As society evolves, so do the mass media. Particular to sexual and/or violent media, there has been a trend in increased visual intensity and graphicness. Indeed, what was considered too graphic to receive an “R” rating a few decades ago is now considered tempered or mainstream by many. For example, just a few decades ago films like The Exorcist and Midnight Cowboy garnered X ratings due to violent and/or sexual content. Today, those movies have been adjusted to an R rating. Perhaps we, as viewers, have become more desensitized. In a word, material apparently needs to be increasingly shocking and graphic for many of us to consider it to be out of the ordinary. One must ask, however, about the impact of such graphic media on viewers’ attitudes and behaviors.

Often such material—particularly sexually violent material—involves depictions of women who are degraded and objectified (Check & Malamuth, 1985). Previous research suggests a correlation between exposure to sexually explicit and sexually violent material and increased negative attitudes toward women. Allen, Emmers, Gebhardt, and Giery (1995), for example, found a relationship between exposure to pornography and rape myth acceptance, particularly for violent pornography. An interpretation of this finding is that the female victim in the sexually violent depictions was viewed as responsible or partially responsible for her treatment. The effect was stronger for men than for women. More recent work has shown similar results (Emmers-Sommer, Triplett, Pauley, Hanzal, & Rhea, 2005). Other researchers, however, have argued that sexually explicit material, such as pornography, does not cause negative attitudes toward women (Padgett, Brislin-Slutz, & Neal, 1989).

As individuals mature and are exposed to various types of media, they develop film preferences or predilections. That is, some individuals have a penchant for storylines that involve, for example, a love, suspense, or sex and violence theme. It is of interest how these prevailing preferences relate to attitudes and behaviors. Particular to sexually violent media, it is worthwhile to explore how these film predilections affect negative attitudes toward women and, more important, if manipulating these types of media (i.e., various levels of film editing) ameliorate or exacerbate such pre-existing attitudes. Although much research exists on sexual and sexually violent entertainment media, less research exists on such material presented in a more realistic format (e.g., documentary or a film based on a true life experience). Particular to sexually violent material, the portrayal of consequences for the perpetrator (e.g., arrest, jail time) is often not present in the material. It is of theoretical interest whether or not viewers learn negative attitudes and behaviors from sexually explicit and violent material and whether such attitudes and behaviors can be tempered or intensified through exposure to the material (e.g., Allen & D’Alessio, 1991; Check & Malamuth, 1983). The purpose of the present investigation is to address that and related questions.

Preferences for Film Type and Gender Differences

Although motion pictures and related entertainment media are created for the mainstream, gender differences appear to emerge particular to content area. Specifically, men appear to hold a greater penchant for media that involve sex and violence, whereas women tend to react more adversely to such fare. Several studies support this contention. For example, in a study of the relationship between sensation seeking and related variables of interest in morbid and sexual events, Zuckerman and Litle (1986) found that men scored higher than women on scales of curiosity about morbid and sexual events in media, self-attendance ratings at horror and X-rated movies, psychoticism, and sensation seeking. The authors found that high sensation seekers, most of whom were men, are interested in stimuli that increase activity in central catecholamine systems. Zuckerman and Litle’s work extended previous work that showed a connection between exposure to media that involve violence, fear-inducement, and eroticism and an increased peripheral catecholamine activity (Levi, 1969). Catecholamines act as hormones or neurotransmitters in the system. The common catecholamines are epinephrine, nonepinephrine, and dopamine.

Related to the impact of sex and violence on men’s and women’s heightened awareness, Bahk (2000) examined college students’ likelihood of watching television shows and theatrical movies that carried with them mature, content-specific advisories. Results indicated that men’s likelihood of viewership increased with all presented mature advisories. For women, however, likelihood of viewing increased with the sexual advisory but decreased with the violence advisory.

Koukounas and McCabe (2001) examined gender differences in emotional response to violent film. Male participants reported more positive feelings, curiosity, and entertainment in response to the violent film, whereas women reported more disgust, anger, boredom, and greater startle reaction to the violent material. On a related topic, Emmers-Sommer et al. (2005) examined the impact of more reality based sexually violent film on attitudes toward women and various degrees of film editing and found that, as violence increased, men’s favorable attitudes toward editing decreased, whereas the opposite held true for women.

