004 sample.
3.1. H1: Number of sexual partners
H1 was largely supported. Three of the four multiple regression analyses for the association between internet pornography exposure and number of sexual partners in the last year revealed a significant or marginally significant positive association (see Table 1). Internet pornography exposure explained between 1% and 3% of the variability in number of sexual partners in the last year in these analyses. Most importantly, the composite sample analysis showed a significant positive association between internet pornography exposure and number of sexual partners in the last year. Only the 2002 association was not at least marginally significant, b = 0.05, t (394) = 0.96, p = .34. Second, three of the four multiple regression analyses (including the composite sample analysis) for the association between internet pornography exposure and number of sexual partners in the last 5 years showed a positive and significant association. All four analyses demonstrated an association that was positive and at least marginally significant (see Table 2). Internet pornography exposure explained between 1% and 3% of the variability in number of sexual partners in the last 5 years in these analyses
3.2. RQ1: condom use
RQ1 asked if there was an association between internet pornography exposure and condom use. No association was found in any year. Correspondingly, an association was not found in the composite sample analysis (see Table 3).1
3.3. H2: extramarital sex
H2 was largely supported. Three of the four logistic regression analyses for the association between internet pornography exposure and ever having engaged in extramarital sex were positive and significant. Most importantly, a one unit change in internet pornography exposure was associated with a 1.65 (95% CI: 1.26– 2.15) increase in the odds of ever having engaged in extramarital sex in the composite sample analysis (see Table 4). Only the 2004 association was not significant, Exp(B) = 1.57, p = .20.
3.4. H3: paid sex
H3 was supported. Logistic regression analyses for the association between internet pornography exposure and ever having engaged in paid sex revealed a positive association for all analyses (see Table 5). Most importantly, a one unit change in internet pornography exposure was associated with a 1.84 (95% CI: 1.47–2.32) increase in the odds of ever having engaged in paid sex in the composite sample analysis.
4. Discussion
Much progress has been made in the battle against STIs in the United States. Individuals have more opportunities for STI education, medical professionals are more cognizant of their role in the prevention of STI transmission, interventions to prevent the spread of STIs have been developed, and behavioral research has shed light on many factors that increase the likelihood of STI transmission. Nevertheless, STIs continue to threaten the public health (Campsmith et al., 2008; Hall et al., 2008; Planned Parenthood, 2011) and research into factors that increase the likelihood of risky sexual behavior remains important (Hall et al., 2011). Certain sexual behaviors are more risky than others in terms of the probability of STI contraction. Having sex without a condom, having sex with many partners, and having sex for pay or paying for sex are especially risky (Campsmith et al., 2008; Workowski & Berman, 2010). For purposes of intervention, it is important to pinpoint modifiable factors that increase the probability that individuals will engage in these behaviors. One such factor among males may be exposure to pornographic media content. Consumption of pornographic media may increase males’ probability of engaging in risky sexual behavior by providing or activating recreational sex scripts (Wright, 2011) that encourage activities such as group sex, paid sex, and extramarital sex. A small but growing body of literature suggests that exposure to pornographic media may affect the risky sexual behavior of US adolescents (Braun-Courville & Rojas, 2009; Brown & L’Engle, 2009; Wingood et al., 2001). The results of the present study imply that consuming internet pornography may be an STI risk factor for adult US males. Specifically, using data from males who responded to the 2000, 2002, and 2004 GSS, the present study found consistent, positive associations between viewing internet pornography and number of past sexual partners, ever having engaged in extramarital sex, and ever having engaged in paid sex. Subsequent GSSs have not asked participants about exposure to internet pornography. As the GSS is the only ongoing, full-probability,
national survey assessing social beliefs and behaviors, the present report provides unique insight into the risky sexual behavior patterns of adult male internet pornography consumers in the United States. The associations between males’ internet pornography consumption and risky sexual behaviors found in the present study were modest, but rivaled and often exceeded the association between these behaviors and other variables known to predict sexual behavior. For example, religiosity is often thought of as an important predictor of sexual behavior, but exposure to internet pornography was more powerfully related to paid sex behavior and number of sexual partners than religiosity in composite sample analyses. The present study found no association between exposure to internet pornography and use of a condom at last sexual encounter. Findings regarding this same outcome variable have been mixed in past studies (Braun-Courville & Rojas, 2009; Wingood et al., 2001). A study conducted in the 1990s indicated that use of condoms in pornography was quite infrequent (Davis & Smith, 2009). It is possible that condom use in sexually explicit fare has increased over time, however, as our society has become more STI conscious. From a script perspective, inconsistent correlations between pornography exposure and condom use would be expected if adult actors inconsistently use condoms. News reports covering condom use in adult films indicates this might be case, as some studios insist on condom use while others let individual performer decide (Strauss, 2008). Contemporary content analytic work exploring condom use in popular pornography should be a top priority.
4.1. Limitations
Three limitations temper the present study’s findings. First, the GSS data sets employed were generated by cross-sectional surveys. As such, it is impossible to ascertain the directionality of the internet pornography exposure/risky sexual behavior relationships uncovered. Does internet pornography consumption lead to risky sexual behavior or do people who engage in risky sexual behavior gravitate to internet pornography? The present report cannot answer this question. Recent research renders this conundrum less vexing than it would have been a few years ago, however, as a number of recent longitudinal studies suggest that sexual media consumption is a precursor of sexual beliefs and behaviors (Bleakley, Hennessy, Fishbein, & Jordan, 2008; Brown & L’Engle, 2009; Chandra et al., 2008; Collins et al., 2004; Peter & Valkenurg, 2009; Wingood et al., 2001). A related limitation is the independent variable, which measured participants’ exposure to Internet pornography in the last 30 days but was used to predict sexual behavior that could have happened in the past. This is an important limitation. It is a flaw less critical than would have been the case a few years ago, though, as recent longitudinal studies indicate that higher levels of sexual media consumption at present are highly correlated with higher levels of sexual media consumption in the past (Bleakley et al., 2008; Peter & Valkenurg, 2009). Second, although the present study did control for a number of potential confounds, the secondary data, cross sectional survey design employed here allows for the possibility that some unmeasured third-variable explains the associations that were found. To overcome this limitation, future researchers should (a) employ longitudinal designs with confounds selected specifically for a study of pornography effects and (b) engage in experimental studies that manipulate pornography exposure and measure sexual behavior intentions. Finally, the results of the present study are generalizable only to adult males in the United States who use the internet, as the 2000, 2002, and 2004 GSS asked only internet users about their internet pornography consumption. It must be remembered, however, that in each of these years a sizeable portion of adult US males had Internet access. According to the Pew Internet and American Life Project, 50% of adults had access in 2000 (Fox, 2010), 59% had access in 2002 (Lenhart, 2006) and 68% had access in 2004 (Lenhart, 2006).
5. Conclusions
The present study explored associations between adult US males’ internet pornography exposure and risky sexual behavior. Internet pornography exposure was positively associated with engaging in paid sex, number of sexual partners, and having extramarital sex. Additional research is needed to determine the directionality of these associations, to more adequately control for confounding variables, to assess psychological processes that mediate these links, and to identify factors that may attenuate the effects of exposure to internet pornography on males’ risky sexual behavior.