Sexual behavior implicates important public health concerns in the United States. Youth between the ages of 15 and 24 have the highest rates of STDs (Fox, 2004) and represent about half of the estimated 19 million new STD infections each year (CDC, 2006b). Moreover, the proportion of youth between the ages of 13 and 24 diagnosed with AIDS continues to increase (CDC, 2006a). Seventy percent of sexually active young adults report having taken a pregnancy test or had a partner who did so. Likely contributing to these alarming data are risky sexual behaviors in which emerging adults have been found to engage. More than half of college students have reported engaging in casual sex--that is, sex outside of established romantic relationships--which is often spontaneous and involves drugs and alcohol (Grello, Welsh, & Harper, 2006). In one study, nearly one-third of 18-19 year olds reported having engaged in sexual behaviors with two or more partners in the past 12 months (Mosher, Chandra, & Jones, 2005). Despite the risks associated with these encounters, which generally involve little to no familiarity with a partner's sexual health history, less than half of young adults who engage in sexual activity report using contraception either regularly or at all when engaging in sex (Hoff, Greene, & Davis, 2003). These data highlight the importance of examining the factors that play a role in sexual socialization, among these the mass media and particularly television, which is recognized as a key socializing agent with regard to sex for young people (Huston, Wartella, & Donnerstein, 1998).
A growing body of studies documents a relationship between exposure to televised sexual content and perceptions of peer norms, expectations about sex, permissive sexual attitudes, and engagement in premarital sexual intercourse (Aubrey, Harrison, Kramer, & Yellin, 2003; Brown & Newcomer, 1991; Collins et al., 2004; Ward, 2003; Ward & Rivadeneyra, 1999). Most of the evidence to date about the effects of sexual portrayals in mainstream entertainment television has emerged from cross-sectional, correlational research.
Only a small handful of studies have experimentally examined the media's influence on sexual outcomes. Bryant and Rockwell (1994) found that young adolescents exposed to 15 hours of sexual television content that "focused on pre, extra, and nonmarital sexual relations" (p. 188) subsequently judged such situations as less morally wrong than adolescents exposed to content depicting intercourse between married individuals and those exposed to shows that did not depict sexual relationships. Farrar (2006) found that exposure to portrayals of safe sex on television drama shows impacted female college students' attitudes toward condom use. And Taylor (2005) found that exposure to verbal televised sexual content perceived to be realistic led female college students to overestimate the level of sexual activity among their peer group. This modest base of experimental evidence is consistent with the findings from correlational studies, but more work is needed to afford confidence in drawing causal conclusions. Moreover, experimental studies are needed for the identification and control of specific processes that operate in the exposure-outcome relationship (Huston et al., 1998).
The current study extends previous research by examining the effects of exposure to a specific contextual variable associated with the portrayal of sex on television on sexual attitudes and moral judgments. Specifically, the study tests the effects of exposure to differentially valenced consequences of premarital sexual intercourse. It tests effects immediately after exposure to two episodes of 1-hour dramas and the persistence of effects 2 weeks later. The study examines sexual media effects during the important development period of emerging adulthood, between the ages of 18 and 25 (Arnett & Tanner, 2006).
Researchers have established that important developmental progressions continue to take place even at the end of the traditional "late-adolescence" period (Arnett, 2000). As identity development is rarely achieved by late-adolescence, the period of emerging adulthood is characterized by increased opportunities for self-exploration, especially in the areas of love, work, and worldviews. This period is further characterized by increased independence from, and access to, traditional socializing agents such as parents. Romantic and sexual relationships in emerging adulthood tend to last longer and involve greater intimacy than in adolescence. There is also much sexual activity during this age period (CDC, 1997), with nearly 80% of emerging adult college students engaging in sexual intercourse. Understanding the media's role in such behavior is crucial, given the public health risks associated with sex.
