|Title:||High school students' adherence to rape myths and the effectiveness of high school rape-awareness...|
|Abstract:||Presents a study on rape myths and the effectiveness of rape-awareness programs in high schools in the United States. Assessment of students' adherence to rape myths; Reference to previous research on this topic; Information on a survey which was conducted; Analysis of the data obtained; Limitations of the study; Conclusions reached based on the study.|
|Full Text Word Count:||7396|
|Database:||Academic Search Premier|
Dateline: Ohio State University, University of Cincinati
AUTHORS' NOTE: This article was presented at the 1994 Annual Meetings of the American Society of Criminology in MIami. The authors would like to thank Professors Amy Elder and Patricia O'Reilly, both of the University of Cincinnati, for comments on an earlier draft of this article.
Over the past 15 years, the emergence of date rape as a social problem has resulted in college campuses implementing rape education or awareness programs. Simultaneously, researchers have worked to uncover the rates of rape on college campuses, college students' adherence to rape myths, and to a lesser extent, the effectiveness of rape-awareness programs on college campuses. This study departs from most of this prior research by focusing on high school students. The focus is twofold. First, the high school students' adherence to rape myths is assessed. Second, the effectiveness of a rape-awareness program for high school students presented by a local female victims' organization is assessed (using an experimental design). Consistent with prior research, the study found reasonably high adherence to rape myths, and that the rape-awareness program was effective in educating students about these myths.
Attitudes about rape have been increasingly examined since the late 1970s and early 1980s. Specifically, feminists have been concerned that the majority of rapes--acquaintance rapes, rapes committed by Anglo men, and rapes committed without weapons--have not been considered "real" rapes (see Estrich, 1987).
Studies have accounted for a number of groups' beliefs about rape myths and what constitutes real rapes. These groups have included citizens in general (M. Burt, 1980; R. Burt & Albin, 1981; Feild, 1978), jurors (LaFree, Reskin, & Visher, 1985), social workers and rape crisis counselors (Feild, 1978; Feldman-Summers & Palmer, 1980), and criminal justice system personnel, such as police officers and judges (Feild, 1978; Feldman-Summers & Palmer, 1980). The group constituting the vast majority of these studies, however, is college students (Barnett & Feild, 1977; Blumberg & Lester, 1991; Check & Malamuth, 1983; Costin, 1985; Deitz, Littman, & Bentley, 1984; Dull & Giacopassi, 1987; Fenstermaker, 1989; Fischer, 1986, 1987; Garrett-Gooding & Senter, 1987; Giacopassi & Dull, 1986; Gilmartin-Zena, 1987, 1988; Holcomb, Holcomb, Sondag, & Williams, 1991; Kanekar & Kolsawalla, 1980; Marx & Gross, 1995; Muehlenhard, 1989; Muehlenhard, Friedman, & Thomas, 1985; Muehlenhard & Linton, 1987; Rapaport & Burkhart, 1984; Sandberg, Jackson, & Petretic-Jackson, 1987; Shotland & Goodstein, 1983; Van Wie, Gross, & Marx, 1995). A noticeable lack of rape attitude studies on high school students exists (exceptions include Blumberg & Lester, 1991; Davis, Peck, & Storment, 1993; Feltey, Ainslie, & Geib, 1991; Goodchilds & Zellman, 1984; Zellman & Goodchilds, 1983).
In an attempt to defy rape myths through education and thus, ideally, deter rapes, a growing number of colleges and high schools have incorporated programs on rape awareness (see Hueston & Burkhart, 1986; Parrot, 1986,1989). One could logically ask then: Are rape attitudes related to participation in rape-awareness programs? Unfortunately, little effort has been made to determine the effectiveness of these programs. The exceptions are two evaluations of rape-awareness programs for college students (Fonow, Richardson, & Wemmerus, 1992; Holcomb, Sarvela, Sondag, & Holcomb, 1993) and one evaluation of a high school program (Feltey et al., 1991). The purpose of the current study is to examine the effectiveness of high school rape-awareness programs as well as to determine high school students' adherence to rape myths.
A review of research on the sexual victimization of college women stated that between 8% and 15% experience forced intercourse, and the rates of less serious sexual victimization (e.g., molesting and verbal sexual harassment) are even higher (see Belknap & Erez, 1995). Furthermore, a study conducted on middle and high school students found that 18% of the girls reported unwanted sexual experiences, including sexual intercourse (Erickson & Rapkin, 1991). Although numerous factors, such as misogyny and a desire to overpower, help explain the occurrence of rape, it is likely that adherence to rape myths also increases the commission of rape. For example, belief in rape myths could justify the rapist's actions to himself and others. Moreover, believing that rape only happens to certain types of girls or women not only defies reality, but it also gives a sense of false security to those who avoid "deserving victim" characteristics (i.e., wearing skimpy clothing, drinking alcohol, being sexually active, and so on). Finally, the focus on the victim fails to attribute blame where it belongs---on the rapist (M. Burt 1980; Giacopassi & Dull, 1986; Parrot, 1989). To more fully understand the rape culture and to combat sexual victimization, it is necessary to assess both the adherence to rape myths and the effectiveness of rape-awareness programs. The current study does both. First, we will briefly summarize previous findings about adherence to rape myths and the effectiveness of rape-awareness programs.