Thus, it is quite clear that men appear to have a greater interest in and desire to watch violent film than women do. Sexual film, on the other hand, is more convoluted, as a sexual film can portray sexual behavior that is less gratuitous in nature, such that it fits in the context of a love story. Conversely, sexuality and sexual behavior in film can be gratuitous in nature, as in many sexually violent, pornographic, snuff, and “slasher films.” Nolan and Ryan (2000) found that men reported more rural terror (e.g., fear of strangers and rural landscapes), whereas women reported more family terror (e.g., betrayed intimacy, spiritual possession, and stalking). Women also reported a higher level and a greater number of fear reactions than men did in response to slasher films, whereas men reported greater anger and frustration responses.

Cowan and O’Brien (1990, as cited in Linz & Malamuth, 1993) did a content analysis of more than 50 slasher films that involved more than 470 victims. They examined acts of violence committed by both men and women on both men and women. Findings indicated that the perpetrators were typically men, and the victims were typically women; sexuality was often coupled with the victimization. Weaver (1991) also content analyzed slasher films to assess if there were gender differences in victimization, but did not find gender differences in victim deaths. However, he did find that the death scenes of female characters were significantly longer than those of male characters and thus evoked more terror and violence.

The literature appears to be quite clear that men have a greater interest and desire to view violent and sexually violent film than women do (Bushman, 1995; Emmers-Sommer et al., 2005; Linz, Donnerstein, & Penrod, 1988; McIlwraith & Schallow, 1983; Oliver, Sargent, & Weaver, 1998). Women, on the other hand, appear not to be adverse to sexuality, as demonstrated by Bahk’s (2000) work; however, context is important in terms of moderating that interest. Specifically, women do not care for sexually violent fare. However, sexuality within the context of a love story appears to be acceptable and more appealing as it is perceived as less gratuitous and more germane to the nature of the storyline. This contention is supported by much gender research, which shows women to be more relationally oriented than men are (e.g., Byers, 1996; Canary & Emmers-Sommer, 1997). In more applied terms, box office receipts demonstrate that more women than men go to the movies to see love stories (e.g., BBC News, 2005).

Theoretical Approaches to Sexually Explicit and Violent Media

A variety of theoretical approaches exist that address the impact of sexually explicit and violent media on viewer attitudes and behaviors. Particularly relevant to this study are excitation transfer theory and social learning theory. Each perspective is addressed in turn below.

Excitation transfer theory

Excitation transfer theory argues that arousal from one context can be transferred to another context (Zillmann, 1971). At its inception, the theory argued that individuals engaged in aggressive behavior without thinking, but was later revised to address cognitive aspects. Specific to sexually violent media, this theory contends that viewers might become “excited” by sexual and sexually violent material and transfer that excitement to another context, for example, by committing a sexual or sexually violent act on a woman or, to a lesser extreme, by becoming accepting of such attitudes and behaviors toward women. Numerous experimental studies have supported this argument (e.g., Donnerstein & Wilson, 1976; Mueller & Donnerstein, 1977). For example, Zillmann (1971) found evidence of the transfer of physiological arousal in numerous studies. In one study, male participants were angered by a confederate and then shown film clips that varied in sexuality and aggression. One film was not arousing; another film was more arousing and also contained non-sexual violence; the third film was the most arousing and contained sexual behavior and nudity, but no violence. Zillmann found that the men were most aggressive when they viewed the sexual clips, rather than the aggressive clips. The explanation for this outcome is that individuals transfer their arousal from one context to another. To account for studies in which participants who were upset and then exposed to sexual material but did not demonstrate increased aggression, the arousal/hedonic valence model was created (Zillmann & Sapolsky, 1977). Specifically, the model contends that, when upset participants are exposed to material that they find to be pleasant, but not arousing, the upset is ameliorated to an extent. From this theoretical perspective, it is expected that men’s level of arousal upon viewing sexual violence is transferred to rape supportive attitudes toward women.

