PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/01 cJCIFIC/DateDuast—pJ 5 REACTIONS TO SEXUAL HARASSMENT: A COMPARISON OF POLICEWOMEN IN THAILAND AND THE UNITED STATES By Somvadee Chaiyavej A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY School of Criminal Justice 2003 ABSTRACT REACTIONS TO SEXUAL HARASSMENT: A COMPARISON OF POLICEWOMEN IN THAILAND AND THE UNITED STATES By Somvadee Chaiyavej The study examined how female officers in law enforcement agencies in Thailand and the U.S. react to sexual harassment. It was anticipated that Thai women would hold more traditional beliefs about gender roles, have higher interdependent views of self- construa1(collectivistic orientation) and lower independent views of self (individualistic orientation). They were also expected to have lower participation in decision-making and to be less likely to agree that their work organization’s policy clearly prohibits harassment. These features were anticipated to lead to lower perceived severity of sexual harassment and higher concerns about negative social reactions in using assertive reactions, which in turn, would decrease assertive reactions in responding to harassment. The population studied was comprised of 106 female law enforcement officers from 5 police agencies in a mid-western state in the U.S. and 109 female law enforcement officers from 4 police agencies in Thailand. Each study participant had experienced an incident of sexual harassment by male coworkers within their organization at least once within the past two years. A survey instrument was used to obtain the data. T-tests were used to analyze national differences on study variables, while regression analyses were conducted to explore the relationship between independent and dependent variables separately for each group. Results suggest that Thai women had more traditional gender role stereotype beliefs, higher interdependent views of self, lower independent views of self, were less likely to view that their organizational policy prohibits harassment, but there was no difference in participation in decision-making. Although Thai women had higher concern about social reactions as hypothesized, contrary to the predictions, in handling harassment, they viewed harassment as more severe and relied more on assertive reactions than American officers. Thus, findings give limited support to the idea that Asian women would be more likely to trivialize harassment and to rely less on assertive reactions. For the American sample, those who perceived sexual harassment events as a violation of organizational policy viewed those events as more severe. Perceived severity, in turn, was accompanied with assertive reactions. Those with a high independent view of self and who felt they had high participation in decision-making felt less concerned about social reactions. Concern about social reactions, in turn, was associated with the use of passive responses. For the Thai sample, women with traditional beliefs about gender roles viewed sexual harassment as less severe and reported higher concern about social reactions. In addition, women with an interdependent view of self showed higher concern about social reactions. Nevertheless, concern about social reactions was not related to assertive reactions. The results were interpreted by drawing on qualitative data on differences in circumstances in each setting which fi’ame how sexual harassment is perceived and resolved. Copyright by SOMVADEE CHAIYAVEJ 2003 DEDICATION Dedicated to three people who have kept me focused on where I wanted to be: My mom, Dr. Morash, and Som, my sister. ACKNOWLEDGEMENTS For the completion of this dissertation, I own the greatest dept of gratitude to my chairperson, Dr. Merry Morash. Without her help, I couldn’t have completed my graduation. She has kept on encouraging and supporting me in every step all along. I continue to be amazed that there is so much patience, unwavering support and understanding within her. This is something that I will never forget. I would like to also thank all of my other three committee members, Dr. Mahesh Nalla, Dr. Linda Jackson and Dr. Ellen Kossek for their guidance and comments. Special thanks to all of the participants as well as the participating agencies in both Thailand and the U.S. for their cooperation in this research. In addition, I am especially indebted to the Royal Thai Police Cadet Academy for the financial support for my graduate program. Appreciation is expressed to Police Colonel Dr. Tamasak Wicharaya for introducing me to School of Criminal Justice, MSU. Finally, I wish to thank my parents, my two sisters, my brother, and my aunt for their patience and sacrifice. Particularly, my mom and my elder sister. They have continuously helped me in everyway they can so that I can focus solely on my dissertation. vi TABLE OF CONTENTS LIST OF TABLES ..................................................... - ................................. x LIST OF FIGURES .................................................................................. xii CHAPTER 1 INTRODUCTION .................................................................................... 1 Research Needed on Women’s Reactions to Sexual Harassment ......................... 5 Purpose of the Study .............................................................................. 6 Significance of the Study ......................................................................... 8 Basic Assumption ................................................................................ 12 Organization ofthe Dissertation............ 14 CHAPTER 2 LITERATURE REVIEW ........................................................................... 16 Historical Review ofSexual Harassmentl6 Sexual Harassment Definitions ................................................................ 19 Legal Definitions ....................................................................... 20 Scholarly Definitions .................................................................. 23 The Amount of Sexual Harassment of Women Working in Organizations. . . . .28. Victim Characteristics ................. . ........................................................ 29 Perpetrator Characteristics .................................................... ‘. . . . . . . . . ....31 Occupation/Organizational Characteristics ................................................... 33 Sexual Harassment of Policewomen .......................................................... 35 Gender Harassment .................................................................... 4O Unwanted Sexual Attention .......................................................... 43 Sexual Coercion ........................................................................ 46 Policewomen’s and Other Victims’ Responses to Sexual Harassment. . . . . . ..47 The Sociocultural Model: Gender Role Stereotype Beliefs ............................... 53 The Self-Construal Model: Interdependent View of Self and Independent View of Self ...................................................................... 61 The Organizational Model: Participation in Decision-Making. . . . . . . . . . . . 72 The Organizational Sexual Harassment Policy Model: Perceptions of Sexual Harassment as a Policy Violation .................................................. 79 Perceived Severity ............................................................................... 83 Concern About Negative Social reactions ................................................... 88 National Differences on Reactions To Harassment ........................................ 92 CHAPTER 3 RESEARCH METHODOLOGY ............................................................... 94 Participants ....................................................................................... 94 Procedure ......................................................................................... l 01 vii Instrumentation and Translation ............................................................ 102 Measures ....................................................................................... 104 Control Variables ............................................................................. l 13 Data Analysis .................................................................................. 1 15 CHAPTER 4 RESULTS ........................................................................................... 119 The Experiences of Sexual Harassment .................................................... 120 Nature of Worst Case Sexual Harassment Used For Data Analysis. . . . . . .. . ....125 Reactions to Sexual Harassment ............................................................ 129 Findings From Test of Hypotheses Regarding National Differences ................. 131 Assessment of Predictors of Perceived Severity, Concern About Negative Social Reactions and Reactions to Sexual Harassment. . . . . . . . . . ....134 Summary of Regression Results ............................................................ 146 Qualitative Findings ............................................................................ 149 Summary of Findings ........................................................................ 159 CHAPTER 5 DISCUSSION ..................................................................................... 161 Study Limitations ............................................................................. 171 Organization Implications ................................................................... 173 Recommendation for Future Research ................................................... 177 APPENDICES .................................................................................... l 82 Appendix A Informed Consent Statement (English Version) .......................................... 183 Appendix B Informed Consent Statement (Thai version) .............................................. 186 Appendix C Instruction of Sexual Harassment Experiences Questionnaire (EnglishVersion) ............................................................................. 1 90 Appendix D Instruction of Sexual Harassment Experiences Questionnaire (Thai Version) ................................................................................. 193 Appendix E Sexual Harassment Experiences Questionnaire (English Version). . . . . . .. .. . . . . ..196 Appendix F Sexual Harassment Experiences Questionnaire (Thai Version) ........................ 204 viii Appendix G . Items Assessing Gender Role Stereotype Beliefs (English Version). . .. . . . . . ..217 Appendix H Items Assessing Gender Role Stereotype Beliefs (Thai Version) ..................... 219 Appendix I Items Assessing Self-Construals (English Version) ..................................... 221 Appendix J Items Assessing Self-Construals (Thai Version) ......................................... 224 Appendix K Items Assessing Participation in Decision-Making (English Version) ............... 228 Appendix L Items Assessing Participation in Decision-Making (Thai Version). . . . . . . . . . ..230 Appendix M. An Open-Ended Question .................................................................... 232 Appendix N Demographic Questionnaire .................................................................. 234 Appendix 0 Debriefing Statements ....................................................................... 239 References ......................................................................................... 242 ix LIST OF TABLES Table 1 Percentage of US and Thai female officers experiencing eachtype of harassment at least once in two years .............................................. Table 2 Percentages of cases for each type of sexual harassment in final case ........... 121 composition by nation .............................................................................. 125 Table 3 Percentages of perpetrators of each type of harassment in final case compositionbynation......................... .. Table 4 Percentages of women who used specific type of responses to harassment ................................................................................. Table 5 127 129 Tests of differences between variable means for the two countries ........................ 132 Table 6 Intercorrelations among study variables for the U.S. sample. . . . Table 7 Predicting perceived severity and concern for social reactions for the US sample ......................................................................... Table 8 Predicting assertive and passive reactions to sexual harassment for the U.S. sample ....................................................................... Table 9 Intercorrelations among study variables for the Thai sample. . . Table 10 Predicting perceived severity and concern for social reactions for the Thai sample ........................................................................ Table 11 Predicting assertive reactions to sexual harassment for the Thai sample ....................................................................... 136 138 ........... 140 .142 ........... 144 ............ 145 Table 12 Sexual Harassment Experiences Questionnaire (English Version). . . . . . . . . . . . 1 97 Table 13 Sexual Harassment Experiences Questionnaire (Thai Version). . . .. . . . .....206 Table 14 Items Assessing Gender Role Stereotype Beliefs (English Version). . . . . . . . . . . . . . . .....217 Table 15 Items Assessing Gender Role Stereotype Beliefs (Thai Version). . .. . . . . . . .....220 Table 16 Items Assessing Self—Construals (English Version) ........................................... 222 Table 17 Items Assessing Self-Construals (Thai Version) ............................................... 225 Table 18 Items Assessing Participation in Decision-Making (English Version) ..................... 229 Table 19 Items Assessing Participation in Decision-Making (Thai Version)....... . . . . . . . . . ...23l xi LIST OF FIGURES Figure 1 . A theoretical model in the present study .......................................................... 52 xii Chapter 1 Introduction During the past four decades, there has been an increased participation of women in the labor force in both the U.S. and Thailand. Though more slowly than others, law enforcement is one of the sectors in which the female utilization trend is growing (Brown & Heidenshon, 2000). In the U.S., constant pressure fi'om the women’s liberation movement targeting sex based discrimination in employment, court mandates, and the assumption that women can bring positive qualities to policing have resulted in efforts of police departments nationwide to recruit female officers (Martin, 1993). In 1975, women represented 2.2% of the sworn personnel in municipal police department (Martin, 1980). By 1990, women’s representation was 10.6%. In 1998, 13.8% of all sworn law enforcement positions in the largest law enforcement agencies are occupied by women (National Center for Women & Policing, 1999). In Thailand, although somewhat slower, women’s representation in the police force has also increased. In 1991 , 5% of police officers were women (Savestanan, 1991). To date, approximately 7% of the entire police force are women (The Royal Thai Police Department, 2000). Against the backdrop of the increased participation of women in the police workforce, police administrators have faced a new set of problems associated with the changing demographic make up of its employees. Among the problems for women at work, sexual harassment has become an issue that has received much attention among organizational administrators as well as social scientists. It has become one of the “hottest” issue in human resource management. Although sexual harassment concerns originated in the U.S., recently many other countries have started to recognize harassment as a social "L‘s 'fif problem that warrants legal and managerial interventions. In fact, some countries go further than the U.S.; sexual harassment is treated as a criminal offense in the Philippines, Taiwan, Italy and Venezuela (Maatman, 2000a) and Mexico (Conway, 1998). In Venezuela, the passage of the Violence against Women and F arnily act in 1999 opened the possibility that harassers could go to prison from 3 to 12 months (Maatrnan, 2000b). While sexual harassment is widespread in different types of work settings, it is believed to be most common in male dominated occupations such as policing. Baker (1989) examined women employed in non-traditionally female jobs and women employed in traditionally female jobs and found that the former reported more occurrences of sexual harassment experiences. In policing, several sources suggested that the problem is pervasive and difficult to control. In the U.S., a study conducted by Martin (1990) revealed that 63% of 72 women officers reported sexual harassment on the job. In Los Angeles, women officers are harassed by organized groups of male officers called “men against women.” The groups use intimidation and criminal activity against policewomen. The department’s refusal to punish the perpetrators has resulted in the drop out of many women (National Center for Women & Policing, 1999). In Great Britain, nearly every female officer experienced sexual harassment from male officers. The data showed that 90% reported having heard sexually explicit comments or jokes about women, 60% reported being a target of offensive comments based on appearance, 30% experienced unwanted touching, and 20% reported pressure to become involved in unwanted dating (Anderson, Brown & Campbell, 1993). Many female officers who stay are forced to accept sexual harassment as a fact of life (Wong, 1984). Despite the increased attention to sexual harassment as a shared problem in policing across jurisdictions and around the world, little attention has been paid to the problem of sexual harassment in Asian police organizations. Thailand is one Asian country where it would be important to examine sexual harassment. While what has been defined as sexual harassment in the North American context and European context may occur in other countries, the absence of recognition of it as an organizational problem could act as a barrier to or inhibit social changes that challenge the legitimacy of traditional gender subordination practices. In 1997, Thailand announced the acceptance of the new constitution, which indicates a commitment to the protection of civil rights of all people (East Asian Executive Reports, 1998). Under this new constitution, a Thai citizen, regardless of his/her origin, sex or religion, shall be protected, and those whose constitutional rights and freedom have been violated can apply the provisions of the constitution in exercising judiciary rights in a court case. If future court rulings follow the path taken in the U.S., whereby sexual harassment is interpreted as sex discrimination, sexual harassment will increasingly be defined as an important issue that the Thai police organization must address. Recently, to conform with international standards in the workplace, the Labor protection Law, an amendment to the 103 Revolutionary Committee order, was enacted in 1997 (The Thai Labor Protection Act, 1998). The Thai labor law prohibits sexual harassment against women and children in the public sector. Although this law does not apply to governmental organizations, feminists’ effort to seek modification of civil servants law is currently underway (Bhatiasevi, 1998). In the meantime, being aware of the legal definitions of unwanted sexual attention and legal protection seems to have already resulted in increased awareness of civil rights from the Thai citizen including :‘\\l rim-r 9 TIE. .l I‘. ha: QAIKA 2‘- up\‘ er can» employee and non-employee. Recently, the Thai Airways International labor union pressured the government to take a boarder view of sexual harassment by passing legislation to protect flight attendants from sexual harassment by passengers (Mahithirook, 2000). Thongbai Thongpao, a legal activist, called for government officials to set a model by stopping asking female officials to serve tea and coffee (Thongpao, 1998). Kathleen Gillogly (2000), a foreigner who had resided in Thailand, upon observing the Thai society, indirectly addressed the importance of examining sexual harassment in the Thai government agencies, when she wrote to a “post bag” column in Bangkok post newspaper: Who thinks that Thailand is not know for sexual harassment? Thousands of Thai women know this, and they are as helpless to do anything about their attackers... ...Interestingly the police and the military were the most dangerous to deal with. . .The only protection a women has is to always be surrounded by other friends and relatives, but this is not feasible for women who choose to work in non-traditional jobs (p. 26). Increased awareness of women’s rights outside the Thai police organization is likely to make Thai female officers more disturbed about incidents that they often assumed as normal part of their career. Being able to label sexual harassment as deviant behavior can have many negative consequences, such as increasing the perception that there is injustice within the police organization. Because the Thai police organization has not developed any formal policies targeting sexual harassment, it is crucial to document the nature of the problem and to assess how policewomen view and respond to sexual harassment. Research Needed on Women’s Reactions to Sexual Harassment Previous research has taken several different approaches in examining sexual harassment. Studies have examined the prevalence of sexual harassment, individual’s interpretation of sexual harassing situations, individual and organizational consequences of sexual harassment, individual responses to sexual harassment experiences and organizational responses to sexual harassment. Research on individual responses to sexual harassment has lagged behind studies in the other areas. While we know with confidence from past research that sexual harassment is pervasive, distressing and subjected to wide interpretation, we know very little about how individuals react to their sexual harassment experiences (Bingham & Sherer, 1993). Most of these research on victim’s reactions was fiom the 19805 and 19905 (e.g., Thacker, 1996; Sullivan & Bybee, 1987; Summers, 1991). This is partially because earlier research has paid considerable attention to the issue of sexual harassment victimization and subsequent research has built on prior knowledge to develop a more sophisticated model of antecedents to sexual harassment victimization and its effects on work related attitudes. Moreover, due to the lack of precise definitions of sexual harassment, while there are certain behaviors that most would agree are sexually harassing, there are also certain behaviors that are subject to interpretation. Accordingly, many studies on sexual harassment still build upon how individuals come to define certain behaviors as sexually harassing and limited research has extended to individual’s reactions to these behaviors. Unfortunately, what is known about how victims react to incidents of harassment is that assertive reactions (e.g., reporting to a supervisor, confronting the harasser), which are believed to offer solutions to the problem of victims, are an exception rather than a o m: We. M.¢ Q a .4 air. 4. rule. Sexual harassment victims often feel reluctant to report or disclose the incidents because of fear of retaliation, self-blame, or sympathy with their harassers. Many female supervisors are reluctant to report harassment because of a belief that reporting might undermine others’ perception of their capacity to lead. In many cases, sexual harassment locks the targets in a losing battle. Minimization of incidents becomes a means to bear the pervasiveness of sexual harassment among victims. As a result, the literature found that individuals rarely report their experiences to authorities. For example, a study conducted by Cochran, Frazier and Olson (1997) found that only 2% of the victims reported sexual harassment to an authority, while 60% ignored the behavior. The passive responses were especially common when the harasser was someone of higher authority over the victim. However, research in this area is limited in the U.S. and has not yet extended to Thailand. Purpose of the Study The purpose of the present study is to examine whether there are any differences in the way the Thai and American female officers interpret sexual harassment and react to it. Specifically, the study explores how two individual differences, gender role stereotype beliefs and self-construals, and two organizational context factors, participation in decision-making and perceptions of sexual harassment acts as a violation of organizational policy, influence women’s reactions to sexual harassment by a male oflicer. These four variables are expected to differ for women in Thailand and the U.S. and influence their perceptions of incidents as serious and their concern about negative social reactions for responding negatively, which in turn are expected to influence their behavioral reactions, which include direct actions (e. g., reporting to an authority, confronting the person) and indirect actions (e. g., ignoring the incident, going along). Although the literature on victim’s responses to sexual harassment has not yet examined differences in how Asians and Americans evaluate and react to sexual harassment, pieces of information based on previous literature suggested the strong possibility that their interpretations and reactions to the event would differ significantly. Theoretical explanations of sexual harassment tend to highlight features sometimes associated with Asian culture and that would make it unlikely that the women would react assertively to sexual harassment. These features include values and norms that discourage women fiom acting assertively, such as the clear differentiation of women’s from men’s social roles (Crittenden, 1991) and the cultural value of being humble and not being assertive in order to preserve harmonious relationships with others (Chu, 1987; Markus & Kitayama, 1991). Moreover, organizational arrangements in the two countries differ with regard to the hierarchical structure of decision-making and the existence of sexual harassment policy. These two organizational differences are likely to influence women’s reactions to sexual harassment. In Thailand, the organization is characterized by a high degree of vertical differentiation, in which a rigid conception of order is emphasized, and there is no formal prohibition against sexual harassment in organizational policy. Thus, it is reasonable to expect that these two features of the organization managerial practices additionally may work to disadvantages the Thai female officers. In sum, the main research question is whether Thai and U.S. women police differ in how they handle sexual harassment. The subsidiary research questions are as follows: 1. Do the Thai and the American female officers differ in their levels of beliefs in gender role stereotypes? 2. Do the Thai and the American female officers differ in self-construals in terms of interdependent and independent views of self ? 3. Do the Thai and the American female officers differ in their perceptions of participation in decision-making? 4. Do the Thai and the American female officers differ in their perceptions of sexual harassment acts as organizational policy violations ? 5. Ifthere are differences in the way Thai and U.S. female officers handle sexual harassment, to what extent can these differences be explained by beliefs about gender role stereotype, self-construals, perceived participation in decision-making and perceptions of sexual harassment as organizational policy violations? 6. Are the effects of the four independent factors-beliefs about gender role stereotypes, self-construals, perceived participation in decision-making and perceived sexual harassment as organizational policy violations-- on individuals’ responses to sexual harassment incidents mediated by perceived severity of the events and concern about negative social reactions? Significance of the Study There are three main contributions from the present study. First, in the police occupation, in particular, there has been limited research on victims’ responses to sexual harassment. Our current knowledge regarding the influence of police organization policies, climate and procedures on the women who have been harassed is unclear. In a setting where female officers are embedded within a male dominated environment, especially when the harasser is of higher organization status and has more extensive social networks, women’s assertive responses to sexual harassment might be much more risky, compared to their assertive responses in a gender mixed context such as university settings or business organizations. Research is needed to explore whether the factors that make it possible for women in other settings to tell others about their experiences of harassment would be sufficient to enable female officers to disclose theirs. As such, understanding women’s perceptions and reactions to sexual harassment can be beneficial for police administrators as a basis for designing curriculum to sensitize internal employees about certain constraints, fi'om women officer’s perspectives, facing the victims in handling sexual harassment. Second, most studies of sexual harassment and policing have been limited to western populations. More recently, there has been an increasing concern among scholars and practitioners about the scarcity of sexual harassment research and women police outside the U.S.. Comparing the American and the Thai female officers should not only be particularly revealing in providing information on whether the explanatory power of these theoretical explanations hold true in both contexts, but also providing information on whether similar or different sets of interventions would be required. Many Asian police departments have adopted western styles of management practices, based on western cultural assumptions. On the one hand, there might be differences in psychological barriers of women with different cultural background. Sensitivity to cultural differences in reactions to sexual harassment is a must for making the problem of victim’s behaviors more understandable and controllable. On the other hand, it would be premature to expect that there should be differences. As Morash, Hoffman, Lee & Shim (1999) stated, ' it is very important to avoid accepting stereotypes or assumptions about Asian women and about the Asian culture, as culture is not static, but over time there are many forces that influence culture, such as legal changes. Without testing stereotypic views of culture, it is impossible to know if assumptions about culture hold true. Police administrators may be misguided and focused on the wrong kinds of intervention in encouraging victims to cope with sexual harassment. The findings will not only be valuable for police organizations in the east and in the west, but will also be informative for organizations that operate overseas and organizations that utilize both Asian and U.S. women in their workforce. Third, the administration of the questionnaires describing different types of sexual harassing experiences is itself a potential way of raising awareness among female officers about many workplace problems they and their fellow officers may encounter collectively. They will have had a chance to ask themselves what they would do if those situations actually happened to them or to their coworkers in the future. Giving them prior knowledge about these situations may help them develop in advance a more prepared way of responding. Because social support in the police environment of female officers often largely depends on female coworkers, due to the human preferences to associate with similar people and women’s exclusion fiom male coworker’s social networks, exposure to these situations may also help them to be more sympathetic to other coworkers who have encountered sexual harassment. Exposure to the experiences of others might also enable them to reflect more on whether certain things they have tolerated in the past should be accepted without challenge. With the increased organizational reliance on female officers, continued research on sexual harassment is important. This is because it involves organizational loss stemming 10 from the inability of organizations to make use of different attitudes, skills and abilities women bring into the workforce and at the same time, the cost of turnover associated with sexual harassment which involves cost in recruiting and training. Sexual harassment has been associated with multiple adverse organizational and individual-level outcomes. At the organizational level, victims of sexual harassment had lower work attendance, as many are more likely to call in sick and absent from work (Gutek & Koss, 1993, Loy & Stewart, 1984). The cost of absenteeism has been estimated to be nearly 9 million dollars in 1988 based on results of a study of U.S. Army personnel and officers (Farley, Knapp, Kustis & Dubois, 1994). For federal employees in the U.S. from 1985 to 1987, it was estimated that the cost of turnover due to sexual harassment is approximately 3 36.7 million, because 36, 647 federal workers quit their jobs because of sexual harassment (USMSPB, 1988). Sexual harassment has also resulted in a reduction of the quality of work and the ability of women to work with others (Gruber & Bjorn, 1982). Victims of sexual harassment exhibited poor attitudes towards the work group (Fitzgerald & Shullrnan, 1993), and many isolate themselves fiom coworkers due to feeling of distrust and betrayal (V aux, 1993). They also experience decreased organizational commitment (Schneider & Swan, 1994) and job satisfaction (Baker, 1989; Watsi, Bergman, Glomb & Drasgow, 2000). Police organizations also suffer the costs of decreased morale and performance when sexual harassment occurs. There is a cost due to limits on the organization’s ability to utilize the skills of female officers. In the U.S., many women officers decreased interest in patrol assignments, management positions and high career goals (Martin, 1980). Litigation costs of sexual harassment are enormous due to the loss of work hours, and disruption of work (Gutek, 1985). Moreover, punitive damages and 11 remedies associated with back pay, and attomey’s fees, are additional consequences of sexual harassment that organizations can suffer. Apart from economical rationale, police organizations could be held morally reprehensible for failure to do something about sexual harassment, since the literature has found several negative psychological and physical consequences associated with harassment. For example, potential psychological impacts on female victims are decreased self-esteem, anxiety, depression, anger, frustration and stress (Crull, 1982; Gutek, 1985; Gutek & Koss, 1993; Koss, 1990). Victims of sexual harassment also reported many physical symptoms including nervousness, teeth grinding, jaw tightness, nervousness, sleep disturbance, nausea, headaches and loss of appetite (Gutek, 1985; Loy & Stewart, 1984, Salisbury, Ginorio,Rernick, & Stringer,l986). These physical symptoms may be correlated with poor coping such as the use of drugs and alcohol, which can further deteriorate women’s medical conditions and involves medical costs. Additionally, many men who work in organizations reported feelings of paranoid that anything they do with female coworkers may be interpreted as sexual harassment (Markert, 1999). Accordingly, many men may minimize their interaction with female coworkers, and limit sponsorship to only male coworkers, due to concerns about sexual harassment accusations. Because of the importance of sexual harassment, more research is needed to understand what police administrator can do to prevent and contain the problem of sexual harassment. Basic Assumption There are many definitions of sexual harassment in the literature. For the purpose of this research, sexual harassment will be defined based on the typologies of the tripartite 12 model of sexual harassment, which has been developed by prior work in the United States (Fitzgerald et al. 1988). In this model, sexual harassment is viewed as consisting of three forms of behaviors: gender harassment, unwanted sexual attention and sexual coercion. Gender harassment involves behaviors, either verbal or nonverbal that convey insulting, hostile and degrading attitudes about women, while unwanted sexual attention involves seductive behaviors and sexual offenses. Finally, sexual coercion involves solicitation of reward or threat of punishment. Although details about different definitions will be presented in the next chapter, it should be noted now that this definition of sexual harassment was developed based on qualitative data of students and employees in the . workplace who have described the sexual harassment events. This definition was selected because first it was developed from victims’ perspectives, so it provides a broad range of verbal and nonverbal actions that have direct link to organizational outcomes such as potential conflicts in potentially sexual harassing situations, and work-related outcomes. Second, this definition also touches upon the legal definition of sexual harassment in the U.S. (Equal Employment Opportunity Commission, 1980), which views sexual harassment as consisting of two types: hostile environment, defined as offensive sex- related behavior that have the effect of leading to hostile environment and quid pro quo, which involves coerced sexual exchange. Gender harassment and unwanted sexual attention, therefore were parallel to the concept of hostile environment, whereas quid pro quo was parallel to the concept of sexual coercion. Accordingly, it is practical in terms of legal implications (Fitzgerald, Swan & Magley, 1997). A weakness of the use of the American’s social consensus definitions is that there may be some forms of socio-sexual interaction defined by the Thai sample as harassing that 13 may be missing from the analysis. However, given that first, the present study can be considered as an exploratory study on sexual harassment in Thailand, without prior work on the Thai women sample, prior conceptualization of sexual. harassment based on the literature developing in the U.S., which had been conducted more extensively should provide an authoritative baseline of information on which fiirther exploration can build. Moreover, theoretically, sexual harassment is expected to be perceived to include a broader range of phenomena in a country where women’s emancipation is higher. Accordingly, the American conceptualization should also allow us to explore a more exhaustive list of behaviors, which will be valuable in exploring sexual harassment in various forms. Organization of the Dissertation This dissertation is organized into five chapters. The first chapter, which is the present chapter,'provides the rationale for investigating policewomen’s responses to sexual harassment. Chapter two is divided into four subsections. The first section provides a discussion about the historical development of sexual harassment and the definitions of sexual harassment in the legal and academic context. The second section provides a discussion of the nature of sexual harassment. Questions addressed are: Who are the victims? Who are the perpetrators? and Where is sexual harassment most likely to occur? The third section discusses what has been done regarding sexual harassment in police setting. The forth section focuses on the issue central to the current study, which is individual’s responses to sexual harassment. In chapter three, the methodology used for investigating female officer’s reactions is presented. The results of the study are reported in the forth chapter. Finally in chapter five, the discussions about the findings, policy 14 implications, limitation of the research and recommendation for firture research are presented. 