(.9 M'SU' RETURNING MATERIALS: P1ace in book drop to ‘0‘ 91:" Is (s. "it ‘I ~ ‘ iii, 5-2:. it 'r “ ugh/tugs remove this checkout from .—c—. your record. FINES-W1.” be charged if book is returned after the date stamped be10w. / "V" .e‘ . v '« ,1 I 2 87% J/ ‘" KC? 2384?: eff; :“ fl 1 ’71529"3 .59 H} . 2 iuAY u 5"’“ "15;! CA n (J SEXUAL HARASSMENT OF UNIVERSITY STUDENTS: STUDENTS' PERCEPTIONS AND RESPONSES BY Mary Sullivan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1985 ABSTRACT SEXUAL HARASSMENT OF UNIVERSITY STUDENTS: STUDENTS' PERCEPTIONS AND RESPONSES BY Mary Sullivan This study addressed undergraduate female students' percep- tions of and responses to sexual harassment in a university setting. Predictors of and attitudes toward reporting procedures were also examined. Two hundred nineteen parti- cipants from Michigan State University were randomly assigned to receive descriptions of a harassment incident in which severity of harassment (mild, moderate, or severe) and status of offender (professor, advisor, teaching assis- tant) were manipulated. Findings indicated that reporting sexual harassment was predicted by severity of the harass- ment, fear of being accused of lying, perceived effective- ness of reporting, and fear of the reporting procedure itself. In addition, participants believed that victims were more likely to report harassment to someone outside of the offender's department and were more likely to report to a woman than to a man. To my parents, Teri Fournier Sullivan and John L. Sullivan ii ACKNOWLEDGMENTS First and foremost I would like to thank Deb Bybee, whose statisticalexpertise,cmeative brainstorming, and mastery of methodological research have greatly contributed not only to this thesis but to my abilities as a researcher. She has truly been mentor, consultant, and friend. Special thanks to my chairperson, Robin Redner--the driving force behind this research. I also extend my appre- ciation to my other committee members, Anne Bogat and Esther Fergus, for their sound advice and unique contri- butions to this study. I would like to thank the women who participated in this study, each of whom contributed to the better under- standing of sexual harassment in university settings. Many thanks to the Feminist Volleyball Team members, past and present, who shared with me their support, love, and laughter. They miraculously helped me maintain my purpose, sanity, and sense of humor through this entire process. To Patti Witte, my never-ending appreciation and love for continually trusting my abilities and encouraging my dreams. iii Finally, I thank my parents, Teri and John Sullivan, whose love, encouragement, and support made it all possible. iv TABLE OF CONTENTS LIST OF TABLES CHAPTER 1 INTRODUCTION What Is Sexual Harassment? Effects of Sexual Harassment Immediate Effects Long-term Effects Physical and Emotional Effects Importance of Reporting Sexual Harassment Institutional Responses to Sexual Harassment Two Types of Grievance Procedures Informal Grievance Procedure Formal Grievance Procedure Summary The Current Research Research Questions and Hypotheses Exclusion of Male Subjects CHAPTER 2 METHOD Sample Design Independent Variables Experimental Booklet Dependent Variables Page viii 10 10 11 11 14 16 18 19 19 20 20 22 23 Procedure Pilot Study Training Research Assistants Experimental Sessions CHAPTER 3 RESULTS Characteristics of the Sample Descriptive Tabulations Perceptions of Sexual Harassment Manipulation Check (Pilot Study) Manipulation Check (Final Study) Major Findings Reporting as a Function of Severity Predictors of Reporting To Whom Students Report- Summary of Major Findings CHAPTER 4 DISCUSSION Major Findings Reporting as a Function of Severity Predictors of Reporting To Whom Students Report Methodological Issues Sample Advisor as Offender Predictor Variables vi Page 24 24 25 25 28 28 29 29 30 35 39 39 44 51 61 63 63 63 63 66 66 66 68 69 Implications of the Research Necessity of Grievance Procedures Future Research APPENDICES Appendix Appendix Appendix Appendix Appendix Appendix REFERENCES Nine Experimental Scenarios Sexual Harassment Questionnaire Pilot Study Questions Research Assistant's Instructions Debriefing Sheet Proctor Sheet Vii Page 70 70 73 75 75 84 101 103 105 106 107 Table 10 11 12 13 14 LIST OF TABLES Definitions of Sexual Harassment as Found by Adams et a1. (1983) 3 x 3 Factorial Design Analysis of Variance Summary Table for Perceived Level of Harassment (Pilot Study) Cell Means for Perceived Level of Harassment (Pilot Study) Analysis of Variance Summary Table for Perceived Power of Harasser (Pilot Study) Cell Means for Perceived Power of Harasser (Pilot Study) Cell Means for the Believability of Scenarios Analysis of Variance Summary Table for Perceived Level of Harassment (Final Study) Cell Means for Perceived Level of Harassment (Final Study) Cell Means for Perceived Power of Harasser (Final Study) Internal Consistency Analysis of Likelihood of Reporting Scale Analysis of Variance Summary Table for Likelihood of Reporting by Victim Cell Means for Likelihood of Reporting by Victim Internal Consistency Analysis of Effectiveness of Reporting Scale viii 15 21 31 32 33 34 36 37 38 40 41 42 43 45 Table 15 16 17 18 19 20 21 22 23 Internal Consistency Analysis Reporting Scale Correlations between Scales Step-Wise Regression Analysis of Reporting Internal Consistency Analysis a Woman Scale Internal Consistency Analysis a Man Scale Internal Consistency Analysis Either Sex Scale Internal Consistency Analysis Within the Department Scale of on of of of of Fear of Predictors Report to Report to Report to Report Internal Consistency Analysis of Report Outside of the Department Scale Internal Consistency Analysis or Out Scale ix of Report In Page 47 48 50 52 53 55 56 57 58 CHAPTER 1 INTRODUCTION Although sexual harassment is not a new problem, research conducted on the issue before 1976 is practically nonexistent (Sandler et al., 1981). Since that time, most research which has documented the occurrence of sexual harassment has focused on the workplace (MacKinnon, 1979; Safran, 1976; U.S. Merit Systems Protection Board, 1981). However, sexual harassment of students in the university setting is a severe and widespread problem (Washington Egg; estimates, 1981). In 1978, a study at California State Uni— versity at Long Beach also found that students were being subjected to sexual harassment in the university setting. That study found that 37% of the student sample there had experienced some form of sexual harassment--harassment being defined as the unwanted imposition of sexual requirements in the context of a relationship of unequal power. Eighty per- cent of the victims were women (Reilly, 1978). Additionally, Benson and Thomson (1980) found that, in a random sample of the senior female students at the University of California, Berkeley, 19.7% were victims of sexual harassment. In that study, the definition of harassment was borrowed from the Working Women United Institute (WWUI, 1978): "Any unwanted sexual leers, suggestions, comments or physical contact which you find objectionable in the context of a teacher- 1 2 student relationship." These statistics were corroborated when a study at Michigan State University reported that 25.1% of a stratified random sample of 998 female students had experienced at least one incident of sexual harassment (Maihoff and Forrest, 1983). In this study, students were asked questions about experiences ranging from jokes about women's anatomy to sexual assault. The above-mentioned examples are but a sampling of reports, surveys, and studies which corroborate the gravity and enormity of the problem of sexual harassment in the uni- versity setting (Adams et al., 1983; Benson, 1977; Munich, 1978; Nelson, 1978; Till, 1980). Unfortunately, many people--including many university policy makers--view sexual harassment as an uncommon occurrence because of the low rate of reported incidents (Sandler et al., 1981). However, studies which have investigated the number of harassees who have filed complaints have found a large discrepancy between the frequency of occurrence and the number of complaints lodged. For example, in a study consisting of a stratified, representative sample of 250 university women, only 20% of the harassees had tried to lodge a complaint (Metha & Niggs, 1983). In addition, another survey sent out to a dispropor- tionately stratified sample of 1,000 male and female students found that no one who had experienced an incident of harass- ment had filed a complaint about it (Adams et al., 1983). In that study, however, the researchers received a response rate of only 37%--a rate too low to be generalizable. 3 However, other research is in accordance with these results, finding that students felt they would be misperceived or not believed, that they would be accused of somehow inviting the harassment, or that retaliation would occur if they reported it (Meek & Lynch, 1983; Sandler et al., 1981). Therefore, evidence suggests that a lack of complaints does not mean the lack of a problem; rather, students feel that reporting harassment would not be the most viable solution to their dilemma. In addition to examining the various definitions of sexual harassment, this introduction will review the follow- ing three aspects of the issue: (a) the physiological, emotional, and psychological effects of this phenomenon on its victims and on women in general; (b) the reasons so many women choose not to report harassment; (c) the institutional procedures designed to deal with this issue. This review of the literature will then be followed by the rationale for conducting a research study on students' perceptions of and responses to sexual harassment in the academic setting. What Is Sexual Harassment? In its broadest sense, sexual harassment is "the unwanted imposition of sexual requirements in the context of a relationship of unequal power" (MacKinnon, 1979). While this definition is certainly accurate when speaking of the more severe levels of harassment (i.e.,b1atant threats, coerced relations), it does not address sexist 4 jokes or lewd remarks, which serve to reinforce the status of women as primarily sexual stimuli for the harasser. A more all-encompassing definition would be as follows: behavior ranging from "sexual innuendos made at inappro- priate times, perhaps in the guise of humor, to coerced sexual relations" (Sandler et al., 1981). Sexual harassment has further been referred to as sexual attention imposed on someone who is not in a posi- tion to refuse that attention (Mac Kinnon, 1979). Along this same line, a more sociological approach to defining sexual harassment is "unsolicited nonreciprocal male beha- vior that asserts a woman's sex role over her function as a worker. It can be any or all of the following: staring at, commenting upon, or touching a woman's body; requests for acquiescence in sexual behavior; repeated nonrecipro- cated propositions for dates; demands for sexual inter- course; and rape" (Farley, 1978) . This definition clearly identifies the harasser as male and the harassee as female, in contrast to many definitions which do not. Mac Kinnon (1979) also supports this view, arguing that the "severe disproportion test" maintains that if the vast majority of harassees are women, it is "customary gender-specific treatment sexually to harass women." In addition, there is little, if any, empirical evidence to show that female harassment of male students, male harassment of male students, or female harassment of female students are any more than extremely isolated incidents (Sandler et al., 5 1981). Sexual harassment, like rape, domestic assault, and incest, is overwhelmingly perpetuated by men against women (Armstrong, 1978; Brownmiller, 1975; Martin, 1981). Effects of Sexual Harassment Immediate Effects. One immediate effect of sexual harassment is that the victim is either implicitly or expli- citly threatened with the loss of academic/economic success and livelihood should she refuse an advance of a sexual nature. For a woman to refuse to acquiesce to the sexual attention of someone who has control over her grades and/or letters of recommendation, is to possibly risk her entire academic future (Sandler et al., 1981). To elaborate, a letter of recommendation could mean the difference between obtaining and losing an academic or work position. A grade in a required course could mean the difference between graduating or not. Many students perceive that to tell their professor, department chair, teaching assistant, or advisor that he is offensive and that his behavior is unac- ceptable is for that student to be putting her entire career in jeopardy. For this reason, feminists conclude that a sexual advance of ggy kind in an academic setting in which a man grades a woman constitutes sexual harassment (Munich, 1978) . Longeterm effects. Another detrimental effect of sexual harassment is that victims tend to avoid situations which may involve being in contact with the harasser. Out of a sample of eighty women who had been sexually harassed, 6 22 had no further contact with the perpetrator. Twenty-four practiced some form of avoidance, and 17 became disillu- sioned with male faculty in general (Benson & Thomson, 1982). While avoiding and mistrusting male faculty may be an understandable response to harassment, this reaction may lead to women being afforded less mentoring opportunities than their male counterparts, as well as fewer opportunities to obtain necessary recommendations or financial support. In the past as well as today, few women have been successful in entering male-dominated fields (Epstein, 1970). For example, 95% of full university professors are male (Patterson & Engelberg, 1978). Social psychological analy- ses report the numerous difficulties women have in seeking careers heretofore dominated by men (Mednich et al., 