It is too cumbersome for the purposes of this article to list all of the studies that find support for rape myths and the tendency to blame victims for their rapes. Suffice it to say that all of the studies on rape attitudes discussed in this article found considerable support for rape myths. These studies have also suggested, however, that adherence to rape myths may vary based on the respondent's gender, race, and age. The remainder of this section will summarize college and high school students' characteristics that have been studied to determine their relationship to rape attitudes.
As expected, although both males and females hold generally high levels of adherence to rape myths, males are more likely than females to accept rape myths as valid (Barnett & Feild, 1977; Blumberg & Lester, 1991; Costin, 1985; Davis et al., 1993; Deitz et al., 1984; Dull & Giacopassi, 1987; Feild, 1978; Feltey et al., 1991; Garrett-Gooding & Senter, 1987; Giacopassi & Dull, 1986; Gilmartin-Zena, 1987, 1988; Goodchilds & Zellman, 1984; Holcomb et al., 1991; Holcomb et al., 1993; Kanekar & Kolsawalla, 1980; Sandberg et al., 1987). Findings on the effect of race are less clear One study found that a student's race is unrelated to his or her adherence to rape myths (Holcomb et al., 1993), whereas another found more adherence to rape myths among African Americans than Anglos (Davis et al., 1993).
Yet other studies reported the existence of important gender distinctions within race and racial distinctions within gender For example, Giacopassi and Dull (1986) found that Black males are more accepting of rape myths than are any other groups. Black females are more likely than Anglo females to see both victim and offender as blameworthy, and females of either race are more likely than males to reject the belief that females falsely accuse men of rape. Dull and Giacopassi (1987) reported that Blacks are less adherent than Anglos on some rape myth items and more adherent than Anglos on other items.
Research on the relationship between the age of the respondent and adherence to rape myths has been similarly inconsistent. One study found no relationship (Dull & Giacopassi, 1987), another found older students less likely to adhere to rape myths (Feltey et al., 1991), and yet another found younger students less likely to adhere to rape myths (Davis et al., 1993). Another study found no relationship between the age of the student and rape myth acceptance (Dull & Giacopassi, 1987).
The limited research assessing the effectiveness of rape-awareness programs in diminishing adherence to rape myths consistently reported that these programs work (Feltey et al., 1991; Fonow et al., 1992; Holcomb et al., 1993). As predicted and desired, educational interventions significantly improved high school and college students' knowledge and attitudes about rape. Thus, despite documentation of the high risk of rape and a generally strong adherence to rape myths, rape-awareness programs appear to promote education about the reality of rape.
For more than a year (1993-1994), the first author of this study worked as a volunteer at Women Helping Women (WHW), a Cincinnati, Ohio, agency that provides services for victims of incest, rape, and battering. In addition to providing a hotline, counseling, and court advocates for victims of sexual abuse and battering, WHW presents rape-awareness programs in the area high schools. The first two authors developed a plan to: (a) assess high school students' adherence to rape myths, (b) determine the effectiveness of a rape-awareness program, and (c) determine if characteristics about the respondents were related to their responses.
Given the results of prior research, we hypothesize that high school students' adherence to rape myths is influenced by age and the intersection of race and gender (whether they are African American females, Anglo females, African American males, and so on). Moreover, unlike prior studies, we included the variables: dating experience (none vs. some), grade in school (freshman, sophomore, and so on), and type of household (whether they lived in a two-parent household) and school they attended. We also hypothesize that participation in the rape-awareness programs decreases high school students' adherence to rape myths. ST.-PROCEDURE
To assess the three goals stated above, the first two authors developed a two-page survey (that served as both the pre- and posttests) on rape attitudes for distribution in high schools where WHW provided rape-awareness programs. An experimental design with pre- and posttests and experimental and control groups was employed to determine the effectiveness of the program.
The survey was administered to students in three Cincinnati high schools (School A, School B, and School C) before and after rape-awareness programs. The pretest and posttest surveys were identical. The experimental groups in the three schools were tested before and after they participated in the rape-awareness program. The control groups did not participate in the rapeawareness program during the course of the study. (For ethical reasons, these control groups received the program after they had taken both the pre- and posttests.) The pretests, treatment (rapeawareness programs), and posttests in Schools A, B, and C occurred during November and December, 1993.
Permission slips concerning the rape-awareness program and participation in the evaluation of the program were created by the staff at WHW. They were distributed to the students approximately 1 week before the pretest was administered. The test distribution method was approximately the same for all three schools, but the time differed slightly (by a few days) for logistical reasons. This minor limitation of the study was unavoidable because of the researchers' conformity to the various schools' timetables and schedules. The first author administered all of the testing and was present to answer questions that the students had about the survey. The students were told that they were part of a study to evaluate WHW's rape-awareness program and that there were experimental and control groups. Students did not write their names on surveys per instructions; rather, they were anonymous. All testing was conducted in a classroom setting, except for the control group at School A where the students were tested in physical education classes.