15 Chapter 2 Literature Review Historical Review of Sexual Harassment Although the term sexual harassment has been adopted not long ago, indeed it is believed that sexual harassment existed and was experienced throughout human history (Geare, 1998). Nevertheless, in modern times, sexual harassment as a turified concept has emerged due to concerns that it is a key obstacle to women’s work motivation. The recognition of heterogeneous phenomenon covered by the rmified term “sexual harassment” first appeared in the U.S. in the context of the women’s movement in the 1970’s (Mackinnon, 1979). Despite the increased numbers of women in the workforce, women still received the lowest paying and lowest status jobs. Many social activists probed into various workplace difficulties that are responsible for sex segregation at work. Among many obstacles, unwanted sexual attention was recognized as one of the key barriers, as it is found to be widespread and extensive throughout the history of women’s participation in the labor force (Bularzik, 1978). However, it was not until 1975 that The Working Women United Institute, an organization created to promote the well-being of women in the workforce, adopted the term “sexual harassment” for the first time (MacKinnon, 1979). The challenge to behavior that had been considered as normative occurred in connection with a case of a woman named Carrnita Wood, who sought redress because she had to leave her job due to sexual advances from her superior (Fitzgerald, 1990). In 1979, Catharine Mackinnon released a book providing arguments that sexual harassment should be viewed as a form of sex discrimination for working women (Mackinnon, 1979). The adoption of the term also resulted in some initial efforts 16 in identifying the prevalence of sexual harassment in the workplace. The pioneering survey was conducted by Working Women United Institute in New York (MacKinnon, 1979). In this survey, 55 working women in the Ithaca area were asked about their sexual harassment experience at work and their relationship with the harasser. It was found that 40% reported being harassed by a superior, 22% reported being harassed by a coworker, 29% were harassed by a client or person who was not in their organization, 1% was harassed by subordinates, and the remaining 8% were harassed by others. This study, while not providing extensive details, did highlight the prevalence of sexual harassment, particularly acts in which superiors harassed subordinates. Another study conducted by the Redbook magazine based on the responses fi'om more than 9000 female readers found that 88% had experienced some form of sexual harassment and 92% viewed it as a serious problems (Beirne & Messerschmidt, 1995; Lindsey, 1977; Oshinsky, .1980). Although this sampling was based on self- selection, the result of the study was particularly revealing about the extent to which the problem has become visible among women in the U.S.. In 1976, the United Nations Ad Hoc Group on Equal Rights for Women examined sex discrimination of its employees. The survey asked 875 employees whether they had experienced overt or subtle sexual pressures by persons in positions of authority. About half of the female respondents answered affirrnatively (Beirne & Messerschmidt, 1995; Farley, 1978). Along with the development of societal awareness of sexual harassment as a social problem, previously passed legislation, which had emerged earlier, became grounds for sexual harassment prohibition in the Civil rights Act. In 1964, Congress passed new employment legislation (Conte, 1997). Title VII of the Civil Rights Act of 1964 was 17 enacted to prohibit all forms of discrimination on the basis of sex, race, color, religion and national of origin in the workplace. This title stated that it is unlawful to refirse to hire or discharge, or discriminate against any individual in compensation, conditions, or privileges of employment or to limit, segregate or classify individuals in a way which may deprive them of employment opportunities. However, this law was not applied to sexual harassment right at the beginning. In the early 19703, the courts viewed sexual harassment as a personal matter, which did not constitute a civil rights violation based on sex and did not lie under the responsibility of the employer (Livingston, 1982). It was in 1976, that societal awareness of sexual harassment was heightened by the recognition of sexual harassment as a form of sexual discrimination in an actual court case. In Williams, Vs. Saxbe (1976) the federal district court first ruled that a female justice department employee who was fired due to her refusal of her supervisor’s sexual advances could bring the case to trial under sexual discrimination allegations (Paley, 1982). In 1986, in Meritor Savings Bank, FSB Vs. Vinson, a case where a black female bank teller was fondled and raped during and after work by a supervisor, the Supreme Court ruled that sexual harassment as sexual discrimination was guaranteed by Title VII of the 1964 Civil Rights Acts and proof of economic loss is not necessary (Paetzold & O’Leary—Kelly, 1993). In Robinson Vs. Jacksonville Shipyards, the presence of obscene pictures and degrading remarks by male employees resulted in the court-imposed sexual harassment policy (Connell, 1991). Ever since, sexual harassment has become a topic of interest to many parties including organizational administrators, psychologists, sociologists, and the legal professions, because legal ramifications had confirmed the significance of sexual harassment. 18 Although the first legal ruling regarding sexual harassment in the mid 19705 had resulted in an interest in the topic in the late 19705, to date interest in sexual harassment doesn’t seem to be diminished but on the contrary, continues to grow. In the early 19805, the Journal of Social Issues devoted an entire volume to sexual harassment. The significance of the topic still remained in the following decade. In 1993, the Journal of Vocational Behavior also devoted an entire volume to sexual harassment. The interest in sexual harassment in the last decade has extended to concerns by the academic world when certain notorious cases have been portrayed by the media. In 1991, the Anita Hill/Clearance Thomas hearings, a case involving a justice of the U.S. Supreme Court allegedly sexually harassing Anita Hill, his former coworker, raised many questions about sexual misconduct even among highly educated, well-trained professions (Shelton, 1999). In a later year, in 1992, a scandal involving several male naval officers who pinched, grabbed, and fondled female naval aviators at a convention, sexual harassment again became a focus of the public (Beirne & Messerschmidt, 1995). The escalating numbers of complaints filed for sexual harassment in the last few years and the increased legal costs associated with compensation awarded by juries for sexual harassment victims demonstrates that sexual harassment has not yet been a settled issue, and still requires attention from organizational administrators. Sexual Harassment Definitions Despite the facts that there is a consensus that sexual harassment is a serious social and organizational problem and it has received an explosion of inquiry, it appears that one important difficulty in examining the sexual harassment phenomena lies in the definition of sexual harassment itself. Fitzgerald (1990) noted that the definition of sexual l9 harassment on the one hand, must be broad so that it could cover a variety of experiences to which the construct refers, but on the other hand it requires specificity for practical use. In her view, the conceptualization and measurement of sexual harassment is still in a rudimentary stage. Unsurprisingly, several studies have adopted different definitions of sexual harassment, which has generated varied estimates of sexual harassment prevalence, and has limited theoretical development. This also stimulated many studies to search for lay definitions of sexual harassment. These varied views nevertheless have been part of the challenging aspects of attempts to tease out the complexities of the real world phenomena that depend on individual’s subjective interpretation based on their dispositions and situational context. Nevertheless, the shared consensus seems to be that no matter how disarrayed our knowledge in this area is, it is better than leaving the issue unnamed and unexplored. Legal Definitions One approach to defining sexual harassment is to use a legal definition. In compliance with the Civil Right Act, The Equal Employment Opportunity Commission established a formal guideline to prohibit sexual harassment, which is considered as unlawful, sex- based discrimination (Equal Employment Opportunity Commission, 1992). It stated that employers have to maintain a workplace free of sexual harassment and that employers can be held liable for sexual harassment. This guideline made a distinction between sexual harassment by supervisory personnel and other parties such as coworkers, visitors or clients. It indicated that if sexual harassment involves the acts of supervisory employees, the employer can be liable regardless of whether the acts were authorized or forbidden by the employee and regardless of whether the employer knew or should have known about 20 the acts. Ifthe harasser is a coworker or a non-employee, employers are held responsible only when they knew or should have known of the harassment and failed to take appropriate corrective action. In order to clarify what is sexual harassment, the commission also provided a formal definition. This definition is important because the Supreme Court has adopted this view as an interpretive guideline. Specifically, the Equal Employment Opportunity Commission has established guidelines on sexual harassment on part 1604.11, Title 29 of the U.S. code, which states that sexual harassment is Unwelcome sexual advances, requests for sexual favors, and other verbal, or physical conduct of a sexual nature constitute sexual harassment when (l) submission to such conduct of a sexual nature is made either explicitly or implicitly a term or condition of an individual’s employment, (2) submission to or rejection of such conduct by an individual is used as the basis for employment decisions affecting such individual, or (3) such conduct has the purpose or effect of substantially interfering with an individual’s work performance or creating an intimidating, hostile, or offensive working environment” (Equal Employment Opportunity Commission, 1992, p. 203). There are three main points that may be gleaned from this definition. First, sexual harassment can be in either verbal or physical forms. Second, the term “unwelcome” highlights that there exist welcome sexual advances, welcome sexual favor and welcome verbal and physical conduct of a sexual nature which would not be considered as sexual harassment. Accordingly, the definition seems to emphasize the requirements that the target’s subjective interpretation that the experience as unwanted together with their reactions that reflect such interpretation is necessary. The third point is that this guideline 21 has been widely interpreted as viewing sexual harassment as consisting of two forms, in which section (1) and section (2) is labeled as “quid pro quo sexual harassment, while section (3) has been labeled as “hostile environment sexual harassment” (Swanson, Territo, & Taylor, 1993). Quid pro quo involves harassment that is more overt and directed at particular individuals, consisting of a sexual act that is imposed on individuals for the exchange of a job benefit. In this type of sexual harassment, sexual favors may he demanded or proposed in exchange for selection, assignments, promotions, dismissals, or withholding job benefits. This type of sexual harassment often involves supervisor and subordinate because supervisors can use power associated with their positions to perform these acts. For example, if the applicant is not hired because of her/his refusal to engage in sexual acts with the interviewer, this could be considered as quid pro quo liability. In this case it is a plaintiff’s burden to establish links between such a condition and sexual demands. Hostile working environment sexual harassment involves boarder forms of sexual harassment such as lewd jokes, sexist comments, displaying of sexually suggestive materials or repeated requests for sexual favors. In order to establish this form of harassment, the act needs to be pervasive and repetitive (Conte, 1997; Van Wormer & Bartollas, 2000). While the concept of sexual harassment based on a legal definition is worded at a high level of abstraction so that it could cover many types of incidents that may occur at work, this guideline is too limited for practical purposes. Because in real life the incident is ambiguous and allows opinions to influence perception of the event, different court decisions have interpreted the same case differently. This limitation may be less salient in quid pro quo sexual harassment, but becomes more evident in hostile environment 22 sexual harassment (Y oder & Aniakudo, 1996). In the former, there seems to be shared consensus of what experiences can be evaluated as sexual harassment because there is specificity in the nature of social exchange between the actor and the target, with career repercussions if targets do not comply. On the other hand, the guideline language on hostile environmental climate provides too vague a definition for interpreting what specific behaviors can be viewed as constituting a hostile environment. Scholarly Definitions Because there is no clear cut criteria regarding how sexual harassment should be defined, in examining sexual harassment phenomena as a part of workplace problems, many researchers have developed their own definitions of sexual harassment. While most of these definitions were similar to the legal definition in adopting a general view that harassment must be unwanted, unsolicited, and one-sided (Farley, 1978; Lafontaine & Tredeau, 1986), several authors also emphasized the unequal power in the relationship of the harasser and the victim. For example, putting the emphasis on the nature of sexual harassment that involves women as the harassed, Farley (1978) viewed sexual harassment as “unsolicited non-reciprocal male behavior that asserts a women’s sex role over her function as worker’ ’(p. 14). Consistently, Lafontaine and Tredeau (1986) viewed sexual harassment as “any action in the workplace whereby women are treated as objects of the male sexual prerogative and all such treatment is seen to constitute sexual harassment, regardless of whether the victim labels it as problematic” (p. 435). Other scholars emphasized the consequences of sexual harassment to the victims. For example, Skrocki (1978) viewed sexual harassment as “any sexually oriented practice that endangers a women’s job-that undermines her job performance and threatens her economic 23 livelihood. . .” (p. 43). These types of definitions describe the nature of the behavior of sexual harassment which is formed based on the theoretical orientation of the researcher, and accordingly they trade off the opportunity to take a social consensus view for the specificity of the phenomena investigated. Because the definition of sexual harassment at a high level of abstraction may be ambiguous for practical purpose in investigating sexual harassment, many studies have explored how lay people come to perceive “socio-sexual behaviors” as sexually harassing (Shelton, 1999; Sperry & Powell, 1999). Social-sexual behavior has been defined as non- work related interaction having a sexual component (Gutek, Cohen & Konrad, 1990). These behaviors may range from ambiguous to blatant acts that few people would consider as sexual harassment such as positive verbal comments about appearance to gross sexual imposition (Sperry & Powell, 1999). In this approach, the researcher provides respondents with a series of brief stories or vignettes of social-sexual behaviors of interests and asks them to make a judgment regarding observer’s perceptions of the behaviors (e.g., offensiveness, appropriateness), and/or whether they view the behavior as sexually harassing. In some cases, vignettes were manipulated so that the effects of situational factors can be examined. This approach has the advantage of providing information on what situational factors or individual factors might influence perceptions of sexual harassment. Because self-report sexual harassment victimization data is tied to subjective interpretations of the experiences, it is unknown to what extent respondents who might have had been recipient of what others considered as “sexual harassment” did not report it because they did not even recognize it as such. Fitzgerald et al. (1988) found that many women who had experienced blatant instances of sexual harassment failed to 24 It I," (Iv recognize their experience as sexual harassment. Thus, the scenario based approach provides complimentary information about the decision making process of the targets as well as observers that may not otherwise be captured by a survey of actual experiences. However, the weakness of this approach is that because it is not possible to examine all types of incidents within any single study, the degree of generalizability to others forms of sexual harassment may be limited to the vignettes examined. Moreover, because of the nature of people’s responses to written scenarios, the extent to which it could be generalized to real life situations is viewed as another disadvantage. Beside the use of social situations, others scholars prefer to use lists of more specific and more concrete behaviors to define sexual harassment. Initially, research that used behavior-based definitions started out by directly asking the respondents about their sexual harassment experience, and left it to victims to identify harassment. Later on, there was concern that some experiences that could be viewed as harassment, but were not construed so by the respondents would be left unexplored. In dealing with the problem, two approaches have been taken. In the first approach, the researcher develops a list of acts that individuals may encounter at work that are sexual in nature, all of which are assumed to be undesirable and are considered sexual harassment by researcher. These behaviors were then provided to respondents. For example, Adams, Kottke, and Padgitt (1983) categorized sexual harassment into 8 categories: sexist comments, undue attention, body language, verbal sexual advances, invitations, physical advances, explicit propositions and sexual bribery, and determined whether the respondents had experienced the behaviors on the list, rather than directly using the term sexual harassment. Another approach is to develop a social consensus view of sexual harassment 25 behaviors. In this approach, a sexual harassment definition is derived from empirical data based on the public’s definitions. Till’s pioneering efforts (Till, 1980) are a major contribution to this approach. Using data based on qualitative interviews about the experiences of college women students and university staff, Till (1980) developed very extensive categories of sexual harassment. In this work, the respondents were asked if they had ever experienced sexual harassment and then if they said so they were asked to describe the events. The content analysis of the data based of the events reported yielded 5 categories of sexual harassment: gender harassment, seductive behaviors, sexual bribery, sexual coercion and sexual imposition. Gender harassment involves sexist remarks, and behaviors involving insulting, hostile and degrading attitudes towards women. Seductive behaviors includes inappropriate and offensive behaviors such as pressure for dates, suggestive looks, and whistles but no threats for noncompliance. Sexual bribery involves sexual activity demanded in exchanges for rewards such as promotion. Sexual coercion involves sexual activity demanded with a threat of punishment, such as demotion, in order to gain compliance. Finally sexual imposition involves attempted or completed acts of touching, fondling, grabbing, kissing and rape. Till’s classification system has been widely accepted and adopted by many studies (Fizgerald et a1, 1988; Welsh, 1997). In an attempt to establish the content validity of an instrument to measure sexual harassment, building on Till’s five typologies of sexual harassment, Fitzgerald and et al. (1988), developed a SEQ questionnaire (Sexual Harassment Experience Questionnaire) and ran a pilot study with 468 students at a university in the mid-west. Based on the feedback of these students, the instrument was then revised to include 28 items. The 26 inventory was later administered to a new sample of 1395 university students. Their results identified three rather than five factors. Sexual bribery and sexual coercion were collapsed into one factor, while seduction and sexual imposition was also collapsed into one factor. This resulted in three factors labeled as gender harassment, unwanted sexual attention and sexual coercion respectively. Recently, the SEQ has also been validated cross-culturally and across settings (Fitzgerald, Gelfand & Drasgow, 1995). The tripartite model of sexual harassment to date is the most widely use in the sexual harassment studies. It seems evident that to date conceptual definitions and operational definitions of sexual harassment are diverse and far from satisfactory, as each has their own limitations. For the purpose of the present study, the definition of sexual harassment will be adopted from Fitzgerald (1990) and Fitzgerald, Swan and Magley (1997), which states that. Sexual harassment consists of sexualization of an instrumental relationship through the introduction or imposition of sexist or sexual remarks, requests or requirements, in the context of a formal power differential. Harassment can also occur where no such formal differential exists, if the behavior is unwanted by or offensive to the women (Fitzgerald, 1990, p.38). Instances of harassment can be classified into the following general categories: gender harassment, unwanted sexual attention (seductive behavior and sexual offenses) and sexual coercion (solicitation of sexual activity by promise of reward or threat of punishment and sexual imposition or assault) (Fitzgerald et al., 1997). While this definition of sexual harassment may not be perfect in capturing all forms of sexual harassment (e. g. it does not address male victims), it integrates both theoretically 27 based-definitions and empirical-based definitions of sexual harassment that result in a middle range approach to define sexual harassment. Accordingly, it is not too board or too specific for research purposes. The definition allows integration of the social-sexual behavior approach by providing guidance about the different types of behaviors that should be considered in research. Attention will be now turned to the circumstances surrounding sexual harassing phenomena in the workplace. The Amount of Sexual Harassment of Women Working in organizations Although researchers have not yet reached consensus about definitions of sexual harassment, several studies have attempted to establish the pervasiveness of sexual harassment in the workplace among working women in various organizations. Generally, the pervasiveness of sexual harassment victimization is difficult to establish because of the wide variety of measuring technique and different definitions of sexual harassment. Unsurprisingly, there are conflicting views in estimating the incidence. According to Gruber (1990), the prevalence of sexual harassment among women, including employees and students ranges between 40% to 90% in different studies. The components which often differ across studies are the operational definitions of sexual harassment used (e. g., providing victims with abstract definitions of sexual harassment, the mention of the term sexual harassment or providing lists of specific behaviors), sampled population (blue collar workers, pink collar workers), sampling methods (self-selection or not), time frame (e.g., 12 months, 24 months, working career), and the perpetrator’s status (e. g. , whether supervisor, coworker, clients included). Nevertheless, these studies identify some common findings about who is at risk, who are the offenders, and the settings in which harassment is likely to occur. 28 Victim Characteristics A consistent finding in sexual harassment research is that women are most likely to be targets (Gutek, 1985; Maypole & Skaine, 1982). A study conducted by Gutek (1985) interviewed 827 working women and 405 working men whose residential phone numbers were listed in Los Angeles county, and asked them whether they had experienced sexual harassment in their entire working careers. It was reported that 53.1% of female respondents and 37% of male respondents had experience sexual harassment. Maypole and Skaine (1982) examined the sexual harassment experiences of 164 male and 160 female workers in manufacturing plants by asking them if they had been sexually harassed. Thirty-six percent of the female respondents and 8% of the male respondents responded positively the question. In general, the relationship between other demographic characteristics and sexual harassment vulnerability revealed certain profiles of female victims, as women in lower positions, single and young and are at higher risks. For example Bularzik (1978) found that during the 19th and 20th centuries, most victims were unskilled workers who held low status jobs. Another study noted that single uneducated females holding menial jobs who had a child to support are prone to sexual harassment victimization (Tangri, Burt & Johnson, 1982). In typical profiles, victims were between the ages of 20-35 and were single or divorced, and held non-supervisory positions (Baroni & Petrini, 1992). However, women in authority also experienced sexual harassment, but they were subjected to less blatant forms of sexual harassment such as verbal remarks, while those who held lower positions may be subjected to both physical and verbal harassment (Backhouse & Cohen, 1981). It should be noted that sexual harassment appears to be 29 common, regardless of women’s physical attractiveness; many working women believed that unattractiveness did not decrease the likelihood of being a target of sexual advances on the job (Collins & Blodgett, 1981 ). Other demographic characteristics such as race/ ethnicity, level of education and organizational tenure showed inconsistent patterns. Some studies indicated that harassment was found to be more common among black women, than white women (Gruber & Bjorn, 1982; Mansfield et a1, 1991), but others found that whites are more at risks than blacks (Niebuhr & Boyles, 1991). Some forms of sexual harassment may be influenced by a groups’ racial composition that produces complex intergroup dynamics unique to the setting. Tenure effects also produced mixed findings, which may be due to the different types of population examined. A study of sexual harassment among military personnel and police officers found that those female officers who had lower tenure were more likely to be harassed (Wong, 1984), but a study conducted with women who worked in the U.S. telephone industry reported that women with higher tenure were more at risk than women with lower tenure (De Coster, Estes & Mueller, 1999). Given that military- like organizations tend to value seniority, it is possible that women with low organizational tenure may be at high risk for being harassed. Education also provides mixed results; Bails (1994) found that education had no effect on sexual harassment victimization, while De Coster, Estes & Mueller (1999) reported that women who had high education are more likely to be at risk for sexual harassment. Part of the inconsistency may stem fi'om the fact that education may influence risk of both victimization and perceptions of sexual harassment (if one interprets the situations as sexual harassment) at the same time, which makes it difficult to estimate sexual 30 harassment victimization. Women with high levels of education may be at less risk for being harassed, but may be more sensitive to sexual harassment than women with less education, which may have resulted in the null effect. Nevertheless, the inconsistency of these demographic characteristics and likelihood of victimization seems to suggest the possibility that women, regardless of their demographic characteristics, seems to be at risks for sexual harassment. Perpetrator Characteristics In an effort to better understand sexual harassment, researchers have also attempted to identify perpetrator characteristics. The study by the Merit Systems Federal Protection Board found that 95% of women who were sexual harassed reported that their harassers as a man and that most harassers tend to act alone, rather than in group. These harassers were somewhat older; approximately 68% of women sexually harassed in federal workplace are younger than their harassers (U SMSPB, 1981). The modal age of the harassers was between 40—49 (Gutek, KaKamura, Gahart & Handschurnacher, 1980) and had below average physical attractiveness fi'om the perspective of the victims (Gutek, 1985). A majority of harassers are married and are more likely to have a reputation for harassing other women at work (Gutek, 1985; USMSPB, 1981), as 43% indicated being aware of their harasser’s bothering others, while 3% of the women respondents were not sure if their harassers also harass someone else (U SMSPB, 1981). Most victims had known the harassers for quite some time before the incident; as Gutek’s data (1985) indicated, approximately half of the victims have known the harassers for over six months before sexual harassment took place. The findings regarding the relationship of the harassers and the victims appeared to be 31 mixed and provide a less clear cut picture. Earlier study found that supervisor harassment constituted the primary type of sexual harassment. For example, the study of sexual harassment in New York conducted by The Working Women United Institute in New York (MacKinnon, 1979) found that 40% reported being harassed by a superior, 22% reported being harassed by a coworker. Later studies tend to reveal that a majority of the harassers do not tend to have formal authority over the victims. A study conducted by the National Merit Systems Protection Broad (1981) reported that only 37% reported being sexually harassed by their immediate supervisors or higher level supervisors, while the rest involved other types of relationships. Consistently, Gutek (1985) found that coworkers constitute a majority group of harassers; 56% of the victims reported being sexually harassed by a coworker. Perhaps, the relationship between status of the harassers and sexual harassment may reflect the changing societal definitions of sexual harassment. In earlier period, sexual harassment interpretation may be based on narrow definitions of sexual harassment, in which more severe forms of sexual harassment constitute a core definition. Quid pro quo sexual harassment, for example, tends to involve supervisor initiated sexual harassment because positions of formal authority play a significant role in the use of threats or premises that affect job conditions. In the later period, when sexual harassrnent’s definitions have been broadened to include different kinds of behaviors including gender harassment, respondents may report more behaviors that were enacted by coworkers as sexually harassing. Moreover, from the probability statistics, there are more coworkers than supervisors, hence, under a boarder definitions of sexual harassment, coworkers may be more likely to constitute a majority of sexual harassment perpetrators. However, it should be pointed out that although sexual 32 harassment by supervisors did not constitute a majority, sexual harassment by supervisors tends to constitute a more severe form of sexual harassment which includes actual or attempted rape and sexual assault (U SMSPB, 1981). Consistently, the severity may also have more repercussions for victims. A review of the data on sexual harassment complaints (Coles, 1986) and the history of women who sought therapy found that most of these women who file complaints and sought therapy respectively had been harassed by a male supervisor (Crull,l982). Occupation/Organizational Characteristics Studies have also explored sexual harassment victimization across settings. Women whose work is in male setting where there is increased opportunity for contacts with men reported higher experiences of sexual harassment than women in gender neutral setting. A study at Pennsylvania State University examined sexual harassment of women in three jobs; tradeswomen, transit workers and school secretaries. Tradeswomen, which is a traditionally male job, reported experiencing higher sexual harassment than transit workers and school secretaries (Mansfield et al. 1991). Work routine also affects sexual harassment victimization. Gutek , Cohen and Konrad (1990), in an examination of working women in Los Angeles county, found that as working women’s opportunity for contact with opposite-sex coworkers increases, their likelihood of encountering sexual harassment increases. In an examination of women and men in the military, it was reported that gender composition of the work group, supervisor gender context and nature of the work space significantly affected employees’ risk of harassment, with those who worked in male dominated work groups, who had male supervisors, and who were required to work in a common space rather than private spaces reported higher levels of 33 sexual harassment than their counterparts (Firestone & Harris, 1994). Consistently, using a composite score of job gender context based on gender tradition of the job, gender mix of the workgroup and supervisor gender, Mclnnis and Fitzgerald (1997) reported that men dominated job gender context increased the likelihood of victimization among federal female workers. Another classification of jobs has been based on the combination of job prestige and whether the job is traditionally female or male jobs. White collar jobs such as management have been viewed as traditionally men’s job and highly prestigious. Blue- collar jobs are jobs that are thought to require masculinity and have relatively low prestige. Pink collar jobs are low in prestige and are associated with feminine roles (librarian, teacher, secretary, waitress). Ragins and Scandura (1995) compared women in white-collar jobs and blue-collars job and found that women in blue-collar jobs reported encountering more sexual harassment than women in white-collar jobs. The former group was also found to respond to sexual harassment more passively. The authors attributed passivity of women in blue collar jobs in the efforts to prove themselves to be one of the boys. Baker (1989) compared blue-collar and pink-collar women and found that those in pink collar jobs reported greater levels of sexual harassment than in blue collar jobs. Nevertheless, Fiske and Glick (1995) argued that for women in pink-collar jobs, sexual harassment may be more motivated by intimacy-seeking or sexual attention, while for women in blue collar jobs, sexual harassment may be more motivated by hostility towards women. 