1975); women's avoidance of certain male professors who have the power to help them attain promotions and positions heightens their disadvantage. In addition, an indirect consequence of sexual harass- ment is that some male instructors, fearful of the sexual implications arising from friendships with female students, do not afford women the same friendships/mentoring experi- ences with which they supply their male students (Sandler et al., 1981). Mentoring has been found to be an important aspect of the educational and career development process by various researchers (Hall & Sandler, 1983; Merriam, 1983). Finally, some women attempt to end sexual harassment by claiming to have other emotional/sexual ties (i.e., "I 7 have a boyfriend who gets insanely jealous"). While this may serve to end the harassment--though it may not--it also perpetuates the assumption that women are defined by their relationships with men and are "owned" by men (Benson & Thomson, 1982). Hence, actions utilized to end sexual harassment on a wide scale need to assert women's basic rights to be treated with respect. Physical and emotional effects. There are also detri- mental physical as well as emotional effects on victims of sexual harassment. These include headaches, insomnia, reduced ambition and self-esteem, diminished concentration, severe depression, and a sense of helplessness (Benson & Thomson, 1982; Sandler et al., 1981). Importance of Reporting Sexual Harassment There are both individual and societal reasons support- ing the importance of reporting sexual harassment. On an individual level, it has been found that students who com- plain to appropriate channels or to the harasser about being harassed are more effective in ending the incidents than those who do not. Sandler (1981) found that 75% of the women in her study who ignored the offensive behavior reported that the harassment continued or worsened. Benson and Thomson (1982) reported that in 86% of the cases where the behavior was ignored, the harassment persisted. The respondents who gig communicate to the perpetrator in some way that his attention was unwanted were much more success— ful in ending it. 8 On a larger scale, many university officials believe that lack of complaints indicates the lack of a problem (Sandler et al., 1981). Time and energy are not likely to be put into programs dealing with sexual harassment if officials believe this problem to be minimal or nonexistent. Additionally, if students are aware that they are not the only ones dealing with being harassed, they may be more likely to step forward with complaints and not blame them- selves for the occurrence. Institutional Responses to Sexual Harassment As with rape and domestic assault (Rose, 1977; Schechtman, 1983), the exposure of sexual harassment as a widespread, serious problem has been largely due to the efforts of feminists (Benson & Thomson, 1980). Victims have been encouraged by feminists to report harassment, and institutions have been pressured to institute adequate grievance procedures. Unfortunately, it wasn't until some women at Yale sued that university for not having any means of dealing with reports of harassment (Alexander v. Yale, 1980) that institutions were forced to deal with this issue. The decision in this case led to the 1981 memorandum from the Office of Civil Rights of the U.S. Department of Educa* tion, which stated that sexual harassment of students is a violation of Title IX of the 1972 Educational Amendments, leaving universities liable if they do not have adequate grievance procedures to handle students' complaints (Adams et al., 1983; Perry, 1983). 9 Although universities have great latitude in deciding how to deal with reports of sexual harassment, the U.S. Merit Systems Protection Board (1981), as well as numerous other sources (Deane & Tillar, 1981; Sandler et al., 1981; Till, 1980), recommend a memorandum from the university president condemning sexual harassment on campus. They believe this disseminates a university attitude of support for students' rights as well as permitting harassees to hold harassers accountable for their actions. A second recommendation of the U.S. Merit Systems Protection Board was for the development of techniques and materials designed to inform students of their rights. These materials were also to warn potential perpetrators of the consequences of sexually harassing students (Deane & Tillar, 1981; Sandler et al., 1981; U.S. Merit Systems Protection Board, 1981). These materials could consist of in-service training sessions, pamphlets, slide shows, films, and/or reports. Most importantly, it is essential that the academic institution have the means to adequately deal with com- plaints when they arise. Deane and Tillar (1981), Sandler et al., (1981), and Till (1980) all stress that one main reason victims of sexual harassment do not report incidents is that they believe the institution will do nothing about it. Students must not only be reassured that an adequate procedure is available to them--they must have this proce- dure demonstrated to them should the need arise. In order 10 to ensure this, monitoring systems must be designed to evaluate this procedure at all levels (Sandler et al., 1981; U.S. Merit Systems Protection Board, 1981). In summary, it has been the recommendation of numerous sources (Deane & Tillar, 1981; Sandler et al., 1981; Till, 1981; U.S. Merit Systems Protection Board, 1981) that the following procedures be instituted at universities to ensure adequate means of dealing with sexual harassment. First, a memorandum from the university president condemning sexual harassment should be distributed so that the university's stance is clearly understood. Second, materials should be designed to educate people with regard to their rights, definitions of sexual harassment, and procedures for dealing with complaints. Third, adequate procedures must be devel* oped to deal with complaints; that is, specific offices and administrators whose duty it is to see that justice is done must be identified. Fourth, a system to monitor and evalu- ate this procedure at all levels needs to be implemented. Two Types of Grievance Procedures Informal grievance procedure. In an anecdotal report, Meek and Lynch (1983) recommended the availability of an informal grievance procedure for students. In response to survey results at the University of Florida, an informal procedure was designed and instituted. Since then, more students have reported sexual harassment instances at this university. 11 Meek and Lynch recommend that certain specific profes- sionals in student affairs be responsible for dealing with complaints of sexual harassment. They believe this is less threatening for students as the setting is more neutral than speaking with someone in the offender's department. They also strongly suggest that these professionals be women, under the assumption that female students would be more comfortable talking with women about incidents of sexual harassment. The experiences of the professionals at the University of Florida support the contention that many students simply want to make their experience known and to ask someone to force the perpetrator to stop his offensive behavior. Most do not want to file a formal complaint. Therefore it is important that students have this informal channel available to them. Formal grievance procedure. For those students who do wish to file a formal complaint, they can do so through the appropriate channels at their university. This typically consists of filing written charges against the person and requesting an investigation. If the investigation finds the allegations to be true, a hearing is held and proper disciplinary measures are taken against the offender. Summary Sexual harassment is not only humiliating and offensive to the victim at the time it is occurring, it has long-range consequences as well. Many women complain of insomnia, 12 headaches, and other physiological ailments as a result of these incidents. Also, many victims avoid further contact with the perpetrator as well as other male faculty, possibly jeopardizing their academic futures. Researchers in the field have found that women who communicate to the harasser in some way that they want the attention stopped are more successful in ending the harass- ment than are women who try to ignore the behavior. How- ever, many students are hesitant to report harassment because they fear they will not be believed, that nothing will be done, or that the situation will worsen. Conse- quently, Meek and Lynch (1983) argue for the implementation of an informal grievance procedure to handle complaints of sexual harassment. This allows students to anonymously talk to someone who is sympathetic and who has the authority to bring the matter to the attention of the harasser and/or his supervisor. In summary, sexual harassment is a widespread problem with serious consequences for the victim and for women in general. Existingrcsearch has looked at the occurrence rate of sexual harassment, the effects of sexual harassment, and the reasons why students do not report incidences of sexual harassment. These studies have been instrumental in understanding this phenomenon and are helpful in guiding the course of further research. However, they raise innu- merable issues surrounding sexual harassment which need to be examined or clarified. There are four major issues 13 which have not as yet been dealt with by research and which are essential to address at this time. For example, many researchers contend that one reason research findings of occurrence rates differ between studies is that researchers use different criteria when identifying sexual harassment (Maihoff & Forrest, 1983; Till, 1980). Students' percep- tions of what constitutes sexual harassment needs to be examined. Second, no experimental research has been con- ducted to show that informal grievance procedures are indeed more desirable or effective in handling sexual harassment complaints. This hypothesis needs to be tested before one can make a case for implementing informal grievance proce- dures at universities. Third, it has yet to be seen if the status of the offender or students' perceptions of the negative consequences of the incident influence students' decision to report. For instance, are students more likely to report a teaching assistant than they are a professor? And fourth, is there an interaction between the status of the offender and the seriousness of the harassment regarding the likelihood of reporting or the negative consequences of the incident? Current research seems to support the hypo- thesis that students are more likely to report serious offenses of sexual harassment (Till, 1981; U.S. Merit Systems Protection Board, 1981). However, does the status of the harasser effect this in any way? The current research was designed to address these four major issues. 14 The Current Research The purpose of this research was to examine various issues which have not as yet been addressed by the litera- ture. First, defining behaviors which constitute varying forms of sexual harassment was seen as an essential step in sexual harassment research. For instance, Maihoff and Forrest (1983) included sexist comments as a form of sexual harassment in their study. However, Adams et al. (1983) found that less than half of the respondents of their study considered sexist comments or undue attention (flirting, being too personal) to be instances of sexual harassment. Therefore, in the present research severity of harassment was manipulated to ascertain subjects' responses to varying degrees of harassment and to examine the effects of severity on reporting behavior. Severity categories were adapted from previously rationally categorized behaviors (Adams et al., 1983). See Table 1. Adams et a1. (1983) also found that harassers are most likely to be professors or teachers (50%), followed by graduate teaching assistants (25%), and academic advisors (17%). Hence, status of offender (professor, teaching assistant, advisor) in the present study was manipulated in order to discover any power differential which may affect students' perceptions of and/or responses to incidents of sexual harassment. In addition, this research asked subjects how likely they thought ftwas for a victim of sexual harassment to 15 wumnfiun Hmsxmm mmocm>©m Hmowmhnm mcoflufimomoum Davaamxm mwocm>om Hmsxom Hmnuo> mmmsmcmH hoom mcoflumuH>GH mucoEEoo umflxom coflucouum maps: ucofimmmumn Hosxwm ma whomouoo ma ousmcs moamfimm mo w ucmfimmmumc Hmsxom mm whommumo mcacflmmo mmHmEom mo w H0H>mnon mo wuowmumu Ammmav .Hm no mEMU< an pcsom mm usoEmmmumm Hmsxom mo mcofluflawwoo H wanna 16 report the harassment to various persons. These people included female faculty, male faculty, female professionals in charge of sexual harassment complaints, male profes- sionals in the same area, male as well as female supervisors of the offender, and department chairwomen and chairmen. These questions were designed to test Meek and Lynch's hypotheses that students are more likely to tell women than men and that they are more likely to tell someone outside of the harasser's department (1983). Finally, the present research study allowed for the examination of the interaction between levels of severity and type of harasser. Research questions and hypotheses. In their studies, Sandler et a1. (1981) as well as Meek and Lynch (1983) found that students did not report instances of sexual harassment because of fear of some type of retaliation. Another reason for low reporting rates was that students feared it would make the situation worse or, at best, that the reporting would be ineffective in stopping the harass- ment (Benson & Thomson, 1982). Therefore the following hypothesis was tested: Reporting behavior will be influ- enced by its perceived effectiveness, the perceived fear of reportingL and