The experimental and control groups took the pretest a few days before the rape-awareness presentations. The experimental groups received the posttest approximately I week after having participated in the rape-awareness program. For consistency; control groups' posttests were timed as closely as possible to the time span of the experimental groups' posttests.
The rape-awareness program presented in the schools by WHW was the treatment in this experimental study. The objectives of the rape-awareness program included dispelling common societal rape myths, learning the warning signs of potentially dangerous situations, educating participants about the need for effective and self-assertive communication in dating relationships, examining the media influences that affect society's and individuals' attitudes about rape, and providing participants with information on community resources available to survivors of sexual assault.
The rape-awareness program consisted of a one-session class that lasted I hour and was presented by a worker from WHW. The course structure was a combination of lecture and interaction by verbal communication between the presenter and the students. The students were given factual, thought-provoking information and then were asked to respond and to participate in a discussion with the presenter. No videotapes or other visual aids were used.
The objectives of the program, as listed above, were met through presentations and discussions about the legal definition of rape, motivation of rape, statistics about rape (concerning frequency), and myths about rape. Class discussion included socialization about rape, gender roles, and sexuality by family, friends, and the media. The students were also informed about the many physical and emotional effects and reactions a person who has been raped may have. The class ended with a discussion about how to prevent rape (individually and socially) and what an individual who has been raped can do to seek help and support. ST.-MEASUREMENT
The survey consisted of 24 true and false statements to test adherence to rape myths and seven items regarding students' descriptive information. The development of the survey items was guided by prior measurement instruments on rape attitudes (e.g., M. Burt, 1980; Fonow et al., 1992; Gilmartin-Zena, 1988; Warshaw, 1988) as well as an attempt to address the specific points WHW covered in their rape-awareness program. Thus, some items from prior studies were used directly, some were modified, and some new items missing from prior studies were added. The survey was pilot tested on 6 high school students (4 males and 2 females). These students reported no problems with the survey, so no adaptations were made.
WHW scheduled rape-awareness presentations at three high schools prior to the researchers' involvement in assessing high school students' adherence to rape myths and the effectiveness of the rape-awareness programs. Thus, the researchers had to work around the scheduled programs. Workers at WHW also contacted school administrators to determine whether they were willing to have the programs evaluated. All three schools agreed and students in classes scheduled to view the awareness program became the subjects for the study The three high schools used in the study were public institutions. School A is a predominantly Euro-American middle-class suburban high school. School B is a mostly African American lower- to working-class urban high school. School C is an ethnically mixed, lower- to working-class, urban high school.
Although a random sample would have been ideal, subject access would have been impossible without the first author's affiliation with WHW. To overcome the absence of random sampling, multivariate statistical control was used in the analysis to statistically control for the characteristics that would have been experimentally controlled. Despite the limitation of using an availability sample, the use of statistical control in conjunction with a large, diverse sample (over high school students) from three high schools in varied neighborhoods is suitable to produce meaningful data.
Random sampling would have ensured that the control and experimental groups were comparable to each other in terms of characteristics that could influence responses to the survey questions (e.g., gender, age, year in school, etc.). However, the control and experimental groups differed somewhat in their proportions of these types of characteristics. To prevent a spurious relationship between whether a student participated in the program and the dependent variable tapping rape awareness, variations in these characteristics were statistically controlled by entering them into a model (along with group membership) predicting the dependent variable.
Students in the experimental group at School A came from six health classes (consisting of mostly freshmen) and one sociology class (consisting mostly of sophomores, juniors, and seniors). The control group at School A included four physical-education classes (consisting of mostly freshmen). The experimental group at School B included two health classes and the control group consisted of one health class; all classes consisted of a variety of students in Grades 9 through 12. Finally, at School C, 12 homeeconomics classes that consisted of a majority of ninth graders participated in the study. The classes were randomly split in half at School C: Six classes served as the experimental group and the other six classes served as the control group. Classes were assigned to be experimental or control groups according to the time that they had originally been scheduled to participate in WHW's rape-awareness program. Those classes scheduled at the beginning became the experimental group and those classes that were scheduled toward the end became the control group. Analysis comparing the control and experimental groups' pretests supported the similarity of these test groups; no differences in mean test scores existed between the two groups before the experimental group received the treatment.[sup1]
Slightly more than half of the sample was male (53%) and the sample was largely Euro-American (61%), although more than one quarter of the sample was African American (29%). Almost half of the students were 13 or 14 years old, 30% were age 15, and 15% were age 16. About 10% of the students were age 17 or older. More than three quarters of the students were freshmen, 11% were sophomores, and only about 5% and 6% were juniors and seniors, respectively. Of the sample, 63% came from School A, 9% from School B, and 28% from School C. Those from two-parent households were 58% of the sample, and 84% had dating experience. The total sample consisted of 60% in the experimental groups with the remaining 40% in the control groups.