34 Sexual Harassment of Policewomen The literature on sexual harassment suggests that those who are subordinates, those who are women, regardless of their ethnicity, or educational level, are at risks of sexual harassment. Moreover, those who are in male-dominated professions tend to be more bothered by sexual harassment due to the nature of sexual harassment in the forms of hostility towards women, in addition to unwanted sexual attention. Policing is one occupation where sexual harassment would be problematic because women in this occupation tend to be predominantly in non-supervisory positions, and work in the presence of many men who have hostility towards them. Currently what we know about sexual harassment in policing comes from the literature on western nations in which women were utilized in firnctions that are traditionally male work, such as detective or patrol work. However, most studies have not focused on sexual harassment as a key issue. Accordingly, information about sexual harassment in policing is‘still largely limited. In the U.S., there is substantial evidence that policewomen have become a target of sexual harassment since the inception of the utilization of women in policing (Brown & Heidensohn, 2000; Schulz, 1995). Earlier accounts of women in policing have identified the media as a main culprit in sexual harassment against policewomen, beginning with comic book and news reports in the early 19005. The forms of sexual harassment during the early era revealed that society’s rejection of the professionalization of women or the defernininization of policewomen was blatant and well publicized. Policewomen were portrayed as masculine, ugly, or simply as sexual objects. Officer Wells, a pioneering police officer experienced sexual harassment. Upon appointment to be a policewomen 35 by the Los Angeles Police Department, she was depicted by journalists as “a bony, masculine person, grasping a revolver, dressed in anything but feminine apparel, hair drawn tightly in a hard little knot at the back of the head, huge unbecoming spectacles, small stiff round disfiguring hat, presenting the idea in a most repellent and unlovely guise” (Schulz, 1995, p.103). Other women were represented as Amazons wearing helmets and holding clubs (Brown & Heidensohn, 2000). Policewomen are referred to as “police beauties” (Appier, 1998). This portrayal of female officers was not unique to the American Experience. In England as well, sexist joke about policewomen were common. For example, in 1933, there was a cartoon in Punch, in which two male offenders were making off with fur coats, while a policewomen with a smiley face stood and watched. The caption states “and whatever will you rascals be up to next, eh?” (Brown & Heidensohn, 2000). Over the years, when the utilization of women in policing became more familiar in western society, these forms of sexist jokes and images in the media lessened, but sexual harassment against policewomen continued to be a problem and become even more visible than in the past, partly due to the increased routine encounters of women with their colleagues, including male coworkers and male supervisors (Brown, 1998; Martin, 1990; Wong, 1984). Harassment has emerged as a unique workplace problem that influences stress at work among women (Morash & Haarr, 1995). With the emergence of legislations, court rulings and social condemnation of sexual harassment, several studies since the 19805 began to examine the sexual harassment experiences of women in policing. Currently, most studies of policewomen that addressed sexual harassment have largely 36 focused on prevalence. An examination of 135 female police officers fi'om several places including the New York City Police department, New York State troopers, New York area, Connecticut, New Jersey and San Diego found that 68% of women indicated having experienced sexual harassment on the job and that young officers are particularly at risks for sexual harassment (Janus et a1, 1988). Martin (1990) examined the experiences of sexual harassment by policewomen in five major police agencies in the U.S.. When asked whether they have experienced sexual harassment, based on a sample of 71 female officers, approximately 75% of female officers indicated that they had experienced sexual harassment, and 49% of sexual harassment involved having been pressured for sexual favors by either a supervisor or fellow officers. Another study examining sexual harassment among females in small town police agencies found that approximately 53% of policewomen in the study reported having been victims of sexual harassment (Bartrol et al., 1992). Sexual harassment is a highly salient issue for policewomen in various organizational settings, regardless of size and locations, as the proportion of women who identified themselves as targets is relatively high compared to other organizations. Women officers in the U.S. also seem to identify themselves as victims of sexual harassment more than elsewhere. Brown and Heidensohn (2000) compared sexual harassment experiences of 804 female officers from 35 countries, which are largely in the Western and Eastern Europe, and North American countries. The majority of the sample is white women (89%) in their 305 (49%), who work in patrol (35%), and includes equal proportion of constable and supervisor levels. This study asked the respondents to answer the questions, “Have you experienced any sexual harassment from officer colleagues at work from male officer colleagues?” and “Have you experienced any 37 sexual harassment from officer colleagues at work fiom female officer colleagues?” Seventy seven percent of the sample reported having been harassed by male coworkers and 24% reported sexual harassment by female coworkers. It was also found that women officers in a country where there is greater emancipation in the police role (e. g.; England and the U.S.) and countries with colonial histories which adopted the ideas of policing from the U.S. and England (e.g., France, Canada, New Zealand), reported having experienced sexual harassment on the job more frequently than women in those countries where policing is rooted in an authoritarian system (e.g., Germans, Belgium and Luxembourg) and those countries with a totalitarian system (e.g., Poland, Hungary). However, because sexual harassment is subjected to interpretation, it is unclear whether higher reports of women in the former two groups come from higher awareness in these countries, actual resistance of male coworkers in their respective organizations, or both. Being in police settings seems to contribute to sexual harassment for both female officers and female civilians. However, the nature of closer contacts of female officers with male officers appears to be associated with even highest risk of sexual harassment. Brown (1998), compared the experiences of 1802 women officers and 164 civilian women from five forces in the United Kingdom in terms of personal experience of sexual harassment either as a target or a bystander. The female respondents were asked whether they ever had been exposed to the following acts (a) heard comments about women’s physical appearance, (b) heard sexually explicit jokes, (c) heard comments about their own physical appearance, ((1) been persistently asked out on unwanted dates, (e) been touched, stroked or pinched, and (0 been subjected to a serious sexual assault in the preceding six months. Policewomen experienced these acts more often than female 38 civilians. About 99.1% of policewomen and about 88% of civilian women reported having heard comments about women’s physical appearance. Hearing sexually explicit jokes was also common in this sample. For this category, 98.9% of policewomen and approximately 87% of civilian women reported that they had encountered such experiences in the preceding six months. Having heard comments about their own physical appearance revealed a large gap between policewomen and civilian women. About 72% of policewomen and 56% of female civilians reported having heard comments about their own physical appearance. Unwanted sexually physical contacts appeared to be the next most common types of sexual harassment for women in police organization; about 62% of policewomen and 38.2% of female civilians reported having been touched, stroked or pinched on their job. Pressure for dates was experienced by 53.2% of policewomen, and 35.8% of female civilians. Finally, sexual assault happened to 5.1% of policewomen, and 1.3% of civilian women. Further evidence of policewomen’s harassment is found in ‘patchy” verbal accounts of policewomen from qualitative research, which provide more descriptive details. In these studies, three types of harassment have been described: gender harassment, unwanted sexual attention and sexual coercion. Accordingly, based on both quantitative and qualitative findings, the discussion which follows will address sexual harassment based on these three categories (Fitzgerald, et al., 1997). The gender harassment category involves sexist and hostile behaviors and attitudes, while unwanted sexual attention involves sexual advances such as staring and leering, touching, and pressure for dating and sexual impositions such as sexual assault and rape. Finally, sexual coercion involves behavior like extortion, and bribery to obtain sexual cooperation. 39 Gender Harassment Gender harassment includes “generalized sexist remarks and behavior not necessarily designed to elicit sexual cooperation, but to convey insulting, degrading or sexist attitudes about women” (Fitzgerald, 1990, p. 33). This form of sexual harassment has been extensively reported in the western women policing literature and found to be more common than unwanted sexual attention or sexual coercion (Brown, 1998; Martin, 1978). The motive for gender harassment on the part of male officers is more hostility- motivated, rather than motivated by paternalism. Martin (1980) explained men’s motives for gender harassment in counties like the U.S., where women’s role in policing is extended to include traditionally men’s work. She explains that policemen were concerned that the utilization of women would reveal to the public that the “man of action” image of policing work in the media and the public was not true. Because police officers primarily have been working class men, there is some adherence to a macho image, and the presence of women presents a formidable threat to the glamorous view of the occupation. Moreover, some men are concerned that the increased representation of women would disadvantage male officers. Because of women’s potential to use their sexuality to gain what they want, women who enter policing may change the rule of equity held by male officers. Well intended programs to increase representation of women in police work and in supervisory positions have led many male officers to feel that they are victims of discrimination in police work (Martin, 1994). Martin (1980) argued that the concerns that some policewomen might do less work but receive greater organizational rewards also upset some men. In addition, in policing, the sense of solidarity is based on common experiences based on both shared prior backgrounds and 40 based on shared work situations. Some men think this might disappear because female officers can create conflict among male officers who compete for sexual attention. Accordingly, the presence of female officers creates potential anxiety among male officers and the men respond by using various means to limit women’ role. The potential loss of personal resources within the organization and the use of biological sex as a basis of categorization seem to further facilitate the devaluation of women and make gender harassment trivial and habitual. Other scholars have echoed the view that male officers feel threatened by the presence of female officers and therefore react to women with hostility. Wexler and Logan (1983) argued that the policemen were aware of negative public attitudes towards police, and when policewomen performed well, this posed a further threat to their own masculinity. The loss of status led policemen to express their frustration by scapegoating female officers. Several studies have pointed out that verbal forms of gender harassment appeared throughout women’s police career. The terms used to address women in policing have been associated with words that could elicit discomfort for women, by isolating them from the group, signifying their incompetence or highlighting their sexuality. In the police academy, the instructors often referred to cadets as guys or men and ignored women cadets (Pike, 1992). Martin (197 8) found that men also used different labels to categorize women into various types such as lady, broads, girls, lesbian, bitch, which connote the degree to which men can control these policewomen and to lower women’s worthiness as human being. Terms like “lady" cast women as dainty, demure and proper women, who deserve respects from male officers because they act consistently with sex- role expectation. Policemen who view policewomen as ladies often limit their 41 opportunities to learn skills necessary to their jobs. The term “girl” has been used to signify women officers as irresponsible and incompetence. Those who do not fall into these categories were those whom policemen cannot control, hence, they are called lesbian or bitch or broad. Apart from sexist name calling, policewomen also constantly encounter sexist statements and behaviors that question their level of competency and challenge their presence in policing. Women who gained some small favors from supervisors have been subjected to crude comments from male officers such as “Who’d you sleep with to get Sunday and Monday off ?” (Martin, 1978, p. 52). On the job, some policemen were straightforward in directly telling female officers that they are not wanted. Overtly hostile statements when a male police officer found out that he had a female partner such as “Oh Shit, another fucking female” were not uncommon (Martin, 1994, p. 390). When individual women failed to act professionally on the job, men police officers are quick to blame women as a group by pointing out that the failure is due to being a woman (W exler & Logan, 1983). Display of sexual. explicit materials and suggestive materials is also common in police culture. Such display is usually done anonymously in an accessible place. Wong (1984) described harassing cartoons on bulletin boards of many station houses. Beside the display of inappropriate materials, when women officers shared locker rooms with men, they often discovered cartoons or suggestive notes posted on their locker doors. Contraceptives and women’s stocking in locker rooms may be left in mailboxes. Occasionally, female and male officer’s names would be attached to these cartoons to characterize a relationship. While most female officers feel annoyed by these acts, they 42 could not pinpoint a perpetrator. Nevertheless, the experience of gender harassment of policewomen is varied, depending on the type of assignment. Gender harassmentmay be more prevalent among women who are assigned to work in the area that has been viewed as men’s territory. In police organizations, feminine work has been associated with formal rules, management, paper work, and masculinity has been associated with informal rules, street cops and the crime-fighting frmctions (Hunt, 1990). It seems that women who violate men’s sanctuaries are pushed back to their zone by gender harassment. Unwanted Sexual Attention Unwanted sexual attention has been defined exactly as sexual attention through verbal and non-verbal behaviors that is unwelcome and offensive and unreciprocated (Fitzgerald et al.1997). This form of sexual harassment occurs when there are no threats or the offers of job-related rewards. It tends to co-occur with gender harassment (Fitzgerald, et 31,1997). This category ranges from mild acts of sexual advances such as whistling, cheers, and howls to severe forms of sexual advances including sexual assault. Making sexual advances is one way to display masculine prowess and validate a man’s sense of masculinity. In police organization, men ofien flirt with women. An instructor may make embarrassing comments, such as telling a female recruit in front of class that they have to stop making eye contact because others may think that there is something going on. Groups of men in a male dominated department may direct sexual attention at women for entertainment during the dull and boring hours. Martin (1980), found that leering at anatomical features of women with sexual connotations was also common. Sometimes, the forms of teasing go so far as a crude comment in presence of 43 many people. What is troubling is that in most cases, the harasser did not seem to be inhibited by the presence of other officers, and harassers are sometimes supervisors. Unwanted sexual advances in a form of propositions and pressure of dates has become an inevitable part of lives of female officers and become a game among male officers. Most women felt that since their first few days at work, policemen were testing them to see how easily they could be conquered (Martin, 1980). In one court case, a female officers sued the precinct because male officers betted on who would be the first one to have sex with her (Detroit news, 1990). Wong (1984) indicated that policemen thought that women who come to police work should also welcome personal challenge to men as well, accordingly most women would be given propositions for many months or even years, until they are successful or until it is clear to them that it would never be successfirl. Some women who were openly homosexual indicated that despite lcnowing her sexual orientation, men keep on flirting with her (Wexler & Logan, 1983). In Wong’s study (1984), most of the women do not generally face retaliation for refusing sexual advances, nevertheless, most of the young, and inexperienced women officers were so afiaid of retaliation that they felt little control over advances. Concerns - about being cut off fiom the social network of the department made them tolerant. Female officers often found that their fellow female officers were unreliable, because some of them thought that women subjected to sexual advances had asked for it. Commanding officers often do not help, and may be harassers themselves. In Detroit, a women officer was repeatedly subjected to unwanted sexual advances from ranking officers. She was shown an erotic book and kissed by her superior, and later told her that they should have an affair (Detroit News, 1990). Deliberate sexual physical contact is also common in the police workplace. Martin (1980) argued that physical touching is a way to convey intimacy while at the same time to symbolically represent the power of the actor over the person being touched. In her study in Washington DC, she found that men’s putting an arm around women was quite common, especially among rookie female officers. Sometimes, this also occurs as a mean to undermine women’s authority. In one instance, a woman who was typing in the office was kissed on the back of her neck by a male coworker in front of the prisoner. Most of these female officers although feel uncomfortable with these experiences, were reluctant to report the incident to an authority. While severe forms of unwanted sexual attention such as sexual assault and rape are considered as rare, unfortunately, female officers also reported this type of experience. A rookie women officer in New York City police department was raped at grmpoint by a male officer (Segrave, 1995). In one study, it was found that at least in England, approximately 5 % of female officers reported having been sexually assault by fellow officers (Brown & Heidensohn, 2000). Although in these incidents, there was no indication of the circumstances surrounding sexual assault of policewomen, it is possible that many of these cases occur in the context of a social gathering outside police settings, where women may have prior relationship with the person, which makes female victims concerned about being blamed for the assaults by other coworkers. Because police officers often socialize in bars or homes after work in order to exchange job-related information, many female officers, especially rookie officers are eager to participate in order to be accepted in networks and learn more about the organization. Unfortunately, this kind of gathering is dangerous for policewomen. Wong (1984) reported that in the 45 U.S., there was this type of incident that a male officer and female officer went home together after both were joining after work drinks and then a female officer placed charged against the male officer for raping her. Although, the male officer admitted that he had raped her, the supervisor viewed that it is more of the case of relationship turned sour and they perceived that the policeman should not have admitted the wrongdoing and should have said that it was just an affair gone bad. Sexual Coercion Sexual coercion has been used to refer to quid pro quo type of sexual harassment. It involves the use of premises or threats relating to job conditions in exchange of sexual cooperation (Fitzgerald et al. 1997). Most supervisors in a position to influence the conditions of work are men. Thus, the typical instances of this type of harassment tend to include harassment of a male supervisor to a female subordinate. This type of sexual harassment is considered as related to unwanted sexual attention, because it requires sexualcooperation on the part of female officers. The difference, however, lies in whether the harasser implicitly or explicitly indicates to the victim that the lack of cooperation may result in a reward or a punishment related to work. While it is believed that sexual invitations in exchange for promises of rewards and punishment exist in police organization, the actual reporting of this type of harassment in the literature is rare. First, because immediate supervisors are involved, women may be reluctant to reveal these incidents. Second, some women who experienced these types of harassment may comply with the harasser’s requests and accept the offer. Third, this type of harassment would not usually occur in the presence of the audience, and the lack of witnesses may make female officers reluctant to claim this type of harassment. 46 Still, there is some evidence that quid pro—quo harassment occurs in policing. As already indicated, Martin (1990) found that approximately 49% of female officers from five police agencies reported that they had been pressured. for sexual favors by a supervisor or fellow officers. In the Santa Clara County Sheriff department, a male officer demanded that women in the unit engage in sexual relationship with him in exchange for positive work reviews. This leads to a federal district court award of $2.7 million to three female sheriff’s deputies in 1990 (Segrave, 1995). In some cases, by turning down a supervisor’s and coworker’s sexual attention, policewomen face numerous forms of gender harassment at work. Wong (1984) described a young, black officer who had been propositioned by both her commanding officers and fellow officers. Upon refusing these sexual advances, she constantly faced sexual remarks and innuendoes. Another study indicated that a woman who had not respond positively to her sergeant’s sexual attention was later subjected to reprimand while she was talking with another male coworker (Martin, 1978). Policewomen’s and Other Victims’ Responses to Sexual Harassment While it appears that policewomen encounter many forms of harassment at work, upon encountering sexual harassment in the workplace, women do not response to sexual harassment in the same way. In the observation of women in policing globally, Brown and Heidensohn (2000) indicated that while most women use more problem-oriented methods (e. g., tell the harasser, report to a supervisor), a significant proportion of female officers adopted “emotion-focused coping responses” (e. g., joking, avoiding). They asked female officers to answer the question “how have you personally coped with discrimination/harassment ?” It was found that 64.2% used assertive responses, 59% 47 used joking, 54.6% said nothing, 27.8% used complaining, 23.6% respond aggressively, and 12.2% relied on the union. Noting that some women do not leave but tolerate harassment, the authors argued that in some cases, the policewomen are similar to battered women who adopt passive strategies despite the fact that they feel that the acts were hostile. Many of these women displayed what they called “Antagonistic co- operation” (p. 125). Martin (1990) examined how policewomen perceived sexual harassment and reported that while there is a wide variety of responses, there were two main patterns. There are women who felt powerless and pressured to tolerate the behaviors of their male counterparts. These women accepted that their circumstances happened as part of being in the men’s world and came to accept it and focused on modifying their own responses to improve the situation. For example, many women indicated that they avoided sexual harassment by minimizing their interactions with male coworkers. This holds true for both work and non-work contacts. Wong (1984) indicated that many policewomen avoided informal gatherings. If they had to, some would take their own cars so they would not have to leave with anyone. Many women avoided contacts with male officers and limit their contacts to just work-related, and tried not to act warm and friendly. For other women, the responses involved resentment and seeking help to modify the circumstances. Many policewomen refused to “put up with it ” and verbally indicated their strong negative attitudes toward their harasser. For example, some women used strategies such as slapping when being touched by a male partner and some made comments like, “Don’t speak to me, and I won’t speak to you” (Martin, 1980, p. 145). In several cases, self-identified victims went to external authorities to seek remedial 48 damages and to ask courts to force the police department to do something about it (Detroit News, 1990). The literature on victim behavioral responses to sexual harassment in organizations in general and in academic settings has examined several strategies of responses. These responses include ignoring the incident, joking, confronting the harasser, confiding to others or reporting to formal authorities, (Charney & Russell, 1994; Gruber & Bjorn, 1986). Research has tried to conceptualize responses on two dimensions; passivity and assertiveness. Passive strategies involve indirect actions that focus on modifying the circumstances but not modifying the behavior of the harasser, while assertive strategies involved direct actions with a purpose of targeting the behavior of the harasser. A particularly robust finding is that passive responses are more common than assertive responses. Ignoring the behavior has been found to be the most common responses to sexual harassment (Gruber & Bjorn, 1986; Loy & Steward, 1984). In 'Gruber’s and Bjom’s study (1986), less than 26% of those female workers who were victims of sexual harassment reported using assertive responses, while 29% adopted passive responses (ignoring, walking away, pretending that it was not happening, and 45% used defective responses (humor, stalling, telling friends). Passive responses appear to be the most common even in the academic settings. A majority of students, faculty and staff who reported unwanted sexual attention said they used passive responses; approximately 60% reported ignoring the behavior, 45% avoided the harasser, 45% talked to others about the harassers. Assertive responses such as confronting the harassers were used by only 25% of the victims and only 2% indicated that they formally reported the incident to authorities (Cochran et al.1997). Although one study of policewomen found that 51% 49 , used assertive responses (Brown & Heidensohn, 2000), there is a reason to believe that passive responses are most common and assertive responses may be overestimated. In the Brown and Heidensohn’s study (2000), the respondents were directly asked how they respond to harassment and discrimination, which might suggest only severe and salient events. Passive responses might have been more common if subjects were thinking of a greater variety of harassing behaviors. The “Silence Reaction of Sexual Harassment Syndrome” is believed to perpetuate the destructive cycle of sexual harassment, and to contribute to possible severe psychological symptoms for all organizational members, including primary victims and bystanders who are secondary victims (Marin & Gudagno, 1999). Gosselin (1984) argued that victims of sexual harassment are caught in multiple incompatible emotional reactions, which could result in poor mental health outcomes. When the victims feel bothered by the harassment, and at the same time do not have confidence in adopting assertive c0ping, they may feel caught between the feeling of anger and fear. Inability to do something about a stressful environment can lead to the experience of powerlessness and influence an individual’s perception of control and interfere with job performance. Bystanders too suffer from harassment; as they have reported experiencing depression (Luzio, 1993). Those who see a fellow female coworker’s suffering but are not in a position to provide help may feel bad about themselves and see the workplace as an unjust world that makes no sense. Women’s lack of assertiveness can perpetuate the cycle, by sending a message to inexperienced female coworkers that keeping silent is the expected behavior. To identify factors that could impact individuals’ perceptions and reactions to sexual harassment, the next section explores plausible determinants of victims’ responses. Figure 50 1 depicts the theoretical model. National group membership is expected to influence gender role stereotype beliefs, self-construals, participation in decision making and perceptions of sexual harassment acts as a violation of organizational policy, which in turn influence victim’s perceived severity of sexual harassment events and concern about negative reactions in responding to those events. These two factors are then expected to influence reactions to sexual harassment. The presentation will begin with the discussions of antecedents of perceived severity and concern about social reactions along with why these factors are expected to differ between Thai and U.S. female officers. Then, the discussion about how perceived severity and concern about social reactions might influence victim’s reactions to harassment will be presented. 