the perceived repercussions of such reporting. Specifically, the more effective the reporting is judged to be, the more likely the student is to report. Also, the less fear the student has of reporting, the more likely she 17 will be to report, and the fewer repercussions the student sees as occurring, the more likely she will be to report. To date there have been no studies addressing whether or not the severity level of the harassment influences one's decision to report it. It is not known whether students are more likely to report mild harassment over severe harassment, moderate harassment over mild harass- ment, etc. However, documented cases suggest that students report more serious harassment (Till, 1981; U.S. Merit Systems Protection Board, 1981). To examine whether or not this is the case, the study tested the following hypothesis: Reporting behavior will be a function of the severity of the harassment, in that the more serious the harassmentL the more likely reporting will be. In an anecdotal report, Meek and Lynch (1983) argued that it was less threatening for a student to lodge a complaint with a staff person in charge of sexual harass- ment issues than someone in the offender's department. Further, they recommended that this person be a woman, as female students would be more comfortable talking with a woman about sexual harassment. Meek and Lynch based their assumptions on survey results as well as on the success of the informal grievance procedure recently implemented at the University of Florida. Their experiences suggest that sex of the person to whom students report and whether they are inside or outside of the offender's department are important. It would seem that students would be most likely 18 to report to a professional woman specifically in charge of handling sexual harassment complaints. To experimentally test these assertions, another hypothesis was tested: Students would be more likely to report harassment to a professional specifically in charge of harassment complaints than to someone in the offender's department, and they would be more likely to report to a woman than to a man. Thus, students would be most likely to report to a profes- sional staff woman in charge of harassment complaints. In addition, this research study examined how common students perceive sexual harassment on campus to be, and how serious a problem they feel it is. This information was necessary in order to assess the accuracy of students' perceptions of sexual harassment and consequently whether or not some type of educational program would be needed. Exclusion of Male Subjects Only women were chosen to participate in this research. There were three factors governing this decision: 1. Sexual harassment is a problem overwhelmingly experienced by women. 2. Men who do claim to have experienced behavior from supervisors which could be classified as sexually harassing tend to find it flattering or humerous (Adams et al., 1983). 3. So few men have reported being sexually harassed that the information gathered is not generalizable (Adams et al., 1983; Maihoff & Forrest, 1983). CHAPTER 2 METHOD Sample The sample for this study consisted of 219 undergradu- ate women from Michigan State University. During the first half of the academic term, notices were posted in introduc- tory psychology classrooms explaining when and where inter- ested students could meet with the researcher or one of her assistants to participate in the study. Students received an extra credit point in class for participating in the research (3 = 150). The notice informed potential subjects that female students were desired to answer a 15 to 30 minute survey. It did not mention that the questionnaire dealt with sexual harassment in order to decrease the risk of subjects' selecting this research on the basis of their interest (or lack thereof) in the research topic. By the middle of the term it was evident that not enough subjects would be obtained through this method. Therefore, the researcher went to lower level undergraduate humanities, social science, criminal justice, English, and natural science classes to recruit subjects. Females were asked to stay after class for 15 to 30 minutes to partici- pate in a questionnaire on women's campus experiences. None of the students in either group knew of the research 19 20 hypotheses, and these latter students did not receive extra credit for their participation (g = 69). Design The design was a 3 x 3 factorial, examining three levels of sexual harassment (mild, moderate, severe) crossed with three types of harasser (professor, advisor, teaching assistant). Subjects were presented with standardized scenarios describing one of nine sexual harassment inci- cents manipulating these two variables. See Table 2 for the design. Independent variables. For the purpose of this study, the first factor (sexual harassment) was categorized into mild, moderate, and severe forms. Mild harassment was categorized by the offender reading a pornographic magazine, staring at the woman's breasts, and having an offensive poster on his door. Moderate harassment included those behaviors which were specifically directed at the woman, but which fell short of implicit or explicit threat to the woman. These behaviors included unwanted touching, sexual remarks to the woman about her physical appearance or private life, and uninvited suggestive looks. In this scenario, the perpe- trator also touched the woman's hair. Finally, the severe level of sexual harassment was 'defined by behaviors which the victim perceived to be implicitly or explicitly threatening. The woman was also physically touched in this scenario. Table 2 21 3 x 3 Factorial Design Status of offender Teaching Severity Advisor assistant Professor Marginal Mild g = 26 g = 25 g = 22 73 Moderate g = 25 g = 24 g = 24 73 Severe g = 25 g = 24 g = 24 73 Marginal 76 73 70 N = 219 22 Therefore, for the purpose of this study the categories of sexual harassment were divided into the following behav- iors: 1. Mild: possessing materials which are offensive to women as a class and/or treating an individual woman in a sexual manner, but falling short of touching her or being verbally offensive (i.e., leering). 2. Moderate: sexual verbal and/or physical behavior directed toward an individual woman, but falling short of implicitly or explicitly threatening/propositioning her. 3. Severe: sexual behaviors directed toward an indi- vidual woman which she perceives to be implicitly or expli- citly coercive or threatening. The second factor in this design (status of offender) was categorized into professor, advisor, and teaching assistant. This categorization was adapted from Adams et al. (1983) in which it was found that 50% of harassers were professors or instructors, 25% were teaching assist- ants, and 17% were advisors. (See Appendix A for the scenarios used.) Experimental booklet. The experimental booklet which each subject received contained (a) a single experimental scenario manipulating the independent variables, (b) several dependent variable items and manipulation check items, and (c) several demographic questions. Each scenario described a situation in which a female undergraduate student had been harassed by either a male 23 professor, a male advisor, or a male teaching assistant. Each situation incorporated one of three levels of harass- ment (mild, moderate, or severe). With one exception, all other factors in the scenarios were held constant, including the victim's name that was used. The exception to this was that in the severe harassment/advisor scenario, the victim was threatened with not being able to get into a graduate school; in the other two severe harassment conditions, victims were threatened with failing their course. Dependent variables. Each scenario was followed by a number of questions pertaining to the scenario and to sexual harassment in general. Subjects were presented with Likert- type scales with values ranging from one to four. Specifi- cally, they were asked questions regarding the following: 1. The likelihood of the action occurring. 2. The severity of the action. 3. The likelihood that the victim would report the incident, given that the victim was a typical university student. 4. The likelihood of reporting if this incident happened to the subject. 5. The perceived effectiveness of the reporting procedures presented. 6. How frightening each reporting procedure seemed. 7. The likelihood of various repercussions after reporting. 24 In addition, subjects were asked a number of open- ended questions dealing with their opinions of how the inci- dent would affect the victim and what they would do if this situation happened to them. Students were also asked two multiple-choice items regarding what they thought most women would do in this situation and what they themselves would do given this situation. They were given six options from which to choose. The second section dealt with the subjects' familiarity with the rate/severity of sexual harassment on campus, know- ledge of reporting procedures on campus, acquaintance with Victims of harassment, and the subjects' own experiences with sexual harassment. These questions were open-ended in order to elicit as much information as possible from each participant. (See Appendix B for complete question— naire.) Procedure Pilot study. In order to determine the effectiveness of the manipulation, the questionnaire was administered to 81 undergraduate women at Michigan State University. The subjects were students in upper-level psychology courses. Four instructors randomly presented the question- naires to the women students, who then completed the ques- tions before class began and returned them to the instructor. Questions addressing subjects' opinions of (a) the severity level of the perceived harassment, (b) the 25 believability of the situation, and (c) how much power the harasser had compared to the victim served as manipulation checks. (See Appendix C for the pilot study questions.) Training research assistants. Sixteen research assist- ants were trained before the beginning of the study to more quickly expedite this project. The researcher recruited 16 female undergraduates to administer the questionnaire throughout the term in exchange for college credits. An initial three-hour meeting was scheduled, at which the research study was explained to them in detail. Their role in the research was explained, and they were given written instructions to take with them to each session (see Appendix D). In addition, the principal researcher also met with each research assistant weekly to review procedures and to address questions and difficulties. Experimental sessions. Subjects were randomly assigned to experimental conditions. For individual subject ses- sions, the researcher or assistant handed the subject the questionnaire, had her read the cover sheet explaining the study, sign this sheet, and hand back this sheet. The researcher placed the sheet face down on the desk and explained to the subject that her name was no longer on the booklet and that complete anonymity was guaranteed. The researcher recorded when the subject began and finished the questionnaire, and made any notation of ques- tions asked or anything unusual happening during the session. See Appendix E for the proctor sheet. 26 If the subject had a question, the researcher clarified any wording but did not address any issues of content. The subject was instructed to use her best judgment in deciding how to answer each question. It was stressed that there were no right or wrong answers and that the subject's honest opinion was desired. When the subjects finished the questionnaires, the researchers gave them a debriefing sheet (see Appendix F) to read while the researchers quickly glanced through the questionnaire, checking to see that all questions had been answered and could be understood. The debriefing sheet thanked them for their participation and stressed the impor- tance of not telling anyone what the study entailed. Each day that a research assistant administered the questionnaire, she returned the completed booklet(s) to the principal researcher. In the group sessions, the researcher and three assist- ants would go to a class taught by an instructor who agreed to allow the researcher to ask for volunteers from their class. Students were told after class that female volun- teers were needed to stay after for fifteen minutes to answer a survey about women's experiences on campus. Those who were interested in participating would wait until everyone else had left the room, and they were seated with at least one seat between each person. From that point on, the session was identical to the individual sessions except 27 that in the group session subjects were asked not to speak to each other or say anything aloud while they were answer- ing the questionnaire. CHAPTER 3 RESULTS Characteristics of the Sample Only women from lower level undergraduate classes were chosen to participate in this study. Over half of the women (58%) were 18 or 19 years old. They were primarily white (88%), single (95%), and living on campus (79%). Almost half of the subjects were from an upper-middle class background (49%). One hundred three of the women had been in college one year or less (47%). Forty percent (80 of the women) had between a 2.5 and a 3.0 grade point average (the scale at this university ranging from 0.0 to 4.0, with 4.0 = A+). Nineteen percent had a grade point average below a 2.5, and 41% had higher than a 3.0 GPA. Because 150 subjects participated in individual sessions and 69 subjects participated in group sessions, the two groups were compared to ensure that they were not significantly different demographically. 