JIDV - IJOV

Auteur(s)

Review of the Literature

ocial scientists have repeatedly expressed a need for a comprehensive examination of the data linking alcohol use and sexual victimization.  Many correlates of sexual aggression experiences have been identified, providing greater understanding of factors associated with increased risk.  Testa and Livingston (2000) called for researchers to further investigate the " . . . [c]omorbidity between alcohol problems and sexual aggression . . . , " since their findings supported the notion that alcohol was "the best predictor" of sexual aggression in cross-sectional analyses (p. 241).

These researchers note that a more thorough understanding of the mechanisms by which alcohol is related to violent (and other, as is so-called "acquaintance rape") victimization can facilitate the development of effective prevention programs for women.  Further, that rape and other forms of sexual violence are increasingly recognized as significant and fairly common experiences in the lives of young women.  Finally, although Testa and Livingston suggest that women's alcohol consumption may increase vulnerability to sexual aggression, "direct evidence for this relationship is sparse as most studies are cross-sectional" (Testa & Livingston, 2000, p. 413).

            Martin & Bachman (1998) emphasized the need for research to study the contribution of alcohol consumption to the severity of rape outcomes; Testa & Dermen (1999) noted that empirical research examining whether risk factors (including alcohol) that are differentially associated with types of sexual aggression are quite rare. 

            When examining the relationship between drinking behaviors and date rape experiences, Canterbury et al. (1993) emphasized the need for multivariate research to establish the relationship of drinking behavior in date rape (see also Greene & Navarro, 1998; Koski, 2003 and authorities cited therein), which link alcohol use to repeat date rape victimization.  Muehlenhard & Linton’s (1987) conclusions point to the weaknesses in studies found in the existing literature, owing to the inability of researchers to randomly assign participants to treatment groups.  They suggested that future research be designed to combine the results from a variety of studies to provide insight into the linkage between alcohol and sexual aggression.  Since alcohol consumption by the victim during a "hetersocial" situation was a significant predictor of alcohol consumption during the assault (see also Abbey et al. (1996)), additional research is needed to establish the causal direction of the relationships that have been found in these and other cross-sectional studies.

            Finally, the need for further research was corroborated by Ullman et al. (1999) when they suggested the need for work that examines the victim preassualt alcohol use as it affects victim vulnerability.  Clearly the need for a study of this nature is well-documented in the existing literature as the current research provides only a fragmented picture.

Hypothesis 

            Thus, this first published meta-analysis analysis exploring the relationship between these variables.  The question specifically addressed here is whether an increase in (or the presence of) alcohol consumption by the victim leads to an increase in the likelihood of sexual violence against her.

Methodology

            A standard practice for determining if a relationship exists between and among variables across studies is to statistically combine independent studies to determine whether a given effect appears in the literature at a rate greater than would be expected by chance.  By examining the reported probability ("p.") levels and sample sizes of several investigations, the probability level of weighted samples can be estimated to confirm if a relationship occurred by chance. Therefore, although not necessarily politically popular under current standards (see Summary and Conclusions, below) the meta-analytic procedure employed here may provide useful information on the covariance of victim alcohol use and sexual violence.  Specifically, our procedure provides both an estimate of the strength of the relationship between alcohol and rape and a test of statistical significance.

           

            An extensive review of the literature addressing the question uncovered 27 peer-reviewed journal articles, four book chapters and two monographs that clearly drew associations between victim risk factors that included victim alcohol consumption and sexual violence.  The studies were conducted in a variety of locations in the US.  Thirteen of these articles included a test of the relationship between rape / sexual violence and risk factors specifically linked to alcohol consumption.  One of the 13 separately presented findings from multiple samples, producing a total of 16 meta-analytic cases from the original 13 studies.

            The studies included in the analysis (and their associated data) are presented in Table 1.  All 16 studies used between-subjects research designs. The sample sizes in each study (denoted as "n") represent the number of subjects in the test condition.  When there was a combination of risk factors reported and the alcohol condition was not specifically enumerated, the n’s were estimated from the total number of subjects.  The p. values listed in Table 1 are the reported values from the main effect -- alcohol on the rape / sexual violence measure. 