51 32m 2535 2.: 5 wow: .302 .mozoeoonh < _ Semi 20.582 368 Scan 5850 22583: 2 acetone”. btofim 32080“— zozeq 35.53.:an 33020; mic—«8-866% E :ozaafifsm mfiameeflfim £239 25886 £8 .0950 3:20sz 52 The Sociolcultural Model: Gender Role Stereotype Beliefs Because women are usually sexual harassment victims and men are the perpetrators, several authors have explained that existing patterns of male dominance in society influence victims’ responses (Brooks & Perot, 1991; Tangri et al., 1982). The sociocultural model of sexual harassment explains that victims’ responses to sexual harassment reflect the differential status of women and men in society in both the political and economic sphere (Tangri et al., 1982). In this view, society reinforces male dominance and female subordination by rewarding aggressive behaviors for men and passivity and acquiescence for women. This would make most women powerless to respond to sexual harassment and less likely to expect that an organization would help them. This hypothesis was tested with the U.S. Merit Systems Protection broad data collected in 1980 from a sample of 10,644 women and 9,439 men. Tangri et a1. speculated that most women would report powerlessness due to their sexual harassment. However, they found that while almost half of the women adopted avoiding responses, only 10% reported a sense of powerlessness, and 61% saw no need to report it. The authors concluded that the evidence provided mixed support for the model. A limitation of the sociocultural model is that all women are not subjected to equal forces of socialization to cultural values of male dominance (Brooks & Perot, 1991). Several processes explain the link between being in a male dominance culture and harassment victims’ passivity. Gender role stereotype beliefs are one predictor that is consistently related to victims’ perception of sexual harassment and reactions to sexual harassment (Brooks & Perot, 1991; Malovich & Stake, 1990). Gender role stereotype beliefs result from gender role socialization, defined as “rule governed learning about 53 how to be an appropriate female or male within society” (W orell & Remer, 1992, p. 55). Sexual script theory provides a basis for expecting that gender role stereotype beliefs are related to reactions to sexual harassment. In this view, sexual scripts are “cognitive representations of the typical sequence of events in a sexual encounter which act as guidelines for an individual’s expectations and behaviors” (Krahe, Scheinberger-Olwig & Kolpin, 2000, p. 313-314). Women who endorsed traditional gender role beliefs, when encountering sexual harassment situations will use different sexual scripts than women who hold liberal gender role attitudes. These scripts influence interpretations about events in terms of inferences about what is occurring, why is it occurring, and what should be happening. Accordingly, it can influence individual’s evaluation of what they should do following harassment. According to traditional sexual scripts, men should be the initiator of sexual contacts, and women should be the recipients. Also, women should not behave assertively in socio-sexual situations and should remain reserved about their true feelings, while men can behave assertively and open about their desires. Thus, women are not expected to hold the male initiator of sexual coercion accountable for what happened. In socio-sexual situations, including sexual harassment, women who endorsed traditional gender role stereotypes beliefs will expect that it is natural for men to harass women and may feel responsible for what happens, and at the same times, are more likely to feel bad about behaving assertively with men. This places the women even more at risk of various forms of sexual coercion (Byers, 1996). Although prior research on sexual harassment has not yet examined how national differences in gender role stereotypes beliefs might affect reactions to sexual harassment, gender role stereotypes beliefs have been viewed as different across cultures due to 54 different kinds of gender relations specific to each culture (Belk et a1, 1989; Millham & Smith, 1981; Smith & Midlarsky, 1985). It is possible that women in different culture such as Thailand and the U.S. may use different sexual scripts in interpreting sexual harassment. Thailand and the United States have different social arrangements that may generate differential role stereotypes of policewomen in the two societies. The U.S. has been cited as generally more liberal than many societies in Asia (Ganguly, 1997; Matsui, 1995). While the arrangements and practices of most societies, to a certain extent, reinforce male dominance and female inferiority, Asian societies often differ in the expectations regarding the roles and responsibility of women. In an examination of Asian Pacific women, Lai (1985) explained that these women were socialized in to subservient roles and that family responsibilities and husband were the main dimensions that women used to form their sense of self. They placed high emphasis on self-effacement and tend to be non-assertive. Consistently, as other Asian cultures, Thai culture is typically depicted as a more patriarchal culture, with a hierarchical social organization in which men have more power and privilege than women. Asymmetrical gender role differentiation among men and women is attributed to the influence of Chinese and Indian cultures, with their emphasis on Buddhism, have exerted an influence on Thai women’s role throughout history (CWAWRD, 1981). The Indian influence is reflected on the dependency of women on men throughout their life cycles. In this view, when women are young they are protected by a father, once they become adults, they are protected by the husband and as they are older, their sons become their guardians. The Chinese cultural influence is based on the idea that women are considered as disgraceful objects hanging on the house roof, which 55 can bring disgrace to the family. Wasigasin and Hemaprasit (1996) examined perceptions of Thai citizens regarding the values of Thai society and found that most respondents viewed that the Thai society still holds many double standards in terms of expectations of women and men. The implication is that if Thai women and American women have differential beliefs about gender role stereotypes, their interpretation of sexual harassment may be different. Research on the acceptance of violence against women provided some evidence that as these people socialized into American society, they are more likely to shift fiorn traditional gender role beliefs to more liberal views and become less tolerant of violence against women. For example, in wife battering research, Ganguly (1997) found that the longer the Indian respondents in the United States received education in the United States, the more they endorsed egalitarian gender role attitudes, which in turn contributed to less permissive attitudes towards wife-beating. In another study, Mori' et al (1995) compared rape attitudes of 302 Asian and Caucasian students and reported that Asians were more likely to endorse rape supportive beliefs than American students. In addition, they found that acculturation of the Asian respondents also contributed to the decrease in rape supportive attitudes by decreasing gender role stereotype beliefs. Several studies on sexual harassment in the United States found that gender role stereotypes influence how sexual harassment was perceived. For example, when given sexual harassment scenarios, the respondents who endorsed traditional views of women were less likely to blame the perpetrator, were more likely to blame the victims, and were less likely to recognize the negative consequences of sexual harassment (Malovich & Stake, 1990). Rubin (1991) found that respondents with traditional views were more 56 likely to believe that (a) sexual harassment is not serious enough to report, (b) reporting would not be taken seriously and would not be believed, (c) they would feel ashamed, ((1) they did something to caused it, and (e) sexual harassment is a routine part of life. Some field studies have started to examine the link of gender role stereotype beliefs to victims’ reactions to sexual harassment. For example, Brooks and Perot (1991) found that compared to liberal women, those who endorsed traditional gender role beliefs were less likely to evaluate sexual harassment experiences as offensive, which later made them less likely to report sexual harassment to organizational authorities. Jensen and Gutek (1982) examined the relationship between gender role stereotypes and victims’ behavioral reactions to sexual harassment fi'om a self-blame attribution framework proposed by Janoff-Bulman ’5 research on rape (1979). They hypothesized that women with traditional gender role stereotype beliefs would be more likely to interpret sexual harassment as their own fault, and then less likely to respond to sexual harassment assertively. Following prior work, they made the distinction between characterological self-blame and behavioral self-blame. The former refers to the victims’ evaluations in terms of whether the incident is caused by their own character and the latter refers to the extent to which victims think their own behaviors contribute to the incidents. While both are viewed as harmful and positively associated with traditional gender role beliefs, the former is expected to be more strongly associated with inward directed emotional responses such as sadness, and guilt and lead to more passivity, than behavioral self-blame. Behavioral self-blame can be viewed as counterfactual thinking which has been used to reestablished control (What could I have been done to avoid this incident?) Hence, it is considered healthier than characterological self-blame. However, 57 assertive responses are expected to occur most often when a victim of harassment uses neither type of blame. The results reveal that traditional women were more likely to use behavioral self-blame and characterological self-blame. Victims who blamed their own behaviors were less likely to report sexual harassment to an authority or to discuss the harassment with a friend, a coworker or family, than women who do not blame their own behaviors. It was argued that the lack of a significant effect of characterological self- blame may stem fiom high correlation of these two types of blame. Although gender-role attitudes may influence perceptions of sexual harassment and responses, the magnitude of the impact also depends on the type of harassment examined. When harassment scenarios involved innocuous incidents, doing nothing may be more likely, regardless of gender role stereotype beliefs. Baker et al. (1990) presented their respondents with 18 sexual harassment scenarios, ranging from mild forms such as graffiti not directed at the individual to severe forms such as rape. For acts that few individuals would consider as harassment, respondents reported that if they became the recipient of these acts, they would adopt passive responses (e. g., doing nothing), regardless of their gender role attitudes. For more intrusive situations such as proposition without threats, women with liberal views were more likely than traditional women to use physical or verbal responses. However, the effect was not found in terms of severe forms of harassment such as rape. In this case, most of the respondents indicated that they would leave the field. It is possible that for obvious and intrusive situations or very mild situations, individuals may behave similarly, regardless of gender role stereotype beliefs. It is when the acts are ambiguous that gender role attitudes come into play. Thus far, only one study has examined the effect of gender-role stereotypes on 58 reactions to sexual harassment in an Asian sample. The effect of gender role stereotypes is consistent with findings in the United States. Matsui et al. (1995) had Japanese college women read scenarios of men to women’s sexual harassment. The scenario involves a man who walked behind a women and physically touched her, while saying “Hurry up, you will never get everything done today. The scenarios varied in the status of the harassers (a supervisor or a coworker), prior relationship between the two (closed or distant) and the body part touched (back or fanny). The respondents were asked about the degree of inappropriateness and intimidation and then to indicate what they thought would be the most appropriate reactions of the target. It was found that the part of the body touched is the only contextual factors that influenced victims’ responses; when the man touches a woman’s fanny, the respondents were more likely to view the act as inappropriate, sexually intimidating and anticipated more assertive responses from the woman than when the man touches the woman’s back. Additionally, liberal women were more likely than traditional women to evaluate the act as inappropriate and anticipated assertive coping from the women (e.g., protest implicitly, protest explicitly and report to a supervisor or others). While research has suggested that gender role stereotype beliefs influence victims’ responses based on their evaluations of the sexual harassment event, prior work has also identified the possibility that gender role stereotype beliefs can lead victims to become more concerned about negative social reactions associated with confronting sexual harassment. As mentioned earlier, sexual script explanation also posits that traditional gender role scripts proscribed women should feel bad about behaving assertively in sexual coercion situation (Krahe, et a1. 2000). When women feel bad about asserting 59 themselves, they also projected that others would not approve of those acts as well. Smith and Self (1981) argued that women who endorse traditional views may differ in their beliefs about their personal relationships with men. For egalitarian women, they are more willing to sacrifice the relationship for social inequality, while traditional women may be reluctant to sacrifice the loss of the relationship. Consistently, Jack (1991) explained that when culture formulates prescriptions of female behaviors in terms of passiveness, inhibiting self-expression, and placing high value on pleasing others, these women may be willing to sacrifice their own feelings for the sake of maintaining the relationship. Accordingly, they deny their own needs and feeling in the process of self-silencing and experience a loss of self and self-denial, which can generate hopelessness, helplessness and depression. Because assertive behaviors such as reporting to authorities, confronting the harassers and telling coworkers can be viewed as responses that can disrupt the degree of harmony in the relationship betweenthe victim and the harasser and may generate a poor reputation of the victim in the eyes of other coworkers, when individuals place high value on relationship maintenance, especially with men, these victims may feel ambivalent towards the use of assertive strategies due to fear of loosing social relationships. Based on the above arguments, these hypotheses are formulated: Hla. The Thai female officers will endorse more traditional gender role beliefs than the American female officers. Hlb. Beliefs about gender role stereotype will be positively related to perceived severity. Specifically, female officers who endorse traditional gender role beliefs will more often view sexual harassment as more severe, than women who endorse egalitarian gender role beliefs. 60 ch. Female officers who endorse traditional gender role beliefs will show more concern about negative social reactions than female officers who endorse egalitarian gender role beliefs. The Self-Construal Model: Interdependent View of Self and Independent View of Self Because assertive reactions to sexual harassment victimization can have potential effects in creating loss of reputations and jeopardize interpersonal relationships of the victims and the harassers, and between the victims and others, one perspective that has not been yet explored, but that may be useful in explaining victims’ reaction to sexual harassment in Thailand and in the United States is the self-construal fiamework (Markus & Kitayarna, 1991). Self-construal is a psychological construct that focuses on self- orientation as opposed to other-orientation and may be useful for examining victims’reactions. The concept has its root in cross—cultural research that has considered collectivism and individualism. Hofstede (1980) has described collectivistic culture as placing emphasis on group rather than individual goals. On the contrary, individualistic culture is characterized by concern for personal fate and emphasis on personal over group interest. Although cultural level variables explain aggregate outcomes, like the crime rate or the rate of economic growth, they do not explain individual behavior. Therefore, individual-level variables that mediate the effect of culture have been adopted to provide explanatory power for individual level behaviors (Oetzel, 1998a). Self-construal has been viewed as a psychological construct that was intended to serve that function. Given that the Thai society has been defined as high on collectivism and low on individualism and Americans have been viewed as low on collectivism and high on individualism (Sorod, 61 1991), self-construals should be a useful concept in comparing reactions to harassment of Thai and U.S. women. The basic assumption of self-construal theory is that individuals’ sense of self in relation to others is crucial in determining social behaviors and that the derivation of the sense of self is heavily influenced by national culture. In collectivistic culture such as Japan and China, an interdependent view of self tends to be fostered. Interdependent view of self involves a configuration of self which focuses on interconnections with others, the blending of the self/other boundary, and affiliating with others (Singelis, 1994). Markus and Kitayarna (1991) argued that for interdependent view of self, the meaning of self is modified by social context, according to the social situation one is in. In this type of self, the emphasis is on promoting social harmony, compliance, loyalty to groups, and downplaying individual’s uniqueness. On the contrary, in individualistic culture such as the United States, the socialization process and social arrangements tend to promote the independent view of self. This self-construal, involves a cognitive configuration, which highlights the uniqueness of personal trait attributes and less sensitivity to external situations. Internal abilities, thoughts, and desire to express oneself, promote one’s goals and value personal freedom, rather than social obligations. People with independent self-constr'ual are autonomous, self-contained and their behaviors are influenced by their own internal thoughts and feelings, rather than social situations. Previous research has found that people who come from collective cultures have a higher interdependent view of self but weaker independent view of self than people in individualistic cultures (Singelis et al., 1999). Therefore, given that the Thai society has been defined as high on collectivism and low on individualism (Sorod, 1991), 62 while Americans have been viewed as low on collectivism and high on individualism, one would expect to find that Thai women will show higher interdependent view of self, while American women will show higher independent view of self. Prior works have addressed several consequences of self-construal. For example, Lee, Gardner and Aaker (2000) explain that because people with interdependent view of self define themselves based on social relationships, in regulating the behaviors, external standards will be a focus and lead individuals to be constantly attentive to negative information about oneself so that they can modify themselves to fit in with the needs and desires of others in their social unit. Hence, positive self-perceptions will not be fostered or even attenuated for people with this type of self-construal. On the other hand, in the individualistic society, due to the emphasis on uniqueness and separateness, individuals ' will generally attempt to positively distinguish themselves from others, resulting in the focus on positive aspects of oneself. In this self-construal, negative information about self may be negated because the goal is to emphasize one’s uniqueness and positive qualities. As a result, members of a collectivist culture, for example in Japan, have been viewed as constantly engaging in self-depreciation, while Americans are more likely to engage in self-enhancement (Kitayama et al., 1997). Apart fipm the tendencies of members of an individualistic culture to view themselves in a positive light, independent self-construal appears to be associated with a high degree of control over various life events. Since an independent view of self establishes a person as separate from others and autonomous, this also reinforces the perception of seeing themselves as less bounded by situational constraints, allowing them to perceive themselves as they having a high sense of personal control (Taylor & Brown, 63 1988). On the other hand, people with interdependent view of the self, generally do not feel that they are always capable of modifying circumstances. Consistently, research on locus of control also found that native born Chinese were higher on external locus of control than Chinese Americans, while Americans have the highest scores on internal locus of control (Hsieh, Shybut and Lotsof, 1969). In the context of negative life event in general, such as the aging process, American elderly reported more perceived control than Asian elders (Wong et a1, 1999). Taking a different perspective on the traditional concept of control used in western literature, Chang, Chua and Toh (1997) explained that when encountering stressfirl life events, individuals may perform two types of control appraisals; one has to do with the beliefs that one can enhance their welfare by influencing existing circumstances, while the other one involves the beliefs that one can change one’s own cognition, affect and behaviors to accommodate existing realities. They ~ argued that while people in the western world view control as modifying the external circumstances to fit one’s own needs, the Asian world view tends to favor the latter concept of control, which is consistent with the nature of hierarchical social structure in which things are determined by situational circumstances, rather than factors internal to individuals. The concept and practices of Buddhism also contribute to the latter concept of control by focusing on a person’s manipulating her or his own thoughts. As such, in adjusting to stressful life event, Westerners may experience a high sense of ability to cope by actively changing the situation, while Asians feel confident in their abilities to modifying themselves to fit in with the situations (e. g., hoping firings will change, thinking about what they learned out of the situation). In addition to the sense of personal control, self-construals appear to be associated with different assumptions about humans capabilities in controlling their own behaviors in general, which have implications for inferences drawn in interpreting life events. Markus and Kitayarna (1991) argued that for people with interdependent view of self, behaviors are often determined by social context. Accordingly, people with interdependent view of self are more likely to be sensitive to social context cues when they try to make sense of various social events. On the other hand, in the independent centered world, behavior is determined by one’s own internal attributes, therefore, observers in this culture will attribute the cause of events to the actor’s internal characteristics. Research conducted in attribution of negative events supports this explanation. Americans and members of more collectivistic culture differ in the way they explain causes of negative events; Americans focus on the personal characteristics of the actor, while members of collectivist culture are more likely to focus on contextual cues. For example, Morris and Peng (1994) analyzed articles about murders in a Chinese and an American newspapers. They found that American reporters more often attributed murders to personal dispositions of the murders, while the Chinese reporters attributed them to external factors. Miller (1984) had American and Indian respondents read about a motorcycle accident, in which the passenger’s head was struck on the pavement. The driver was on his way to work, and just took the passenger to court and then went to work himself, after which the passenger died. The respondents were asked about the reasons for the driver’s behavior. Americans attributed the driver’s actions to personal qualities such as irresponsibility, while Indians explained the events based on contextual factors. A study conducted on family problems also found that American students were more likely to attribute their family problems to factors that are stable, and internal to the 65 character of the person, while western Samoan students are more likely to make attributions based on situational circumstances such as emotional state of the actor at the time (Poasa, Mallinckrodt & Suzuki, 2000). Having posited that at least three psychological processes differ among individuals with different self-construals, (a) self-criticism tendency, (b) the sense of self- deterrnination, and (c) the assumption of human’s ability to control their own behaviors, there are theoretical reasons and some empirical data to expect that self-construals as linked to these processes may influence evaluations of seriousness of sexual harassment, which in turn influence reactions to sexual harassment. First, due to their sensitivity on negative feedback about oneself, which comes as a result of the need of social approval, peOple with an interdependent view of self, might have more readiness to perceive sexual harassment as less offensive since their self-critical tendency to fit in with others may lead them to internalized negative messages about oneself due to sexual harassment. Landrine and Klonoff, (1997) explains that avoidance coping to sexist discrimination such as quitting, the job, happened because victims has internalized its insidious messages. On the other hand, people with an independent view of self, due to their tendencies to view themselves in a positive light, may be more likely to engage in defensive reactions to negative feedback about oneself. Second, because self-construals appear to influence individuals’ confidence in their ability to control external circumstances, it is possible that their styles of coping with problematic events such as sexual harassment may be different. Psychological reaction theory suggests that expectations of control would influence perception of loss of control (Greenberger & Strasser, 1991). Individuals with high expectations of control may 66 experience more 1055 of control due to the discrepancies between their initial expectations of control and the realities of being harassed. Accordingly, they may experience higher level of dissatisfaction and be likely to engage in self-protective behaviors to restore their sense of control. Hence, if people with an independent view of self have a high level of control, they may experience higher loss of control when they are harassed. Alternatively, the interdependent view of self is associated with the belief that external events exert much influence on one’s life. Then, negative life events such as sexual harassment may be evaluated as not much different from some other life events that cannot be controlled, which should lead people to react less defensively. Consistently, prior research found that locus of control, which refers to the extent to which individuals believe that what happens to them is determined by factors within by factors within their control or is the result of external circumstances such as faith or behaviors of others (Rotter, 1990) appears to influence victims’ reactions to sexual harassment. A study conducted in a clinical setting found that women with high perceptions of control were more likely to recognize their experiences as sexual harassment and dealt with it, while those who have external locus of control were more likely to deny it (Goldfarb, 1985). This provides another reason to expect that self-construals will be related to perceived severity of sexual harassment. Moreover, if self-construals influence how people generate causal explanations about the social events, it is possible that when encountering sexual harassment, women with interdependent view of self may attribute the event to factors that are situation specific, rather than to the harasser’s internal attributes (e. g. “it is because he was moody today, rather than because he is sexist”). This might reduce the perceived seriousness of the 67 event, due to the fact that the negative impact created by the event would be temporal (it’s just today as opposed to more would be coming), that the event is uncontrollable by the actor and may not have been caused by the actor (e.g., “It is because of his socialization or it is because of my fate”). Weiner’s attribution theory (1985) also explains that when encountering negative events, people search for the explanations of the cause of the events by considering three dimensions. They consider locus of causality by evaluating the extent to which the event is caused by factors internal or external to the actor. They consider stability or the extent to which the event is viewed as fixed or likely to change over time. Finally, they consider controllability, or the extent to which the act could have been controlled by the person. When individuals attribute responsibility to the actor’s character, it is possible that they are more likely to view the event as stable, as caused by factors internal to the actor, and as possibly controlled by the person. According to Williams et al. (1995), these judgments could lead to negative emotional reactions such as anger. Emotional state such as anger in turn could influence accessibility of available social constructs relevant to interpreting the sexual harassment event, such as viewing it as offensive. People who judge acts as due to the person are more likely to evaluate harassment as more severe. Alternatively, people with an interdependent self-construal focus on contextual factors, so they may evaluate harassment as transient, stemming fiom factors beyond the actor’s control, and therefore react less negatively. Beyond the influence of self-construals on perceptions of sexual harassment, self- construals could influence victims’ concern about social reactions directly, which in turn influence victims’ assertive responses. Concern about social reactions can be viewed as outcome expectations individuals have regarding assertive responses in the situation, 68 which are jointly determined by both individuals’ beliefs about the likelihood of achieving certain outcomes and valence attached to those outcomes. Expectancy theories of motivation (V room, 1964) postulate that individuals will choose to behave so that the instrumentality of such actions will be maximized. Applying this principle to reactions to sexual harassment, Stockdale (1996) explains that victims would not act if they believe that the course of action would be futile. Possibly, people with positive self-concept also believe that others would also view themselves favorably and hence responsive to their needs. For people with an interdependent view of self, their evaluations of self as problematic, their lack of personal control beliefs and assumptions about the inability of humans to control their own behaviors may lead them to be pessimistic that their assertive responses will lead to desirable outcomes. Negativeself-evaluations seems to influence individuals awareness of a cure for harassment, by focusing on changing oneself to correct the problems, while the lack of personal control may limit individuals’ beliefs about their ability to influence others to improve the situations. The assumptions that people have limited self-determination may also makes the prospects of achieving success based on acknowledging sexual harassment become less likely, as others may also have limited control over these circmnstances. Therefore, for them, modifying their own behaviors and thoughts to fit the external realities, are associated with more benefits than attempting to seek support form others. On the other hand, for people with an independent view of self, positive self-evaluations, the beliefs in personal control, and the beliefs that others including the perpetrators and organizational members can control external circumstances should also make them feel optimistic that their assertive responses would lead to desirable outcomes. 69 Although research has not yet considered the possibility that sexual harassment victims with different types of self-construals might respond to sexual harassment differently, prior research on self-construals and communication also supports the notion that self- construals influence interpersonal communication strategies. Singelis et al. (1999) explained that because people with interdependent self-construal were sensitive to others’ views, they constantly felt concerned for negative self evaluations, which resulted in high awareness of one’s own public image. Kim and Sharkey (1995) also suggest that self- construals influence relationship maintenance goals and therefore influence preference for different communication strategies in social situations. They argued that in Asian society, where the collective goal is emphasized, the goals of saving one’s face and the other’s face is more important in interpersonal commtmication than in Western society. In individualistic cultures, due to high emphasis on one’s own internal attributes and self- definitions as autonomous and unique, the need to display one’s own internal thoughts and feelings should lead to the need for being direct in communication rather than suppressing their view for the sake of the social relationship. In support of this argument, their study found that in social dilemma situations, such as when someone who did poorly asked individuals how they did, an interdependent view of self leads individuals to place high importance on communication goals of avoiding negative evaluation by one’s conversation’s partner, and avoiding hurting the hearer’s feelings, while an independent view of self leads to the emphasis on communication clarity, rather than face maintenance. Self-construals also influenced how individuals managed interpersonal conflicts. People with an independent self-construal preferred dominating styles, while those with an interdependent self-construal preferred conflict avoidance, obliging and 70 compromising (Oetzel, 1998a). Independent view of self was associated with competitive tactics, while the interdependent view of self was positively correlated with cooperative tactics in group processes (Oetzel, 1998b). Given that assertive reactions to sexual harassment can lead to social rejections of the victims, people with interdependent view of self, due to the need of social approval, may feel more apprehensive about being rejected and hence become less willing to adopt assertive responses. On the other hand, people with a high independent view of self validate their sense of self-worth by their internal evaluations rather than through social approval. Therefore, to them negative social reactions may be less important, while other goals such as expressing their actual feelings about the situations and seeking justice may be perceived as important. Based on the above arguments, it is anticipated that H2a. The Thai female officers will be higher on interdependent view of self than the American female officers. On the contrary, the Thai female officers will be lower on independent view of self than the American female officers. H2b. Self-construals will be related to perceived severity of sexual harassment such that female officers who are high on interdependent view of self will view sexual harassment as less severe than female officers who are low on interdependent view of self. On the contrary, female officers who are high on independent view of self will view sexual harassment as more severe than female officers who are low on independent view of self. H2c. Female officers who are high on interdependent view of self will show more concern about negative social reactions in responding to sexual harassment than female officers who are low on interdependent view of self. On the contrary, female officers 71 who are high on independent view of self will show less concern about negative social reactions in responding to sexual harassment than female officers who are low on independent view of self. The Organizational Model: Participation in Decision-Making Another framework which will be used in explaining reactions to sexual harassment is the organizational model of sexual harassment (Gruber & Bjorn, 1986; Tangri, Burt and Johnson, 1982). In this view, reactions to sexual harassment are influenced by differential status and power of women and men in the workplace. Structural aspects of organization including the hierarchical nature, sex-ratio and the nature of job tasks and requirements have been viewed as crucial in placing women in situations where they have limited resources (Rospenda, Richman & Nawyn 1988). Women who are in these low status positions, due to their dependence on others, could not afford to resist harassment (Thacker, 1996). Most studies that are conducted under this model have focused on examining the relationship between structural positions of women, and findings seemed to reveal mixed results. In support of this view, prior research that found that women who worked in a male dominated work group and women with low job skills tend to experience sexual harassment more frequently and are more likely to respond to sexual harassment in a passive manner (Gruber & Bjorn, 1986). Research in academic settings also found that both students and staff who were harassed by someone with higher authority over them were less likely to report to organizational authorities (Benson & Thompson, 1982; Cochran et al., 1997; Till, 1980). Thacker (1996) reported that passive reactions such as avoidance and going along were often used in responding to harassment from higher-level 72 supervisors and coworker harassers. However, a study conducted by Gruber and Bjorn (1986) did not find the effects of harasser’s status, victim’s job status, nor victim’s organizational tenure on victim’s assertive reactions. Research on contrapower sexual harassment as well suggested that sexual harassment of women by lewer status perpetrators was not uncommon (Rospenda et al., 1998). Although most organizations have many structural features that limit the status and the influence of women in organization, it is possible that other organization factors such as managerial practices may play a role in modifying power dynamics in the workplace. Some organizations that take interest in low status employees, for example giving them job autonomy and emphasizing fairness over hierarchical positions, and informal coalitions, may be able to reduce status discrepancies of men and women in organization, and help victims feel less concerned about career repercussions associated with reactions. One dimension of managerial practices seemed to be different in the East and the West and that may be able to explain differential reactions to sexual harassment of U.S. and Thai female officers is participation in decision-making. Given that this concept can provide direct practical implication for managers, it might be useful to expand the framework of the organizational model by examining this concept and its relations to reactions to harassment. The literature has conceptualized participation in decision-making in two ways: the managerial practices approach and the motivational approach. In the first, perspective, decision-making is seen as management behavior, which involves the extent to which an organization gives employees the ability, authority and power in making task-related decision and allows them to have freedom in coping with the workplace environment. 73 For example, Neumann (1989, p.184) defines participation in decision making as "structures and processes for organizing individual autonomy in the context of group responsibility and linked to system-wide influence". Spreitzer (1996) viewed participation in decision-making as organization environments that encourage innovation rather than organizational control. In the second view, participation in decision-making is seen as individuals’ interpretations of their opportunities to participate in making decisions relevant to organizational outcomes or their own individual work outcomes. For example, Denton and Zeytinoglue (1993), in examining perceptions of faculty in a university setting, viewed participation in decision making as individuals’ awareness of their opportunity to influence critical issues in their department and the university. White and Ruh (1973) conceptualized participation in decision making as individuals’ autonomy in making decisions affecting their work. For the purpose of the present study, the fist approach will be taken because victims’ passiveness has been posited to be influenced by organizational practices that fail to disperse power and authority in decision-making (Knapp et al., 1997). Participation in decision-making will be conceptualized as intended effort of people at higher levels of the organization to provide opportunities for peOple at lower level to have a voice in performing their jobs. Participation in decision-making may influence how the recipients form the expectations and react to harassment. The idea that participation in decision-making may differ in Western and Asian organizations have been explained by the concept of national differences on power distance. Hofstede (1980; 1986) argued that one important dimension that cultural differ is power distance, which is “the extent to which the less powerful persons in a society accept inequality in power and consider it as normal” (Hofstede, 1986, p. 307). 