The survey served as both the pretest and posttest. A total of 866 surveys were usable for the descriptive part of the study, described below, because they did not contain any missing responses to the attitude questions related to rape. In the pretest, 172 control-group surveys and 257 experimental-group surveys contained no missing information on attitudes. In the posttest, 174 control-group surveys and 263 experimental-group surveys also did not lack this information. A total of 837 surveys were usable for the mutltivariate analysis, also described below, because they did not contain missing responses to any of the survey questions used in the construction of the variables in the analysis. This number included 161 pretest control-group surveys, 256 pretest experimental-group surveys, 164 posttest control-group surveys, and 256 posttest experimental-group surveys.
The 24 true-and-false statements in the survey were used to construct the dependent variable for the study. Specifically, the dependent variable was measured as the total number of correct responses to these statements provided by a student. This meant that the dependent variable could potentially range from 0 to 24. This measure constitutes a limited ratio variable because of the lower and upper boundaries of the scale. It can still be considered a ratio level of measurement, however, because of the characteristics of equal intervals and an absolute zero point (i.e., the value of 0 indicates no correct responses).
Cronbach's alpha was calculated to check the interitem reliability of the 24 dichotomous survey items. The alpha coefficient for the pretest was 0.69 and the value for the posttest was 0.83. Readers should be aware that values of alpha equal to or greater than 0.80 are ideal for widely used scales because such values ensure that the correlations between scale items have not been influenced significantly by random measurement error (Carmines & Zeller, 1979).
Data analysis was conducted in two phases. First, frequencies were run to describe the sample and to determine the high school students' overall adherence to rape myths, using the pretests. Second, multivariate analysis was conducted to determine whether posttest performance was related to group membership, controlling for respondents' demographic and social demographic characteristics. Ordinary least squares (OLS) regression was used to examine the relationship between the independent variable of interest (i.e., whether a student participated in the rape-awareness program) and the ratio-level dependent variable (i.e., the number of test questions answered correctly by a student), controlling for the other variables (aforementioned) that could confound the relationship. Inclusion of the control variables permitted an analysis of whether a respondent's race, gender, age, grade, dating experience, and type of household (whether twoparent) were related to both pre- and posttest responses to rape myths. Given that prior research has noted an important interaction between race and gender (Fischer, 1987; Giacopassi & Dull, 1986), the analysis included dummy variables tapping comparisons between different race or gender groups.[sup2]
Two multivariate OLS analyses were conducted and the differences in the results were compared. The first analysis involved predicting the number of items answered correctly by a student during the pretest. The second analysis involved predicting the number of items answered correctly by a student during the posttest. Both analyses included group membership (i.e., control versus experimental) and the statistical control variables. It was hypothesized that the difference in test performance between the control and experimental groups would be nonsignificant before the experimental group members underwent the program and that the difference between the two groups would become significant after the experimental group members took the class. These results would support the idea that the rape-awareness program significantly improved students' knowledge of issues related to rape.
To assess the importance of group membership for predicting test performance in each of the two models, all of the statistical control variables were entered into the model first as a block, followed by the independent variable of interest (i.e., group membership).
It should also be noted that there is some variation in the numbers of students in the pre- and posttest groups. This is likely a factor of students being absent on the days that the pre- or posttests were distributed. Additionally, it should be noted that one dummy variable comparing School B with Schools A and C was included in the analysis to control for environmental differences between the schools. School B is located in a census tract with significantly higher proportions of the following (compared to the census tracts of Schools A and c): households living at or below poverty, unemployed adults, and one-parent households (U.S. Bureau of the Census, 1992). If a lower socioeconomic environment is related to rape awareness, then whether a student attends School B could be related to a student's level of rape awareness. Because of a discrepancy in the numbers of usable surveys from School B between the pretest and posttest periods, this variable was statistically controlled in the multivariate analysis.
Table 1 presents the univariate descriptives for the sample. The focus of the table is on posttest scores, although the pretest mean scores are included in parentheses. As expected, the groups were similar except for a higher mean for number of test questions answered correctly for the posttest groups. (This table is necessary for reading Tables 3 and 4.)
Table 2 presents the overall pretest scores for both the control and experimental groups (to assess high school students' general adherence to rape myths before participation in a rape-awareness program), the posttest scores for the control group, and the post-test scores for the experimental group. Simply eyeballing these results suggests that the experimental group benefited from the treatment of participating in a rape-awareness program.
The reader may note very high percentages of correct responses for both the control and experimental groups on some of the items presented in Table 2 (such as Items 2 and 3). Although these particular items may not be important for determining significant differences between the two groups, they are insightful in that they reveal the areas in which students are most well-informed. Conversely, a close look at the remaining items reveals the issues to which students need more exposure.