           

Although several different types of alcohol consumption were included in the various studies, only those studies that reported the results of a measure that specifically assessed the extent to which the victim engaged in voluntary alcohol consumption were included.

           

One of the difficulties with this procedure involves the question of estimating the p. values for non-significant results that are typically left unreported.  These results follow the common practice of utilizing a p. value of .50 in that instance.

Results

           

As is apparent in Table 1, nine (9) studies reported a p. level that was significant at the p. < .05 level, and one that approached statistical significance, reporting a p. = .067.  Four of the six studies with non-significant results with a p. value of .50 were inserted.  To estimate the probability that an alcohol / sexual violence relationship occurs by chance, an overall Z score for the 16 studies was calculated.  This resulted in a weighted Z-score of 1.65 for the combined studies.  This combined Z-score is statistically significant with the probability that these results occurred by chance equaling p. = .0495. 

           

Thus, given the frequency with which alcohol is found to be a significantly contributing risk factor in the occurrence of rape or sexual violence, it is most certainly unlikely that these findings can be attributed to chance alone at the p. < .05 level.  The statistical combination of relevant studies thus provides support for the association between victim alcohol us and sexual victimization.  The overall effect size estimate of alcohol use by the victim is .119.

           

One finding of some interest was that of Testa and Collins (1997).  Women were divided into two groups: one group included data descriptions of a (1) sexual encounter with a new or occasional partner involving alcohol only; the other group included data descriptions of a (2) sexual encounter with a new or occasional partner involving no alcohol or drug use.  Among other information obtained, the women were surveyed to assess the degree to which they felt pressured to engage in sexual activity, using a seven-point scale (1 = not at all, 7 = very much so).  The intent was to rate the degree to which their partner had pressured these women to have sex.

           

The findings showed that that women who did not use alcohol felt more pressure to engage in sex than those who did use alcohol.  This finding was statistically significant, but not in the predicted direction, because it was theorized that women who used alcohol would feel more pressure to engage in sex.  Therefore, a p. value of .5 was substituted for the probability that was statistically significant but did not support the expected results.

Summary and Conclusions

            Our meta-analysis, the first conducted and presented in the sociological literature, has confirmed the single hypothesis presented -- that victim alcohol use and sexual violence are not just cross-sectionally, but multivariately associated -- and has done so across not just our own but multiple analyses.  While again admittedly an (arguably) unpopular conclusion, it is too among feminist and other "advocacy" analyses, which largely rely on politically motivated and qualitative "results": "[f]eminist knowledge, theory and research is very much rooted in female experience, and has tended to utilize qualitative research methods.  This, along with its association with politics, has often proved too uncomfortable for those in the mainstream scientific community" (Ward 1996). 

            The instant analysis, which looks not simply into alcohol use by the offender but also that of the victim, is consistent with our analysis in Part I of this article: that once the rhetoric of "victim blame" is transcended, hope for change -- rape avoidance behavior -- is possible among victims. 

            Rarely is this true of sexual offenders, who recidivate at rates estimated at 80% and above. 

 

Table 1

Main Effect for Alcohol Use by Victim

_______________________________________________________________________

 Author            Year     n         p             z          (z)(n)          r

_______________________________________________________________________

Testa    

Dermen            1999    166      0.025     1.96     325.36        0.152

Greene           

Navarro

   Exp 1            1998    274      0.001     3.08     843.92        0.186

   Exp 2            1998    88        0.001     3.08     271.04        0.328

   Exp 3            1998    105      0.020     2.05     215.25        0.200

Leventhal         1998   1019     0.5         0.0        0.0            0.000

Canterbury                             

Grossman                             

Lloyd               1993   1765      0.01      2.33    4112.45        0.055

Martin                             

Bachman          1998    200       0.066     1.5      300.0           0.106

Gorman                                                                                                        

Speer

Labouvie

Subaiya            1998    223      0.5           0.0        0.0             0.000    

Muehlenhard                             

Linton              1987    577      0.024       1.98    1142.46        0.082

Abbey                              

Ross                             

McDuffie                             

McAuslan        1996    609      0.001       3.08    1875.72        0.125   

                                                    (Table Continues)