74 According to this view, in a culture with high power distance, relationships are shaped by hierarchy, and status differentiation. Chen & Ditomaso (1996) argued that supervisors with high power distance were more likely to see their role as commander, authority and expert, and supervisors with low power distance are more likely to view their role as coordinator, facilitator or an advisor. Empirical data suggests that these differences were perceived by employees of organization in different countries. One study found that leaders from high power distance cultures were seen as less approachable, less likely to communicate with subordinates, and less likely to delegate decision latitude to subordinates (Offermann & Hellmann, 1997). It is highly likely that cultural differences based on power distance value may affect opportunities for female subordinates to participate in organizational decision-making. Current practices of organizations in the United States and Thailand seemed to reflect differences in the degree of participation in decision making organizational superiors are willing to give to subordinates, which may be due to the influence of differences in national culture. In the U.S., police organizations are changing to a community policing philosophy that requires low ranking subordinates to participate in making decisions and to develop role innovations (Radelet & Carter, 1994). Additionally, court rulings against gender discrimination (1972 Title VII Amendment to the Civil rights Act of 1964) increased U.S. police departments’ awareness about the working conditions of female officers (Van Wormer & Bartollas, 2000). On the contrary, in Thailand, centralized decision-making still predominates. Innovations are not encouraged and disciplinary emphasis based on the chain of command is particularly strong. Although women occupy some higher ranking positions, in practice, for women, promotion simply means 75 increased rank and salary and has had little bearing on women’s opportunity to influence organizational outcomes. Participation in decision-making seemed to be associated with positive outcomes for employees. Spreitzer (1996) argued that in participative organizational environments where acknowledgment, creation and liberation are valued in contrast to top-down control, employees will develop these four cogrritions of empowerment; (a) meaning, (b) competence, (c) self-determination, and (d) impact. Spreitzer (1995) explained that meaning refers to perceived fit between work requirements and a person’s values, beliefs, and behaviors. Competence involves self-efficacy, that is the belief that one can do the job. Self-determination involves the ability to choose in initiating and regulating actions. Finally, impact refers to perceptions that one can influence strategic administrative or Operating outcomes at work. Once formed, empowerment cognition helps individuals cope with the demands of a work situation and increase the ability of . individuals to perform work effectively (Mishra, & Spreitzer, 1998; Spreitzer, 1996). By encouraging participation in decision-making, an organization can also increase the sense of commitment of individuals to the organization. Because participation in decision-making can be viewed as one form of social support from the organization, it can signify organizational concerns about an employee’s point of view. From a social exchange perspective, an organization that shows commitment to employees will lead individual employees to feel obligated to reciprocate by engaging in activities beneficial to the organization (Armeli, F asolo and Lynch, 1998). Salancik (1977) argued that comrrritrnent can result in felt responsibility. In the whistle blowing literature (Victor, Trevino & Shapiro, 1993), perceived role responsibility in turn, has been found to be a 76 good predictor of reporting wrongdoing of peers (Victor, Trevino & Shapiro, 1993), which can be viewed as a form of prosocial behaviors for the sake of one’s group. Moreover, participation in decision-making can also contribute to the perception of fairness of an organization. Organizational justice theorists have recognized two kinds of justice: distributive justice and procedural justice. Distributive justice involves perceived fairness of outcomes, while procedural justice involves perceived fairness of process used to generate the outcomes (Witt and Myers, 1992). Procedural justice has been linked to participation in decision-making. Thibaut and Walker, (1975) wrote that the “voice” or the opportunity for control in the decision making process can enhance perceptions of procedural justice. Witt and Myers (1992) argued that because participation in decision making itself can be viewed as an opportunity for employees to have input or control over process, participation in decision-making contributes to perception of fairness of the organization system. Employees who had an opportunity to participate perceived the organization as fair in making personnel decisions. Perceived fairness is viewed as important for the intention to report unethical behaviors (Victor et a1. 1993). Participation in decision-making may influence how victims define sexual harassment. First, because participation in decision-making is correlated with psychological empowerment, which reflects the sense of control in the environment, those who participate more in decision making should also experience greater loss of control in sexual harassment situations, as psychological reactance theory suggests that higher perceived control leads to higher perceived loss of control (Greenberger & Strasser, 1991). Given that participation in decision-making can help signify how much an organization allows individuals to influence their environment, those who have high 77 participation in decision-making may use this information to generate expectations that others would treat one with respect, resulting in higher expectations than their counterparts. When experiencing sexual harassment, individuals who have high participation may think that the treatment they received could have been better compared to individuals who have lower participation in decision-making, who may think that it could have been worse. Thus, those who participate in decision—making would view harassment as more serious because of a larger difference between their expectations and their actual experiences. Moreover, participation in decision-making fiom superiors should also decrease victims’concem about negative reactions fiom other coworkers. Because participation in decision—making can create a sense of empowerment (Spreitzer, 1996), empowerment . should lead individuals to reduce their dependency on other members in sustaining successful work outcomes. Second, as participation in decision-making can increase felt obligation (Salancik, 1977), those who do participate would feel obligated to report organizational misconduct and thereby to preserve the organization. This should lead individuals to prioritize goals by putting organizational benefits (exposure of wrong doing) above and before concern about how others would view themselves. Hence, they should feel less concerned about being evaluated negatively. Moreover, because participation in decision-making contributes to perceived organizational fairness (Witt & Myers, 1992), perceived fairness should also motivate individuals to expect that organizational authorities would be responsive to their complaints and prevent them from retaliation. Based on the above arguments, it was expected that H 3a. The Thai female officers will be more likely to perceive that they have lower 78 level participation in decision-making, compared to the American female officers H 3b. Perceived participation in decision-making will be positively related perceptions of sexual harassment such that female officers who perceived higher level of participation in decision-making will view sexual harassment as more severe, compared to female officers who perceived lower level of participation in decision-making. H 3c. Female officers who perceived more participation in decision-making will report less concern about negative social reactions, compared to female officers who perceive lower level of decision-making in organization. The Organizational Sexual Harassment Policy Model: Perceptions of Sexual Harassment as a Policy Violation ' Another source of variability of perceptions of sexual harassment may be formal organizational norms regarding sexual harassment. Previous research has identified important features of sexual harassment differently. For example, Livingston (1982) identified four core components of responsiveness policy, (a) definition of unacceptable behavior, ranging from general to specific instances of sexual harassment, (b) a statement conveying disapproval of the acts, (c) a discussion of the negative consequences of sexual harassment, and (d) a corrective or disciplinary sanctions that will be followed. Connell (1991) argued that components of effective sexual harassment policies should include (a) statement of prohibited conduct, which provides examples of sexual harassment, (b) specific schedules of penalties for each behavior, (c) clear complaint and investigation procedures, and (d) establishing training and education that acknowledges sexual harassment. Rowe (1996) recommended that policies should define sexual harassment, describe management responsibilities, address options of the victims and address the role 79 of complainants, respondents, supervisors and bystanders. Although there are many components of organizational policy that may affect victims’ responses, most researchers emphasized the need to buildshared definitions of the acts as wrong, through providing meaningful sanctions to the perpetrator, which helps employees to develop less tolerant attitudes toward sexual harassment and inhibits them from committing sexual harassment (Connell, 1991; Livingston, 1982). Erickson (1966) wrote that formal prohibition of the acts can establish and maintain social norms and help regulate members’ behaviors. From the standpoint of the victims, the prohibition of the acts and specification of sanctions would contradict “myths of sexual harassment” and “victims blame attitudes” and motivate victims to confront sexual harassment. O’Meara (1989) argued that cross-sex interaction rules are generally ambiguous, when norms are unclear, vague, confusing and opened to interpretation. The ambiguity can lead many victirrrs to feel concerned for how others might take it (Popovich, 1988; Livingston, 1982). The inability to determine precisely what acts deserve attention may increase victims’ unwillingness to trust their own feelings. In this respect, organizational initiatives that attempt to establish shared definitions are important for facilitating victims’ disclosure behaviors (Popovich, 1988; Livingston, 1982). Results from research that examined the role of policy was consistent with the view that policy may regulate group norms regarding sexual harassment and lead to more favorable organizational outcomes. For example, Zlotnick (1994) analyzed data obtained from 91 schools in the U.S. during 1990 to 1991 and found that schools where organizational policies had been in existence the longest, had the most means of disseminating sexual harassment education and information to members and used more 80 disciplinary sanctions in formal procedures in sexual harassment cases, had higher complaint rates. Similarly, Gruber and Smith (1995) found that when the organization developed more extensive methods for dealing with sexual harassment, women were more likely to respond assertively. The police organizations in the United States and Thailand differ clearly in terms of sexual harassment policy. In the United States, the vast changes in women’s opportrmity in the workplace and awareness of women’s rights have resulted in legal changes that require organizations to consider the needs of women, including the consideration of sexual harassment. As a result of EEOC guideline which place the responsibility of sexual harassment on organization and previous lawsuits regarding sexual harassment in a series of court cases, most organizations in the U.S. have been forced to establish organization sexual harassment policies and grievance procedures in accordance with the law. Accordingly, these organizations are required to establish definitions of sexual harassment and communicate them to employees. In Thailand, although sexual harassment has received some attention, it has not yet been defined explicitly and has not yet become an issue in government agencies. Individuals are left to define those harassing behaviors by using their own views. Therefore, it is very likely that organizational response to cross—sex interactions will differ for Thailand and the U.S. Even though previous research on organizational policy found that policy may help facilitate victims’ disclosure behaviors, no research has examined whether policy might Operate through victim’s perceptions. Jocobi (1999) argued that victims oftentimes experienced role conflict and role ambiguity between their own feelings and organizational roles with conflicting expectations (e.g., I feel angry, but my role as a 81 subordinate is to be polite to my supervisor). Such conflicts, then can lead victims to feel confused about what to make out of the situations. Thus, when organization policy exists, it may help victims validate their own feelings andlevaluate the behaviors more critically. In addition, given that policy is designed to limit the problems of harassment, victims in organization where policy exists may feel strongly bothered by sexually harassing behaviors in their organization since the existence of policy should be associated with the expectations that these behaviors should not have happened. Consistently, in the ethical decision making framework (Jone, 1991), formal prohibitions of behaviors can affect social behaviors due to the fact that formal prohibition can heighten moral intensity cognitions. Moral intensity cognitions can further lead to the likelihood that one will evaluate certain acts as morally reprehensible, which is viewed as important for deterrrrining the likelihood that one would refrain from these acts, would intervene with those acts, and would help the recipients and condemn. the wrongdoer. Moral intensity involves judgments based on six dimensions; (a) social consensus, which refers to perceptions of social disapproval of the acts as wrong, (b) proximity, which refers to the feeling of nearness with victims of the act, (c) the magnitude of consequences of the act, which refers to harm done to the victims, ((1) concentration of effect, which refers to the numbers of people affect by the acts, (e) probability of effect, which refers to probability that the act will actually occur and that it will cause harm, and (Otemporal immediacy, which refers to the length of time between the present and the onset of consequences of the acts. When the acts are formally prohibited, moral cognitions become salient and individuals are more likely to disapprove of the behaviors. Vi ctirns’ awareness that organizational policy prohibits sexual harassment might make 82 them more likely to interpret the event as severe. There is also a reason to expect the link between organizational policy and victims’ concern about social reactions. Organizational policy has been viewed as helping others to view sexual harassment from the victim’s perspective (Kulik, Perry & Smidtke, 1997). Thus, supportive reactions may be anticipated in the place where organizational policy exists. In prior research conducted on military settings, when organization policies and procedures took several steps to deal with sexual harassment (e.g., having penalties against the harassers, providing training for personnel, publicizing how to file complaints procedures), sexual harassment complainants also indicated that organizational authorities were more responsive to their complaints (Dubois et a1, 1999). This suggests that policy can lead to more positive social reactions to the victim and help reduce concern about negative social reactions. Thus, it was hypothesized that. H 4a. The Thai female officers will be less likely to perceive sexual harassment as prohibited by organizational policy, compared to the American female officers. H 4b. Female officers who perceive sexual harassment acts as prohibited by organizational policy will view sexual harassment as more severe than those who do not perceive sexual harassment acts as prohibited by organizational policy. H 4c. Perceptions of sexual harassment acts as prohibited by organizational policy will be negatively related to concern about negative social reactions. Perceived Severity Given that varying attitudes of women and the organizational context in which they are located are expected to influence individual’s interpretations of sexual harassment in terms of its severity, the remaining question is how perceived severity might influence 83 victim’s reactions to sexual harassment. The explanations for the links of the two can be viewed from the stress-coping literature. Research that has considered perceived severity of sexual harassment in explaining victim’s responses generally behavioral responses of the victims as reactions to stressful event, or what is called coping responses (Fitzgerald et al., 1997; Lazarus & Folkrnan, 1984). In this view, subjective experience with the event that is responsible for how individuals react to it. Upon encountering negative events, including events in the context of social relationships, individuals spend time calculating how problematic the event is for them. In so doing, individuals do not make the decision in a vacuum, but bring with them goals, assumptions and expectations (Nurius et a1, 2000). Lazarus and Folkrnan (1984) explained that in this process, individuals may engage in three types ofjudgrnents including (a) to what extent the event facing them is relevant or non relevant, (b) to what extent the event is benign-positive, and (c) to what extent the event is stressful. First, when individuals. have no stake in the outcomes, the stimuli are viewed as irrelevant. In this case, reactions to the events may not occur. Second, when individuals appraise the event as positive, the stimuli are viewed as benign. In this case coping might not occur since the stimuli is not viewed as aversive. Finally, in making a decision about how stressful the stimuli is, individuals engage in three sub-process of evaluating harm/loss, threat and challenge. We calculate the impact of costs that results from the event in terms of whether the situation touches on social-esteem, self-esteem, and moral values. In harms/loss analysis, we evaluate damage that had been done. For threat appraisal, we calculate future costs. Ifthe event has not yet damaged an individual’s self or social esteem but may create potential damage to the person in the firture, the event would be construed as constituting a threat. Finally, 84 when threat appraisals occur but individuals view that it could offer them the opportunity to learn and increases their sense of masterly, the event could be interpreted as a challenge. In this case, positive emotions might be experienced. Along with these processes, Nurius et al (2000) explained that when negative outward focused emotion is experienced, it tends be associated with the decision to deal with the violators defensively. Psychological reactance theory also has been used to explain how perceived severity might influence victim’s responses to sexual harassment (Moore, 1995). Psychological reactance theory maintains that peOple prefer a sense of control (W orchel & Brehm, 1971). Greenberger & Strasser (1991) explained that control refers to “an individual’s belief at a given point in time, in his or her ability to effect a change in a desired direction” (p. 11). Loss of personal control can be externally induced. Externally induced loss of control may occur when one is denied freedom of action, when the envirOnment posts constraints upon the individuals and when the event is salient. In these situations, when personal freedom is threatened, individuals will evaluate the magnitude of outcomes that are forced upon them. When loss of control is experience as severe, motivational arousal will be strong, and individuals will be more motivated to engage in behaviors aimed to reestablish control in the situation. However, this theory also argues that when we enter the situation, we bring in differential expectations of control. Those who have higher expectations of control may experience more 1055 of control due to the discrepancy between expectations and the limited control induced by the situation. For example, people with trait anger and self-esteem are more likely to experience psychological reactance than their counterparts, and hence more likely to engage in 85 » defensive reactions (Hong & Giannakopoulos, 1994). IfThai women are anticipated to have more traditional gender role beliefs, higher interdependent view of self, lower independent view of self, less participation in decision making and less likely to perceive sexual harassment acts as prohibited by their organizational policy and these features are anticipated to be associated with perceived severity, it is not unreasonable to expect that Thai women will view harassment as less severe. Moreover, since in the Thai society, social awareness about sexual harassment is Iirnited, it is possible that Thai women will feel less negative towards harassment as they may bring with them lower standard of men’s behavior when making judgment about the behaviors. In support of this view, the literature suggested that people in countries where social awareness regarding sexual harassment are less extensive appear to be least sensitive to the harmfulness of sexual harassment. For example, Brazilian students more Often viewed several acts that could be construed as harassment by professors as harmless than did U.S. students (DeSouza, Pryor & Hutz, 1998). Neale (1991) compared Perception of sexual harassment of Asian students (students fi'om China, Taiwan, Korea, Vietnam and Japan) and American students and found that when presented with scenarios 0n a video, American male and female students more often found the scenes to be more inappropriate, threatening and difficult to handle. Studies that examined reactions of sexual harassment to harassment showed that PErceptions of the behaviors was an important factor in determining reactions to harassment (Brooks and Perot, 1991; Welsh, 1997). Using a scenario approach, Sullivan and Bybee (1987) examined the connection of the severity of an event to victims’ definition of the event, which in turn was expected to influence victims’ responses. They 86 constructed scenarios depicting sexual harassment of a male instructor against a female undergraduate student. Scenarios described mild harassment (reading a pornographic magazine, staring at the student’s breasts, posting offensive poster on the door), moderate harassment (unwanted touching, making sexual remarks about the student’s appearance and private life and suggestive looks), and severe forms of harassment (physically touching, threats of failing the students or not accepting the student in graduate schools if she was not sexually cooperative). Rather than asking the student respondents how they would react, the respondents were asked about the likelihood that a female student would report the incident to an authority. Objective severity was correlated with perceived severity, which in turn influence the respondent’s expectation that the female student would report sexual harassment to an authority. A study conducted by Brooks and Perot (1999) also highlights the importance of perceived severity of harassment in explaining the effects of many victim’s characteristics on victim’s responses to sexual harassment. The researcher hypothesized that women who are older, married, and who endorse feminist ideology (holding more liberal views of women) and women who had been frequent targets of sexual harassment, would perceive their sexual harassment experience as more severe, which in turn leads to a higher likelihood of reporting. It was found that perceived severity, measured in term of offensiveness, accounted for most of the variance in the model. Perceived offensiveness of the experience also mediated the relationship between feminist ideology and reporting, as well as victimization frequency and reporting. Although perceived offensiveness could not explain the effects of age and martial status on victim’s responses as expected, the authors speculated that it is perhaps because these two factors were highly correlated with 87 feminist ideology. Based on the theoretical rationale and the existing empirical data, it is expected that H5a. Thai female officers will evaluate sexual harassment as less severe, compare to American female officers. H5b. Perceived severity of sexual harassment is positively related to female officers’ use of assertive responses and negatively related to female officer’s use of passive responses. Concern about negative social reactions In addition to perceived offensiveness, concern about social reactions should also be crucial in determining reactions to harassment. Knapp et al. (1997) suggested that individuals’ expectancies concerning outcomes associated with reporting sexual harassment should influence reporting. Costs stemming form negative social reactions has been cited as an importance factor that inhibits victims from adopting direCt strategies for handling sexual harassment (Martin, 1978; Wong, 1984). Unfortunately, assertive responses tend to be associated with various costs or no benefits for victims. About 33% of victims who took formal action in the USMSPB’s study indicated that it made things worse for them, while 66% indicated that talking to someone made no difference (Livingston, 1982; USMSPB, 1981). Actually, several victims who had reported sexual harassment to an authority were subjected to further harassment (Benson & Thompson, 1982) and perceived the organization as even less fair than those victims who did not report (Adams-Roy & Barling, 1998). Several authors explained why people may not be responsive to the victims and evaluate the victims negatively. Lerner (1971) wrote that observers are motivated to 88 assign blame to victims in order to preserve their view that the world is in order and that people get what they deserve, which could let them maintain the sense that the world is predictable. Other scholars (e. g., Somers & Clementson-Mohr, 1979) explain many people endorsed sexual harassment myths. Sexual harassment myths are beliefs that; (a) sexual harassment is a personal matter, (b) women asked for it, (c) sexual harassment charges were used to get back at men when relationships turn sour, ((1) sexual harassment does not happen frequently, (c) it is a natural attraction between men, and (0 women can simply ignore it. Myths have been viewed as contributing to lack of sensitivity towards victims of sexual harassment. Marin & Guadagrro (1999) argued that when women responded assertively they were blamed, because gender role expectations are that women are passive and affiliative, not assertive and independent. Accordingly, women who behave assertively are generally subjected to public scrutiny, in addition to being blamed for being harassed. Littler -Bishop et al.(1982) in a simulated study provided the respondents with scenarios involving a women who had been subjected to harassment and found that women who had experienced more severe forms of harassment were evaluated by the respondents as less likable and less desirable than those who had experienced milder forms of harassment. Observers also perceived women who reported sexual harassment as less trustworthy, less feminine and less likable (Marin & Guadagno, 1999). Especially, when the parties involved have been engaged in competition for a job, and when the complainant was viewed as having a feminist orientation, the female complainants were judged as having negative motivation for reporting sexual harassment (Summers, 1991). In a male dominated organization such as policing, it is not unreasonable to expect that 89 members would not be supportive of the victims. According to Shaver’s defensive attribution theory (1970), observers will attribute causality to events in a way that minimizes the possibility that they would be held at fault. Because sexual harassment usually involves female victims and male perpetrators, men, who are typically accused of harassment, would tend to use defensive attribution. For example, male observers were more likely than women to assign blame to the victim and to perceive that the complainers are overreacting (Kenig & Ryan, 1986). Gruber (1997) explains that in male dominated occupation such as policing, where male-related attributes such as aggression, risk taking, and manual dexterity are valued, hostility and sexist and aggressive behaviors towards women flourish. In this type of environments, men tend to use their masculinity .. to create male bonding. Sexuality and dominanceare enmeshed and used to display their ' domination. Additionally, the nature of police organization, where there are many rules ’ to obey and officers need to rely on one another for self-protection both fiom administrators and danger from the job, have contributed to the strong influence of group norms in determining the day to day experience with group members. The need to depend on work group support may make women particularly concerned about how coworkers view them, and their responses to sexual harassment. However, the extent to which coworkers respond negatively to victims may be different in Thailand and the U.S. In Thailand, social awareness of sexual harassment is more limited due to the limited discussion of harassment either in general society and within police organizations. Under this condition, coworkers may view harassment as trivial and view women who used assertive reactions as overreacting. On the contrary, in the U.S., female officers may experience less negative social reactions since other 90 members may understand the seriousness of harassment. Prior research on observers’ perceptions of harassment reported that American observers view harassment as more serious than Asian observers (Neale, 1991) and when the behavior is evaluated as serious, observers more often disapprove of the perpetrator (Jones & Remland, 1992), anticipate that a victim will report sexual harassment (Sullivan & Bybee, 1987), recommend a severe penalty for the perpetrator (Mangione—Lambie, 1994) and intervene with the situation by providing emotional support, testifying as a witness, and telling the actor to stop (Sperry & Powell, 1999). Thus, negative reactions associated with the lacks of social awareness of the seriousness of harassment in Thailand may lead to higher concern about social reactions among Thai victims. Research on rape and sexual harassment in other settings found that perceptions of the beliefs of significant others are an important factor in decision making about reporting (Collings, 1987; Feldman-Summers & Ashworth, 1981; Feldman-surnmers & Norris, 1 984). In sexual harassment research, Adams et al. (1983) examined victims’ responses and found that there are many reasons that non-response is explained by different factors. Many victims indicated that they did not report sexual harassment because they believed that they might suffer retaliation (13%), because they might not be believed (19%) and that they would be treated as if they did something to cause the harassment (41%). Prior research in police settings (Wong, 1984) found that although the harassers do not signify threats, many female officers, particularly, those who are young, and are rookie officers feel reluctant to communicate their displeasure. These women sometimes feel that they Will not receive support from other women because group sanctions in this environment are so strong that other female officers would not want to take a risk in being isolated 91 further from the organizational networks. Furthermore, the code of silence prohibiting reporting misconduct of other officers, which developed because police work requires interdependency, also increases the chances for negative reactions to a woman who calls attention to sexual harassment. Martin (1978) found that many women have to find ways to resist sexual harassment without creating resentment. Therefore, it is expected that H6a. The Thai women Officers will have higher concern about negative social reactions than the American female officers. H 6b. Concern about negative social reactions are negatively related to female officers’ use of assertive responses and will be positively related to female officers’ use of passive responses. National Differences on Reactions to Sexual Harassment Given that perceived severity of harassment and concern about negative social reactions are expected to differ among women in the two countries such that female officers in Thailand are expected to perceived sexual harassment as less severe and have higher concern about social reactions, it is believed that reactions to sexual harassment of female officers in the two countries will be different, and Thai women will be less likely than American to adopt assertive strategies and more like to adopt passive strategies in handling harassment, compared to American women. H7a. Thai women will be more likely to indicate that they adopted assertive strategies in dealing with harassment and less likely to indicate that they adopted passive strategies in dealing with sexual harassment. In sum, prior literature has depicted Asian cultures as different from western culture in many aspects. Asian cultures have been viewed as having social arrangement practices 92 that promote gender role stereotype beliefs (Ganguly,l997; Mori et a1, 1995), favor the development of interdependent view of self, rather than independent view of self (Kitayama et al., 1997), limit participation in decision making of subordinates (Chen & Ditomaso, 1996; Hofstede, 1986) and have limited social awareness of sexual harassment in organizations (Bhatiasevi, 1998). Given that judgments of problematic event and the decision to handle it can be determined by goals, assumptions and expectations individuals bring in interpersonal situations (Nurius et al, 2000), it was expected that these features will influence the ways individuals interpret and react to harassment such that these features will lead female officers to trivialize sexual harassment and have high concern about negative social reactions. The low perceptions of severity and higher concern about social reactions were then expected to decrease assertive reactions and increase passive reactions to sexual harassment. In addition, compared to American female officers, Thai female officers were anticipated to find sexual harassment less offensive, report higher concern about social reactions, less incline to use assertive reactions and more incline to use passive reactions in responding to harassment. 