2 tests were per- formed on the following ordinal demographic variables: age, class level, grade point average, years in college, and socioeconomic status. In addition, chi square tests were performed on the four non-ordinal demographic variables: race, relationship status, hometown, and whether the sub- jects lived on or off campus. 28 29 There were no significant differences at the p = .05 level for any of these variables, confirming that the two groups were sufficiently comparable. Descriptive Tabulations Perceptions of Sexual Harassment Frequency. When asked what percentage of women students experience harassment similar to the scenario, 92 of the respondents saidCHsto 25% (42%), and 99 said 26% to 50% (45%). Only 13% thought the frequency rate was 51% or higher. Effect. Of the 73 women who received the mild harass- ment scenario, 53% indicated the situation would annoy them. This was followed by 21% responding that they would be angry, 19% responding that they would be frightened, and only 7% indicating the situation would not bother them at all. Of the 73 women in the moderate condition, 45% answered that they would be frightened, while 36% responded they would be angry, and the remainder (19%) indicated they would be annoyed. Fifty eight percent of the 73 women in the severe condition answered that they would be angry if the situation occurred. The second most frequent response was that they would be frightened (33%), and 10% of the women said they would be annoyed. None of the women in the moderate or severe condition said that the situation would not bother them. 30 Manipulation Check (Pilot Study) To examine the effectiveness of the manipulation, a 3 x 3 analysis of variance was performed on the pilot study with perceived level of harassment as the dependent variable. As hypothesized, a main effect for severity was found (F(2,72) = 32.86, p < .001). No significant interaction between severity and status of offender was found; nor was there a main effect for status of offender. See Table 3 for the analysis of variance summary table and Table 4 for cell means. The Scheffé post hoc test was used to determine which of the levels of severity differed from each other. It was found that each manipulated level (mild, moderate, severe) differed significantly from each other (p < .05) on perceived level of harassment. A 3 x 3 analysis of variance was also performed with power of harasser as the dependent variable. A significant 2-way interaction between severity and status of offender was found (F(4,72) = 3.14, p < .05). See Tables 5 and 6 for the analysis of variance summary table and cell means. The Scheffé post hoc test was once again used to further explain this interaction effect. However, because individual cell sizes were so small (range: 4 to 15) and so varied, cell mean differences were too complex to be amenable to straightforward interpretation. Therefore, it was decided to wait until the final study was completed to see if this finding held. Table 3 31 Analysis of Variance Summary Table for Perceived Level of Harassment (Pilot Study) Source §§ DE MS F Sig. of E Main effects 22.68 4 5.67 17.03 .001 Severity 21.88 2 10.94 32.86 .001 Status 0.61 2 0.31 0.92 .404 Interaction 1.35 4 0.34 1.01 .406 Explained 24.03 8 3.00 9.02 .001 Residual 23.97 72 0.33 Total 48.00 80 0.60 Table 4 Cell Means 32 for Perceived Level of Harassment (Pilot Study) Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 2.50 2.00 2.17 2.21 (_n_ = 4) (g = 4) (_n_ = 6) (g = 14) Moderate 2.90 3.27 3.22 3.13 (g = 10) (p = 11) (g = 9) (g = 30) Severe 3.50 3.83 3.67 3.68 (p = 10) (n = 12) (p = 15) (g = 37) Marginal 3.08 3.33 3.23 3.22 (3 = 24) (g = 27) (fl = 30) (E = 81) Note. Response options for dependent variable were as follows: 1 2 3 4 no, not an instance of sexual harassment yes, a mild form of sexual harassment yes, a moderate form of sexual harassment yes, a severe form of sexual harassment Table 5 33 Analysis of Variance Summary Table for Perceived Power of Harasser (Pilot Study) Source SS 2: MS F Sig of E Main effects 11.49 4 2.87 4.19 .004 Severity 6.53 2 3.27 4.76 .011 Status 5.06 2 2.53 3.68 .030 Interaction 8.63 4 2.16 3.14 .019 Explained 20.13 8 2.52 3.67 .001 Residual 49.43 72 0.69 Total 69.56 80 0.87 Table 6 34 Cell Means for Perceived Power of Harasser (Pilot Study) Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 2.50 3.40 2.71 (g = 10) (g = 10) (Q = 24) Moderate 3.36 2.92 3.04 (B = 11) (fl = 12) (n = 27) Severe 3.33 3.47 3.40 (E = 9) (g = 15) (n = 30) Marginal 3.07 3.27 3.07 (p = 30) (g = 37) (g = 81) Note. Response options for dependent variable were as follows: 1 2 3 4 same degree of power as "Sara" a little more power than "Sara" moderately more power than "Sara" quite a bit more power than "Sara" 35 An analysis of variance with believability of the scenarios as the dependent variable showed that there were no significant interactions or main effects present. Also, all of the subjects rated the scenarios as more than some- what believable; therefore, the scenarios were not altered. See Table 7 for cell means. Manipulation Check (Final Study) To double-check the effectiveness of the scenario mani- pulations, 3 x 3 analyses of variance were performed on the data from the final research study as well. On these data, when level of harassment was entered as the dependent vari- able, a significant two-way interaction between severity and status of offender was found (F (2,210) = 2.60, p < .05). See Tables 8 and 9 for the analysis of variance summary table and cell means. Upon completing Scheffé post hoc tests on the indivi- dual 3 x 3 cell means, it was found that there was a clear difference between mild and severe harassment for both professor and teaching assistant. There were no significant differences between mild and moderate or moderate and severe for either group. With regard to the severity levels for "advisor,' subjects saw no differences among any of the three levels (M = 2.71). A 3 x 3 analysis of variance was then performed with power of harasser as the dependent variable. Contrary to findings from the pilot study, no significant interactions or main effects were found. The apparent interaction found Table 7 36 Cell Means for the Believability of Scenarios Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 3.00 3.25 3.50 3.29 (g = 4) (g = 4) (p = 6) (Q = 14) Moderate 3.60 3.55 3.78 3.63 (g = 10) (fl = 11) (g = 9) (g = 30) Severe 3.20 3.50 3.47 3.41 (n = 10) (n = 12) (p = 15) (p = 37) Marginal 3.33 3.48 3.57 3.47 (g = 24) (n = 27) (p = 30) (3‘1 = 81) Note. Response options for dependent variable were as follows: 1 = unbelievable 2 = somewhat unvelievable 3 = somewhat believable 4 = believable Table 8 37 Analysis of Variance Summarerable for Perceived Level of Harassment (Final Study) Source SS 93 MS S Sig. of S Main effects 40.99 4 10.25 21.29 .001 Severity 38.09 2 19.04 39.55 .001 Status 2.46 2 1.23 2.56 .080 Interaction 5.00 4 1.25 2.60 .037 Explained 46.00 8 5.75 11.94 .001 Residual 101.11 210 0.48 Total 147.11 218 0.68 Table 9 Cell ”Means 38 for Perceived Level of Harassment (Final STudy) Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 2.46 1.96 2.23 2.22 (g = 26) (g = 25) (g = 22) (g = 73) Moderate 2.48 2.67 3.04 2.73 (g = 25) (g = 24) (g = 24) (g = 73) Severe 3.20 3.21 3.33 3.25 (g = 25) (p = 24) (Q = 24) (p = 73) Marginal 2.71 2.60 2.89 2.73 (g = 76 (g = 73) (g = 70) (N = 219) Note. Response options for dependent variable were as follows: 1 2 3 4 no, not an instance of sexual harassment yes, a mild form of sexual harassment yes, a moderate form of sexual harassment yes, a severe form of sexual harassment 39 in the pilot study may have been due to the low as well as varied number of subjects per cell. See Table 10 for the final study cell means. Major Findings Reporting as a Function of Severity In order to test the hypothesis that the more severe the harassment, the more likely a subject would be to believe that a victim would report the incident, a scale was first constructed which measured likelihood of report- ing. This ten-item scale measured the extent to which subjects believed that the victim (described as a typical university student) was likely to make use of various reporting procedures available to her (Alpha = .91; see Table 11). A 3 x 3 analysis of variance was then performed with the Likelihood of Reporting (LOR) Scale as the depen- dent variable. I A main effect for severity was found (S(2,210) = 6.94, p < .001). No main effect for status of offender was found; nor was there a significant interaction present. See Table 12 for the complete analysis of variance summary table and Table 13 for the cell means. A multiple comparison procedure was performed in order to determine which cells (mild, moderate, severe) differed significantly from each other. Using the Scheffé post hoc procedure at the p = .05 level, it was found that cells one and two (the mild and moderate levels) differed 40 Table 10 Cell Means for Perceived Power of Harasser (Final Study) Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 2.58 2.68 2.86 2.70 (g = 26) (fl = 25) (g = 22) (g = 73) Moderate 2.80 3.08 3.21 3.03 (g = 25) (g = 24) (g = 24) (Q = 73) Severe 2.92 2.63 2.92 2.82 (p = 25) (g = 24) (g = 24) (g = 73) Marginal 2.76 2.77 3.00 2.85 (E = 76) (g = 73) (Q = 70 (E = 219) Note. Response options for dependent variable were as = same degree of power as "Sara" a little more power than "Sara" moderately more power than "Sara" quite a bit more power than "Sara" follows: bWNI—A 41 Table 11 Internal Consistency Analysis of Likelihood of Reporting §E§l§ Corrected Alpha item-total if item Item # Item content correlation deleted 4 Do something about it .70 .89 7 Tell woman faculty member .56 .90 8 Tell male faculty member .51 .90 10 Tell staff woman in charge of harassment complaints .76 .89 11 Tell staff man in charge of harassment complaints .66 .89 12 Tell offender's chairwoman .74 .89 13 Tell offender's chairman .72 .89 14 File written complaint with offender's supervisor .68 .89 15 File written complaint with university channels .68 .89 16 File criminal charges .62 .90 Note. Alpha = .90. Standardized item Alpha = .91. Table 12 42 Analysis of Variance Summarerable for Likelihood of Reporting by Victim Source SS 23 MS S Sig. of S Main effects 4.14 4 1.03 3.48 .009 Severity 4.13 2 2.06 6.94 .001 Status 0.01 2 0.01 0.02 .982 Interaction 2.75 4 0.89 2.31 .059 Explained 6.88 8 0.86 2.89 .004 Residual 62.49 210 0.30 Total 69.37 218 0.32 43 Table 13 Cell Means for Likelihood of Reporting by Victim Status of offender Graduate Severity Advisor assistant Professor Marginal Mild 1.95 1.74 1.94 1.88 (p = 26) (p = 25) (p = 22) (p = 73) Moderate 2.09 2.13 1.84 2.03 (g = 25) (p = 24) (p = 24) (p = 73) Severe 2.05 2.27 2.31 2.21 (p = 25) (p = 24) (p = 24) (p = 73) Marginal 2.03 2.05 2.04 2.04 (p = 76) (g = 73) (I; = 70) (91 = 219) 44 significantly from cells two and three (the moderate and severe levels). In other words, those subjects receiving the scenario of severe harassment thought it more likely that the victim in the scenario would report the harassment than those subjects who received the mild level of harass- ment scenario. However, there were no significant differ- ences between mild and moderate or moderate and severe scenarios for how likely subjects thought it was for a student to report harassment. This is consistent with the finding that, while subjects clearly differentiated between the mild and severe scenarios, they did not differentiate between the mild and moderate scenarios or the moderate and severe scenarios. Predictors of Reporting A major hypothesis of this study was that reporting behavior would be predicted by perceived effectiveness of reporting, perceived fear of reporting, and perceived repercussions or reporting. To reduce the number of items measuring the fear of reporting and the perceived effective- ness of reporting, two scales were constructed. Additional scales were constructed in order to create unidimensional variables measuring respondents' perceptions of negative consequences of reporting. Effectiveness of Reporting (EOR) Scale. This scale consisted of four items which assessed subjects' perceptions of the effectiveness of various reporting procedures in stopping harassment (Alpha = .62; see Table 14). 45 Table 14 Internal Consistency Analysis of Effectiveness of Reporting Scale Corrected Alpha item-total if item Item # Item content correlation deleted 25 Tell a peer of offender .37 .54 26 Tell offender's supervisor .51 .48 27 File complaint in writing through university .52 .47 28 File criminal charges .29 .60 Note. Alpha = .61. Standardized item Alpha = .62. 46 Fear of Reporting (FOR) Scale. The four items in this scale assessed how frightening subjects thought it would be to avail themselves of various reporting procedures (Alpha = .79; see Table 15). Independence of Scales. Intercorrelations between the LOR, EOR and FOR scales were examined in order to determine independence. Each scale was acceptably independent, although some were slightly correlated (with correlations ranging from -.13 to .19). Therefore, adequate evidence was provided to assure that each scale was measuring a distinct construct. See Table 16 for the intercorrelations. Negative Consequences of Reporting. In regard to creating unidimensional variables measuring respondents' perceptions of negative consequences of reporting, several variables (items 36 to 39) were recoded. Students had originally been asked whether they thought a typical univer- sity student would experience negative consequences (being blamed for it, accused of lying, or suffering retaliation) if they reported the harassment to each of several authority figures (harasser's peer, harasser's supervisor, written complaint at the university level, criminal charges). Because these response options were not ordinal, it was necessary to recode these variables. In order to determine if subjects responded differently to the four different authority figures, correlations for how frightening subjects thought it would be to report to each of them were examined. Correlations among these four authority options were 47 Table 15 Internal Consistency Analysis of Fear of Reporting Scale Corrected Alpha item-total if item Item # Item content correlation deleted 31 Tell a peer of offender .56 .75 32 Tell offender's supervisor .72 .70 33 File complaint in writing through university .64 .72 34 File criminal charges .58 .75 Note. Alpha = .79. Standardized item Alpha = .79. 