Social learning theory

From a media perspective, social learning theory argues that viewers learn about reality via symbols (Bandura, 1977, 1994). From a sexual standpoint, social learning theory contends that individuals learn sexual information from films that include sexual content. Specifically, viewers are more likely to model behavior that they see being positively reinforced, as opposed to behavior they see being punished. If the male aggressor in a sexually violent film is positively reinforced (e.g., receives pleasure) or the female target succumbs to the advances and ultimately enjoys the encounter, then the viewer might model the behavior. According to this perspective, then, men and women could learn information regarding the norms of sexual relationships and might model behaviors they have viewed (Allen et al., 1995; Duncan, 1990; Emmers-Sommer & Burns, 2005), particularly if the behaviors were positively reinforced or rewarded. From a social learning perspective, viewers of sexually explicit and/or violent material believe that the relationships portrayed in the film reflect those of real life (Check & Malamuth, 1985). Typically, sexually violent material portrays men’s aggression against women, but the consequences of such actions are often not shown (Brown & Bryant, 1989). As a result, the viewer might believe that the behaviors exhibited in the materials constitute typical sexual relationships, and they might eventually adopt such behaviors themselves. Particular to the current investigation, it is of interest how individuals’ pre-existing film predilections and the behavior modeled in the material play a role in how those individuals are affected by sexually violent media presented in a more realistic (i.e., based on a true story) versus an entertaining format. It was expected that those with sex and violence predilections would be more supportive of RMA than those who prefer non-violent fare, such as a love story.

Rape Myths and Rape Myth Acceptance

Burt (1980) coined the term “rape myth.” Rape myth acceptance (RMA) involves beliefs and values associated with sexual assault, the perpetrators, and the victims. The RMA measure was designed to identify an individual’s adoption of such beliefs (Burt & Albin, 1981). One who adopts RMA beliefs believes that a woman who is victimized by rape is partially or completely responsible for the act. Items from the RMA scale that demonstrate this notion include “A woman who goes to the home or apartment of a man on their first date implies that she is willing to have sex,” “Any healthy woman can successfully resist rape if she really wants to,” “Women who get raped while hitchhiking get what they deserve,” and “Many women have an unconscious desire to be raped, and may then unconsciously set up a situation in which they are likely to be attacked” (p. 223).

Burt’s (1980) measure has been tested often as it relates to pornography (e.g., Allen et al., 1995). Less known is the outcome of Burt’s measure as it relates to situations that involve sexual violence in a less entertaining context (e.g., documentaries, true-life portrayals). Overall, whether viewers adopt adverse attitudes from sexual media presented in this context is of interest. Further, the impact of adopting such attitudes on behaviors is also of interest. Previous research has demonstrated a strong relation between attitude and behavior (Sheppard, Hartwick, & Warshaw, 1988). Donnerstein (1983, as cited in Russell, 1998) argued that the link between sexual violence and aggression is stronger than the link between smoking and lung cancer. Yet, the studies Donnerstein to which referred to were largely entertainment-based.

Men, Women, and Sexual Coercion

Research indicates that both men and women instigate sexually coercive behaviors (Sigelman, Berry, & Wiles, 1984) and are victimized by sexual coercion (Muehlenhard & Cook, 1988). Nevertheless, the majority of victims of sex crimes are women (Becchofer & Parrot, 1991; United States Bureau of Justice Statistics, 2004). Various theoretical perspectives address perpetration and victimization. One perspective suggests that men who are sexually coercive are meeting the social (e.g., men are the dominant sex) and relational (e.g., men are the pursuers in relationships) expectations of them (Emmers-Sommer, 2002). Another argument is that inequity exists between the sexes, and coercion is one manner by which men sustain that inequity (Stock, 1991). Finally, the traditional sexual script (TSS) imposes a sexually aggressive expectation on men and a coy, submissive expectation on women (Byers, 1996). This perspective, as noted earlier, is also perpetuated in the media. Research suggests that men tend to hold more traditional attitudes (Byers, 1996; Emmers-Sommer et al., 2005) and greater RMA beliefs than women do (Anderson, Cooper, & Okamura, 1997), and those attitudes contribute to coercion (Muehlenhard & Falcon, 1991; Emmers-Sommer & Allen, 1999). Research also indicates that traditionalism relates to RMA (Emmers-Sommer et al., 2005) and that RMA more strongly relates to coercive than to non-coercive pornography (Emmers-Sommer & Burns, 2005).