93 Chapter 3 Research Methodology Participants Participants in this study involved a convenience sample of 106 female law enforcement Officers in United States and 109 female Officers Thailand who are full-time sworn officers during Summer and Early Fall 2002. After receiving a list of work schedule and the locations of female officers who would be available from the police departments that agree to participate in the study, the researcher planned a survey administration session by meeting with these female officers before or after their work schedules at their respective police agencies. Because there were some female officers Who were having a long vacation or were in training, the police departments had arranged for the investigators to meet with only those female officers who were available at the time during the survey administration process, which lasted approxirnately one to two weeks for each of these law enforcement agencies. These meetings provided an opportunity for the researcher to meet with female officers and invite them to participate in the study voluntarily. Therefore, the participants in this study were self-selected participants who came from law enforcement agencies whose administrators had agreed to participate in the study. Given that the focus is on policewomen’ reactions to sexual harassment by male Officers, the target population is female officers who work in law enforcement agencies in U.S. and Thailand who are victims of sexual harassment. Ideally, it would be desirable to I‘andomly select women officers from different law enforcement agencies throughout the “no countries who are victims of harassment or randomly select the police departments 94 that will be representative of the police departments in the two countries to maximize the chance that the sample is representative of the population. However, as Kauppinen- toropainen & Gruber (1993) suggested, in practice, random sampling techniques were impractical when the researchers were unable to gather an exhaustive lists of potential respondents. Due to the fact that those techniques may necessitate that the participants complete the survey on their own, as selected participants may be working in different parts of the countries, such techniques may limit the opportunity to clarify the meaning of a particular question when the participants have problems in terms of interpretation and generate low response rates. Therefore, it is considered that nonrandom technique, which allows the study to collect the data from the place where there is concentration of female officers would be more appropriate for practical purposes. In the U.S., originally, female officers were drawn from 5 law enforcement agencies in a mid western state. These participating agencies consisted of 2 Sherist departments, 2 police departments and one state owned agency. Based on a total of 121 female officers, 4 female officers or 3.31 % of these officers refused to participate in the study, due to two reasons; (a) time limitations and (b) concerns for anonymity of the respondents. Eventually, paper and pencil questionnaires were administered to the remaining 117 female officers who were willing to participate in the study. Thus, the overall response rate was high, as 96.69% the potential participants completed the survey. Nevertheless, based on the total of 117 returned questionnaires, 11 were removed from analysis because the participants indicated that they have never experienced sexual harassment within the past 2 years. Therefore, the analysis reported here therefore rely only on data from the remaining 106 participants, who can provide the information about their past reactions to 95 sexual harassment. The American sample ranged in age from 21 to 63 years, with a mean age of 35.66 years (_S_D_ = 8.08). The ethnic composition for the American sample was 69.8 % White, 23.6%, Afiican American, 3.8 % Hispanics, 1.9 % American Indian, and l % other. About half of the participants were married (49.1 %), while those who are single makes up 35.8% and those who are divorced or separated constitute 15.1%. In term of education, 37 % had educational level less than undergraduate level, while about half or 54.7% had undergraduate degree and 8.5% had completed graduate degree. These female officers performed a wide variety types of work including patrol work, supervising inmates and administrative functions. F orty-three percent of the participants were assigned to perform traditional police work including patrol, detective work and as community liaisons, while 23.6% were assigned to supervising inmates in the jail. About 27% performed administrative functions including managerial works and clerical works. Finally, about 4 % were court officers. Most of these court officers indicated that their current responsibilities involved maintaining courtroom security and transporting inmates to courts, and indicated that they had been previously assigned to street work or correctional work. The ranking composition of the sample also mirrors the pyramid shape of the Organizations. Seventy-five percent of the respondents are in the positions of officers or deputy, 14 % were sergeant, while 11.3 percent had a position of lieutenant or above. Nearly half (44 %) of female officers reported having supervisory experience. The mean tenure working for a current department was 10.23 years (S_D = 6.3 8) and the range was 1 Year to 29 years. A majority of the respondents planned to stay in police work until 96 retirement (76.4%). Most of female officers indicated that their work requires them to interact primarily with male coworkers rather than female coworkers (88.7%), while small proportions of female officers indicated that their work requires them to interact equally with male and female coworkers (5.7%) or interact more with female coworkers rather than male coworkers (5.7%). For the Thai participants, female officers were drawn from four police departments. Two of these police departments were located in Bangkok, one department was located in the suburb of Bangkok and one department is located in the northeast of Thailand. Initially, a total of 129 female police officers in these departments who were available at the time of the survey administration were invited to participate in the study. Of these 1 29 female Officers, 7 participants or 5.42 % disagreed to participate in the project. The reasons given for not participating were similar to those of the Americans, while an additional reason provided was that they did not think they could provide relevant information, as two female officers had been assigned in the current department less than a week. Hence, the response rate of this sample was also high, given that 94.57 percent of the potential participants have agreed to participate in the study. Of the remaining 122 female officers who completed the surveys, 2 female officers did not provide complete l‘eSponses on the survey, while 11 participants indicated that they have never experienced harassment by male officers within the past two years. Accordingly, the analysis here focuses on the remaining 109 participants. All of the Thai participants performed supportive functions, which are similar to Work performed by civilians in American law enforcement agencies. Their work responsibilities ranged fi'om teaching, writing specific projects assigned to their unit, 97 clerical work, accounting, printing services. This is because female officers in Thailand have not yet been employed to perform traditional police work, therefore, their job responsibilities are primarily different from female officers in the United States who performed men’s work. Unsurprisingly, their job requires them to perform in a more female dominated environment. About 57.8% indicated that their work routine requires them to interact primarily with female coworkers, while 16.5% reported that their work routine requires them to interact equally with both female and male coworkers. The remaining 25.7% indicated that they interact more with male coworkers rather than female coworkers. The rank composition of the Thai sample does not differ much from the American sample. The rank composition of female officers was 62.4% low ranking officers (non- commissioned officers), 23.9 % medium level officers (Sublieutenant to Captain), and l 5% high ranking officers (Major or above). Only about 29.4% of the Thai sample reported that they had some supervisory experience. The mean tenure working for a current department was 6.88 years (SD = 5.14) and the range was .25 to 28 years. This is lower than the mean tenure found in the American group. Perhaps it may be because in Thailand female officers were also subjected to rotation to different police agencies. As such, their departmental tenure tend to be lower, compared to the American counterparts who tended to work in the same agency throughout their career. About 65.1% of the Thai participants plan to remain in police work until retirement and 10.1 percent did not plan to do so. Twenty—five percent indicated that they were undecided. Therefore, for both samples, many participants planned to remain in police work until retirement, although fewer Thai female officers indicated so than the Americans (65.1% as opposed to 88.7%). 98 Although in both samples, most of the participants were in their thirties, the Thai sample were slightly younger than the American participants, as the mean age for the Thai sample 33.07 years, SD = 5.87), with a range of ages of 24-49 years. The Thai officers did not differ from American officers in terms of marital status, except that there are fewer participants who were divorced. The data revealed that nearly half of the participants were married (45.90%), with 46.89 % single and 7.3 % divorce or separated. Regarding their educational background, more Thai female Officers, than American female officers have education at undergraduate degree level; about 17.4 percent had educational level lower than undergraduate degree level, while an overwhelming majority (74.3 %) had undergraduate degrees and about 8.3% had a graduate degree. The overrepresentation of female officers with undergraduate degrees is unexpected, given that college degree education is not a requirement for becoming a non-commissioned officer, who constitutes a majority of this sample. This may be that due to the job stability of governmental agencies, many female officers who are overqualified in terms of educational requirements may be willing to seek employment in police agencies. Ideally, the characteristics of the samples should be similar in order to compare their reactions to harassment. It should be noted that in the present study, some demographic differences were found between the Thai and the U.S. samples, specifically in educational level, age, organizational tenure, job type, and supervisory experience and men composition in the work group. The Thai sample tended to have an undergraduate degree, lower average age and lower organizational tenure. They also perform non-police work in a more female dominated environment and fewer of them had supervisory experience. However, these differences seemed to simply reflect different characteristics 99 of the population of the female officers in the two countries. In the U.S., although policewomen were first employed in special functions, for example support frmctions, investigating crimes against other women or handling juvenile and women offenders, after women sued police agencies in efforts to have equal employment officers were they hired for road patrol. In Thailand, women have been utilized in the police force less than 4 decades and their responsibility involves non-traditional police work. Therefore, they tend to be younger, have lower organizational tenure, compared to the American sample and worked in administrative functions. The lack of policy on women’ issues has also limited their experience as supervisors, and the stability of jobs in the government these days have led many to be overqualified in terms of education. However, it is believed that these differences do not interact with the variables in the model in influencing the results. Watsi et al. (2000) recommended the inspections of correlations of demographic characteristics with outcome variables as the rule of thumb in ruling out the influence of demographic variables on the results of the study. If the correlations of demographic characteristics that differ between the two sample and outcome variables are minimal, the researcher can be more confident that these variables have limited influence on the results of the study. In the current study, it was found that there was only one significant correlation of demographic and outcome variables (the correlation between supervisory experience and assertive reactions was -.21 for the U.S. group). These correlations were much weaker than the variables in the framework hypothesized to influence each outcome variables. Therefore, it is believed that the results of the study are not due to the demographic differences. lOO Procedure Initial contacts with police administrators of the participating agencies were made during Summer 2002. The participating agencies received all of the documents pertaining to the study including the survey, as well as anticipated risks and benefits of the participants from participating in the study. After permission fi'om each of the police department had been granted, the survey administrators met with female officers in a small group (2-10) at their organizations during their regular working hours. The potential participants were then invited to participate in the study and, if they agreed, they filled out the survey in their respective organizational settings, which took about 15 to 25 minutes. In the U.S., the researcher administered the survey at each of the police departments. In Thailand, a female officer at each of the four police departments was contacted to administer the survey. These officers were chosen because of their familiarity with other female officers within the department, which was expected to facilitate the recruitment process and reduce the participants concerns about privacy of their data. Prior to collection of the data, these pre-selected female officers were asked to sign the contracts stating that they will keep all of the obtained information confidential, and that the subjects they solicited would participate in this study on a voluntary basis. They were also required to mail the packets back to the researcher in the United States directly, without examining those materials. Prior to the administration of the survey, all of these four pre- selected female officers were instructed to follow the standardized instructions to ensure that all of the respondents understand the procedures in the same way. In both countries, the survey administration process begins with the introduction that 101 the researcher was conducting a study on social interactions of female officers and male officers in the work place and would like to invite the participants to participate in the study. The survey administrator informed the potential participants that their participation needed to be voluntary and that all of their information would be kept confidential. The participants were asked to read the informed consent statement (See Appendix A and B) and go over the survey in order that they could decide whether they wanted to participate. Those who agreed to participate were then asked to stay and fill out the survey, while those who did not agree to participate were dismissed. After the participants completed the survey, they were asked to put it in the envelop provided, sealed it and handed it to the adrrrinistrators. Then, a sheet of paper containing debriefing statements were presented to the participants to inform them that the actual purpose of the study is to compare Thai and the U.S. policewomen’s reactions to sexual harassment. Additionally, the participants were informed about private organization they may contact, if they want to talk to someone about sexual harassment. Instrumentation and Translation The participants were asked to complete a questionnaire, which was used to assess the variables of interest in the study including (a) reactions to sexual harassment, (b) gender role stereotype beliefs, (c) self-construals, (d) participation in decision-making, (e) perceptions of organizational policy, (0 perceived severity, (g) concern about social reactions, and (h) sexual harassment experiences. However, because the original questionnaire was in English, the original survey was first translated into Thai. All items in the questionnaires were translated from English into Thai by the author and modifications have been done based on suggestions of three Thai female graduate 102 students at Michigan State University. These three students are majoring in Packaging, Architecture, Linguistics. The diversity in educational background was an advantage because skilled translators often fail to understand words and phrases used by lay people. Due to the nature of the questionnaire, especially the Sexual Harassment Experiences questionnaire, which contains some items that‘have no equivalent vocabulary, the translation was based on consideration of the meaningfulness of the questions in the Thai context rather than word by word translation. Another commonly used method, back- translation, would yield a translation similar to the English but potentially would be meaningless to the participants. As Herrera, DelCampo & Ames (1993) suggested, back- translation “can have reasonable conceptual similarity to the original version, although the original translation may be of poor quality” (p.357). Accordingly, the first task was to make the items meaningful to the Thai population. For example, in Thailand, for lay people there is no term for sexist language. However, a person who used sexist language is considered as a person who used language in a manner that degrades women. Accordingly, the English word was replaced with phrases so that the meaning is clearer to the Thai participants. Another example is in Thailand the word “suggestive materials” is depicted by materials that are precarious or dubious in nature. In order that the questionnaire items will make sense to the respondents, the translation cannot be word by word. Although, ideally it might be important to maintain the original version of the questionnaires, it is considered here that in some cases, this goal needs to be compromised, otherwise, items that do not make sense to the respondents may yield results that are ambiguous and not meaningful to interpret. Following the procedure recommended by Wallace and Brislin (1973) of back 103 translation, the Thai version of the questionnaire was presented to a bilingual person, who was asked to translate it back into English. At this stage, all of the original interpreters were asked to compare the original survey and the translated survey to see if the translation yields acceptable equivalent meanings. Most of the items fi'om preexisting scales yield acceptable similar meanings. However, there were certain discrepancies between the original English version and the second English version. Most of the discrepancies stem form the fact that the instruction concerning reactions to sexual harassment from the original English version for this study was not worded clearly, which resulted in some misunderstanding of the instructions, when the Thai version was translated back into English. Accordingly, both the original English version and the Thai version were slightly revised. Measures Reactions to Sexual Harassment The dependent variable in this study is reactions to sexual harassment. Therefore, to assess this variable, it is necessary to assess sexual harassment experience so that the respondents can use a particular incident to indicate their reactions. Sexual harassment experience was assessed in terms of how individuals reacted to the worst case experience of sexual harassment for each of the three types of harassment (gender harassment, unwanted sexual harassment and sexual coercion) that happened to the participants within the past two years. In keeping with other studies which assessed sexual harassment events by adopting the tripartite model of sexual harassment, the sexual harassment experiences questionnaire (Fitzgerald, Gelfand & Drasgow, 1995) was selected to elicit victims’ 104 responses. The original SEQ (Fitzgerald et al., 1988) was developed based on Till’s (1980) five categories of sexual harassment, which include gender harassment, seductive behavior, sexual bribery, sexual coercion and sexual imposition. It was developed to assess how Often the respondents experience each behavioral instance of sexual harassment. For example, the first question asked the respondents “have you ever been in a situation where a professor habitually told you suggestive jokes or told suggestive stories or offensive jokes?.” The respondent then indicates the choice of never, once or twice, sometimes, often or most of the times. The SEQ has been widely used in previous research on victims’ responses to sexual harassment (Morse, 1995; Welsh, 1997). Its psychometric properties have also been established. Fitzgerald et al. (1988) piloted this early version of SEQ with 468 college students in a nridwestem university and found that the Cronbach’s coefficient alpha was .92 . The test-retest reliability of the questionnaire with 46 students over a two-week period was .86. The items load on three factors, gender harassment, unwanted sexual attention and sexual coercion. Later, Fitzgerald et al. (1995) shortened the original scale for practical use but kept more sensitive items, resulting in 19 items for gender harassment, rmwanted sexual attention and sexual coercion plus one criterion item (Have you ever been sexually harassed?) The reliability yields a Cronbach’s alpha of .89. Reliability coefficient were .86, .75 and .87 for gender harassment, unwanted sexual attention and sexual coercion respectively. The new scale is called “SEQ”. The SEQ version used in the present study is drawn from the one used in Jacobi’s study (1999), which was slightly modified for assessing sexual harassment experience of women in the organizational context (See Appendix C, D, E and F). The SEQ has been cross validated. 105 Gelfand, Fitzgrald & Drasgow (1995) examined the structure of sexual harassment by the SEQ among three different samples; female college students in the U.S., working women in the U.S. and female student in Brazil, and reported that the three factor solutions based on the bipartite model was confirmed across settings (work versus education), and countries. Previous research has not yet established conclusively how responses to sexual harassment should be assessed. Different studies developed their own taxonomies for victims’ responses. For example, Bingharn and Scherer (1993) focused on 6 categories: (a) filing a formal complaint, (b) informal discussion with external authority, (c) informal discussion with an internal authority ((1) talking to co-workers, (e) talking to friends/family and (f) talking the perpetrators. Gruber and Bjorn (1986) categorized responses into three types: passive, deflective and aggressive. Following Matsui et al. (1995), the current study focuses on behavioral responses that range from non-assertive behaviors to assertive responses. Non-assertive responses focus on changing one’s own behavior to modify the situations; assertive responses focus on modifying the behaviors of the actor. The measures of behavioral responses include responses used in prior research. Non-assertive responses examined here involve (a) going along, (b) ignoring the behavior, (c) deflecting the perpetrator’s attention and (d) changing one’s own habitual ways of doing things to avoid similar situations. Assertive responses include (a) reporting the incident to an authority, (b) hinting displeasure to the person, (c) protesting explicitly and (d) threatening the person with some negative consequences (See Appendix E and F). For each of the response strategies, the respondents were asked to indicate to what extent they agree that each of these strategies listed was used in the situation. The possible 106 responses can range from 5 (Strongly Agree) to 1 (Strongly disagree). Although responses to sexual harassment were conceptualized as a categorical variable by some prior research, the view taken here is that it could be conceptualized as a continuous measure, in which the responses are considered as similar or deviate fiom assertive or non-assertive responses. This view has been taken by previous studies on women’s responses to sexual threats on campus (Nurius et al., 2000) and increase the distinction between responses that are highly assertive from those that are somewhat to minimally assertive. In assessing reactions to harassment, the respondents were asked to report how they reacted to the worst case of sexual harassment for each of the three types of sexual harassment that happened to them within the preceding two years. Ifthe participant reported having experienced one type of harassment, their reactions to that particular ' incident were used in the analysis. If the participants experienced more than one type of harassment, their reactions to one type of incident were randomly selected to use in the analysis. This is done because the alternative of using case base analysis may be problematic due to the fact that different numbers of unit of observations (cases) are nested under subjects. Therefore, the results may have been influenced by the patterns of reactions of those participants who have several cases. In addition, the respondents were asked to provide the identity of the perpetrators in these cases. The categories of perpetrators are (a) direct supervisor, (b) higher superior, (c) coworker, or (d) subordinate. In assessing the internal consistency of the two scales, assertive reactions and passive reactions, it was found that the assertive reaction scale has acceptable internal consistency, while passive reactions have unacceptable internal consistency. For the assertive scale, the Cronbrach’s alpha for the American and the Thai sample was .76, and 107 .68 respectively. For the non-assertive scale, the internal consistency for the American and Thai sample was .34 and .19 respectively. When “going along” item was deleted from the scale, the internal reliability coefficient for the American group was .61 and the Thai group was .31. Due to low internal consistency of this scale for the Thai group, non- assertive scores were formed and used in the analysis only for the American group. Gender Role Stereotype Beliefs Gender role stereotype beliefs were assessed by the 5-item measure used in thel993 General Social Survey (See Appendix G and H). This scale has been designed to measure gender role behaviors. It assesses attitudes towards women’s role in the home, in the office and in politics. A prior study conducted by Wright and Young (1998) found that the Cronbach’s alpha of the scale is .81. This scale has been used to examine the relationship between family structure and gender role attitudes in the adult American population sample. The respondents coming from father headed single parent families endorsed more traditional gender role attitudes than did the respondents coming from families where both parents were present, while the respondents coming from female headed single families endorsed more egalitarian gender role attitudes, than did those who come fiom the families where both parents were present. In assessing this measure, the respondents were asked to indicate whether they agree or disagree with each of the five statements. Response options ranges from 5 (Strongly Agree) to l (Strongly Disagree). The total scores were formed by summing responses across items, with higher scores corresponding to higher traditional gender role beliefs. The Cronbrach’s alpha was .45 and .58 for the American and Thai samples, respectively. The inspection of the item-scale correlations indicated that for the Thai group, item 3 has 108 low correlations with the scale. Once this item is dropped from the scale, the alpha for Thai improved to .67. However, for the American sample, the internal consistency of this scale with or without this item was still too low to warrant further analysis (45 and .39 respectively). Accordingly, the effect of gender role stereotype beliefs were assessed only for the Thais but not Americans. Self-Construals Self-construals were assessed by using items constructed by Kim and Sharkey (1995). The interdependent scale consists of 10 items. It tapped into the importance of group goals, modesty and connectedness with others. The independent scale consists of 8 items and focuses on a sense of uniqueness and independence, the tendency to behave consistently across settings and concern for one’s own internal characteristics. The responses can range 13cm 5 (Strongly Agree) to l (Strongly Disagree). These two scales are presented in Appendix I and J. The scales were developed by revising items fiom the original version created by Singelis (1994). The reliability of the revised version has also been established, with the Cronbach’s Coefficient Alpha of .72 for interdependent scale and .65 for the independent scale respectively. For the present study, the overall Cronbach’s alpha for interdependent view of self is .72 for Americans and .65 for Thais. The internal consistency for independent view of self scale for American and Thai is .61 and .65 respectively. Participation in Decision-Making Participation in decision-making was assessed by adapting the 5-item measure developed by White and Ruh (197 3) for assessing participation in decision-making. This scale was originally developed to measure an individual’s perceptions of opportunities in 109 making decisions affecting their work. J eanquart~Barone (1996) reported that the scale reliability was acceptable, with the value of Cronbach’s alpha of .88. In examining the relationship between participation in decision-making and job attitudes among 2,755 employees in 6 manufacturing organizations, it was found that the two factors were positively correlated. The effect was found to be robust independent of individual values measured by the Rokeach value survey (White & Ruh, 1973). For the purpose of the present study, this scale was adapted to indicate behaviors of superiors as a group, rather than perceptions of individuals regarding their opportunities to participation in decision-making. This is because supervisory practices are expected to differ between the two countries. Designing the measure to tap perceptions of supervisory practices in the department in general to represent organizational power in a more global and stable sense, rather than focusing on practices of an immediate supervisor also had an additional advantage. In some cases, the immediate supervisor may become harassers themselves; when they allow subordinates the opportunities to influence their jobs, participation in decision-making may lead to non-assertive responses. Due to the fact that behaviors of other supervisors towards oneself may better capture the influence individuals have with people who include supervisors, peers and subordinates within organization, it makes sense to focus on supervisors in general, rather than current supervisors. For the measure of participation in decision-making, the respondents were asked to give global ratings of the extent to which their supervisors in the organization behaved in a way that encouraged them to have influence in their work. The response in this scale can range from 5 (A great deal) to 1 (Not at all). Then, scores were formed by summing 110 responses across items, with higher scores indicating higher participation in decision- making. The adapted version is presented in Appendix K and L. The internal consistency of this item was higher than previous research. The overall Cronbrach’s alpha for the American and Thai sample was .91 and .94 respectively. Perception of Organizational Policy Violations Perception of organizational policy violations was assessed by asking the respondents to indicate if they agree that three worst cases they identified of sexual harassment, including gender harassment, unwanted sexual attention and sexual coercion, are clearly prohibited by their organizational policy (See Appendix E and F). For each type of harassment, the respondents were asked to recall the worst case incident that happened to them within the past two years and use that incident to decide if they agree or disagree that their organizational policy clearly prohibited such act. If they have never experienced any types of harassment, they were asked to recall the worst case experience that happened to other female coworkers that they were aware of, which could have happened at any time, and rate to what extent those behaviors were clearly prohibited by their organizational policy. The responses can range from 5 (Strongly Agree) to l (Strongly Disagree). The total scores were summed across three types of situations, yielding a composite score for this measure ranging from 1 to 15. The internal reliability for American and Thai groups was .80 and .84 respectively. Perceived Severity of Sexual Harassment Perceived severity of sexual harassment was assessed by asking the respondents to think of the worst case of sexual harassment that happened to them within the past two years, for each type of harassment, and indicate how much they were offended by those 111 behaviors (See Appendix E and F). In the case that the respondents had never experienced any types of incident, they were asked to think of the worse case incident that happened to their female coworkers, and imagine that if they were in that situations, how much they would feel offended by it. This measure consists of one item for each of the three types of harassment, resulting in 3 items for each subject. These three items were then combined to indicate the measure of psychological definitions of sexual harassment. The responses can range from 5 (Strongly Agree) to 1 (Strongly agree). The internal consistency reliability coefficient alpha for this measure for the American sample and the Thai sample is .63 and .64 respectively. Concern about social reactions Similar to perceived severity of sexual harassment, concern about social reactions were assessed by asking the respondents to rate on a five-point Likert scale, for the three cases of sexual harassment experience (See Appendix E and F). For each incident, the respondents were asked if they called attention to the issue, to what extent were they concerned that if they called attention to the situations, their coworkers would think of them more negatively. In the case that they had never experienced certain types of harassment, they were asked to think of the worse case that happened to their coworkers. The responses could range from 5 (Strongly Agree) to 1 (Strongly Disagree). The total score based on one item measure across three types of sexual harassment situations was then combined to represent individuals’concems tendency in evaluating sexual harassment experience. This measure was adopted to maintain consistency with perceived severity measures, except that the dimension being assessed is concern about 112 negative social reactions. The alpha for this scale for American and for Thai sample is .82 and .86 respectively. Control Variables Frequency of being harassed and sex ratio in the work group have been viewed as important factors that influence victim’s responses (Bingharn & Scherer, 1993; Wong, 1984). To ascertain that the findings are not due to these factors, flequency of sexual harassment experiences, and workgroup sex ratio were also assessed and entered into the analysis as control variables. Frequency of harassment This variable was intended to measure prior victimization experience Of female officers. It was assessed by the SEQ, described earlier by using the time flame of the preceding two years. The version of the SEQ used in the present study consisted of 19 items covering the three types of harassing behaviors (See Appendix E and F). The respondents were asked to indicate how often, within the past two years, they have encounter the 19 behaviors listed. Their responses may be never, once/twice, sometimes, often or most of the time. The total scores for this scale were then summed up and used to represent flequency of sexual harassment victimization. The total scores could range flom 19 to 135. The Cronbach’s alpha coefficient is .84 and .88 for the American and the Thai sample respectively. Work group sex ratio The sex ratio of the work group was assessed by one item measure, asking the respondents to rate on a five-point Likert scale the extent to which their work routine 113 requires them to interact primarily with men in the absence of women. The responses can range flom 5 (Mostly men) to 1 (Mostly women). Sexual harassment experiences An open-ended question designed to ask the participants about some of their sexual harassment experiences in the organizational was also included (See Appendix M). This is because it is still unknown what kinds of situations might be viewed as sexual harassing for the participants in Thailand. The SEQ as developed based in the Western context may overlook some of those situations. Therefore, it might be useful to compare responses flom women in the two countries. In order to get this information, the respondents were asked to describe certain undesirable situations involving interactions with male coworkers and their being women. They were asked to explain how they reacted to those situations and why they behaved the way they did. 114 Data Analysis 1 Before testing all of the hypotheses, chi-square tests and t-test were conducted to provide the overall characteristics of the sexual harassment that happened in both countries. This is done to evaluate if there would be any differences in terms of sexual harassment victimization in the two countries and any differences between final case compositions of sexual harassment that subjects used to indicate their reactions. To evaluate all of the hypothesis regarding national differences between American and Thai participants for hypothesized variables including (a) gender role stereotype beliefs, (b) self-construals, (c) participation in decision-making, (d) perceptions of sexual harassment as a violation of organizational policy, (e) perceived severity of harassment, (0 concern about social reactions, (g) assertive reactions to harassment and (h) passive reactions to harassment, a t-test was conducted on each of these variables and the level of significant was set at .05. However, because the internal reliability of gender role stereotype beliefs for the American sample and that of non-assertive reactions had unacceptable reliabilities for the Thai group, these two measures were dropped flom the 1 For data analysis, a Cronbach's alpha score of at least .70 is generally used as an acceptable standard for establishing internal consistency of the measures. However, in prior doctoral dissertation that are exploratory in nature, the reliabilities of .55—60 were reported and employed in model assessment, particularly when examining foreign populations (Raksiltham, 1984; Wickliffe, 1988). This limitation in the present study should be acknowledged. However, since the current study can be viewed as exploratory, the decisions to retain variables for further analysis were done by using a Cronbach's alpha of .60 as a cut-off point. 