|Title:||The effects of exposure to filmed sexual violence on attitudes toward rape.|
|Abstract:||Investigates the effects of sexual violence presented in feature-length films. Showing of the films `Deliverance,' `Straw Dogs,' `Die Hard 2,' and `Days of Thunder' to male and female respondents; Males' acceptance of interpersonal violence and rape myths, and attraction to sexual aggression; Females' non-affectation by film type.|
|Full Text Word Count:||4521|
|Database:||Academic Search Premier|
Depictions of women "enjoying" the application of force or as the deserved victims of sexual and nonsexual violence are pervasive in the mass media (Berkowitz, 1984; Donnerstein, Berkowitz, & Linz, 1986; Linz, Donnerstein, & Adams, 1989; Linz, Donnerstein, Bross, & Chapin, 1986; Linz, Donnerstein, & Penrod, 1988). Although many people may recognize the fictional nature of this type of information, there is evidence that suggests such depictions nonetheless may have a significant impact (Hans, 1980; Tversky & Kahneman, 1973). Studies have shown that viewing sexually aggressive films negatively influenced viewers' attitudes toward women (Linz et al., 1989; Malamuth & Check, 1981).
The available literature also suggests that violence against women need not occur in a pornographic or sexually explicit context to have a negative effect upon viewer attitudes and behavior. More important, it appears that violent images, rather than sexual ones, are more potent in changing attitudes about women and rape (Donnerstein & Linz, 1986; Scott & Schwalm, 1988): A number of independent studies have shown that following exposure to R-rated aggressive and sexually aggressive films, men treat female victims with less sensitivity, attribute less injury, are less empathic and helpful, are generally more callous and aggressive, and are more accepting of interpersonal violence and cultural stereotypes (Donnerstein et al., 1986; Donnerstein & Linz, 1986; Linz et al., 1989).
In one of the few studies to include both male and female subjects, Malamuth and Check (1981) found that exposure to films portraying women who benefit from sexual violence significantly increased males' but not females' acceptance of interpersonal violence. In fact, there was some indication of an opposite effect for women: Women were less accepting of rape myths and interpersonal violence after exposure to sexually violent films.
Various explanations have been offered in an attempt to explain the difference between men and women with respect to the effect of exposure to media violence and sexual violence. For example, Malamuth and Check (1981) suggested a possible "attitude polarization" effect (see also Lord, Ross, & Lepper, 1979) where people with differing views on a particular issue are presented with mixed data so that each side can focus on the information consistent with their own views. Another explanation is a "reactance phenomenon" (Hellman, 1976; Sensenig & Brehm, 1968) in which women identify with female victims and activate various defense strategies to consider why the information conveyed is false. The victim portrayed in the sexually violent films used in research is always a female. Thus it is reasonable to assume that the men watching these films are unable or unwilling to identify with these victims. If women are less likely to be affected by exposure to these films (because they more easily identify with the victims), then perhaps the same would be true of men who would view other men as victims of assault, specifically sexual assault.
In order to test this hypothesis, the present experiment exposed both males and females to one of four feature-length films: a neutral film depicting no sex or violence; a film depicting sexual violence directed against a woman; a film showing the violent rape of a man; and a film containing physical, but no sexual, violence.
A total of 193 university students served as voluntary participants. The subjects consisted of 87 males aged 18-42 (M = 20, SD = 2.5), and 106 females aged 18-32 (M = 19, SD = 2.0). Students were recruited through bulletin board advertisements for psychology studies. Most participants received course credit in return for participation.
All participants were asked to sign a consent form and to complete a demographic data questionnaire. The following paper-and-pencil self-report measures were administered to all subjects: the Acceptance of Interpersonal Violence Scale (Burt, 1980); the Attraction to Sexual Aggression Scale (Malamuth, 1989); the Buss-Durkee Hostility Inventory (Buss & Durkee, 1957); the Marlowe-Crowne Social Desirability Scale--short form (Strahan & Gerbasi, 1972); the Mehrabian-Epstein Empathy Scale (Mehrabian & Epstein, 1972); and the Rape Myth Acceptance Scale (Burt, 1980). In addition, a movie rating questionnaire (cf. Ramirez, Bryant, & Zillmann, 1982) and a 23-item rape trial questionnaire developed by Linz et at. (1988) also were administered to participants. The combination of these tests resulted in a questionnaire having a total of 275 items. All questionnaires were kept intact, but the sequence of the scales was randomized, resulting in four different orders of presentation.