Before comparing the answers of the two groups to the specific items, it is worthwhile to first examine the pretest responses as an indication of the high school students' general adherence to rape myths. Fifty percent or more of the students answered three items (1, 5, and 18) incorrectly. Specifically, students seem most unaware that rapists commit rape for power, that rapists usually choose their victims, and that most rapists are not mentally ill (or sick). Generally, students also had trouble understanding (between 51% and 65% answered correctly) that:. Victims cannot always prevent rape, that intimidating someone into having sex relations is rape, that victims do not claim rape simply to get back at men, and that rape is never the victim's fault (Items 6, 20, 21, and 23). The following information was answered correctly by 65.9% to 85% of the students on the pretests: Women do not provoke rape; rape is prevalent; rape victims do not always fight back; sexual relations with women in an altered state is rape; the rape penalty is second to murder; rape victims are not simply women with bad reputations; a drunk woman is not responsible if she is raped; rape is intended to control victims; and persons in altered states are not responsible for being raped (Items 4, 7, 9,10,12,13,15,17, and 22).
High school students were generally aware of the following information (more than 85% answered accurately): Rape can happen to anyone; forcing a sexual act is rape; no does not mean yes or maybe; we should trust our gut feelings in potentially violent situations; even if someone has consented to have sexual relations before, they can still be raped; a woman does not owe a sexual act simply because a man pays for dinner or movies; women can be raped by their husbands; and males can avoid rape accusations if their partners willingly give consent (Items 2, 3, 8, 11, 14, 16, 19, and 24).
Tables 3 and 4 present the results from the multivariate models predicting test performance (measured as the number of correct responses on the survey items) for the pretest and posttest groups, respectively. Two variables were not significantly related to test performance in either the pretest or posttest: which high school the student attended and whether the student had dating experience. Analysis (not shown in the tables) examining the overall effects of race and gender showed that these variables were significantly related to test performance in both the pretests and posttests. Consistent with prior research, the results indicated that males were significantly more likely than females to adhere to rape myths. This analysis also found that African American students were more likely to adhere to rape myths than Anglo students? The results shown in Tables 3 and 4, however, include the interactive effects of race and gender. These findings suggest that Anglo females are the least likely to adhere to rape myths, followed by African American females, Anglo males, and African American males, respectively. This was the case in both the preand posttests.
One variable was significant in the multivariate pretest model but became nonsignificant in the multivariate posttest model: type of household (i.e., whether the student lived in a two-parent home). Notably, students from two-parent homes were more likely to adhere to rape myths in the pretests, but this relationship was washed out with the rape education. It is possible that youth in female-headed households have more respect for women in general or receive less information supporting rape.
Three variables that were nonsignificant in the pretest model became significant in the posttest model: experimental-or controlgroup membership, grade level in school, and age of student. Most important, the results for group membership indicate that students in the experimental and control groups did not differ in their adherence to rape myths prior to the experimental group's exposure to the rape-awareness programs, but participation in the rape-awareness program led members of the experimental group to perform significantly better on the posttest compared to members of the control group.
To check the findings, the mean test scores for all four groups were calculated, as were f tests, to determine whether significant differences occurred between the pre- and posttests of the control and experimental groups. As expected, no significant change occurred in the control group's mean from the pretest (X = 17.509) to the posttest (X = 17.774), t = --0.255. The change in the group mean from the experimental group's pretest (X = 17.774) to the posttest (X = 19.832), however, was significant (t = -5.582, p < .0001), providing further support for the effectiveness of the rapeawareness program. (These calculations are not in tables.)
Finally, analyses were conducted to determine changes between the pre-and posttests on individual items for the experimental and control groups. The percentage for each true-false was calculated in 2 x 2 frequency tables. If the chi-square value comparing the control and treatment groups for the posttest scores presented for each item in Table 2 was 3.841 or greater, the difference for that particular item was significant (p<.05). The results revealed that the experimental group performed significantly better than the control group on 15 of the 24 questions (Items 1 to 5, 8 to 10, 12, 15, 17, 18, 20, 21, and 23). Conversely, the control group performed significantly better on one question (Item 6) and no significant differences occurred between the two groups for the remaining items (7, 13, 14, 16, 19, 22, and 24).[sup3] This information provides fairly powerful evidence of a significant effect for the rape-awareness program, but also shows that the program was more successful in some areas than others.
In the posttest analysis, the higher the grade level, the less likely a student was to report adherence to rape myths. The coefficient for the age of the student, however, is negative, implying that the younger the student, the more likely it is that she or he answered the rape myth item correctly. The reason for this seemingly contradictory finding is probably because of the nature of the analysis that controls for grade level and age simultaneously. Thus, within each grade, the older students are less likely to answer the rape myth items correctly. Older students in each dass are likely to be those who have failed one or more grades, are not good students, and thus respond less favorably to education and do not perform well in testing. This is a reasonable explanation for the finding related to age.
The adjusted R[sup2] increased from 0.200 in the pretest multivariate analysis to 0.285 in the posttest multivariate analysis. This suggests that rape education, as stated before, is an important piece of the puzzle explaining students' accuracy of information on rape. However, it also suggests that a significant portion of rape awareness is explained by variables not included in the model.