115 report of the results. Regression analyses were used to assess the effects of gender role stereotypes, self- construals, participation in decision-making and perceptions of sexual harassment as policy violations on perceived severity of harassment and concern about social reactions, as well as the effects of perceived severity of harassment and concern about social reactions on reactions to sexual harassment. Structural equation modeling (e. g., with Lisrel, EQS or Emos) was considered as alternative approach for testing the model since the model includes intervening variables. However, structural equation modeling requires that the ratio of the parameter estimates and sample size to parameter of five or more (Benlter, 1985). The sample size of the current study was not sufficient'to for the latter approach, which would involve higher numbers of parameters to be estimated. It is usually recommended that structural equation modeling be done in two steps, the first being the test of the measurement model and the second the test of the theoretical model. The first step involves estimation of each path between indicators and their latent constructs (Judge & Ferris, 1993). As a result, there is a significant increase in the number of parameters to be estimated, and the sample size limitations for the present study may yield unreliable estimates for the results of the measurement model and the theoretical model. Therefore, regression analysis, which treats each latent construct as one variable, is more suitable for the data analysis. These analyses were conducted separately for the two samples to determine if the hypothesized effects would be found in both groups. One reason for the separate analyses is that the findings might suggest that different interventions are required in the two settings. Second, there are certain characteristics of the sample that are apparently 116 different, and the analysis based on a combined sample could be misleading. For example, for the combined data, interdependent view of self is positively correlated with perceived severity. However, the correlations for sub-samples are low for each group. The initial high correlation is explained by the connection of both interdependent view of self and perceived severity with nationality. Therefore, sub-sample analysis can provide more meaningful, country specific results. Regression analyses were performed separately for each of the outcome variables, including perceived severity, concern about social reactions and reactions to sexual harassment. In the first two regressions, beliefs about gender role stereotypes, self- construals, participation in decision-making and perceptions of sexual harassment as a violation of organizational policy were used as predictors, with perceived severity of harassment was used as a dependent variable in the first regression and concern about social reactions were entered as a dependent variable in the second regression. However, as mentioned earlier, because the internal reliability for beliefs about gender role stereotype for the American group was too low to warrant firrther analysis, in the first and second regression analyses, this variable was dropped flom these two regression analyses. In the third and forth regression analyses, perceived severity and concern about social reactions were used as predictors, in predicting assertive and passive reactions to sexual harassment. Yet, due to the fact that the internal consistency of passive reactions for the Thai group was unacceptable, the forth regression in which perceived severity and concern about negative social reactions were used to predict passive reactions to harassment was omitted for the Thai group. For all of the regression analyses performed, the level of significant was set at .05 and standardized regression coefficients were used 117 to interpret relationship among the hypothesized variables. As recommended by Tsui, Egan and O’Reilly (1992), all of these regression analyses were conducted in two steps. In the first step, the variables in the model were used as a predictor, while in the second model, the controlled variables were added to examine the stability of the effects of the predictors. In sum, this study examined reactions to sexual harassment of female officers in Thailand and the United States by obtaining the data flom a convenient sample of 106 American and 109 Thai female officers who are full time officers in law enforcement agencies in the United States and Thailand. These women were those who had at least experienced one type of harassment by male coworkers within the past two years. The survey technique was used to gather the information regarding reactions to sexual harassment, beliefs about gender role stereotypes, self-construals, participation in decision-making, perceptions of sexual harassment policy, perceiVed severity of harassment, concern about social reactions. Chi-square, t-tests, and regressions analyses were used to analyze the data. 118 Chapter 4 Results This study examined reactions to sexual harassment by male officers among female police officers in the U.S. and Thailand. Specifically, it was intended to assess whether there would be any differences between female officers in the two countries in terms of gender role stereorype beliefs, self-construals, participation in decision-making, and perceptions of organizational policy prohibiting sexually harassing behaviors. It was expected that Thai female officers would endorse more gender role stereotype beliefs, have higher interdependent view of self, have lower independent view of self; report limited participation in decision-making and be less likely to agree that their organizational policy clearly prohibits sexual harassment, compared to the American counterparts. In addition, it was hypothesized that female officers who have these characteristics would be more likely than their counterparts to evaluate sexual harassment that happened to them as less severe and feel more concerned with negative social reactions in adopting assertive reactions. These perceptions, in turn, were expected to be positively associated with the use of assertive reactions and negatively associated with the use of passive reactions in handing sexual harassment. Thai women were also anticipated to view sexual harassment as less severe, feel more concerned with negative social reactions and be less likely to adopt assertive reactions in handling sexual harassment. This chapter begins with a general description of the experiences of sexual harassment of female officers and their reaction to sexual harassment. Then, the results regarding mean differences of each of the predictors and their relationship to outcome variables will be presented. 119 The Experiences of Sexual Harassment The data revealed that flom the total number of 117 American participants, 106 participants or 90.59% reported at least once experiencing sexual harassment within the past two years flom male officers in their organization. For the Thai sample, flom the total numbers of 120 participants, 109 or 90.83% reported experiencing harassment. Thus, the proportions of female officers in the two countries who experienced at least one sexually harassing behavior within the past two years were nearly identical. Table 1 depicts the data of those participants who reported having experienced at least one potentially sexually harassing behavior flom male officers within the past two years. Within the gender harassment category, suggestive jokes/offensive stories were experienced by a majority of female officers; about 92% of the U.S. sample and 95% of the Thai sample reported having experienced suggestive jokes and offensive stories. The display, use and distribution of sexist/suggestive materials were experienced by about half of the respondents in each group (50.9 and 48.6% for U.S. and Thai sample respectively). However, overall, relative to Thai officers, U.S. Officers were more likely to report experiencing several behaviors in the gender harassment category, including crude sexual remarks, being treated differently due to sex, and being put down or condescended to due to sex. Among the U.S. participants, 75.5% reported having been subjected to crude sexual remarks, but only about 47.7% of the Thai sample did. Likewise, 76.4% of the American group reported being treated differently due to sex, but only half of the Thai respondents reported being subjected to such behavior. Finally, 59.4% of American officers reported being put down or condescended to due to sex, but only 36% of the Thai sample were subjected to such behavior. The only exception is sexist remarks, which 120 Table 1 Percentage of U.S. and Thai female officers experiencing each type of harassment at least once in two years Type of Harassment (U.S. % n % 11 Gender Harassment 1.1 Suggestive stories or offensive jokes 92.5 98 95.4 104 1.2 Crudely sexual remarks 75.5 80 47.7 52 1.3 Treated differently due to sex 76.4 81 50.5 55 1.4 Displayed, used, distributed 50.9 54 48.6 53 1.5 Sexist remarks 56.6 60 64.2 70 1.6 Putdown/condescending due to sex 59.4 63 33.0 36 Unwanted Sexual attention 2.1 Unwanted discussion of personal 50.0 53 53.2 58 /sexual matter 2.2 Unwanted Sexual attention 46.2 49 50.5 55 2.3 Attempts to establish a sexual relation 22.6 24 40.4 44 2.4 Unwanted invitations 21.7 23 44.0 48 2.5 Unwelcome touching 40.6 43 32.1 35 2.6 Unwanted attempts to stroke/fondle 17.0 18 9.2 10 2.7. Sexual Assault 0.9 1 6.4 7 121 Table l (cont’d). Type of Harassment U.S. Thai % % Sexual Coercion 3.1 Subtle Sexual bribery 5.7 17.4 19 3.2 Subtle threats of retaliation for sexual 2.8 10.1 11 noncooperation 3.3 Irnplying better treatment for sexual 5.7 12.8 14 cooperation 3.4 Treated badly for sexual 6.6 11.0 12 noncooperation 3.5 Anticipated poor treatment for sexual 3.8 11.0 12 noncooperation 3.6 Treated badly for refusing to have sex 2.8 4.6 5 were reported by more Thai than Americans (64.2 and 54.6%respectively). The numbers of participants who indicated that they have been subjected to unwanted sexual attention were slightly different, with more Thai participants reporting having experienced such incidents than Americans (50.5% for Thais and 46.2% for Americans). However, information on unwanted sexual attention suggested different patterns for the two groups. Thai participants more often than the American participants reported unwanted attempts to establish sexual/romantic relationship despite discouragement, while unwelcome touching and unwanted attempts to stroke/ fondle happened to more American than Thai participants. A higher proportion of Thai participants experienced extreme forms of physical contacts, such as sexual assault; only 1% of the American respondents reported having been subjected to sexual assault, but 7% of the Thai respondents did 50. Thus, it seems that the Thai culture, which has strong prohibitions against physical contacts during cross sex interactions in public settings, may reduce milder and more public forms of harassment, but at the some time may encourage extreme forms of harassment in private settings. A higher proportion of the Thai participants consistently reported having experienced all forms of sexual coercion than did the American participants. Subtle sexual bribery was reported by 17.4% of the Thai participants but only by 5.7% of the American participants. Similarly, when asked if they have ever been in a situation where any male officers in their organizations “made them feel subtly threatened with some sort of 9’ ‘6 retaliation for not being sexually cooperative , implied faster promotions or better treatment if they were sexually cooperative , made it necessary for them to respond positively to sexual or social invitations in order to be well treated” and “made them 123 aflaid that they would be treated poorly if they did cooperate sexually”, about 10-13% of the Thais reported having been subjected to these forms of harassment. Among the U.S. respondents, the range of percentages for experiencing each situation was flom 2.8 to 6.6%. Finally, although the experience of receiving bad treatment for refusing to have sex happened to fewer participants in both countries than did other forms of sexual coercion, this type of incident also happened to higher proportion of Thai participants than Americans (4.6 as opposed to 2.8%). In addition, the respondents were also asked to indicate how often they have been subjected to each type of harassment within the past two years. When considering the overall frequency of harassment experienced by each group, American participants reported lower frequency of harassment M = 29.78, _S_I_)_ = 7.09) than did Thai participants M= 31.88, SQ ==9.79), but the difference was not statistically significant, 1 (197) = -1.80, p = .07. (The probability of a Type I error was maintained at .05 for all subsequent analyses.) The examination of mean difference in flequency of exposure to harassment for each type of harassment revealed that there were no differences in terms of gender harassment between the two groups M = 13.50, _S_I_)_ = 4.04, M: 13.06, Si); 4.44 for Americans and Thai respectively), t_ (213) =.76, p = .44. However, Thai female officers significantly more Often experienced unwanted sexual attention M = 11.54, §_I_) = 4.73) than Americans M = 9.87 E) =3.38), t (196) = -2.97, 9 =00 and also significantly more often encountered sexual coercion (M = 8.46, _S__Q_= 3.96) than U.S. female officers M = 7.42, S_D_ = 1.64), t (145) =-2.53, p=.01). It appears that the lack of country differences in terms of overall fiequency of harassment is because both groups experience similar levels of the most common form of harassment, gender harassment. However the 124 Thai group more often experiences extreme forms of harassment when that type is considered separately. Nature of Worst Case Sexual Harassment Used For Data Analysis As mentioned earlier, reactions to sexual harassment were assessed by asking the respondents to think of the worst case incident that happened to them within the past two years for each type of harassment and then indicate how they reacted in those situations. Due to the fact that there are unequal number of cases across subjects, because some participants have never experienced some forms of harassment, only one case is used for each individual. For those who had one case, that case was retained for the analysis. For those who have more than one case, one case was randomly selected. As shown in Table 2, based on 106 worse case incidents for the American sample and Table 2 Percentages of cases for each type of sexual harassment in final case composition by nation U.S. Thai % of Cases No.0f cases % of cases No.of cases Gender Harassment 64.2 68 58.7 64 Unwanted Sexual Attention 31.1 33 33.9 37 Sexual Coercion 4.2 5 7.3 8 Total Cases 100 106 100 109 109 worse case incidents for the Thai sample, gender harassment constitutes a majority, while sexual coercion constitutes the lowest percentage of cases from both groups. The final case composition for the American sample was: 64.2% gender harassment, 31.1% 125 unwanted sexual attention and 4.2% sexual coercion. Cases being analyzed for the Thai sample were: 58.7% gender harassment, 33.9% unwanted sexual attention and 7.3% sexual coercion. Thus, it appears that the final case composition for the American group involved a higher proportion of gender harassment cases but lower proportions of unwanted sexual attention and sexual coercion cases than that of Thai. However, a 2 x 3 (Nation x Harassment type) chi-square test did not reveal significant national differences on proportion of case type, )6 (2, 1:1; 215) =1, 2 =60. Therefore, this suggests that the final case compositions flom the two groups are reasonably similar and comparable. Table 3 provides further details about characteristics of the perpetrators in each of these three types of harassment. As can be seen, the data flom the American group reveals that subordinate, coworker, direct supervisor and higher superior were the perpetrator of 4.41%, 73.52%, 14.70% and 7.35% of gender harassment cases respectively. Hence, coworkers constitute most perpetrators of gender harassment. For unwanted sexual attention, subordinates, coworkers, direct supervisors and higher superiors were the perpetrator of 9.09%, 72.72%, 15.15% and 3.03% respectively. Finally, for sexual coercion, there was no subordinate’s involvement in cases being analyzed. Coworkers and direct supervisors were both involved in 40% of the cases, while higher superiors were involved in the remaining 20% of the cases. Hence, coworkers constitute most perpetrators of all types of harassment for Americans. The data flom the Thai group reveals that for gender harassment cases, the percentages of higher superiors and subordinates involved in the cases were identical; higher superiors and subordinates were the perpetrator of 6.3% of those cases, with direct supervisors constituting 15.6% of the perpetrators. Similar to the American group, coworkers were 126 Table 3 Percentages of perpetrators of each type of harassment in final case composition by nation Type of Harassment U.S. . Thai % of cases No. of cases % of cases No of cases for each type for each type of harassment of harassment Gender Harassment 100 68 100 64 Subordinate 4.4 3 58.7 4 Coworker 73.5 50 6.3 46 Higher superior 7.4 5 71.9 4 Direct supervisor 14.7 10 6.3 10 Unwanted sexual attention 100 33 100 37 Subordinate 9.1 3 8.1 3 Coworker 72.7 24 62.2 23 Higher superior 3.0 1 13.5 5 Direct supervisor 15.2 5 16.2 6 Sexual coercion 100 5 100 8 Subordinate 0.0 O 0.0 0 Coworker 40 2 12.5 1 Higher superior 20 1 50.0 4 Direct supervisor 40 2 37.5 3 127 the main perpetrators of these cases (71.9). For unwanted sexual attention cases, the identity of the perpetrator was identified as follows: subordinate 8.1%, coworker 62.6%, higher superior 13.5% and direct supervisor 16.2%. Thus, in Thailand more cases involved a direct supervisor and fewer cases involved coworkers, compared to Americans. Finally, regarding sexual coercion, the identity of the perpetrator was as follows: subordinates 0%, coworker 12.5% higher superior 50% and direct supervisor 37.5. Thus, sexual coercion by higher superior was slightly more common among Thais than among Americans. Overall, in both groups a majority of cases were gender harassment and unwanted sexual attention by a male coworker. Two 2x2 chi-square analyses (Nationality x Perpetrator’s status) were performed to test the independence between nationality and the status of the perpetrators for major types of harassment. Although originally there were 4 categories of the perpetrator, due to the fact that some types of perpetrators have been identified as the perpetrator in less than 5 cases, the analyses were conducted by collapsing perpetrator’s status into 2 categories (higher status than the victim or equal to /lower status than the victim). The analyses were conducted separately for gender harassment cases and for unwanted sexual attention/sexual coercion cases. The unwanted sexual attention and sexual coercion cases were combined because there were too few sexual coercion cases for analysis, and both types of cases reflect unwanted sexual attention. The results revealed that there was no significant relationship between nationality and status of the perpetrators for gender harassment cases, 36 (1, H= 132) = .00, p_= .57, nor unwanted sexual attention/sexual coercion cases, 75’ (l, N_= 83) = 2.49, p_=.08. Thus, the results again confirmed prior results that the final case compositions being analyzed were quite comparable. 128 Reactions to Sexual Harassment Table 4 provides the percentages of U.S. and Thai respondents who agreed that they Table 4 Percentages of women who used specific types of responses to harassment Reactions U.S. Thai % n % n Assertive reafction_s_ Report to authority 19.8 21 54.1 59 Threats 44.3 47 57.8 63 Protest 51.9 55 69.7 76 Hinting dissatisfaction 61.3 65 86.2 94 Passive reactions Go along 31 .l 33 4.6 5 Ignore 46.2 49 37.6 41 Deflect 56.6 60 88.1 96 Avoid 31.1 33 60.6 66 had adopted each strategy in responding to sexual harassment. For the U.S. participants, hinting dissatisfaction was the most common reaction to the worst case of sexual harassment experienced, as 61.3% of the respondents indicated that they used such strategies in handling sexual harassment. The next most common reaction for the American group was deflecting the behaviors of the harassers (56.6%), followed by protesting to the perpetrators directly (51.9%). Informing the perpetrators about negative 129 consequences was used by 44.3% of the respondents. The least common reaction to sexual harassment for the American group was reporting the incident to an authority, while the next least common used strategy was ignoring and going along with the behaviors. For the Thai respondents, the most common reaction to the worst case of sexual harassment was deflecting, which was used by 88.1% of the respondents, followed by hinting dissatisfaction, which was employed by 86.2% of the respondents. Similar to the American groups, the next most common responses were protesting the perpetrators directly and then informing the perpetrators about the negative consequences of the behaviors. The next most common action used by the Thai participants was avoiding the perpetrators (60.6%). Unlike the American group, which relied least on reporting to an authority in dealing with harassment, more than half of the Thai participants (54.1%) indicated that they reported the incident to an organizational authority. The least common reaction for the Thai participants was going along, and ignoring the incident was the next least employed reaction (37.6%). Overall, it seems that a higher proportion of Thai officers used assertive responses than did American officers. However, for passive reactions, the results were mixed. Going along and ignoring harassment were used by higher proportions of American respondents, while avoiding and deflecting were used by higher proportion of Thais. A 2x2 chi-square (Nationality x Strategy Presence) was conducted for each reaction strategy. The results revealed that a significantly higher proportion of Thai participants reported sexual harassment to an authority, )6 (1, lfl= 215) = .27 .08, p_= .00, used threats 130 with the perpetrators, x2 (l, l_\l_= 215) = .3.89, p_= .03, protested the perpetrators, x2 (1, E: 215) = 7.18, p_= .00 and hinted their dissatisfaction to the perpetrator, x2 ( 1, _N_= 215) = .1732, p_= .00. For passive strategies, there was no significant national differences in proportion of female officers who ignored the incident, )6 (l, _N_= 215) = 1.63, p_= .12. However, a significantly higher proportion of U.S. female officers went along with the behavior, 76 (1, fl= 215) = 26.02, p_=\.00, while a significant higher proportion of Thai female officers deflected the perpetrators’ attention, )6 (1, _N= 215) = 26.72, p_= .00 and avoided the perpetrators, x2 (1, _N_= 215) = 18.72, p_= .00). Findings From Tests of Hypotheses Regarding National Differences There are 7 hypothesized national differences in beliefs about gender role stereotypes, self-construals, participation in decision-making, perceptions of organizational policy, perceived severity, concern about social reactions and reactions to sexual harassment. It was expected that Thai female officers would endorse more gender role stereotype beliefis, have higher interdependent view of self; have lower independent view of self; report limited participation in decision-making and be less likely to agree that their organizational policy clearly prohibits sexual harassment, compared to the American counterparts. In addition, they were also expected to view sexual harassment as less severe, have higher concern about social reactions, have lower scores on assertive reactions and have higher scores on passive reactions. However, due to the problem of internal consistency of beliefs about gender role stereotype in the U.S. sample, and the problem of internal consistency of passive reactions for the Thai sample, these two variables were not used to perform the bivariate tests of significance. As can be seen flom Table 5, for other variables that are tested, Thais had significantly higher mean 131 Table 5 Tests of differences between variable means for the two countries Variables U.S. t p M. .52 M. 5.2 Gender role stereotypea 7.08 2.36 9.84 3.40 -- -- Interdependent self 29.32 5.78 36.90 4.81 -10.46 .00** Independent self 31.72 4.06 27.80 4.59 6.60 .00** Participation 16.00 5.00 16.81 5.23 -1.18 .23 Perceived policy 12.55 2.43 9.89 2.62 7.69 .00** Perceived severity 11.47 2.78 12.34 2.68 -2.78 .00** Concern about social 7.84 3.30 8.87 3.05 -2.37 .01** reactions Assertive reactions 11.95 3.94 14.61 2.87 -5.64 .00** Passive reactions 3 11.36 3.20 12.10 2.25 -- -- Note. ' T-tests are omitted due to low internal consistency of the scale score in certain groups "p<.01, * p<.05 scores on interdependent view of self , t_(213) =10.46, p=.00 and had lower scores on independent view of self 1 (213) =-6.60, p=.00. However, there was no differences in participation in decision—making, t_(213) =-1.18, p_=.23. Thais were less likely to view that their organizational policy clearly prohibits sexual harassment, t_(213) =7.69, 9 =00. They viewed harassment as more severe, t (213) =-2.78, p_=.00 and reported higher concern about social reactions. _t (213) =-2.37, 2 =01. Finally, they also had significantly higher scores on assertive reactions than Americans, t (213) = -5.64, p=.00. The results of the t-tests are summarized for each hypothesis. Hypothesis 1 a: As expected, the inspection of the mean scores for this scale suggested that Thai women endorsed more traditional gender role beliefir than Americans (M=9.84, _S_1_)_=3.4 as opposed to _M_=7.08, §_I_)_=2.36). However, the t-test was omitted for this variable, as the internal consistency for the American group was too low to warrant further analysis. Hypothesis 2 a: The predictions made regarding national differences in self-construals were supported. Thai female officers had significantly higher scores on interdependent view of selfbut had significantly lower scores on independent view of self Hypothesis 3 a: Findings were inconsistent with the prediction that Thai female officers will be more likely to perceive that they have lower level of participation in decision- making, compared to the American female officers. This hypothesis was not supported; there was no differences in participation in decision-making Hypothesis 4 a: As predicted, Thai female officers were less likely to agree that their organizational policy clearly prohibits sexual harassment, thus, this hypothesis cannot be rejected. Hypothesis 5a: Contrary to the expectation that Thai female officers would perceive sexual harassment as less severe than American female officers, Thai female officers perceived sexual harassment as more severe. Hypothesis 6 a: Consistent with the prediction that Thai women would have higher concern about social reactions, Thai female officers had higher concern about social reactions than American officers, Hypothesis 7a: Contrary to the prediction that Thai officers will have lower scores on 133 assertive reactions compared to American officers, Thai women were more likely to adopt assertive strategies in dealing with sexual harassment. Additionally, Thai women were expected to have higher scores on passive reactions than American women. The raw data reveals that Thai women also had higher average scores on passive reactions than Americans (M =12.10, _S_Q =2.25 for Thais as opposed to _M =11.36, SQ =3.20 for Americans). Because the passive strategies scale for the Thai group has low internal consistency, the t-test for differences between the means may be misleading. Thai participants who adopt certain passive strategies may rely less on other types of passive strategies. Further inspection of t-tests of mean differences for each individual type of passive reactions reveals that Thai women significantly had low scores on the strategy, going along, 1 (183.80) = 5.97, p=.00, but had higher scores on avoiding the perpetrators, t (213)-4.91, p=.00 and deflecting the perpetrators ’ attention, L(l68.71)=—6.07, p=.00. However, there was no difference on scores for ignoring the incidents, t_(213)=0.6l, p=.53. Assessment of Predictors Of Perceived Severity, Concern about Social Reactions and Reactions To Sexual Harassment For each sample, separate blocked regressions were run for each of the three dependent variables: perceived severity, concern about social reactions and reactions to sexual harassment. As mentioned earlier, the procedure recommended by Tsui, Egan & O’Reilly (1992) in which the predictors were entered first, and then control variables were added was employed. The two step process allows for addressing the primary question, “are the independent variables related to the dependent variables as predicted by theory?’ and then to addressing the question, “could any relationship be the result of a control 134 variable”? This method has been successfully used to ascertain the effect of predictors. It should be noted that due to low internal consistency of some scales, it was necessary to drop some variables for at least one country. For the U.S. group, all hypotheses regarding beliefs about gender role stereotypes were not tested. For the Thai sample, all of the hypotheses involving passive strategies were not tested. U.S. Findings Table 6 depicts intercorrelations among study variables for the American group. As can be seen, perceptions of organizational policy was correlated with perceived severity, while perceived severity was related to assertive strategies. Both dimensions of self contruals were related to concern about social reactions, but in a different direction. Interdependent view of self was positively correlated with concern about social reactions but independent view of self was negatively related to concern about social reactions. The two dimensions of self-construal were also negatively correlated with each other. In addition, participation in decision-making was negatively related to concern about social reactions. Concern about social reactions was positively associated with passive strategies. Apart from that, no other correlation was significant. 135 Table 6 Intercorrelations among study variables for the U.S. sample Variables Correlations 1 2 3 4 g 5 6 7 8 Gender role a -- stereotype Interdependent self -- Independent self -- -.44** Participation -- -.03 .16 Perceived policy -- .06 .05 .08 Perceived severity -- .02 .08 -.O3 .47** Concern about -- .27** -.36** ~.32** .03 .01 social reactions Assertive reactions -- -.13 .10 .05 .21 .46** -.O3 Passive reactions -- .19 -.09 .10 .10 -.O8 .37" .10 M ' All of the correlations between beliefs about gender role stereotype and other study variables are omitted due to low internal consistency of the scale assessing beliefs about gender role stereotype for the U.S. group. *‘p<.01, "‘ p<.05 136 Regression results for the U.S. sample2 For the US group, regression analyses were performed for each of the four dependent variables, including perceived severity, concern about social reactions, assertive reactions, and passive reactions. The results of regression analyses for the U.S. group are presented in Table 7 and Table 8. As seen in Table 7, when self-construals, participation in decision-making and perceived organizational policy were used to predict perceived severity, the entire model accounted for 20.3% of the variance in the perceived severity score, F (4, 101) = 7.67, p=.00. Once control variables were added into the model, the model accounted for 25.1% of the variance, F (6, 99) = 6.87, p =.00. Both before and after the introduction of control variables, the only significant predictor of perceived severity is whether the respondent perceived that there is a strong department policy against harassment ( t = 5.37, b = .47, 2 =00 and 1 = 6.07, LP .53, p=.00 respectively). Additionally, the control variable, working primarily with men, is negatively related to perceived severity of harassment ( t_ = -261, l_)_= .23, p_=.01) In 2 Prior to conducting the analyses, scatterplots of standardized residuals as a function of standardized predicted values were inspected to assess the assumptions of linearity (Osborne & Waters, 2002). The results revealed that each of the outcome and independent variables were linearly related to one another. Additionally, the Durbin-Watson statistics were assessed to ascertain the assumptions of independent of errors (Tabachnick & Fidell, 1989). The Durbin Watson statistics ranged from 1.87-2.04, which fall in the range of 1.5-2.5 (Garson, 2003), suggesting that there was no violation of this assumption. The observedvvariance inflation factor indexes ranged from 1.00 to 1.30, which is lower than the cut-off point of 4 (Garson, 2003), indicating that multicollinearity is not a problem in the analysis. 137 Table 7 Predicting perceived severity and concern about social reactions for the U.S. sample Variables Step1 Step2 B t p B t p Perceived Severitya Interdependent self .02 0.22 .82 -.03 -O.3O .76 Independent self .08 0.80 .42 .05 0.58 .55 Participation -.08 -1.00 .3 1 -.O3 -O.4O .68 Perceived policy .47 5.37 .00** .53 6.07 .00** Frequency of harassment -- -- -- .13 1.49 .13 Work group sex ratio -- -- -- -.23 -2.61 .01** Concern about social reactionsb Interdependent self .15 1.59 .11 .12 1.28 .20 Independent self -.24 -247 .01** -.26 -275 .oo** Participation -.28 -3.16 .00** -.23 -271 .00** Perceived policy ..00 -0.06 94 -.01 on .89 Frequency of harassment -- -- -- .29 3.36 .00" Work group sex ratio -- -- -- .10 1.26 .20 Note. '18 = .203, F (4, 101 ) = 7.67, p_=.oo in step 1 and R2 =.251, F (6, 99) = 6.87, p= .00 in step 2. I’18 =.193 ,1= (4, 101)= 7.29 , p_=.oo in step 1 and R2 =.277, F (6, 99) = 7.69, p=.oo in step 2 138 other words, American women in male dominated work group were more likely to view sexual harassment as less severe. For concern about social reactions, in the first step, when only self-construals, participation in decision-making and perceived organizational policy were used as predictors, the entire model accounted for 19.3% of the variance. In this model, independent view of self and participation in decision-making were negatively related to concern about social reactions ( t = -2.47, 1; =24, p =.Ol and t = -3.16, b = -.28, p = .00 respectively). After the two control variables were added into the model, the model accounted for 27.7% of the variance. In this model, participation in decision-mala'ng and concern about social reactions were still significant predictors of concern about social reactions (t_= - 2.75, b = -.26, 2 =00 and t_= -2.71, b = -.23 , 2 =00 respectively). In addition, frequency of harassment was also positively related to concern about social reactions (1 = 3.36, b =.29, 2 =00). Thus, American women who had high independent view of self, high participation in decision-making and who were subjected to sexual harassment less frequently were less likely to report that they felt concerned about negative social reactions than their counterparts. Table 8 shows that when perceived severity and concern about social reactions were used as predictors of assertive reactions in responding to sexual harassment, the model accounted for 20.6% of the variance, F ( 2, 103) = 14.64, p_= .00. In this step, only perceived severity was positively correlated with assertive reactions (1 = 5.39, b = .46, p=.00). In the second step, once control variables were added, the overall model 139 Table 8 Predicting assertive and passive reactions to harassment for the U.S. sample Variables Step 1 Step2 B t g p B t p Assertive macaw? Perceived severity .46 5.39 .00** .46 5.23 .00** Concern about social reactions -.04 -0.47 .63 -,09 -1 .00 .31 Frequency of harassment __ __ __ .14 1.49 .13 Work group sex ratio __ -- -_ .02- 0,26 .78 Passive reatctionsb ‘— pmeivedsevefity -.01 on .88 .02 .024 .30 Concern about social reactions '3 7 4-08 ~00" -31 3.21 -00** Frequency of harassment " " “ .15 1-54 .12 Work group sex ratio -- -- -- .03 ' 0.35 .72 Note. ‘ R7= .206, F (2, 103 ) = 14.64, 2 =00 in step 1 and R2 =.209, F (4, 101) = 7.92, p = .00 in step 2. "R2 =.123,1=(2,103)= 8.33, p =.oo in step 1 and R2 =.127,r(4,101)= 4.81, p = .00 instep 2. accounted for 20.9% of the variance, F ( 4, 101) = 7.92, p_=.00. Again, perceived severity was the only significant predictor of assertive reactions, (t_= 5.23, p_= .46, p=.00). For passive reactions, the first model account for 12.3% of the variance, F ( 2, 103) = 8.33, p = .00 and the second model accounted for 12.7% of the variance, F ( 4, 101) = 4. 