The following films were used in this experiment: Days of Thunder (neutral, 106 minutes; Simpson, Bruckheimer, & Scott, 1990), Deliverance (male raped by a male, 106 minutes; Boorman, 1970), Die Hard 2 (physical aggression, i.e., typically males physically aggressing toward mostly males and some females resulting in relatively equal damage to both genders, 116 minutes; Gordon, Silver, Gordon, & Harlin, 1990), and Straw Dogs (female raped by a male, 107 minutes; Melnick & Peckinpah, 1970). Two of the films (Deliverance and Straw Dogs) were slightly edited in order to satisfy treatment conditions, that is, certain violent segments were edited from the two sexually aggressive films to avoid confounding the sexually aggressive and physically aggressive variables. A reenactment of a rape trial (i.e., a mock trial of a case that involved the alleged rape of a woman during a fraternity party that contained opening statements from both attorneys, witness testimony, attorney closing arguments, and final instructions from the trial judge; 54 minutes; Linz et al., 1988) also was used.
Subjects were told that the purpose of the experiment was to compare their perceptions of films regarding a variety of topics and their attitudes on various issues, such as sex, friendship, helping behaviors, and legal issues, with those of other students. Then all participants were asked to sign a consent form.
Subjects were randomly assigned to one of the four treatment conditions: (a) sexual aggression against a male; (b) sexual aggression against a female; (c) physical aggression; or (d) neutral. Participants viewed each film in groups ranging in size from 45 to 54 people, in a comfortable classroom setting. After film viewing, subjects were given as much time as they needed to complete the 252-item questionnaire consisting of all scales except the rape trial questionnaire. Participants then viewed the reenacted rape trial on video. Following this, the rape trial questionnaire was administered. Upon completion of the experiment, the actual purpose of the study was explained and subjects were debriefed.
In order to examine the effects of gender and film type, a 2 x 4 between-subjects multivariate analysis of variance (MANOVA) was performed on the five dependent variables--acceptance of interpersonal violence, rape myth acceptance, attraction to sexual aggression, victim sympathy, and verdict--as well as on the hostility and empathy variables.
Using Wilks's criterion, the MANOVA revealed a main effect of gender, F(12, 173) = 7.00, p < .001, and a film type by gender interaction, F(36, 512) = 1.60, p < .05. The results yielded a moderate association between gender and the combined dependent variables (eta[sup 2] = .33), as well as between the interaction of gender and film type and the dependent variables (eta[sup 2] = .27).
The mean scores on the Acceptance of Interpersonal Violence Scale, the Rape Myth Acceptance Scale, and the Attraction to Sexual Aggression Scale as a function of the interaction of film type and gender are shown in Figures 1, 2, and 3, respectively. Examination of these figures suggests large differences in response patterns between men and women. The results of the MANOVA (main effect of gender) confirmed these visual impressions. Specifically, males were significantly more accepting of interpersonal violence (M = 9.08 vs. 5.30; F[1, 184] = 24.13, p < .001), more attracted to sexual aggression (M = 19.11 vs. 12.26; F[1,184] = 10.35,p < .01), and more accepting of rape myths (M = 23.43 vs. 13.61; F[1, 184] = 25.50, p < .001) than females.
The mean scores obtained on the scales assessing victim sympathy and verdict as a function of the interaction of film type and gender are shown in Figures 4 and 5, respectively. The results of the MANOVA showed a significant main effect of gender on both of these dependent variables; when compared to males, females exhibited higher levels of sympathy for the victim in the rape trial (M = 12.14 vs. 9.77; F[1, 184] = 13.19, p < .001), and a greater likelihood of finding the defendant guilty (M = 0.73 vs. 0.40; F[1, 184] = 22.42, p < .001). These results also are presented in Table 1, which shows the mean scores of males and females on the various dependent variables for the main effect of gender.