Although the study findings suggest that the rape-awareness program works, it is not clear for how long, and how much the posttest responses may have been due to participants' desires to give the "right" answer. It is possible that the educational advantages of the program wore off after 2 weeks, 6 months, or some other time period. Furthermore, it is difficult to assess if, and to what degree, students may have written down responses they believed were right, rather than what they actually believed. Even if some students lied about their rape attitudes on the posttest, this may be an important step in the right direction regarding how they perceive the justifiability of rape through the the realization that their attitudes may not be held by others.
The most important finding in this study of high school students is that participation in a rape-awareness program has a significant impact on decreasing students' adherence to rape myths. These findings are consistent with other studies on the effectiveness of such programs in colleges and high schools (Feltey et al., 1991; Fonow et al., 1992; Holcomb et al., 1993). Whether a student is from a two-parent home was a significant predictor of pretest performance but was not a significant predictor of posttest performance. This suggests that rape-awareness programs may be more important than what students learn about rape from their parents or guardians. Race and gender were consistent predictors of adherence to rape myths both before and after the rape-awareness programs. Furthermore, it is necessary to control for the interaction between race and gender to understand their effects. Grade level in school and age were only significant in the posttest multivariate analysis, suggesting the educability of students. In short, the findings from this study provide strong support for the inclusion of rape-awareness programs in high schools.
[sup1.] The pretest group comparison is not included in the tables. Anyone interested in a copy of this analysis may write to the second author.
[sup2.] A series of dummy variables reflecting interactions between a student's race and gender were included as independent variables in the multivariate models (see Table 1). This was done to examine whether particular race/gender groups performed better than others on the survey items tapping rape myths, as suggested by previous research. The race/gender interaction analysis included here dichotomized race into African American and Anglo. Students of other races were excluded from this analysis because they were a varied group and made up such a small part of the sample. Analyses that coded race as African Americans and "all others" had similar findings to those reported in the tables.
[sup3.] The chi-square values are not reported in Table 2.
A=Measurement B=Frequencies for Pretest C=Frequencies for Posttest TABLE 1 Variables and Descriptives Variable A B C Dependent Number of test questions answered Range: 5-24 M = 17.8 M = 19.0 correctly SD = 3.3 SD = 3.9 Independent Whether student underwent dass 1 = no 161 164 2 = yes 256 256 Controls School B student 1 = no 382 382 2 = yes 35 38 Grade level in school 1 = freshman 329 328 2 = sophomore 39 47 3 = junior 25 19 4 = senior 24 26 African American male 1 = no 339 342 2 = yes 78 78 African American female 1 = no 348 340 2 = yes 69 80 Anglo American male 1 = no 296 305 2 = yes 121 115 Age of student (in years) Range: 13-19 M = 14.8 M = 14.9 SD = 1.1 SD = 1.1 Type of household 1 = Two parents 256 245 2 = All else 161 175 Dating experience 1 = None 67 67 2 = Some 350 353 N 417 420 NOTE: Descriptive statistics were calculated based on the samples examined in the multivariate analysis. A=All Pretests B=Control Posttest C=Experimental D=(N-174) E=Posttest (N=263) TABLE 2 Percentage of Correct Responses on Rape Myth Items A B C Item (N=429) D E 1. Most rapists commit rape for sex. 26.8 21.8 55.9 2. Rape can happen to anyone. 97.0 94.3 98.1 3. Physically forcing someone to have sex is rape. 94.2 91.4 97.0 4. Women provoke and invite rape by their appearance and behavior 65.7 63.8 79.8 5. Most rapists choose someone they know as a victim. 44.3 47.1 74.1 6. An unwilling victim could prevent rape if she or he wanted to. 58.7 71.8 61.6 7. Rape is prevalent among high school and college students. 77.2 75.9 81.7 8. If a woman says no to sexual advances she really means yes or maybe. 91.1 86.2 94.3 9. A victim who did not fight back was not raped. 77.9 74.1 85.6 10. Having sex with a woman after deliberately forcing or tricking her into taking drugs or alcohol is rape. 74.6 73.0 86.3 11. A good way to be aware of possible violent situations is to listen to and trust your feelings. 92.1 87.4 92.8 12. The penalty for rape is second to murder as the most serious crime in the state of Ohio. 78.6 79.9 90.9 13. Most women who say they have been raped usually have had reputations and are promiscuous. 74.6 73.6 79.1 14. A woman can he raped by someone she willingly had sex with before. 91.4 89.7 92.4 15. If a woman gets drunk, it's her fault if she's raped. 