48 Table 16 Correlations between Scales Scales LOR FOR Effectiveness of reporting (EOR) .19* -.06 Likelihood of reporting (LOR) -.13* Fear of reporting (FOR) *p < .05. 49 reasonably substantial (ranging from ,30 to .74; p < .001). Three of the six correlated .5 or higher. This justified summing affirmative responses to given questions across these four categories. For instance, a respondent who believed that a victim would be accused of lying if she either filed charges through the university channels or filed criminal charges, but felt the victim would be blamed for the harassment if she told a peer of the offender would receive a score of "2" on the Not Believed Scale and a score of "1" on the Blamed Scale. This procedure resulted in the creation of three additional scales: Blamed, Not Believed, and Retaliation. After these scales were constructed, a forward step- wise multiple regression was performed to predict the participants' perceptions of the likelihood of a typical university student reporting sexual harassment. The results are presented in Table 17. Thirteen variables were included in this regression: subject's home town, age, how long they had been in college, grade point average, status of harasser, level of harass- ment, and the following scales: Effectiveness of Reporting, Fear of Reporting, Blamed, Not Believed, and Retaliation. Four variables made significant unique contributions to the regression equation: severity of the harassment (g = .20), Not Believed (S = -.91), Effectiveness of Reporting (S = .21), and Fear of Reporting (B = -.11). The R2 between these predictors and the dependent variable Likelihood of 50 .Hoco. v m**.* .moo. v mirr .woo. v mir .mmo. v m* .mfim u z .muoz «aaaom.m mac. mma.| mqfi. mmm. «mo.v unmflum mo Hoboq v *«RRBo.oH omo. mma. ema. com. «enm.h mcfluuomon mo mmocm>fluoommm m ¥«*«NO.NH avo. ovH.I OOH. mam. eraaw.m EGHMH mo ommsood m «assho.mfi owo. mvN. ooo. mvm. akakho.MH >HHHO>Om H M moccnu m NM m Hmpco manuauo> moum Hfimum>o NM mflmsflm mfimfluasz 0» M mcfluuomom mo muouoflooum co mflmwamcfi :oflmmoumom omflzlmoum NH OHQMB 51 Reporting (.15) represented a statistically significant relationship (p < .0001). Hence, the decision of a victim to report sexual harassment could be partially predicted by the severity level of the harassment (such that the more severe the harassment, the more likely the reporting), the belief that one would be accused of lying, perceived effectiveness of reporting, and how frightening students believe it would be to report the harassment. To Whom Students Report Subjects responded to many questions addressing to whom they thought victims of harassment would report. This study specifically wanted to investigate whether subjects believed victims would be more likely to report to someone within the harasser's department, someone outside of the harasser's department, a man, and/or a woman. Therefore, six new scales were constructed. Report to a Woman. This scale consisted of items 7, 10, and 12: the likelihood of reporting to a woman faculty member, the likelihood of reporting to a staff woman in charge of harassment complaints, and the likelihood of reporting to the harasser's chairwoman (Alpha = .79; see Table 18). Report to a Man; Items 3, 11, and 13 formed this scale: the likelihood of reporting to a faculty man, a staff man in charge of harassment complaints, and the harasser's chairman (Alpha = .81; see Table 19). 52 Table 18 Internal Consistency Analysis of Report to a Woman Scale Corrected Alpha item-total if item Item # Item content correlation deleted 7 Tell woman faculty member .56 .79 10 Tell staff woman in charge of harassment complaints .67 .67 12 Tell offender's chairwoman .67 .68 Note. Alpha = .79. Standardized item Alpha 53 Table 19 Internal Consistency Analysis of Report to a Man Scale Corrected Alpha item—total if item Item # Item content correlation deleted 8 Tell male faculty member .57 .82 11 Tell staff man in charge of harassment complaints .68 .71 13 Tell offender's chairman .72 .66 Note. Alpha = .81. Standardized item Alpha = .80. 54 Report to Either Sex. This scale consisted of all of the items in both of the preceding scales. The purpose of creating this scale was to assess if reporting to either a man or a woman accounted for any unique variance not already accounted for by reporting in general (Alpha = .87; see Table 20). Report Within the Department. This scale measured the likelihood of reporting within the harasser's department. Items included telling the harasser's chairwoman, telling the harasser's chairman, and filing a written complaint with the harasser's supervisor (Alpha = .80; see Table 21). Report Outside of the Department. This scale consisted of three items addressing the likelihood of reporting sexual harassment outside of the offender's department. The items were: telling a staff man in charge of harassment com- plaints, telling a staff woman in charge of harassment complaints, and filing a written complaint through univer- sity channels (Alpha = .79; see Table 22). Reporting In or Out. This scale was the culmination of the items in both the Report Outside the Department scale and the Report Within the Department scale. Again, this was done to determine if reporting within or outside of the offender's department accounted for a unique portion of the variance not already accounted for by reporting in general (Alpha = .89; see Table 23). Women vs. Men. Subjects believed it to be very unlikely that a typical university student would report 55 Table 20 Internal Consistency Analysis of Report to Either Sex Scale Corrected Alpha item-total if item Item # Item content correlation deleted 7 Tell woman faculty member .63 .86 8 Tell male faculty member .57 .87 10 Tell staff woman in charge of harassment complaints .67 .85 11 Tell staff man in charge of harassment complaints .71 .84 12 Tell offender's chairwoman .70 .85 13 Tell offender's chairman .76 .84 Note. Alpha = .87. Standardized item Alpha = .87 56 Table 21 Internal Consistency Analysis of Report Within the Department Scale Corrected Alpha item-total if item Item # Item content correlation deleted 12 Tell offender's chairwoman .71 .64 13 Tell offender's chairman .65 .71 14 File written complaint with offender's supervisor .56 .80 Note. Alpha = .80. Standardized item Alpha = .80. 57 Table 22 Internal Consistency Analysis of Report Outside of the Department Scale Corrected Alpha item-total if item Item # Item content correlation deleted 10 Tell staff woman in charge of harassment complaints .73 .60 11 Tell staff man in charge of harassment complaints .59 .76 15 File written complaint with university channels .58 .77 Note. Alpha = .79. Standardized item Alpha = .79. 58 Table 23 Internal Consistency Analysis of Reporting In or Out Scale Corrected Alpha item-total it item Item # Item content correlation deleted 10 Tell staff woman in charge of harassment complaints .74 .86 11 Tell staff man in charge of harassment complaints .66 .87 12 Tell offender's chairwoman .74 .86 13 Tell offender's chairman .71 .87 14 File written complaint with offender's supervisor .67 .87 15 File written complaint with university channels .60 .87 Note. Alpha = .89. Standardized item Alpha = .89. 59 sexual harassment to a man, regardless of his position. Only 15% of the respondents thought it likely that a victim would tell a professional man in charge of sexual harassment complaints, and 11% thought it likely that a victim would report to a faculty man. Further, only 17% thought a victim would tell the chairman of the offender's department. In comparison, 45% believed a victim would tell a faculty woman, 40% thought a typical student would tell a professional woman in charge of harassment complaints, and 35% said a victim would tell the chairwoman of the offen- der's department. A 3 x 3 multivariate analysis of variance was performed in order to examine whether these various decisions were affected by the manipulation. To determine if reporting to either a male or a female contributed any unique portion of the variance not already accounted for by reporting to either sex, the following scales were entered as the depen- dent variables: Reporting to Either Sex, Reporting to a Man, and Reporting to a Woman. The multivariate analysis of variance revealed that all differences between conditions of the manipulation were accounted for by the Reporting to Either Sex scale. No additional unique variance among the manipulation conditions was accounted for by either Reporting to a Woman or Report- ing to a Man. Therefore, all severity levels were combined in these two scales, and 3 tests were performed. 60 The 5 tests were performed in order to determine if there was a significant difference between the likelihood of telling women vs. the likelihood of telling men. Students believed it much more likely that a victim would tell a woman faculty person over a male faculty person (3(218) = 15.6, p < .0001), and a professional woman in charge of harassment complaints rather than a man in charge of harassment complaints (3(218) = 12.4, p < .0001). They also believed it more likely for a victim to tell a chair- woman than a chairman (3(218) = 11.2, p < .0001). Outside vs. within department. Not only did subjects believe that it was more likely that victims would report harassment to women than to men, they also found it much more likely that a victim would report to someone outside of the offender's department than within it. A multivariate analysis of variance was performed to discover if reporting to channels inside of or outside of the offender's department was affected by the manipulation. Reporting Within the Department, Reporting Outside of the Department, and Reporting In or Out were the three scales used as the dependent variables. Once again, all differ- ences between conditions were accounted for by Reporting In or Out, with no additional unique variance accounted for by the other two variables. Hence, severity levels were com- bined in these two scales and a 3 test was performed. 61 The 3 test confirmed that respondents thought it more likely that a victim would report to someone outside of the offender's department than within it (3(218) = 2.41, E < .05). Summary of Major Findings In summary, there were three major hypotheses with which this study was concerned. The first of these hypothe- ses was that subjects' perceptions of reporting behavior would be a function of severity such that, the more severe the harassment, the more likely a subject would believe that the victim would report it. The data verified that subjects who received the severe scenario were more likely to believe the harassment would be reported than those subjects who received the mild scenario. The second hypothesis of this study was that subjects' beliefs that a victim would report sexual harassment could be predicted by their perceptions of how effective they thought reporting would be, how frightened they would be to report harassment, and their fear of repercussions from reporting. Again, data analysis bore out that the reporting decision could be partially predicted by four variables: severity of the harassment, subjects' perceptions of effec- tiveness of reporting, subjects' fear of the reporting pro- cedure, and subjects' belief that they would not be believed. Finally, it was hypothesized that students would be more likely to report harassment to a professional in charge of harassment complaints than to someone in the offender's 62 department, and they would be more likely to report to a woman than to a man. Thus, students would be most likely to report to a professional staff woman in charge of sexual harassment complaints. It was found that subjects gepe more likely to think a victim would report to someone outside of the offender's department, as did they believe that a victim would be more likely to report to a woman than to a man. Further, sub- jects found it most likely that a victim of sexual harass- ment would report it to a professional staff woman in charge of harassment complaints. These results and their implications will be further discussed in the following section. CHAPTER 4 DISCUSSION Major Findings Reporting as a Function of Severity It had been hypothesized that reporting behavior would be a function of the severity of the harassment such that the more severe the harassment, the more likely reporting would be. Study results did confirm that subjects believe women who experience severe harassment are more likely to do something about the episode than women who experience mild harassment. However, no significant differences were found between the subjects who received the mild condition and the subjects who received the moderate condition with regard to reporting; nor was there a significant difference between subjects who received the moderate condition and subjects who received the severe condition with regard to reporting. In other words, the only significant difference present was between mild and severe harassment conditions. Subjects who received the severe scenario were more likely to believe the victim would report the harassment than subjects who received the mild condition. This finding is in accordance with the fact that subjects did not differen- tiate between mild, moderate, and severe levels of sexual harassment in the final study; rather, they only considered the mild and severe conditions to be significantly different 63 64 from one