Thus, within the context of the present study, and given the aforementioned review and theoretical frameworks, it was expected that individuals who have film predilections for sex and violence also have more traditional beliefs than those individuals who prefer films with a love story or suspense theme. Furthermore, social learning theory argues that men and women are more likely to model attitudes and behavior that are rewarded positively. Overall, male characters are rewarded more positively than female characters in films that depict sexual violence (Emmers-Sommer & Burns, 2005) and men tend to be more traditional in nature than women (e.g., Emmers-Sommer et al., 2005). Thus, the following hypotheses were posited:

  1. H1

    Men and women would differ significantly in their film predilections, such that men prefer sex and violence and women prefer love stories.

  2. H2

    Individuals with sex and violence film predilections would have more traditional attitudes than individuals with love story or suspense film predilections.

Effects of Viewing Mediated Sexual Violence

According to Carll (2003a), the media maintain control of the dissemination of information about what is occurring in society, and also reinforce violence. More specifically, information about women and violence in the media aids in shaping public opinion and public policy as well as reinforcing stereotypes that can affect the judicial process (e.g., individuals who are exposed to violent media serve on juries). These stereotypes include the previously discussed rape myths, for example, “a healthy woman would be able to resist rape if she wanted to” (Burt, 1980, p. 223). Such myths and stereotypes have led society to focus more on the victim than on the rapist. However, to have a comprehensive understanding of violence against women, it is necessary to examine both the perpetrator and victim (Boeschen, Sales, & Koss, 1998).

Some media, such as news coverage, often report rape and violence against women as less significant than other forms of violence (Carll, 2003a). This representation of women and violence denies the social roots of violence against women and reinforces the idea that violence against women does not deserve the voice and attention afforded other forms of violence. Carll (2003b) stated that the media shape our conceptions of reality and serve as the gatekeeper of how the world is perceived. From a theoretical perspective, it is imperative that the violent acts are depicted truthfully and accurately because the audience forms their beliefs about violence and women based on those representations (Carll, 2003b).

These trends are especially disturbing when we consider the impact of sexually violent media on male viewers. Malamuth and Check (1981) reported that exposure to violent sexual content in films increased male participants’ acceptance of violence against women, whereas female participants’ acceptance of such violence decreased. The discrepancy between their responses to sexual violence against women is representative of a trend evident in similar studies. According to Weisz and Earls (1995), the viewing of sexually violent film produced equal amounts of affect in both male and female participants, but only male participants demonstrated a negative behavioral response. Malamuth and Briere (1986) found that increased aggression is one notable behavioral effect in men that might indirectly result from viewing sexual violence against women.

Research suggests that repeated exposure to sexually-violent films tends to desensitize both men and women to rape and increases the likelihood that both sexes will believe that rape was precipitated by the victim’s actions (Goleman, 1985). Even exposure to still images that depict female sexuality results in male participants being significantly more accepting of rape-supportive attitudes than men exposed to non-sexual stills (Lanis & Covell, 1995). Similarly, women repeatedly exposed to films that depict sexual violence against women were less likely to believe that the victim was distressed by the assault (Dexter, Penrod, & Linz, 1997). Finally, Berry, Gray, and Donnerstein (1999) conducted a study in which they manipulated the length and nature of edits in violent films. Specifically, in a series of experiments, the researchers manipulated the extent of violence portrayed in a film. Experiment 1 involved a 90 s edit of Reservoir Dogs. Version one consisted of one clip that had edited out either the shooting of policemen with the shaking of their bodies and the spraying of blood, or a scene in which a woman is shot in the chest and the bullet hole was edited out. In version two, both scenarios described in version one were used. In version three, the entire clip was shown free of edits. Results of the experiments were consistent with extant research findings that men enjoy viewing violent media. Specifically, men indicated far greater enjoyment than the women did of the unedited versions of the films. The women in the study indicated that the scenes that depict violence were troubling, but the men made no such indication. The results of these studies confirm that men not only prefer themes of sex and violence in the media they view, but they are also more likely to indicate that they do not enjoy media when such themes have been edited out (e.g., Emmers-Sommer et al., 2005).