76.9 71.8 85.2 16. If a woman lets a man buy her dinner or pay for movies, she owes him sex. 88.6 84.5 90.5 17. Rape is an act intended to control, humiliate, dominate and degrade the victim. 74.1 70.1 83.7 18. Most rapists are crazy or psychologically sick. 47.1 47.1 57.0 19. A woman can be raped by her husband. 90.4 83.9 90.1 20. Intimidating someone into having sex is rape. 57.3 66.1 78.3 21. Most women "cry rape" to get hack at men. 58.3 58.6 71.1 22. A person who ropes while under the influence of alcohol or drags is responsible for committing this crime. 80.4 80.5 85.2 23. Rape is never the victim's fault. 57.8 50.0 73.4 24. One way men can avoid being accused of rape is if they have sex only with partners who have clearly and willingly given consent. 88.3 87.4 87.8 NOTE: The following items should be marked true: 2, 3, 5, 7, 10, 11, 10, 11, 12, 14, 17,19, 20, 22, 23, 24. The remaining items should be marked false. TABLE 3 Model Predicting Test Performance (ordinary least-squares coefficients reported) Pretest Groups Variable B SE B Constant 28.916* 3.268 Whether student underwent class 0.288 0.307 School B student 0.546 0.585 Grade level 0.170 0.301 African American male -3.824* 0.461 African American female -1.477* 0.464 Anglo American male -2.261* 0.371 Age of student (in years) -0.199 0:239 Type of household -0.748* 0.341 Dating experience 0.404 0.402 NOTE: Please also refer to Table 1. Model F = 12.61 at .001 level of significane. Multiple R2 = 0.467; adjusted R2 = 0.200. *Significant at the 0.05 level. TABLE 4 Model Predicting Test Performance (ordinary least-squares coefficients reported) Posttest Groups Variable B SE B Constant 34.909* 3.526 Whether student underwent class 1.736** 0.351 School B student -0.685 0.644 Grade level 0.900* 0.321 African American male -4.108* 02,31 African American female -1.707* 0.524 Anglo American male -2.172* 0.422 Age of student (in years) -0.633* 0.251 Type of household -0.732 0.400 Dating experience 0.526 0.455 NOTE: Please also refer to Table 1. Model F = 18.18 at .001 level of significance. Multiple R2 = 0.5,34; adjusted R2 = 0.285. The changes in both the multiple and adjusted R2s from the model in Table 3 to Table 4 were significant at .001. *Significant at the 0.05 level. **Significant at the .001 level.
Barnett, N.J., & Feild, H. S. (1977, March). Gender differences in university students' attitudes toward rape. Journal of College Student Personnel, 93-96.
Belknap, J., & Erez, E. (1995). The victimization of women on college campuses: Courtship violence, date rape, and sexual harassment. In B. S. Fisher & R.J.J. Sloan, III (Eds.), Campus crime (pp. 156-178). Springfield, IL: Charles C Thomas.
Blumberg, M. L., & Lester, D. (1991). High school and college students' attitudes toward rape. Adolescence, 26, 727-729.
Burt, M. (1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology, 28, 217-230.
Burr, R. R., & Albin, R. S. (1981). Rape myths, rape definitions, and probability of conviction. Journal of Applied Social Psychology, 11, 212-230.
Carmines, E.G., & Zeller, R. A. (1979). Reliability and validity assessment. Beverly Hills, CA: Sage.
Check, J.V.E, & Malamuth, N.M. (1983). Gender role stereotyping and reactions to depictions of stranger versus acquaintance rape. Journal of Personality and Social Psychology, 45, 344-356.
Costin, E (1985). Beliefs about rape and women's social roles. Archives of Sexual Behavior, 14, 319-325.
Davis, T. C., Peck, G. Q., & Storment, J. M. (1993). Acquaintance rape and the high school student. Journal of Adolescent Health, 14, 220-224.
Deitz, S. R., Littman, M., & Bentley, B. J. (1984). Attribution of responsibility for rape: The influence of observer empathy, victim resistance, and victim attractiveness. Gender< Roles, 10, 261-280.
Dull, R. T., & Giacopassi, D. J. (1987). Demographic correlates of sexual and dating attitudes: A study of date rape. Criminal Justice and Behalf, 14,175-193.
Erickson, P. I., & Rapkin, A. J. (1991). Unwanted sexual experiences among middle and high school youth. Journal of Adolescent Health, 12, 319-325.
Estrich, S. (1987). Real rape. Cambridge, MA: Harvard University Press.
Feild, H. S. (1978). Attitudes toward rape: A comparative analysis of police, rapists, crisis counselors, and citizens. Journal of Personality and Social Psychology, 36,156-179.
Feldman-Summers, S., & Palmer, G. C. (1980). Rape as viewed by judges, prosecutors, and police officers. Criminal Justice and Behavior, 7,19-40.
Feltey, K. M., Ainslie, J. J., & Geib, A. (1991). Sexual coercion attitudes among high school students. Youth & Society, 23, 229-250.
Fenstermaker, S. (1989). Acquaintance rape on campus: Responsibility and attributions of crime. In M. A. Pirog-Good & J. E. Stets (Eds.), Violence and dating relationships: Emerging social issues (pp. 257-272). New York: Praeger.
Fischer, G. J. (1986). College student attitudes toward forcible date rape. Archives of Sexual< Behavior, 15, 457-466.
Fischer, G. J. (1987). Hispanic majority student attitudes toward forcible date rape as a function of differences in attitudes about women. Gender Roles, 17, 93-101.
Fonow, M. M., Richardson, L., & Wemmerus, V. A. (1992). Feminist rape education: Does it work? Gender & Society, 6, 108-121.