Sex on Television: Content and Context
Though television use tends to decrease after early adolescence, it is still the most popular media pastime among emerging adults (Gunter & McAleer, 1997). Studies have shown that there is a considerable amount of sexual messages across the television landscape (Greenberg & Hofschire, 2000). For example, Kunkel, Farrar, Eyal, Biely, and Donnerstein (2007) found that 64% of television shows contained some sexual content. Thus, most emerging adults will be exposed to many messages about sex when they watch television. Most of these sexual messages are conveyed through conversations (Greenberg & Woods, 1999; Kunkel et al., 2007). Over the years, there has been an increase in the depiction of sexual behaviors (Kunkel et al., 2007), the most common of these being passionate kissing (Greenberg & Busselle, 1994). Sexual intercourse, which is usually implied by narrative device rather than visually depicted, has become more frequent over time, recently occurring in 11% of shows (Kunkel, Eyal, Biely, Finnerty, & Donnerstein, 2005).
It is increasingly clear that the context of portrayals is important in understanding media effects (Kunkel et al., 1995). Context is thought to influence the meaning viewers make of televised content, the extent to which they identify with particular plots or characters, and the lessons they learn from portrayals of behavior. Researchers have identified several contextual variables that are potentially consequential for media effects in the realm of sex, including the age of characters who engage in sexual intercourse and their relationship status (Kunkel et al., 2001).
Central for the current study, an important contextual element in sexual portrayals involves the consequences portrayed for the sexual activities. Sexual intercourse can result in different outcomes, both positive (e.g., increased social status, personal satisfaction) and negative (e.g., social stigma, relationship damage, STD contraction). Content analyses have found that clear depictions of consequences of sexual behaviors are often omitted in television stories (Greenberg & Woods, 1999; Kunkel et al., 2001, 2007) so that televised portrayals tend to present sexual acts as resulting in neither clearly positive nor clearly negative consequences. When outcomes of sexual behaviors are portrayed, they more often tend to be positive. Cope-Farrar and Kunkel (2002) reported that of the 25% of intercourse participants who experienced clear outcomes for their actions, most of these reflected positive consequences. Also, television characters' attitudes toward unmarried intercourse tend to be positive. For example, Greenberg and Busselle (1994) reported that about 40% of male and female characters exhibited positive attitudes toward this activity as compared to less than 20% who exhibited a negative attitude. As will be reviewed below, media violence research has found that the consequences portrayed for aggressive behaviors have important implications for viewer effects. In comparison, the effects of televised messages about sexual intercourse that results in different consequences have not yet been tested. Addressing this lacuna is an important basis for the current study.
Theoretical Framework for the Study--Social Cognitive Theory
According to social cognitive theory (SCT), cognitive processes account for which events in the environment are observed, what meaning is given to them, and how information about them is organized for future use by the individual (Bandura, 2001). Wide support has been established for these theoretical premises, especially in, but not limited to, the realm of media violence (Gentile, 2003; Morrison & Westman, 2001). SCT also accounts for effects on attitudes and emotions related to specific behaviors, as models can vicariously elicit affective dispositions in viewers (Check & Malamuth, 1986).
SCT accounts for several groups of intervening variables in its theoretical mechanism. These include characteristics of the model, the behavior observed (Grusec, 1973), the observer (Perry, Perry, & Boldizar, 1990), and the observed event. Among the most important characteristics of the observed event is the action's consequences. Previous research (Bandura, 1965) has established that viewing televised violent behaviors that result in positive outcomes or rewards leads to an increased viewer tendency to imitate the acts, due to a shift in attitudes toward the violent behavior from negative to positive (i.e., disinhibition). Conversely, viewing violent behaviors punished or negatively reinforced on television leads to a decreased tendency to imitate them, motivated by less favorable attitudes toward them.
Bandura (1977) does not restrict his theorizing about the role of consequences only to violent acts. Rather, the theory clearly applies broadly to all observed behaviors. However, there has been little generalization of these premises to other realms of media effects besides violence where behaviors may be positively or negatively reinforced. This study tests whether similar effects of consequence contingencies...