Measures administered to obtain general levels of empathy and hostility also yielded a main effect of gender for empathy, F(1, 184) = 61.79, p < .001, that is, females were found to be significantly more empathic than males (M = 138.19 vs. 118.93). A significant main effect of film type for the hostility measure was observed, F(3, 184) = 4.89, p < .01. Post hoc mean comparisons using Bonferroni's correction (alpha/3 = .0167) showed that participants who viewed films involving sexual violence against either a male or a female obtained the highest hostility scores (M = 41.86 and 40.61, respectively), when compared to subjects exposed to the physically violent film (M = 34.69; t[1, 184] = 3.38) and the neutral film (M = 35.76; t[1,184] = 2.98).
Table 2 shows the mean scores on the various dependent variables for the interaction effect of Film Type x Gender; a significant difference was obtained on the measure assessing acceptance of interpersonal violence, F(3, 184) = 2.82, p < .05. To be more specific, multiple comparisons using the Tukey-B procedure (significance level of p < .05) revealed that males, who viewed the films depicting sexual violence against a female or a male, were significantly more accepting of interpersonal violence (M = 11.24 and 10.41, respectively) than females exposed to the same films (M = 5.00 and 5.13, respectively), as well as females who viewed either the physically violent film (M = 4.59) or the neutral film (M = 6.16). A tendency toward significance with respect to the Film Type x Gender interaction effect also was revealed on the Rape Myth Acceptance Scale (p = .10). In order to explore this difference, a one-way analysis of variance was conducted separately for each gender group. Using this analysis, a significant difference was revealed for males concerning rape myth acceptance, F(3, 83) = 3.09, p < .05. Post hoc mean comparisons using Bonferroni's correction (alpha/3 = .0167) indicated that males exposed to sexual violence against either a male or a female (M = 26.63 and 28.52, respectively) were significantly more accepting of rape myths than males who viewed either the film depicting physical aggression (M = 17.80) or the neutral content film (M = 19.16; t[3, 83] = 3.04).
When data for males and females were collapsed over type of sexually violent film (i.e., sexual violence against a male and sexual violence against a female), a significant Film Type x Gender effect emerged on the Acceptance of Interpersonal Violence Scale, F(2, 186) = 4.11, p < .05, as well as on the Attraction to Sexual Aggression Scale, F(2, 186) = 3.24, p < .05; in addition, significance was approached on the Rape Myth Acceptance Scale (p = .055). Thus it appears that males who watched either of the sexually violent films were most accepting of interpersonal violence, most attracted to sexual aggression, and showed a tendency toward being most accepting of rape myths, as compared to males who watched either the physically violent film or the neutral film, or females exposed to any film.
The MANOVA also revealed a significant Film Type x Gender effect for victim sympathy, F(3, 184) = 6.60, p < .001 and verdict, F(3, 184 = 5.69, p = .001. Multiple comparisons (using the Tukey-B procedure with a significance level of p < .05) showed that with respect to victim sympathy, males exposed to sexual violence against a male were significantly less sympathetic toward the victim in the rape trial (M = 8.00) than males who viewed physical violence (M = 11.90) or females who viewed either of the sexually violent films (M = 13.83, male rape; M = 11.93, female rape) or the neutral film (M = 12.94). Further, the Tukey-B procedure also demonstrated that males, who viewed either the film depicting sexual violence against a male or the neutral content film, were significantly more lenient toward the perpetrator in the rape trial video (M = 0.26 for each group) than females exposed to either of the sexually aggressive films (M = 0.87, male rape; M = 0.79, female rape) or the neutral film (M = 0.78).
Finally, it is important to note that following viewing of the neutral content film, males and females did not significantly differ from each other on measures assessing acceptance of interpersonal violence, rape myth acceptance, attraction to sexual aggression, or victim sympathy. However, when exposed to films containing sexual violence (against either a male or a female), males became significantly more accepting of interpersonal violence and rape myths when compared to females.