Garrett-Gooding, J., & Senter, R. (1987). Attitudes and acts of sexual aggression on a university campus. Sociological Inquiry, 57, 348-371.
Giacopassi, D. J., & Dull, R. T. (1986). Gender and racial differences in the acceptance of rape myths within a college population. Gender Roles, 15, 63-75.
Gilmartin-Zena, E (1987). Attitudes toward rape: Student characteristics as predictors. Free< Inquiry in Creative Sociology, 15,175-182.
Gilmartin-Zena, P. (1988). Gender differences in students' attitudes toward rape. Sociological Focus, 21, 279-292.
Goodchilds, J. D., & Zellman, G. L. (1984). Sexual signaling and sexual aggression in adolescent relationships. In N.M. Malamuth & E. Donnerstein (Eds.), Pornography and sexual aggression (pp. 233-243). Orlando, FL: Academic Press.
Holcomb, D. R., Holcomb, L. C., Sondag, K., & Williams, N. (1991). Attitudes about date rape: Gender differences among college students. College Student Journal, 25, 434-439.
Holcomb, D. R., Sarvela, ED., Sondag, K. A., & Holcomb, L.C.H. (1993). An evaluation of a mixed-gender date rape prevention workshop. Journal of American College Health, 41, 159-164.
Hueston, H. R., & Burkhart, B. (1986). Acquaintance rape: Developing a comprehensive prevention program. Campus Law Enforcement Journal, 31, 30-33.
Kanekar, S., & Kolsawalla, M. B. (1980). Responsibility of a rape victim in relation to her respectability, attractiveness, and provocativeness. Journal of Social Psychology, 112, 153-154.
LaFree, G. D., Reskin, B. E, & Visher, C. A. (1985). Jurors' responses to victims' behavior and legal issues in sexual assault trials. Social Problems, 32, 389-407.
Marx, B., & Gross, A.M. (1995). A male's conception of date rape: An analysis of two contextual variables. Behavior Modification, 19, 451-463.
Muehlenhard, C. L. (1989). Misinterpreted dating behaviors and the risk of rape. In M. A. Pirog-Good and J. E. Stets (Eds.), Violence and dating relationships: Emerging social issues (pp. 241-256). New York: Praeger.
Muehlenhard, C. L., Friedman, D. E., & Thomas, C. M. (1985). Is date rape justifiable? The effects of dating activity, who initiated, who paid, and men's attitudes toward women. Psychology of Women Quarterly, 43, 86-196.
Muehlenhard, C. L, & Linton, M. A. (1987). Date rape and sexual aggression in dating situations: Incidence and risk factors. Journal of Counseling Psychology, 34,186-196.
Parrot, A. (1986). Acquaintance rape and sexual assault prevention training manual (2rid ed.). Ithaca, NY: Department of Human Service Studies, Cornell University.
Parrot, A. (1989). Acquaintance rape among adolescents: Identifying risk groups and intervention strategies. Journal of Social Work and Human Sexuality, 8, 47-61.
Rapaport, K., & Burkhart, B. R. (1984). Personality and attitudinal characteristics of sexually coercive college males. Journal of Abnormal Psychology, 93, 216-221.
Sandberg, G., Jackson, T. L., & Petretic-Jackson, E (1987). College students' attitudes regarding sexual coercion and aggresion: Developing educational and preventive strategies. Journal of College Student Personnel, 28, 302-311.
Shotland, R. L., & Goodstein, L. (1983). Just because she doesn't want to doesn't mean it's rape: An experimentally based causal model of the perception of rape in a dating situation. Social Psychology Quarterly, 46, 220-232.
U.S. Bureau of the Census. (1992). U.S. census of population and housing. Washington, DC: Goverment Printing Office.
Van Wie, V. E., Gross, A.M., & Marx, B. E (1995). Females' perception of date rape: An examination of two contextual variables. Violence Against Women, 1, 351-365.
Warshaw, R. (1988). I never called it rape. New York: Ms. Magazine/Sarah Lazin Books.
Zellman, G. L, & Goodchilds, J. D. (1983). Becoming sexual in adolescence. In E. Allgeier & N. McCormick (Eds.), Changing boundaries: Gender roles and sexual behavior (pp. 49-63). Palo Alto, CA: May field.
by LAURA PROTO-CAMPISE, JOANNE BELKNAP, JOHN WOOLDREDGE
Laura Proto-Campise is the coordinator of the Sexual Assault-Domestic Violence Program at the Ohio State University Medical Center. This program is situated in the emergency department and provides crisis intervention and emotional support to sexual assault and domestic violence survivors. She received her M.A. degree in women's studies from the University of Cincinnati in 1994
Joanne Belknap is an associate professor of criminal justice and women's studies at the University of Cincinnati. Her major area of research is feminist criminology. -She has conducted research on women and girls as victims and offenders and women working in the crime-processing system. She recently published a book on this topic, titled The Invisible Woman: Gender, Crime, and Justice
John Wooldredge is an associate professor of criminal justice at the University of