From a social learning theory and an excitation theory standpoint, it is expected that men are more likely than women to react positively to sexually violent depictions. Although the treatment film used in the present study was based on a true story and does demonstrate sanctions for the perpetrators, it was expected that the depictions of sanctions would not override the degree of excitement and arousal experienced by the male participants. However, it was expected that the depictions would affect women’s RMA given that viewing a sexually violent film based on a true story is significantly more meaningful to women than to men (Emmers-Sommer et al., 2005) and that women tend to react adversely to depictions of sexual violence (e.g., Koukounas & McCabe, 2001). Thus, the final hypothesis and research question are presented:

  1. H3

    Exposure to sexually violent media would moderate women’s but not men’s film predilections as they relate to rape myth acceptance.

  2. RQ1

    Do individuals who have sex and violence film predilections also have less favorable attitudes toward film editing than those with predilections for a love story or suspense film, and are those predilections moderated by film type?

Materials and Methods

Sample

The sample consisted of 174 undergraduate communication students at a large southwestern university (54 men, 120 women). The disproportionate number of men to women is reflective of the student population in the Department of Communication. Participants’ ages ranged from 18 to 34 years old (M = 19.86, SD = 1.88). Over one-half of the participants (54%) were 19–20 years of age, and the sample was predominantly White.Footnote 1 Most of the participants were involved in a romantic relationship (65 people were casually dating, and 50 were seriously dating); 50 people were uninvolved at the time of the study. Participants reported that their weekly usage of media ranged from 0.5 to 90 h (M = 16.895, SD = 13.74). Regarding the number of hours spent viewing films each week, participants reported viewing 0 to 55 h (M = 4.14, SD = 4.75). In terms of film rating preference, 65% preferred to view films that are R-rated, and 23% preferred PG-13 films.

Instruments

Rape Myth Acceptance scale (RMA) (Burt, 1980)

The RMA is a 10-item, 1–7 likert scale (1 = strongly agree, 7 = strongly disagree) measure that provides statements about rape-related attitudes, beliefs, and the propensity to blame the victim for the perpetrator’s actions. Examples of items from the scale include “Any healthy woman can successfully resist a rapist if she really wants to” and “In the majority of rapes, the victim is promiscuous or has a bad reputation” (p. 223). Past reliability of the scale has been reported as Cronbach’s alpha = 0.875 (Burt, 1980). Reliabilities in the current study were acceptable, with Cronbach’s alphas = 0.80 at T1 and 0.85 at T2.

Sex role stereotyping scale (Burt, 1980)

This scale was incorporated to measure attitudes toward traditional gender roles. Items from the 9-item likert scale (1 = strongly agree, 7 = strongly disagree) represent support of traditional gender roles. Examples of items from the scale include “A man should fight when the woman he is with is insulted by another man” and “It looks worse for a woman to be drunk than a man to be drunk.” Reliabilities in the present study for the scale were Cronbach’s alphas 0.62 at T1 and 0.68 at T2.

Film editing and integrity scale

This 11-item scale was developed for the purpose of the present study to examine attitudes toward editing film for offensive content and preserving the integrity of the art. Items from the scale include “I think that film should be shown in its original and entire form,” “Violence should be cut out of film entirely,” “Sex in film should be edited (i.e., scene fuzzed/blackened out),” and “Manipulating film compromises artistic integrity.” Three items were reversed on the 1-9 scale (1 = strongly disagree, 9 = strongly agree) to enhance reliability. Higher scores indicate a more favorable position toward editing film for sex, violence, and profanity, whereas lower scores indicate a less favorable position. The items were subjected to a principal components analysis with varimax rotation. One factor was extracted, which explained 47% of the variance. Reliabilities in the current study were Cronbach’s alphas = 0.83 at T1 and 0.81 at T2.