Results showed large and consistent differences between males and females. Overall, when compared to women, men were more accepting of interpersonal violence and rape myths, more attracted to sexual aggression, less sympathetic toward the rape trial victim, less likely to judge the defendant as guilty, and generally less empathic. Further, significant differences were observed based on the interaction effect of film type and gender on the Acceptance of Interpersonal Violence Scale, as well as the measures used to assess victim sympathy and verdict. Specifically, multiple comparisons revealed that males exposed to either film depicting sexual violence (against a male or a female) were significantly more accepting of interpersonal violence than females who viewed any film whatsoever. With respect to victim sympathy, males exposed to sexual violence against a male were least sympathetic when compared to females exposed to films involving sexual aggression (against a male or a female) or neutral content, and males who viewed physical violence. Males who watched sexual violence against a male or the neutral film were significantly less likely to convict the perpetrator than females who viewed either film depicting sexual aggression or the neutral content film.
One unexpected finding was that males were generally not differentially influenced by the sex of the victim in the sexually violent films. Collapsing data for both males and females over type of sexually violent film (i.e., sexual violence against a male and sexual violence against a female) revealed significant film type by gender interaction effects on acceptance of interpersonal violence, attraction to sexual aggression, victim sympathy, and verdict; males viewing a sexually aggressive film, regardless of victim gender, tend to be more accepting of interpersonal violence, more attracted to sexual aggression, and less sympathetic toward a victim of rape when compared to females exposed to the same films or males and females who viewed either physical violence or neutral films.
Most important, this study showed significant and meaningful changes in attitude after viewing commercially available feature films. Although females remain relatively unaffected by film type, males were most affected by the sexually aggressive films resulting in negative changes in certain attitudes toward and perceptions of women indicating that women deserve or secretly desire rape.
Consistent with prior research (Barnett & Field, 1977; Malamuth & Check, 1981; Malamuth, Haber, & Feshbach, 1980; Selby, Calhoun, & Brock, 1977; Tieger, 1981), the present study found male subjects to be more accepting of interpersonal violence and rape myths than females. Malamuth and Check (1981) found that exposure to films portraying violent sexuality (against women) increased male subjects' acceptance of interpersonal violence against women. Similarly, males in the present investigation, who viewed sexual violence against either a man or a woman, obtained higher scores on scales measuring acceptance of interpersonal violence and rape myth acceptance when compared to males who viewed either the physically violent film or the neutral film. Malamuth and Check (1981) also reported that viewing sexually aggressive films significantly increased men's but not women's acceptance of cultural stereotypes indicating that women deserve or secretly desire rape. The present investigation replicated these results.
It is also interesting that in the present experiment females did not seem to be affected by film type. For the moment, it is not clear why females manage to escape the influence of the information contained in either violent or sexually violent films. By including the depiction of a male rape in the present study, we attempted to control possible "attitude polarization" or "reactance phenomenon" effects. However, because of the use of commercially available feature films, it was impossible to manipulate the extent to which male subjects identified with the male victim. Rather, the most likely explanation of the present data is the "just world" theory.
Linz et al. (1989) have reasoned that exposure to many scenes from "slasher" type of films that nearly always portray female victims willingly placing themselves in situations that inevitably lead to injury or death may cause viewers to blame the victim for her own assault (ascribing to the belief in a "just world," the idea that ultimately we all get what we deserve; Lerner, 1965, 1971). Zillmann and Bryant (1982, 1984) also have suggested that prolonged exposure to images of women depicted as sexually promiscuous results in the trivialization of rape and other forms of sexual violence. With respect to the results of the present research, the above-mentioned theory may partially explain the effects of exposure to sexual violence for males. Another possible explanation of these results is the concept of availability. After being exposed to the information presented in the films depicting sexual aggression, these effects are what become more readily available cognitively. Exposure to these stimuli may have encouraged male subjects, who perhaps already upheld specific thought patterns that supported or reinforced sexual violence in others. Finally, male subjects viewing other males being sexually aggressive toward a female may simply become disinhibited against subsequent aggression toward women via desensitization or modeling effects.