another. Therefore, it stands to reason that sub- jects would not perceive reporting differences among all three levels of harassment. Subjects perceived two levels of sexual harassment and, in line with that thinking, believed that victims of the more serious level would be more likely to report the harassment. But ypy are subjects more inclined to believe that a victim of severe harassment will be more likely to report the incident(s)? It could simply be that the severe inci- dent is so frightening and threatening. In the severe scenario the victim is blatantly coerced, which could be enough to make her report the behavior. This explanation is, of course, speculative. Only more extensive research conducted in this area will determine exactly why students perceive it to be more likely that severe harassment will be reported. Predictors of Reporting As was hypothesized, reporting behavior was found to be influenced by the perceived effectiveness of reporting as well as how frightening reporting was perceived to be. It had also been hypothesized that subjects' perceptions of possible repercussions from reporting would predict whether or not they believed a victim would report sexual harassment. When various types of repercussions were examined individually, it was found that the belief that a victim would be accused of lying did negatively predict reporting behavior. Therefore, according to the results of 65 this study, reporting behavior could be partially predicted by how severe the harassment was, how effective subjects thought it would be to report the incident, how frightening it would be to report the incident, and the belief that one would be believed. The belief that one would either be blamed for the harassment or would suffer retaliation from the harasser did not contribute any additional unique vari- ance to the equation. In addition, subject's age, grade point average, number of years in college, and socioeconomic status did not predict whether or not a student would report the sexual harassment. These findings are important for the following three reasons: 1. If it is demonstrated to students that reporting sexual harassment will be effective in ending it, they will be more likely to report incidents. Logically, if it is demonstrated that reporting is ineffective (i.e., it becomes publicly known that an admitted harasser was not punished in any way), theoretically fewer students will bother to report the offensive behaviors. 2. Students are more likely to report harassment if they don't think that the act of reporting will be a frightening experience. Hence, if students know there is someone who is unbiased outside of the offender's depart- ment to whom they can talk without being intimidated, they will theoretically be more willing to report. 66 3. Students must also be assured that they will be believed about sexual harassment complaints if they are to be expected to come forward with information. If a sexual harassment victim perceives that she will be accused of lying, she is certainly not going to report her experi- ence(s). The practical implications of these findings will be discussed later in this chapter. To Whom Students Report This study confirmed the hypothesis that students would be most likely to report sexual harassment to a pro- fessional staff woman in charge of such complaints. There- fore, it is only reasonable that a woman be appointed to handle charges of sexual harassment, and that she be dedi- cated to effectively handling complaints to the satisfaction of the victim. Methodological Issues Sample Originally, only undergraduate women from introductory psychology courses were going to be considered for this study. There were two major reasons for this decision: First, although mailing questionnaires to a random sample of female students may be advantageous in obtaining a rep- resentative sample, given the nature of this study it could have also resulted in an overrepresentation of harassment victims and/or feminists. It could have also resulted in 67 respondents discussing answers with others rather than using their own judgment. The second reason for using the psychology undergraduate pool was that students received extra credit for participating in research studies; this was their motivation for participating, and they did not know what the research topic was prior to participation. Therefore, although this method resulted in obtaining mostly first-year students rather than a stratified sample, using these students seemed to be the best possible option. When it was evident by the middle of the term that not enough subjects would be obtained using this method, students were also recruited from introductory social science, criminal justice, natural science, English, and humanities classes. All of the 219 subjects were told only that the study was about women's experiences on campus; however, subjects obtained by this latter method did not earn extra credit for participating. By the end of the term, 150 of the 219 subjects were from the psychology pool. Although this sampling procedure was less than ideal, it is not believed that the sample was biased in any way because of this. All precautions were taken to minimize discrepancies between groups, and both groups were compa- rable with regard to age, race, years in college, home town, grade point average, class level, relationship status and whether they lived on or off campus. 68 Advisor as Offender The three status levels were chosen because Adams et al. (1983) found that sexual harassers were most likely to be professors (50%), teaching assistants (25%), and academic advisors (17%). One purpose of this study was to explore whether or not the harasser's academic position affected students' perceptions of the severity of the sexually harassing incidences and/or reporting behaviors. Although in this study status of offender did not affect subjects' views of reporting harassment, there was a signi- ficant interaction with regard to subjects' perception of the harassment itself. Subjects did not clearly differen- tiate between mild, moderate, and severe harassment when "advisor" was the harasser's academic position. There are two major reasons why studying students' reactions to academic advisors is difficult. First, advisors are more an integral component of some students' academic programs than of others'. For instance, some students major in programs which are highly structured and in which one deals with one's advisor every term throughout every year. These students' relationships with their advisors would be markedly different from those students involved in large, flexible programs in which they may never meet with an advisor. The second reason that it is so difficult to examine perceptions of advisors as sexual harassers is that the status of academic advisors varies not only among programs 69 but among educational institutions as well. Advisors in major universities may have different functions than advisors at community colleges, and the role of the advisor may differ from college to college and from university to university. Therefore, results obtained in studies re- garding academic advisors are only generalizable to those academic institutions in which the role of advisors is very similar to the original experimental setting. Predictor Variables The majority of questions in this study concerned "Sara,' explained to subjects as a young, female student at another university in the midwest. Subjects were instructed to think of "Sara" as a typical university stu- dent. Therefore, when predicting the likelihood of a typical university student reporting sexual harassment, all variables should technically have been in reference to Sara. One of the scales, however—~Fear of Reporting--dealt with how frightening subjects thought it would be to use the various reporting options presented. The reason that the variables in this scale addressed subjects'cnniresponses was because these variables dealt with emotions rather than perceptions. Whereas subjects could generalize about how effective a reporting procedure might be for students, or how most harassers might react to being reported, they could only know how frightened they themselves might be when it came to reporting harassment. And that was, after all, what the study was interested in discovering. 70 However, it should be made clear here that the multiple regression confirmed that subjects believed that the like- lihood of a typical university student reporting sexual harassment could be partially predicted by (a) how effec- tive the subjects thought such reporting would be in stop- ping the victim's harassment, (b) whether the subject thought the victim would be believed, (0) how frightening the subject thought reporting to be, and (d) how severe the subject thought the harassment was. It should also be mentioned here that the 32 between the predictors and the dependent variable was only .15. While this may represent a statistically significant rela- tionship (p < .0001), it also means that only a small amount of the variance (15%) was actually accounted for. This leads one to wonder what other variables affect one's deci— sion about whether or not to report sexual harassment. Implications of the Research Necessity of Grievance Procedures Demonstrated effectiveness. The forward step-wise multiple regression procedure verified that subjects' beliefs regarding a student's decision to report sexual harassment was predicted in part by how effective they per- ceived such reporting to be. Therefore,:tEaiuniversity's goal is to encourage all victims of sexual harassment to report the occurrences, a logical course of action is to publicly demonstrate the effectiveness of the university's 71 grievance procedure(s). This can be accomplished in a variety of ways. One way to demonstrate the effectiveness of a grievance procedure is to distribute pamphlets outlining exactly what the consequences are for each type of sexual harassment perpetrated. These should be specific about what will happen to someone who makes offensive sexist comments to a student, who touches a student without her/his consent, and who propositions, coerces, or threatens a student. Of course, it goes without saying that these rules must be consistently enforced if students are to trust them. Another method of proving to students that grievance procedures are adequate is to publicze the results of hearings in which someone has been found guilty of sexual harassment. This is also a way of ensuring that the univer- sity administrators are handling sexual harassment com- plaints to the satisfaction of students. Low-fear environment. Subjects' perceptions of a student's decision whether or not to report sexual harass- ment was also predicted by how frightening they thought reporting would be. This finding supports Meek and Lynch's (1983) hypothesis that students need access to a grievance procedure through which they can remain anonymous and through which confidentiality is guaranteed. Further, students should not have reason to be afraid of being accused of lying and/or being blamed for the incidents. 72 Unbiased professional. As was hypothesized, students believed it to be more likely for a victim to report sexual harassment to someone outside of the offender's department than to the offender's colleagues or supervisors. This finding supports the need for a reporting procedure in which students can voice their complaints to an unbiased profes- sional. This staff person should be dedicated to effec- tively handling charges of sexual harassment to the satis- faction of the complainant. Sex of professional. As previously stated, only 15% of the subjects thought it likely that a victim would report to a man in charge of handling sexual harassment complaints. This is in comparison to 40% who thought a victim would tell a woman of the same authority. Since subjects feel it is much more likely to report to a pro- fessional staff woman, it is only reasonable that a woman be appointed to handle charges of sexual harassment. In conclusion, if students are to be expected to report incidences of sexual harassment, they should be provided with an effective grievance procedure in which they can anonymously speak with an unbiased staff woman who will make reporting as least fear-inducing as possible. If all of these program components were implemented, theoretically students would be much more likely to report incidences of sexual harassment. 73 Future Research This research study was merely a stepping stone regard- ing sexual harassment in an academic setting. It answered some very specific questions regarding students' percep- tions of and responses to various levels of sexual harass- ment, as intended. However, these answers only lead to even more questions which have yet to be researched. For instance, future research needs to experimentally explore whether or not an academic grievance procedure, consisting of a dedicated staff woman, which has demon- strated its effectiveness, results in a significant increase of sexual harassment complaints. The results from this type of research could then be used to disseminate the importnace of every academic institution having this type of grievance procedure. However, an adequate grievance procedure is but one component of dealing with reports of sexual harassment. Whether or not academic administrators are actually punish- ing and preventing sexual harassment needs to be researched also. For, no matter how effective a grievance procedure may be in investigating harassment charges, it is the final outcome of the investigation which is of interest to the victim as well as all potential victims of sexual harass- ment. Therefore, steps need to be taken to insure that satisfactory measures are taken against anyone who sexually harasses a student. 