Procedure

Participants were instructed that they needed to be at least 18 years old to participate in the study, that the study would involve approximately 2.5 to 3 h of their time, and that they would receive extra course credit for their participation. When they arrived for the experiment, participants were randomly assigned to one of four groups: Group one received the uncut version of the treatment film, Group two received the mosaiced version of the treatment film (i.e., the sexually violent assault was shown, but bodies were digitized so that viewers were unable to see exactly what was occurring), Group three received the professionally edited version of the treatment film, and Group four received the control film Free Willy, a film about a young boy and his relationship with a whale. The Accused served as the treatment film and had a running time of 119 min. It is an award-winning film about a brutal gang rape in a bar and the resultant trial that brought the perpetrators to justice. Jodie Foster earned an Academy Award for her portrayal of the sexually assaulted woman. The film is based on a true story and that fact is made known to viewers at the onset of the film. Prior to the beginning of the experiment, participants received an informed consent form and completed the Time 1 questionnaire. The Time 1 questionnaire included demographic questions and the various measures (e.g., RMA). Participants then saw their respective film and completed the Time 2 questionnaire. The Time 2 questionnaire did not include the demographic questions, but did include the same measures presented in the Time 1 questionnaire and a few additional questions that asked participants about their reactions to the film. The ordering of the three measures was Sex Role Stereotyping Scale, RMA scale, and Film Editing and Integrity Scale in both questionnaires. Participants then received an oral and written debriefing, were given additional materials on sexual assault, and were thanked for their participation.

Results

Hypothesis 1 stated that men and women would differ significantly in their film predilections, such that men prefer sex and violence and women prefer love stories. H1 was supported, X 2(2) = 54.86, phi = 0.562, p < 0.0001. An examination of the cells indicated that four men preferred love stories, 30 preferred sex and violence, and 20 preferred suspense. Conversely, 75 women preferred love stories, 14 women preferred sex and violence, and 31 preferred suspense.

Hypothesis 2 stated that individuals with sex and violence film predilections would have more traditional attitudes than individuals with love story or suspense film predilections. H2 was supported, F(2, 172) = 8.37, p < 0.0001. Post hoc tests indicated that individuals with sex and violence predilections had more traditional beliefs (M = 4.26, lower mean = more traditional attitudes) than individuals with love story (M = 5.23) or suspense (M = 5.09) predilections.

Hypothesis 3 stated that exposure to sexually violent media moderates women’s but not men’s film predilections as they relate to rape myth acceptance. To test this hypothesis, a repeated measures ANOVA was conducted, with preference of film theme, film group assignment (i.e., three different levels of the treatment and the control), and sex as the between-subjects factors and rape myth acceptance as the within-subjects factor. Results indicated a significant effect for sex, F(1, 147) = 17.28, p < 0.0001, an effect for film preference that approached significance, F(2, 147) = 2.70, p < 0.07, and a significant film preference by film group interaction, F(6, 147) = 3.45, p < 0.003. It is interesting that the manipulation of the film was not as powerful in terms of its effect on RMA. Although RMA followed the pattern expected due to visual intensity (i.e., RMA for uncut > than RMA for mosaic > RMA for edited > RMA for control), the group differences were not statistically significant. However, the relationship between rape myth acceptance and pre-existing film preferences, the relationship between RMA and the sex of the participant, and the relationship between the film predilections and the film group assignment during the experiment provides more insight. The univariate test for sex was significant, F(1, 147) = 28.51, p < 0.0001. Men (M = 5.05) reported stronger beliefs in rape myths than women (M = 5.90) did; lower means indicate greater RMA. The mean difference between men’s and women’s scores was significant, p < 0.0001. The relationship between having a sex and violence film predilection and stronger rape myth beliefs approached significance, F(2, 147) = 2.63, p < 0.07. Specifically, those individuals who prefer sexually violent films (M = 5.25) were more inclined to accept rape myth beliefs than were those who prefer love stories (M = 5.89) or suspense films (M = 5.81). Unfortunately, in terms of further exploration of film preferences by film condition, men’s film preferences were not distributed across conditions of the experiment (i.e., no male participants randomly assigned to the edited or control condition reported a love story predilection). Thus, men’s and women’s predilections in the uncut and mosaic conditions were examined in terms of change in rape myth acceptance. As expected, men’s acceptance of rape myths did not significantly change regardless of their film predilection. Specifically, no significant changes occurred for men with a love story film predilection, t(3) = −0.91, p > 0.05, sex and violence predilection, t(12) = −1.05, p > 0.05, or a suspense predilection, t(11) = −0.98, p > 0.05. However, there was a significant change for women who reported a love story predilection, t(40) = −3.09, p < 0.004, which indicates that women with a preference for love stories were more likely to reduce their rape myth acceptance after exposure to the film. Women who reported a preference for sex and violence also reduced their rape myth acceptance and that change approached significance, t(6) = −2.27, p < 0.06. Finally, women with a predisposition for suspense films remained unchanged in terms of their rape myth acceptance, t(13) = −0.39, p > 0.05. In sum, H3 was partially supported. Men’s and women’s rape myth acceptance means for T1 and T2 by film predilection and film group appear in Table 1.