Obviously, the present research presents some limitations, most of which are endemic to all laboratory studies of this nature. First, the participants in this study were all university students. Second, subjects were asked to fill out questionnaires and to act as "mock jurors" after watching a reenactment of a rape trim immediately following exposure to the various films. Third, the films used in this study contained particular types of violence; it is important to examine who the violence is directed at and how the victims are depicted.
Future research should examine the possible interacting effects of predisposing personality characteristics, family history, consumption of pornography, sexual experiences, and quantity of exposure to televised and filmed violence and/or sexual violence. In addition, it would be interesting to vary both the number and type of films, as well as the time interval between movie viewing and the dependent measure tasks. The use of a more objective measure, such as the Buss-Durkee Hostility Paradigm, as well as measuring physiological arousal during exposure to sexually violent films, also would be beneficial.
Authors' Note: We wish to thank William Tooke for his assistance in conducting the research and Martin Lalumiere for his comments on an earlier version of this manuscript and for his patience and kindness. Reprint requests should be sent to C.M. Earls, Department de Psychologie, Universite de Montreal, Case Postale 6128, succursale "A", Montreal, Quebec, Canada, H3C 3J7.
TABLE 1: Mean Scores for Main Effect of Gender
Legend for Chart: A - Scale B - Males, (n = 87) C - Females, (n = 106) A B C Acceptance of Interpersonal Violence 9.08 5.30[**] Rape Myth Acceptance 23.43 13.61[**] Attraction to Sexual Aggression 19.11 12.26[*] Victim Sympathy 9.77 12.14[**] Verdict 0.40 0.73[**]
* p </= .01; ** p N/= .001.
TABLE 2: Mean Scores for Interaction Effect of Film Type by Gender
Film Type Physical Male Female Violence Rape Rape Neutral Scale (n = 42) (n = 50) (n = 50) (n = 51) Acceptance of Interpersonal Violence M 7.00 10.41 11.24 7.00 F 4.59 5.13 5.00 6.16[*] Rape Myth Acceptance M 17.80 26.63 28.52 19.16 F 13.86 11.57 15.25 13.47 Attraction to Sexual Aggression M 13.30 24.85 20.14 15.95 F 9.27 12.35 10.04 16.19 Victim Sympathy M 11.90 8.00 9.91 9.90 F 9.96 13.83 11.93 12.94[**] Verdict M 0.55 0.26 0.57 0.26 F 0.46 0.87 0.79 0.78[**]
* p </= .05; ** p </= .001.
GRAPH: Figure 1: Mean Scores for Males and Females on the Acceptance of Interpersonal Violence Scale as a Function of FIlm Type and Gender
GRAPH: Figure 2: Mean Scores for Males and Females on the Rape Myth Acceptance Scale as a Function of Film Type and Gender
GRAPH: Figure 3: Mean Scores for Males and Females on the Attraction to Sexual Aggression Scale as a Function of Film Type and Gender
GRAPH: Figure 4: Mean Scores for Males and Females on the Victim Sympathy Measure as a Function of FIlm Type and Gender
GRAPH: Figure 5: Mean Scores for Males and Females on the Verdict Measure as a Function of Film Type and Gender
Barnett, N. J., & Field, H. (1977). Sex differences in university students' attitudes toward rape. Journal of College Student Personnel, 2, 93-96.
Berkowitz, L. (1984). Some effects of thoughts on anti- and prosocial influence of media events: A cognitive neoassociation approach. Psychological Bulletin, 95, 410-427.
Boorman, J. (Producer & Director). (1970). Deliverance [Film]. Burbank, CA: Warner Bros.
Burt, M. R. (1980). Cultural myths and support for rape. Journal of Personality and Social Psychology, 38, 217-230.
Buss, A., & Durkee, A. (1957). An inventory for assessing different kinds of hostility. Journal of Consulting Psychology, 21, 343-349.
Donnerstein, E., Berkowitz, L., & Linz, D. (1986). Role of aggressive and sexual images in violent pornography. in E.
Donnerstein, D. Linz, & S. Penrod (Eds.), The question of pornography. New York: Free Press. Donnerstein, E., & Linz, D. (1986). The question of pornography. Psychology Today, 20(12), 56-59.