74 Thus far, all of the research in this area has focused on dealing with sexual harassment once it has occurred. Studies have examined the extent of the problem, the detri- mental effects on the victims, and the reasons why so few students report offensive incidences. All of these researchers have attempted to alleviate the symptoms rather than cure the disease. To date, no one has scientifically evaluated the most effective means of preventing sexually harassing behaviors. This has been a critical oversight which must be amended if any significant social change is to take place. APPENDICES APPENDIX A NINE EXPERIMENTAL SCENARIOS APPENDIX A NINE EXPERIMENTAL SCENARIOS [Advisor/Mild Sexual Harassment] Sara was enrolled as a social science major at a medium-sized midwestern univeristy. Halfway through her junior year she decided to go to graduate school in her major, so went to see her advisor. Her advisor was a new man on staff whom she had not previously seen. When she arrived at his office she noticed that the advisor was looking at a pornographic magazine. He promptly set it aside and was very helpful in discussing graduate school requirements. As he spoke he looked more at her blouse than at her face. When she was leaving, Sara noticed a poster behind the advisor's door of a woman in a bikini skiing down a mountain. Four men in hunting outfits carrying rifles were skiing after her. The slogan read, "Skiing. Hunting. All winter sports. Come to Mountain." Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 75 76 [Advisor/Moderate Sexual Harassmentl Sara was enrolled as a social science major at a medium-sized midwestern university. Halfway through her junior year she decided to go to graduate school in her major, so went to see her advisor. Her advisor was a new man on staff whom she had not previously seen. When she arrived at his office, he closed the door and then chatted with her a few minutes about graduate schools. He then complimented her on her new jeans. When Sara did not respond, he moved closer to her and stroked her hair. "What's the matter? It's not like I'm coming on to you. I'm a friendly kind of guy. Don't you like being friendly?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 77 [Advisor/Severe Sexual Harassment] Sara was enrolled as a social science major at a medium—sized midwestern university. Halfway through her junior year she decided to go to graduate school in her major, so went to see her advisor. Her advisor was a new man on staff whom she had not previously seen. During their talk her advisor was very discouraging about her possibilities for acceptance into graduate school because of her grades. He then added that there was one school she may be able to get into because of his connections there. Abruptly he put his hand on her shoulder and stroked her hair. When she moved away from him he said, "I could get you into that graduate school." He then leaned over and tried to kiss her. Again Sara moved away from him, at which point he said, "You must not want to get into a graduate school, do you?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your 0WD . 78 [Teaching Assistant/Mild Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to see the class teaching assistant whom she had only seen previously during the class lecture hours. When she went to his office during office hours, Sara noticed that the T.A. was looking at a pornographic maga- zine. He promptly set it aside and was very helpful in explaining the class material that Sara needed help with. As he spoke he looked more at her blouse than at her face. When she was leaving Sara noticed a poster behind the teaching assistant's door of a woman in a bikini skiing down a mountain. Four men in hunting outfits carrying rifles were skiing after her. The slogan read,"Skiing. Hunting. All winter sports. Come to Mountain." Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 79 [Teaching Assistant/Moderate Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to see the class teaching assistant whom she had only seen previously during the class lecture hours. When she went to his office during office hours, he closed the door and then chatted with her for a few minutes about the assignment. He then complimented her on her new jeans. When Sara did not respond he moved closer to her and stroked her hair. "What's the matter? It's not like I'm coming on to you. I'm a friendly kind of guy. Don't you like being friendly?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 80 [Teaching Assistant/Severe Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to the class teaching assistant whom she had only seen previously during the class lecture hours. When she went to his office during office hours, he closed the door and then chatted with her for a few minutes about the assignment. Abruptly he put his hand on her shoulder and stroked her hair. When she moved away from him he said, "I could make your assignment very easy for you." He then leaned over and tried to kiss her. Again Sara moved away from him, at which point he said, "You don't want to fail this course, do you?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 81 [Professor/Mild Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to see the professor whom she had only seen previously during the class lecture hours. When she went to his office during office hours, Sara noticed that the professor was looking at a pornographic magazine. He promptly set it aside and was very helpful in explaining the class material that Sara needed help with. As he spoke he looked more at her blouse than at her face. When she was leaving Sara noticed a poster behind the professor's door of a woman in a bikini skiing down a mountain. Four men in hunting outfits carrying rifles were skiing after her. The slogan read, "Skiing. Hunting. All winter sports. Come to Mountain." Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 82 [Professor/Moderate Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to see the professor whom she had only seen previously during the class lecture hours. When she went to his office during office hours, he closed the door and then chatted with her for a few minutes about the assignment. He then complimented her on her new jeans. When Sara did not respond he moved closer to her and stroked her hair. "What's the matter? It's not like I'm coming on to you. I'm a friendly kind of guy. Don't you like being friendly?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . 83 [Professor/Severe Sexual Harassment] Sara was enrolled in an introductory social science course at a medium-sized midwestern university. Halfway through the academic term, she needed some questions answered about an upcoming paper which was worth 50% of her grade. She went to see the professor whom she had only seen previously during the class lecture hours. When she went to his office during office hours, he closed the door and then chatted with her for a few minutes about the assignment. Abruptly he put his hand on her shoulder and stroked her hair. When she moved away from him, he said, "I could make your assignment very easy for you." He then leaned over and tried to kiss her. Again Sara moved away from him, at which point he said, "You don't want to fail this course, do you?" Please answer the questions beginning on the next page by circling the number whose answer best corresponds with your own . APPENDIX B SEXUAL HARASSMENT QUESTIONNAIRE APPENDIX B SEXUAL HARASSMENT QUESTIONNAIRE Research Booklet Inside this booklet will be a description of an event which happened to a young female undergraduate at another univer- sity in the Midwest. She has consented to its use for research purposes as long as the people involved are not identified; therefore, the names have been changed and identifying details have been eliminated from the descrip- tion. Please read the description of the event and answer the following questions as honestly as possible. There are no right or wrong answers-~your honest opinion is all that is asked. We hope that the results of this research based on actual case histories will benefit undergraduate women in university settings in the future. 1. The research study conducted by Mary Sullivan under the supervision of Dr. Robin Redner has been explained to me and I fully understand it. 2. I freely consent to participate in this research project. I realize that I may discontinue my participation in this study at any time without recrimination. 3. I have been informed that my answers will be treated with the strictest confidence and that anonymity is guaranteed. On request and within these restrictions, results of this project will be made available to me. Participant's Name Researcher's Name Date: 84 85 Please circle the number of the statement which most closely matches your own opinion or feeling. For each question, please remember to pick just ONE answer. 1. How would you feel if this type of situation happened to you? 1. it wouldn't bother me at all 2. I'd be annoyed 3. I'd be frightened 4. I'd be angry 2. What do you think is the likelihood of this situation occurring? 1. very unlikely 2. unlikely 3. likely 4. very likely 3. What percentage of university women do you think experi- ence this type of situation? 1. 0-25% (very small percentage) 2. 26-50% (less than half) 3. 51-75% (more than half) 4. 76-100% (very large percentage) 4. How much power does the man in this incident have in comparison to Sare? 1. the same degree of power as Sara 2. a little more power than Sara 3. moderately more power than Sara 4. much more power than Sara 86 I'd like you to answer the following questions assuming Sara is a typical university student. Let me remind you that there are no right or wrong answers. I would simply like your honest opinion. Please circle the number of the statement which most closely matches your own feeling or opinion. For each question please only circle ppe response. WHAT IS THE LIKELIHOOD THAT SARA WILL: 4. do something about this situation? 1. very unlikely 2. unlikely 3. likely 4. very likely 5. tell him directly that she does not appreciate or desire his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 6. tell her friends about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 7. tell a woman faculty member whom she knows about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 87 WHAT IS THE LIKELIHOOD THAT SARA WILL: 8. tell a male faculty member whom she knows about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 9. avoid this person in the future? 1. very unlikely 2. unlikely 3. likely 4. very likely 10. talk to a professional staff woman who is familiar with university policy (ex.: woman at Student Services, Counseling Center, etc.) about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 11. talk to a professional staff man who is familiar with university policy (ex.: man at Student Services, Counseling Center, etc.) about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely 88 WHAT IS THE LIKELIHOOD THAT SARA WILL: 12. tell the chairwoman of the person's department (assume 13. 14. 15. the chairperson is a woman) about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely tell the chairman of the person's department (assume the chairperson is a man) about his behavior? 1. very unlikely 2. unlikely 3. likely 4. very likely file a complaint in writing with the person's super- visor? 1. very unlikely 2. unlikely 3. likely 4. very likely file a complaint in writing with the appropriate admini- strative channel which deals with possible sexual harassment cases? 1. very unlikely 2. unlikely 3. likely 4. very likely 89 WHAT IS THE LIKELIHOOD THAT SARA WILL: 16. file criminal charges against this person? 1. very unlikely 2. unlikely 3. likely 4. very likely 17. Do you feel that this event is an instance of sexual harassment? 1. no, not at all 2. yes, a mild form 3. yes, a moderate form 4. yes, a severe form Please go on to the next page. 90 Now I'd like to know what you think MOST women would want to have happen after this situation occurred. Please circle the number of the statement which most closely matches your own opinion or feeling. For each question, please choose just ONE response. 18. I think MOST women would: 1. want nothing to be done and nothing to change 2. want him to stop the behavior, but not want him to get into trouble 3. want one of his peers to know and to ask him to stop 4. want his supervisor to know and to force him to stop 5. want a formal complaint to be filed against him 6. want criminal charges to be brought against him 19. What do you think YOU would do if this situation happened to you? (Please think this over carefully before answering.) 20. What would YOU want to have happen after this situation occurred? Please just circle ONE number. I would: 1. want nothing to be done and nothing to change 2. want him to stop the behavior, but not want him to get into trouble 3. want one of his peers to know and to ask him to stop 4. want his supervisor to know and to force him to stop 5. want a formal complaint to be filed against him 6. want criminal charges to be brought against him 21. 22. 91 In what ways, if any, do you think this incident will affect Sara? How serious do you think these effects (in general) will be for Sara? 92 Now I'd like to know how effective you think different methods of dealing with this type of situation are. Effective means the person will not do this again to Sara or to any othe woman. Please circle only one response per question. How effective are the following actions? 23. 24. 25. 26. ignore the behavior 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective tell the person directly that his behavior is undesirable 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective tell a peer of the person who could ask him to change his behavior 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective tell the person's supervisor who could make him change his behavior 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective 93 27. file a formal complaint in writing against the person at the university level 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective 28. file criminal charges against the person 1. completely ineffective 2. mostly ineffective 3. mostly effective 4. completely effective -_._‘—A=. 