Table 1 Men’s and women’s T1 and T2 rape myth acceptance scores by film predilection*.
Full size table

Research Question 1 asked if individuals who had sex and violence film predilections also had less favorable attitudes toward film editing than did those with love story or suspense film predilections, and whether those predilections were moderated by film type. Once again, a repeated measures ANOVA was conducted, with film predilection as the between-subjects factor and attitudes toward film editing as the within-subjects factor. The result was significant, F(2, 123) = 5.87, p < 0.004. Post hoc tests indicated that those who had sex and violence film predilections were significantly less in favor of film editing (M = 3.11, lower mean = less favorable attitude) than were those who had love story (M = 3.89) or suspense predilections (M = 3.89).

Discussion

Results of this study support earlier research that suggests that men are more favorable than women are to media that contains sex and violence. (e.g., Bushman, 1995; Linz et al., 1988). The present investigation offers several extensions of those studies. First, much of the previous research is on entertainment-based sexually violent fare. In the present study we examined reactions to sexual violence based on a true story.

Second, an interesting finding of our work is that women’s sex and violence predilections were moderated to an extent by their exposure to sexually violent media that was based on a true story. Unlike the men, rape myth acceptance was ameliorated to some degree for the women. This finding varies from earlier research on sexually violent media, which suggested that both men and women become desensitized to such fare (e.g., Goleman, 1985). Perhaps the effects vary for women when they are exposed to more “realistic” material based on a true story that also portrays consequences for the perpetrators. This information may be particularly valuable for women from an educational and intervention standpoint. Future research to this end is suggested. Equally important, from an educational and intervention standpoint, is the finding that men’s attitudes are not affected by more realistic depictions of sexual violence. This result has also been found in previous research on sexual violence from an entertainment slant. Such material rarely, if ever, portrays consequences for sexually violent actions (e.g., sexually violent pornography). Although the film shown for the treatment in the current experiment portrayed a trial and consequences for the perpetrators, it appeared to have little impact on men’s attitudes. This finding is somewhat curious, but does appear to be more supportive of the excitation transfer theory than of the social learning theory. Specifically, a social learning theory perspective contends that attitudes and behaviors are learned and that they become part of one’s reality, particularly if the behavior is rewarded. If the behavior is not rewarded, adoption of attitudes and behaviors should be lessened. Thus, one would expect that men who viewed sexually violent material in which there are real consequences for the perpetrators would adopt anti-RMA beliefs. From an excitation transfer standpoint, it was expected that men who held sex and violence predilections would have increased RMA if they were assigned to the uncut version of the film as the arousal or excitement was transferred to rape myth acceptance. Men who were assigned to the mosaic version of the film might have had increased arousal and anger because they could tell something that was going on, but could not see exactly what was occurring due to the digitization. However, men randomly assigned to the uncut and mosaic versions of the film did not change RMA regardless of film predilection. It could be that the arousal from the material held attitudes constant. Overall, it is premature to draw conclusions based on a single exposure to such material. Future researchers should examine the effects of repeated exposure to material offered in a more realistic forum (e.g., consequences for the perpetrator).