_____________________________________________________________________

________________________________________________________________________

Larimer                                            

Lyndum                                            

Anderson                                             

Turner              1999    296      0.05      1.65     488.4         0.096

Testa                             

Livingston        2000     47      0.0 1     2.33     109.51        0.340

McCaw                             

Senn                1998     40      0.5       0.0        0.0              0.000

Ullman                             

Karabatsos                             

Koss                1999    200      0.001     3.08     616           0.218

Koss                             

Dinero                              

Seibel              1988    468      0.001     3.08    1441.44        0.142

Testa                             

Collins             1997    112      0.5       0.0        0.0              0.000

[Testa and Collins's results were opposite to that which were predicted.  Women who did not use alcohol felt more pressure to engage in sex than those who did.]

 

Références

Abbey, A., Ross, L., McDuffie, D., and McAuslan, P. (1996). Alcohol and dating risk factors for sexual assault among college women. Psychology of Women Quarterly (20), 147-169.

Allen, J.  (2003). Calculating risk: A new look at rape prevention.  On Wisconsin (Spring), 30-35.

Canterbury, R., Grossman, S. J., and Lloyd, E. (1993). Drinking behaviors and lifetime incidents of date rape among high school graduates upon entering college. College Student Journal (27), 75-84.

Greene, E., and Navarro, R. L. (1998). Situation-specific assertiveness in the epidemiology of sexual victimization among university women: A prospective path analysis. Psychology of Women Quarterly (22), 589-604.

Gorman, D. M., Speer, P. W., Labouvie, E. W., and Subaiya, A. P. (1998). Risk of assaultive violence and alcohol availability in New Jersey. American Journal of Public Health (88), 1, 97-100.

Koski, D. D. (2003). Jury selection: A social scientific analysis. In Koski, D. D. (Ed.) The Jury Trial in Criminal Justice, pp. 242-259. Durham, NC. Carolina Academic Press.

Koss, M. P., Dinero, T. E., and Seibel, C. A. (1988). Stranger and acquaintance rape: Are there differences in the victim’s experience? Psychology of Women Quarterly (12), 1-24.

Larimer, L. E., Lydum, A. R., Anderson, B. K., and Turner, A. P. (1999). Male and female recipients of unwanted sexual contact in a college student sample: Prevalence rates, alcohol use, and depression symptoms. Sex Roles (40), 3, 295-308.

Leventhal, J. M. (1998). Epidemiology of sexual abuse of children: Old problems, new directions. Child Abuse and Neglect (22), 6, 481-491.

Lauritsen, J. L. and K. F. Davis Quinet.  (1995). Repeat victimization among adolescents and young adults. Journal of Quantitative

Criminology 11, 143-166.

Martin, S., and Bachman, R. (1998). The Contribution of alcohol to the likelihood of completion and severity of injury in rape incidents.  Violence Against Women (4), 6, 694-712.

McCaw, J. M., and Senn, C. Y., (1998). Perception of cues in conflictual dating situations. Violence Against Women (4), 5, 609-624.

Muehlenhard, C. L., and Linton, M. A. (1987). Date rape and sexual aggression in dating situations: Incidence and risk factors. Journal of Counseling Psychology (34), 2, 186-196.

Testa, M., and Collins, L. (1997). Alcohol and risky sexual behavior: Event-based analyses among a sample of high-risk women. Psychology of Addictive Behaviors (11), 3, 190-201.

Testa, M., and Dermen. K. (1999). The differential correlates of sexual coercion and rape. Journal of interpersonal Violence (14), 5,  548-561.

Testa, M., and Livingston, J. (2000). Alcohol and sexual aggression: Reciprocal relationships over time in a sample of high-risk women. Journal of interpersonal Violence (15), 4, 413-427.

Ullman, S. E., Karabatsos, G., and Koss, M. P. (1999). Alcohol and sexual assault in a national sample of college women. Journal of Interpersonal Violence 14, 6, 603-625.

Ward, C. A. (1996). Attitudes toward rape: Feminist and social psychological perspectives. The Journal of Sexual Aggression (2), 2, 143-154.