94 Now I'd like to know how frightening you think it would be to be in the following positions. For each question, please circle just ONE. How frightening are the following actions? 29. 30. 31. 32. ignoring the behavior 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening telling the person directly that his behavior is undesirable 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening telling a peer of the person who could ask him to change his behavior 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening telling the person's supervisor who could tell him to change his behavior 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening 33. 34. 95 filing a formal complaint at the university level against the person 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening filing criminal charges against the person 1. not frightening 2. a little frightening 3. fairly frightening 4. extremely frightening 96 Now I'd like to know what negative effects, if any, you think there would be for Sara should she take the following actions. For each question, please just pick ONE response. What do you think is the most likely consequence or effect for Sara if she: 35. 36. 37. told the person directly that his behavior was unde- sirable there would be no negative effects on Sara she would be accused of lying she would be blamed for "asking for it" the person would somehow get back at her (reduce her grade, write a bad letter of recommendation, spread rumors, etc.) told a peer of the person who could ask him to change the behavior there would be no negative effects on Sara she would be accused of lying she would be blamed for "asking for it" the person would somehow get back at her (reduce her grade, write a bad letter of recommendation, spread rumors, etc.) told the person's supervisor who could tell him to change his behavior there would be no negative effects on Sara she would be accused of lying she would be blamed for "asking for it" the person would somehow get back at her (reduce her grade, write a bad letter of recommendation, spread rumors, etc.) 38. 39. 97 filed a formal complaint in writing against the person at the university level there would be no negative effects on Sara she would be accused of lying she would be blamed for "asking for it" the person would somehow get back at her (reduce her grade, write a bad letter of recommendation, spread rumors, etc.) filed criminal charges against the person there would be no negative effects on Sara she would be accused of lying she would be blamed for "asking for it" the person would somehow get back at her (reduce her grade, write a bad letter of recommendation, spread rumors, etc.) 98 Now I'd just like to ask you a few questions about your own personal experiences. If you feel that any of the questions are too intrusive please do not force yourself to answer them. However, the more responses I get, the better I will be able to get a clear picture of what is going on. Let me remind you that your name does not appear anywhere on this booklet and that your answers are anony- mous and confidential. 40. How many women do you know who have experienced Spy type of sexual harassment in a university setting? Sexual harassment has been defined as any unwanted sexual leers, suggestions, comments or physical contact which a student finds objectionable in the context of a teacher-student relationship. 41. Have you ever experienced an incident in a university setting which YOU consider to be a form of sexual harassment? 1. yes 2. no (please skip to question 45) 42. If yes, how many? (If more than one, I'd like you to think of the one that bothered you the most when answering the next question.) 43. Do you consider this incident to have been: 1. a mild form of sexual harassment 2. a moderate form of sexual harassment 3. a severe form of sexual harassment 44. Would you please explain this incident? (You may use the back.) 45. Please explain what you think the process at MSU is for dealing with sexual harassment claims. 46. Have you ever seen a slide tape at MSU dealing with sexual harassment? 1. yes 2. no 99 Last but not least, I just need to know about the back- ground of the participants of this study. 47. 48. 49. 50. 51. 52. 53. 54. Age: What is your race? 1. white 2. black 3. asian 4. other What is your class level? 1. freshman 2. sophomore 3. junior 4. senior How many years have you attended college? What is your overall grade point average? What is your major? Do you live on or off campus? 1. on 2. off What is your current relationship status? 1. single 2. married, 3. separated 4. divorced 5. living with someone 6. widowed 100 55. Are you originally from: 1. a rural area (like Williamston or Bath) 2. a suburban area (like Okemos) 3. a small city (like East Lansing) 4. a medium-sized city (like Lansing) 5. a large city (like Detroit or Chicago) 56. Judging from your parents' income, do you consider your family to be: 1. lower class 2. working class 3. middle class 4. upper-middle class 5. upper class I just want to thank you again for your help in this important study. Please don't leave without getting and reading the explanation of this study. APPENDIX C PILOT STUDY QUESTIONS APPENDIX C PILOT STUDY QUESTIONS The following incident happened to an undergraduate student at a large midwestern university (one other than MSU). She has consented to let her story be told for research purposes as long as her identity remains confidential. Therefore, her name has been changed for this short survey. Please read the summary of what happened to "Sara" and then circle the response below which best indicates your own opinion. There are no right or wrong answers; just your own feelings. Thank you. 1. Do you feel that what happened to "Sara" is an instance of sexual harassment? 1. no, not at all 2. yes, a mild form 3. yes, a moderate form 4. yes, a severe form 2. How believable is "Sara's" story of what happened? 1. completely unbelievable 2. somewhat unbelievable 3. somewhat believable 4. completely believable 101 3. 102 How much power does the man in the story have in com- parison to "Sara"? 1. the same degree of power as "Sara" 2. a little more power than "Sara" 3. moderately more power than "Sara" 4. a great deal more power than "Sara" APPENDIX D RESEARCH ASSISTANT'S INSTRUCTIONS APPENDIX D RESEARCH ASSISTANT'S INSTRUCTIONS Prgper Testing Procedure *Tell students beforehand that this is for one credit. 1. When student enters, introduce yourself and explain that you are a research assistant for this study. Sign the cover sheet of their booklet before handing it to them. The cover sheet and the booklet should each have the same number on them. Ask the person to read the cover sheet, sign it, and hand you the cover sheet. Explain that we do this so that their name is nowhere on the booklet and that their answers are strictly anonymous. Take the cover sheet (signed) and write the time on your Proctor Sheet under "Time begin." Sit down and read something so that you don't make the person nervous. If the subject asks a question you may clarify only mechanical problems (i.e., can they circle more than one answer for this question? what does this word mean?). DO NOT clarify content by going into any detail. Simply tell them to interpret it the best they can and to write any clarifications and comments in the borders of the booklet. Be sure to write down anything the subject says under "Comments" on your Proctor Sheet. When the person is finished, write the time under "Time end." Ask them to look over the booklet once again to make sure they answered every question. When they are done, take the booklet and put it face down on the desk to show them that you are not going to look at it. Hand them the "Explanation of Study" sheet and ask them to read it before they go so you can answer any questions. Stress the importance of not talking about this study and tell them that, while we can't answer questions about the study at this time, they can contact me sometime in the summer for more information. Please be very open and friendly, especially at the beginning, so that they will take the study seriously and not just rush through it. 103 104 6. Be sure to fill out their "credit card" before they leave. They can get one credit for this study. *Be on time if not EARLY for appointments and treat booklets with care and confidentiality while they are in your possession. APPENDIX E DEBRIEFING SHEET APPENDIX E DEBRIEFING SHEET Explanation of Study: The purpose of this study is to assess women's perceptions of and responses to sexual harassment on campus. This is a very important study--very little is known in this area. Therefore I'd like to thank you right now for your participation. I'd also like to stress that it's vital to this study that you don't tell anyone what this study was about until after spring term. If people answer this questionnaire knowing beforehand that it deals with sexual harassment, the data will be useless and the study will not result in positive changes for women on campus. Let me remind you that when this study is completed, sometime in the summer, you may contact me for more information. And thank you again for your help in this important study. Sincerely, Mary Sullivan 355-7440 105 APPENDIX F PROCTOR SHEET APPENDIX F PROCTOR SHEET Session # Condition # Time Begin: Time End: Comments: Name 3 Date: 106 REFERENCES REFERENCES Adams, J., Kottke, H., & Padgitt, J. (1983). Sexual harass- ment of university students. Journal of College Student Personnel, 484—490. Alexander v. Yale University, 631 F.2d 178 (2nd Cir. 1980). Armstrong, L. (1978). Kiss Daddy Goodnight. New York: Hawthorn Books. Benson, D. (1977). The sexualization of the teacher-student relationship. Unpublished manuscript. Benson, D. & Thomson, G. Sexual harassment on a university campus. Paper presented at the annual meeting of the American Sociological Association, New York. Benson, D., & Thomson, G. (1982). Sexual harassment on a university campus: the confluence of authority relations, sexual interest and gender stratification. Social Problems, 2. Brownmiller, S. (1975). Against Our Will: Men, Women, and Rape. New York: Simon & Schuster. Deane, N. H. & Tiller, D. L. (1981). Sexual harassment: an employment issue. Washington, DC: College and University Personnel Association. Epstein, C. (1970). Woman's Place: Options and Limits in Professional Careers. Berkeley: University of California Press. Farley, L. (1978). Sexual Shakedown: The Sexual Harassment of Women on the Job. New York: McGraw-Hill. Hall, R., & Sandler, B. (1983). Academic mentoring for women students and faculty: a new look at an old way to get ahead. Project on the Status and Education of Women. MacKinnon, C. (1979). Sexual Harassment of Working Women: A Case of Sex Discrimination. New Haven, CT: Yale Univer- sity Press. 107 108 Maihoff, N. & Forrest, L. (1983). Sexual harassment in higher education: an assessment study. Journal of the National Association for Women Deans, Administrators, and Counselors. 3S, 3-8. Martin, D. (1981). Battered Wives. San Francisco: Volcano Press. Mednick, M., Schuch, S., & Hoffman, L. (Eds.). (1975). Women and Achievement: Social and Motivational Analysis. New York: Halsted Press. Meek, P. & Lynch, A. (1983). Establishing an informal grievance procedure for cases of sexual harassment of students. Journal of the National Association for Women Deans, Administratorsy and Counselors. 4S, 30-33. Merriam, S.(1983). Mentors and proteges--a critical review of the literature. Adult EducationyQuarterly. Metha, A., & Nigg, J. (1983). Sexual harassment on campus: an institutional response. Journal of the National Association for Women DeansL Administrators, and Counselors. AS, 9-15. Munich, A. (1978, February). Seduction in academe. Psychology Todey. pp. 82-84, 108. Nelson, A. (1978, January 7-14). Sexual harassment at Yale. Nation, pp. 7-10. Patterson, M. & Engelberg, L. (1979). Women in male- dominated professions. In A. H. Stromberg & S. Harkess (Eds.), Women Working. Palo Alto, CA: Mayfield Publishing. Perry, S. (1983). Sexual harassment on the campuses: deciding where to draw the line. Chronicle of Higher Education, pp. 21-22. Reilly, K. (1978). A study of sexual harassment. Unpublished manuscript. Rose, V. (1977). Rape as a social problem: a by-product of the feminist movement. Social Problems, SS, 75-89. Safran, C. (1976, November). What men do to women on the job: a shocking look at sexual harassment. Redbook, pp. 148-149. Sandler, B. R. & Associates. (1981). Sexual harassment: a hidden problem. Educational Record, SS, 52-57. 109 Schechter, S. (1982). Women and Male Violence: The Visions and Struggles of the Battered Women's Movement. Boston: South End Press. Simon, L. & Forrest, L. (1983). Implementing a sexual harassment program at a large university. Journal of the National Association for Women Deans, Administrators, and Counselors, 4S. Till, F. (1980). Sexual harassment: a report on the sexual harassment of students. Report of the National Advisory Council on Women's Educational Programs. Washington, DC: U.S. Department of Education. U.S. Merit Systems Protection Board. (1981). Sexual harass- ment in the federal workplace: is it alproblem? Washing- ton, DC: U.S. Government Printing Office. Washington Post estimates two percent of students face faculty bribes or threats for sexual activity. (1981). On Campus with Women, SS. Working Women United Institute. (1978). Sexual Harassment On the Job: Questions and Answers. New York: author. .o\bo "7'1])1‘1)]|7])i)]]])))(