As expected, individuals who reported sex and violence predilections also reported more traditional beliefs than did those who reported love story and suspense predilections. This finding can be explained by social learning theory. Specifically, because men prefer films that contain sex and violence, depictions of women who are degraded and objectified are commonly viewed. According to social learning theory, viewers learn information about reality through symbols portrayed within the media (Bandura, 1977, 1994). Research has shown that both men and women can buy into RMA (Allen et al., 1995) and that desensitization can occur for both genders due to repeated exposure (Goleman, 1985). Thus, individuals who are frequently exposed to sexually violent material are likely to learn and abide by such negative behaviors and also to accept the violent treatment women receive as justified. Furthermore, individuals who view sexually violent media might be more willing to accept rape myths and to believe that the relationships portrayed on film reflect those of real life. For example, Burt and Albin (1981) found that a person who adopts RMA beliefs is more likely to see the woman as partially or completely responsible for any sexual acts, including rape or assault. Therefore, individuals’ repeated exposure to media that degrade women would contribute to their more traditional attitudes and the acceptance of degrading behavior. Given that men were more likely than women to have sex and violence predilections, these findings are amplified for men. As found, the majority of women reported a love story predilection and might not be exposed to as many negative attitudes toward women in such fare. In addition, love stories often promote responsibility between both partners in the relationship, and they typically do not place all of the blame on the woman if something goes wrong. This suggests that women are exposed to different norms regarding sexual relationships than men are, and they thus adopt different attitudes than men do. The research is clear that women are typically more other- than self-oriented in their relationships than men are (e.g., Byers, 1996; Emmers-Sommer, 2002).

Finally, findings indicate that those individuals who preferred sex and violence over love stories or suspense were significantly less in favor of altering a film’s original content. According to social learning and excitation transfer theories, people who prefer sex and violence also might prefer to view an unedited film because it more accurately reinforces their attitudes and beliefs about sexual relationships, as well as increases excitement and arousal, and thereby is more satisfying to high sensation seekers.

Policy Implications

According to Mundorf, Allen, D’Alessio, and Emmers-Sommer (2007), several issues need to be taken into consideration as they relate to the implications of sexual and sexually violent material. First, it is necessary to explore the reactions to such material in conjunction with mere exposure to the material, selection of the material, etc. Although exposure’s effect on attitudes is interesting, our interest also lies with how those attitudes are played out socially and interpersonally. Indeed, Allen, D’Alessio, and Emmers-Sommer (1999) argued that “Media effects is, in a sense, a misnomer; the real effect is how media contact affects the behavior of an individual engaged in social and personal relationships” (p. 158, emphasis in original) Second, and related to the present study, film predilection is a voluntary choice. Individuals choose what they want to see and pay to see it. Mundorf et al. argued that exploration of the nature of selection is a viable research focus and that selection of media is affected by the choices available. With the influx of sexual violence more readily available on the Internet or on directors’ cuts presented in DVD rentals, the implications of research findings on public policy is of value as various interest groups grapple with obscenity and avoidance of censorship.

Limitation

A limitation of this study was the number of men represented by each theme. More specifically, the number of men who chose love stories as their favorite theme was extremely small, and this limited the distribution of film preferences by film condition. As noted in earlier, the disproportionate number of men and women is reflective of the Department’s undergraduate communication student population.

Conclusion

The present study offers some insight into how film predilections are affected by sexually violent material based on a true story. Findings are more pronounced for women and suggest that approaching this subject matter from a more realistic standpoint may be effective from an educational or intervention bent. As noted earlier, it is of interest to explore if repeated exposure to material of this nature further ameliorates women’s RMA or if it eventually desensitizes them; the effect of repeated exposure on men’s attitudes also requires further study. Clearly, further exploration of these issues can contribute to our understanding of the impact of sexually violent material on attitudes and behaviors and offer suggestions for public policy (e.g., Kunkel et al., 2003).