Gordon, L., Silver, J., & Gordon, C. (Producers), & Harlin, R. (Director). (1990). Die Hard 2 [Film]. New York, NY: Twentieth Century Fox.
Hans, V. P. (1980, September). Pornography and feminism: Empirical evidence and directions for research. Paper presented at the annual meeting of the American Psychological Association, Montreal.
Heilman, M. (1976). Oppositional behavior as a function of influence attempt intensity and retaliation threat. Journal of Personality and Social Psychology, 33, 574-578.
Lerner, M. J. (1965). The effect of responsibility and choice on a partner's attractiveness following failure. Journal of Personality, 33, 178-187.
Lerner, M. J. (1971). Observer's evaluation of a victim: Justice, guilt, and veridical perception. Journal of Personality and Social Psychology, 20, 127-135.
Linz, D., Donnerstein, E., & Adams, S. M. (1989). Physiological desensitization and judgments about female victims of violence. Human Communication Research, 15(4), 509-522.
Linz, D., Donnerstein, E., Bross, M., & Chapin, M. (1986). Mitigating the influence of violence on television and sexual violence in the media. In R. Blanchard (Ed.), Advances in the study of aggression (Vol. 2, pp. 165-194). New York: Academic Press.
Linz, D., Donnerstein, E., & Penrod, S. (1988). The effects of long-term exposure to violent and sexually degrading depictions of women. Journal of Personality and Social Psychology, 55, 758-768.
Lord, C. G., Ross, L., & Lepper, M. R. (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology, 37, 2098-2109.
Malamuth, N. M. (1989). The Attraction to Sexual Aggression Scale: Part one. Journal of Sex Research, 26(1), 26-49.
Malamuth, N. M., & Check, J.V.P. (1981). The effects of mass media exposure on acceptance of violence against women: A field experiment. Journal of Research in Personality, 15, 436-446.
Malamuth, N, M., Haber, S., & Feshbach, S. (1980). Testing hypotheses regarding rape: Exposure to sexual violence, sex differences, and the normality of rapists. Journal of Research in Personality, 14, 121-137.
Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy. Journal of Personality, 40, 525-543.
Melnick, D. (Producer), & Peckinpah, S. (Director). (1970). Straw Dogs [Film]. Farmington Hills, MI: ABC Pictures Corporation.
Ramirez, J., Bryant, J., & Zillmann, D. (1982). Effects of erotica on retaliatory behavior as a function of level of prior provocation. Journal of Personality and Social Psychology, 43(5), 971-978.
Scott, J. E., & Schwalm, L. A. (1988). Rape rates and the circulation rates of adult magazines. Journal of Sex Research, 24, 241-250.
Selby, J. W., Calhoun, L. G., & Brock, T. (1977). Sex differences in the social perception of rape victims. Personality and Social Psychology Bulletin, 3, 412-415.
Sensenig, J., & Brehm, J. W. (1968). Attitude change from an implied threat in attitudinal freedom. Journal of Personality and Social Psychology, 8, 324-330.
Simpson, D., Bruckheimer, J. (Producers), & Scott, T. (Director). (1990). Days of Thunder [Film]. Hollywood, CA: Paramount Pictures Corporation.
Strahan, R., & Gerbasi, K. C. (1972). Short, homogeneous versions of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology, 28, 191-193.
Tieger, T. (1981). Self-rated likelihood of raping and the social perception of rape. Journal of Research in Personality, 15, 147-154.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207-232.
Zillmann, D., & Bryant, J. (1982). Pornography, sexual callousness, and the trivialization of rape. Journal of Communication, 32(4), 10-21.
Zillmann, D., & Bryant J. (1984). Effects of massive exposure to pornography. In N. Malamuth & E. Donnerstein (Eds.), Pornography and sexual aggression (pp. 115-138). New York: Academic Press.
By MONICA G. WEISZ and CHRISTOPHER M. EARLS
Monica G. Weisz received her doctorate in psychology from the Universite de Montreal and is currently in private practice in Montreal, Quebec. Her current research interests include the quality of life and psychological well-being of menopausal women
Christopher M. Earls is an associate professor at the Universite de Montreal His major research interest is the assessment and treatment of sexual offenders.