81, p = .00. Both before and afier the control variables were added into the model, concern about social reactions is the only variable that is a significant predictor of passive reactions (t_= 4.08, l_)= .37, p = .00 and t = 3.21, l_) = .31 , 1; =00 respectively), 140 while perceived severity, frequency of harassment and work sex ratio were not related to the use of passive strategies in handling harassment. Thai Findings Table 9 reports the means, standard deviations and correlations of the study variables. For the Thai sample, out of four variables hypothesized to be associated with perceived severity, two had a significant correlation with perceived severity. Gender role stereotypes beliefs were negatively associated with perceived severity in a hypothesized direction, and independent view of self was positively correlated with perceived severity in an expected direction. On the contrary, interdependent view of self, participation in decision-making and perceptions of organizational policy were not significantly correlated with perceived severity. Gender role stereotype beliefs and interdependent view of self were correlated with concern about social reactions, while independent view of self, participation in decision- making and perceptions of organizational policy were not associated with concern about social reactions. The correlation results revealed mixed support for the effects of variables hypothesized to be related to reactions to harassment. On the one hand, as expected, perceived severity was positively associated with assertive reactions, but on the other hand it was not associated with passive reactions. In addition, perceptions of organizational policy was positively associated with assertive reactions. 14] Table 9 Intercorrelations among study variables for the Thai sample Variables Correlations l 2 3 4 _ 5 6 7 8 Gender role stereotype Interdependent self .12 Independent self -.26** -.08 Participation -.01 .01 -.06 Perceived policy -.02 -.03 .00 .11 Perceived severity -.28** -.01 -.29** -.02 .05 Concern about .24 .24* .29** -.03 -. 12 .03 social reactions Assertive reactions .12 .01 -.08 .12 .29** .39** .06 Passive reactions 3 - -- -- - - - - -- -— Note ' All of the correlations between passive reactions and other study variables are omitted due to low internal consistency of the scale assessing passive reactions for the Thai group. ”p<.01, * p<.05 142 Regression results for the Thai sample 3 The results of regression analyses for the Thai sample are reported in Table 10 and Table 11. Table 10 reports the results of predicting perceived severity and concern about social reactions. First, when gender role stereotype beliefs, interdependent view of self, independent view of self, participation in decision-making and perceptions of organizational policy were entered as independent variables in order to predict perceived severity, the overall model accounted for 9.4% of the variance, F (5, 103) =3.22, p_=.01. In this model, beliefs about gender role stereotypes and independent view of self were the two variables that are related to perceived severity (t = - 2.40, 1; =23, p_ =.01 and t = 2.38, l;=.22, p =.01 respectively). While women who endorse beliefs about gender role stereotype view sexual harassment as less severe, women with independent view of self view sexual harassment as more severe. Once the control variables, including frequency of harassment and worlgolace sex ratio, were introduced in the model, the overall model accounted for 12.3% of the variance, F (7,101) =3.l6, p=.00. However, in this model, while beliefs about gender role stereotypes were still a significant predictor of perceived severity, independent view of self was no longer a significant predictor of perceived severity (t = -2.4, _l; = -.22, 9 =0] and t_= 1.94, b =.l8, 9 =05). In addition, frequency of harassment was also positively correlated with perceived severity (t_=2.10, l_>=.20, p_= .03). 3 Similar to the American group, prior to conducting regression analyses, scatter plots were inspected and revealed that the relationships between each of the predictors and dependent variables are linear. The Durbin-Watson statistics obtained indicates that error correlations were not a problem, as the values of this index obtained range from 1.94-2.08. In addition, the highest VlF statistics was 1.14, suggesting that multicollinearity is not a problem. 143 Table 10 Predicting perceived severity and concern about social reactions for the Thai V sample Variables Step 1 Step2 B t p B t P Perceived Severity a Gender role stereotype -.23 -024 01* -.22 -240 01* Interdependent self .01 0.17 .86 -.04 —0.48 .62 Independent self .22 2.38 .01* .18 1.94 .05 P ”9913390“ -.01 -0.18 .85 -.00 -0.07 .94 Perceived P°1i°Y .05 0.59 .55 .06 0.74 .45 Frequency of harassment __ __ __ .20 2.10 .03... Work group sex ratio __ __ __ .06 0.70 .48 Concern about social reactionsb Gender role Stereotype .21 2.24 .02* .20 2.15 .03* Interdependent self .26 2.93 .00** .26 2.73 .00** Independent self .01 0.14 .88 .01 .11 .90 Participation -.13 -1.43 .15 -.13 -1.42 .15 Perceived policy .06 0.68 .49 .06 .68 .49 Frequency of harassment -- -- -- -.00 -0.00 .99 Work group sex ratio -- -- -- .03 0.32 .74 Note.‘ R2=.094,F(5, 103)=3.22,p=.01 instep 1 and R2 =.123,1=(7, 101)=3.16,p=.001n step 2. ”R2 = .107, F(5, 103') = 3.59, 2 =00 in step 1 and R2 =.091, F (7, 101) = 2.53, p = .01 in step 2. 144 Regarding concern about social reactions, in the first step, when control variables were not considered, the four predictors combined accounted for 10.7% of the variance, F (5, 103) = 3.59, 1; =00. In this model, two variables, beliefis about gender role stereotype and interdependent view of self were both positively related to concern about social reactions. In the second model, the overall model accounted for 9.1% of the variance, F (7,101) = 2.53, p=.01. Both beliefs about gender role stereotype and interdependent view of self were still related to concern about social reactions (1 = 2.24, b = .21, p = .02 for beliefs about gender role stereotype and t = 2.93, 1; =26 , p_=.00 for interdependent self), but the two control variables were not related to concern about social reactions. Table 1 l Predicting assertive reactions to sexual harassment for the Thai sample Variables Step 1 Step2 B t p B t P Assertive regetioni Perceived severity .39 4.46 .00** .36 4.02 .00** Concern about social .05 0.63 .52 .04 0.48 .62 reactions Frequency of -- -- -- .09 0.96 .33 harassment Work group sex ratio -- -- -- .07 0.85 .39 Note '112 = use (2, 106) = 10.28, p =.00 in step 1 and R2 =.146, 1= (4, 104) = 5.62, p = .00 in step 2. Table 11 presents the results of predicting assertive reactions to sexual harassment for the Thai sample. As seen, in step 1, the overall model accounted for 14.6% of the variance, F (2, 106) =10.23, p=.00. In step 2, once control variables were added, the model still accounted for 14.6% of the variance in assertive scores, F (4, 104) = 5.62, p = 145 .00. In both models, the only significant variable in predicting assertive reactions for the Thai sample was perceived severity (t = 4.46, t_r = .39, p_ =.00 in the first model and t = 4.02, b_= .36, p = .00 in the second model). Summary of Regression Results In sum, for the U.S. sample, only perceptions of sexual harassment policy can predict perceived severity; those female officers who perceived that their organizations prohibits sexual harassment that they encountered were more likely to evaluate the incident as more severe. Perceived severity was in turn was positively related to the use of assertive reactions. Two variables, independent view of self and participation in decision-making were negatively related to concern about social reactions. Concern about social reactions was found to be positively associated the use of passive strategies. The two control variables were also significant; women in male dominated work groups tend to view harassment as less severe, and women who experienced sexual harassment more ofien felt more concerned about negative social reactions. The results from regression analyses for the Thai sample revealed that beliefis about gender role stereotypes were significantly related to perceived severity. Those female officers who had strong beliefs about gender role stereotype viewed sexual harassment as less severe, compared to those who had more liberal views toward women. In addition, women with strong beliefs about gender role stereotype and women with interdependent view of self showed more concerned about social reactions. However, only perceived severity but not concern about social reactions was related to assertive reactions to harassment. The results of regression analyses are summarized for each hypothesis. 146 Hypothesis 1b and 1c: Beliefs about Gender Role Stereotypes The hypothesis that beliefs about gender role stereotype will be positively related to perceived severity of sexual harassment but negatively related to concern about social reactions can only be tested with the data from the Thai sample due to low internal consistency of this measure for the American group. As predicted, beliefs about gender role stereotypes were negatively associated with perceived severity of sexual harassment and positively associated with concern about social reactions. Hypothesis 2b and 2C: Self-Construals Inconsistent with the hypothesis that those who are high on interdependent view of self will view sexual harassment as less severe than female officers who are low on interdependent view of self: the results for both groups suggested that interdependent view of self was not related to perceived severity. However, the hypothesis that interdependent view of self will be positively related to concern about social reactions was partially supported. Interdependent view of self was positively related to concern about social reaction only in the Thai sample, but not the American sample. Contrary to expectation, independent view of self cannot predict perceived severity in both samples, although for the Thai group, its relationship with perceived severity was nearly significant. As expected, independent view of self was negatively related to concern about social reactions. However, this result was found only in the American group. Hypothesis 3b and 3c: Participation in Decision-Making The hypothesis connecting participation in decision-making and perceived severity was not supported in both groups. Participation in decision-making was not found to be 147 predictive of perceptions of severity of sexual harassment in both samples. The hypothesis connecting participation in decision-making and concern about social reactions was partially supported. The data fiom the American sample suggested that as participation in decision-making increases, concern about social reactions decreases, but this relationship was not found in the Thai sample. Hypothesis 4a and 4b: Perceived Organizational Policy The hypothesis that female officers who perceive sexual harassment acts as definitely prohibited by organizational policy will view as sexual harassment as more severe than those who do not perceive sexual harassment as prohibited organizational policy was supported only by the data fiom the American group but not by the data from the Thai group. Thus, this hypothesis was partially supported. Contrary to the expectation that perceived organizational policy will be negatively related to concern about social reactions, the data fi'om both samples revealed no significant relationship between the two. Hypothesis 5b Perceived Severity As expected, the results from both samples suggested that perceived severity of sexual harassment was positively related to female officer’s use of assertive reactions. However, due to low internal consistency of passive reactions for the Thai group, the hypothesis connecting the negative relationship between perceived severity of sexual harassment and female’s officer’s use of passive reactions can only be tested by using the data from the American group. However, this hypothesis was not supported. Perceived severity was not significantly associated with American female officer’s use of passive reactions. 148 Hypothesis 6b: Concern about social reactions For both the Thai and the U.S. groups, findings were inconsistent with the prediction that concern about social reaction will be negatively associated with female officer’s use of assertive reactions. However, consistent with the prediction, American officers who had high concern about social reactions were more likely to adopt passive reactions in dealing with sexual harassment than their counterparts. This hypothesis, however, cannot be tested with the Thai data. Qualitative Findings Although quantitative data has provided some information about how women officers respond to their worst case experience of sexual harassment, it is useful to understand the experiences of police women in the two countries in the broader context of the nature of harassment women are subjected to on a daily basis. Such information may be useful in understanding how women in the two countries experience, perceive and react to harassment in general. Therefore, the participants were asked to identify things that any male officers in their agency had done or said that made them feel uncomfortable and to indicate how they reacted. The data suggested that there are some noteworthy similarities and differences between countries. As found in previous research (Welsh, 1997), gender harassment is a common occurrence in male-dominated environments both in the United States and Thailand. Although quantitative data suggested that offensive jokes and crude remarks happened quite often and were witnessed by nearly every woman, the responses from qualitative data suggested that female officers in both places were not very troubled by these jokes and remarks if they were not targeted directly at a specific victim. Few officers 149 mentioned bothersome incidents when men were simply talking in a nasty way. It seems that many female officers expected such talk as a part of the job in male—dominated environment. Moreover, due to the fact that most of these behaviors are not directed at the participants, these behaviors therefore are evaluated as inevitable routines of the job For example, as one Thai female officer put it. I used to work in a place (police department) where there were more than 100 male officers, with only 2 female officers. Hearing offensive remarks used between them was unavoidable. Since I have been transferred to a new place, where there is an equal proportion of women and men, I have never heard offensive remarks or jokes ever since. Similarly, American female officers also have accepted offensive jokes and crude remarks as part of working in police organizations. One mentioned that There have been many occasions where male coworkers have made sexually explicit jokes or comments. However, I choose not to take the talk personally. I don’t find their behavior offensive, just childish. I don’t put any energy into being concerned about what people are joking about ...... but I don’t dwell on it. For American women, these comments they received were not usually directed at anyone in particular, but instead seemed to be used by male officers to establish the sense of esprit de corps among officers in the work unit. One American female officer indicated that some male officers may tell sexually explicit jokes or pass intemet pictures/jokes around, but they do it discreetly and only with the group they work with. It was explained that the cohesion and the trust within the groups has been viewed as playing an important part in how these behaviors were perceived. Several female officers 150 also indicated that these behaviors provide an opporttmity for them to be part of the group and they responded by kidding male officers back. Most believed that if they were ever in a situation where they felt uncomfortable, they would let the men know and think that the men would listen. One officer remarked Offensive jokes and comments were made quite frequently, however, I participate in the joking and do not find it offensive. I believe that if I did find it offensive and stated it to my coworkers and supervisors, they would stop. Most male coworkers and supervisors consider me “one of the group”. The situations were viewed differently in Thailand. Although Thai female officers also view several jokes and crude remarks as a common part of working in male- dominated environments, such perceptions are limited to the situations where the behaviors are done between men and they happened to witness it because of being there. On several occasions, these behaviors were directed to a specific target for a purpose of embarrassing the women in order to convey male-dominance. While, several American officers explicitly said that they were not offended by most jokes and that they participated in the joking, no one in the Thai sample made such a statement. Instead, most women found the behaviors offensive, annoying, or embarrassing and indicated that they usually tried to ignore them or avoid interacting with particular persons. Sometimes, when they felt that the limits were reached, they decided to take actions against the person, for example by confronting or threatened him. The kind of behaviors that Thai female officers were subjected to suggests that Thai male officers are less sensitive to women’s issues. In one instance, very crude remarks were made. For example, one low-ranking female officer indicated that while she was 151 naveling with three male police officers to other city, upon seeing many young girls in the back of a pick-up, one male lieutenant asserte “ ..(I) would really love to have sex with those children.” Men also frequently made jokes and comments about particular women whom they wanted to embarrass. U.S. women usually described male officers who seemed to know not to cross the line established by their female coworkers. In contrast, many Thai policemen exhibited behaviors that greatly offended their female colleagues. Thai female officers, however, seemed to feel reluctant to do anything to convey their satisfactions and often waited until when harassment was repeated and become severe before confionting their harassers. Perhaps, this may be due to the fact that in Thai culture, people are supposed to be aware of one another’s feelings. Therefore, behaviors that are not evaluated as severe violations were tolerated and rarely brought to the attention of the harassers. This is not to say that the American participants were not subjected to behaviors that they found offensive, it is just simply that it happened to few of the women. For example, one American correctional officer reported that she was humiliated when an officer said to her “She’s sweetheart and She swallows too.” The same officer also reported that one male officer made a comment to her, “ I saw you off duty the other day at a party store and thought, I did not know there were prostitutes working in ....... (the name of the township)” Another American ofiicer who is a patrol officer described that she witnessed an incident where a male trooper was greeting a female trooper by putting the back of her hand on his crotch and said “Good morning” in a room full of troopers. However, in these types of incidents that are evaluated as serious by American officers, often times, the situations are resolved by the fact that male officers were later aware of 152 how their behaviors might have offended the women and then apologized, or the women were able to give a sharp reply and feel better with the situations. For example, a correction officer stated that she reported other male officers calling women “chicks,” and after telling him, he apologized and promised not to use that word around her again. Thus, the willingness of American female officers to communicate their dissatisfaction and the willingness of male officers to accept their mistakes may also help ameliorate the situations. In Thailand, the situations often times become more difficult for women to respond to because of the differential status of the speaker and the women. For example, a Thai sergeant reported having heard a higher ranking officer saying how easy he can get women (into his bed). Being angry with the speaker, she confronted the person but was then blamed by her direct superior for being out of line as her status was lower. This female sergeant reported she felt fi'ustrated with the strong emphasis. on the ranking system and felt that lower status people must unwillingly accept insulting comments. The bothersome display of explicitly sexual materials was mentioned more often by female officers in Thailand. This may be because sexual harassment policy in the United states clearly prohibits these behaviors, which are less ambiguous than other forms of harassment. Respondents mentioned the installation of nude pictures of women as a wall paper on the desktop computer that women officers use. Some female officers also mentioned that male officers would bring in magazines, calendars, or even videotapes of a sexual nature, and discuss the sexual activities in them, and dero gate women during the discussion. However, most of the women stated that they have not done anything to change the situations. One woman did mention telling a male coworker to remove the 153 nude calendar from the wall several times, but went on to say he ignored her requests. The lack of clear rules prohibiting these behaviors in the workplace may leave women feeling reluctant to try to change things. For the American sample, there was only one correctional and one patrol officer who mentioned that their male coworkers were reading playboy at work during the shift. However, it was reported that normally, once the persons witnessed that other people came into the post, the material was put out of view For American female officers, the types of situations that bothered them the most involved both behavioral and verbal sexist behaviors of male officers. Time and time again, American female officers mentioned a double standard at work. They felt their competence was disregarded and opportunities to display their competence were not offered. For example, one female officer wrote “ I was patting down a male inmate and another officer (male) stepped in and took over-looking like I could not handle the situation.” Some other comments are: The females are often given the job of covering the back door during an entry to a building. The men want to go inside and they usually volunteer the female to stay outside. . ..Most often during a high risk situations, female officers are given the least important job. Mainly the only issue that I have had to deal with are men acting like they can handle a situation better than I can. So, for example, if I get a call with a male officer, he would sometimes just step in and assume the authoritative position without letting me do my job because he thinks I can’t handle it as well as he can.. . .Some male officers have indicated that they would not want certain female 154 officers as back up. I find that I personally put forth great effort and display courage and assertiveness in said situations (high risks situations) to try and earn the respect of fellow male officers. I feel we are earning the same wage, share the same job title, I therefore should be subject to the same risks. I feel good female officers have to try twice as hard to earn half respect as male officers. . .. Thai female officers are also subjected to sexist behaviors. However, because their day-to-day job within the organization was relatively more structured than the police work for the U.S. officers, job assignment was not much of an issue. Thai female officers were bothered by the fact that they were ignored and sometimes their orders were treated as inconsequential compared to those of male offices. One female lieutenant who worked at the police academy explained that the police cadets are disrespectful towards her, while they were more respectful towards people in a position to give them rewards or punishments (e. g., commanding officers). When a cadet did something wrong and she mentioned it, he would respond negatively, asking her what she wants fiom him, and acting like he did not have time for this. Although it has been mentioned less often than gender harassment, unwanted, unwelcome verbal and non-verbal sexual attention was reported by some participants in both groups. However, unwanted sexual attention was less often mentioned by Americans than Thais. Perhaps, the nature of Thai policewomen’s job, which many consider to be “feminine,” results in the Thai women being more often subjected to unwanted sexual attention. Some Thai female officers identified comments or questions about personal, and private matters as unwelcome. In a relationship centered society like Thailand, 155 generally, it is quite natural that coworkers would ask questions about one’s personal life out of concern for another’s well-being. However, in some instances, some men may use this as an opportunity to learn more about women and make a decision about pursuing sexual interests. Whether individuals interpret these behaviors as showing interests that go beyond concern for another’s well-beings depends on how it was done. Some Thai women reported feeling uncomfortable about these situations. For example, a female lieutenant in her early thirties felt uncomfortable that she was called by her supervisor to meet him in private in his office and being asked if she had a boyfriend or was married. She felt like the question went beyond showing general concern about her well-being. Particularly for many single female officers, it was bothersome to be asked why they have never been married. Although the Thai police department emphasizes the chain of command, women are not flee from unwanted sexual attention by subordinates. One woman reported that when shewas having trouble opening a bottle of coke, a subordinate stepped into help her and said, “Can I open your virginity, please?” Another female officer reported that one of her male subordinate kept watching her in the office, as well as showed up at her house and at the university she attended. Even though she had communicated to him that she was not interested, he continued to stalk her. One police lieutenant mentioned that she had been in situations where a non-commissioned officer in her unit was drunk and told her that he liked her and asked why she had stayed away from him. Her reaction was to avoid interacting with him. Unwanted sexual attention was reported by some participants in the United States. However, although the incidents happened, U.S. women did not have strong reactions. 156 Perhaps, it was because the harasser usually tried to get a “foot in the door,” but when there was no reciprocation, withdrew. The harassers seemed to be aware that they had crossed the line, which may have lowered the resentment of female officers. For example, one female court officer reported that while at a department party, a married supervisor from another division tried to made sexual advances towards her. He also tried to asked her for a good night kiss and offered to buy her a drink. This female officer refused as politely as possible. After that incident, he acted as if he were afi'aid she would mention the incident, avoided making eye contact, and tried to avoid her. Another female officer mentioned that after a male officer told her she “had a nice butt” and he would like to see her in a pair of thongs, she told him she did not appreciate his comments. The male officer apologized and accepted the fact that he was out of line. In both groups, none of the participants described a situation that could be characterized as sexual coercion through the exchange of rewards or avoidance of punishment for sexual favors, as things that had made them feel uncomfortable. This is inconsistent with the quantitative data that suggested that these situations have happened to some of the participants. There might be several reasons why these types of incidents were not reported. First, male officers may rarely make explicit verbal statements about what they would do if women do not engage with them sexually. This would certainly put male officers in a weak position for defending themselves in the event of, accusations. Therefore, threats may be subtle, rather than explicit, and women may see them as being ordinary unwanted sexual attention. Second, many women who experienced these types of harassment may already have suffered career repercussions and left the organization. Third, although this study has guaranteed the participants that their anonymity would be 157 protected, some may have felt concerned that if they described the incident, it would be unavoidable to provide specific details, which might give some clues about their identity and the harasser’s identity; therefore, fear of retaliation. and rumors might be another consideration that led the victims not to mentioned this type of harassment. Forth, it might be possible that some women who experienced these types of incidents were forced to comply with the harassers’ wish. As a result, over time, it might be more comforting for them to disassociate the connection between any positive benefits they received and their responsive behaviors so that they can maintain a positive image of themselves. In sum, the qualitative data revealed that hostile environment harassment made female officers feel uncomfortable in both countries more often than any other types of harassment. However, women in the U.S. were highly concerned about sex discrimination which is related directly to the job, while Thai female officers were very concerned about sexual remarks. Unwanted sexual attention was also noted by both groups, but the incidents experienced by the U.S. group seemed to be part of a gradual test of women’s interest. Overall, female officers in the United States were more likely to view situations that made them feel uncomfortable as not difficult to handle. They tended to have the impression that, once the undesired behavior of the male officers in question was acknowledged, the men would stop the behaviors. Many female officers reported that once the male officers were notified about the inappropriateness of their behaviors, those male officers were willing to apologize and back down. The only issue that seems to be unresolved for them is systematical sex discrimination on the job that they experienced due to the male officers. Women saw this as reflecting problems with police administrators rather than between themselves. Female officers in Thailand were less 158. likely to believe that acknowledging the inappropriateness of their behaviors would help improve the situations. Accordingly, many of these situations were left unresolved, and women remained unsatisfied by the outcomes. Summary of Findings In conclusion, the main findings for the present study suggested that as hypothesized, Thai female officers had stronger beliefs in gender role, higher interdependent view of self, lower independent view of self and were less likely to agree that their departments’ policy prohibited against sexual harassment. However, there was no difference in participation in decision-making reported by the Thai and American participants. While several features expected to be characteristics of the West and the East were confirmed in this sample, not all of these features had the impact on outcome variables. For the Thai sample, only beliefs about gender role stereotypes and interdependent view of self were important predictors of outcome variables. Gender role stereotype beliefs were negatively related to perceived severity of harassment. Perceived severity of harassment, in turn, was positively related to the use of assertive reactions in responding to harassment. In addition, women with traditional gender role beliefs and women with an interdependent view of self had more concern about social reactions for adopting assertive reactions in handling harassment, compared to their counterparts. However, concern about social reactions was not related to assertive reactions to sexual harassment. For the American group, women who perceived sexual harassment as an organizational policy violation were more likely to view it as severe. Perceived severity was in turn associated with the use of assertive responses. Moreover, independent view of self and participation in decision-making were negatively related to concern about 159 social reactions. Concern about social reactions was in turn associated with the use of passive reactions. The results also revealed that Thai women viewed sexual harassment as more severe, had higher concern about social reactions than Americans. They were also more likely to adopt more assertive strategies than Americans. High concern about negative social reactions among the Thai women did not appear to inhibit them from adopting assertive reactions. In fact, a significantly higher proportion of Thai employed assertive reactions than did the Americans. This appeared counterintuitive, given that Thai women have several features that should be associated with the discomforts of adopting assertive reactions. However, qualitative data revealed that the circumstances surrounding sexual harassment that happened in the two countries might be different. While many American officers viewed sexual harassment by male officers that happened as trivial, easy to handle and as part of establishing the sense of cohesion in the work groups, Thai women viewed sexual harassment by male officers intheir organizations as crude, intentional and did not expect that the harassers would apologize and modify their behaviors. Unsurprisingly, there are more reasons for them to adopt a wide variety of assertive responses to those situations and to act more defensively. 160 Chapter 5 Discussion The purpose of this research was to explore and explain reactions to sexual harassment among female officers. It examines how policewomen in Thailand and the United States interpret sexual harassment and react to sexual harassment. Despite a growing body of literature on sexual harassment within the past few decades, little research to date has focused on reactions to sexual harassment in the police context. This is surprising given that male-dominated organizations such as police agencies have been viewed places where sexual harassment flourishes. Moreover, the examination of reactions of harassment in Asian police organization and the interpretation by victims is virtually uncharted. While several studies conducted in North American have demonstrated that several individual differences and organizational characteristics contribute to victims’ tendencies to become passive toward harassment, these features have been viewed as pervasive in Asian organizations and are assumed to result in many police women’s experiencing obstacles in handling harassment. Thus, this research can be considered beneficial in revealing the extent to which these characteristics play a role in women’s interpretation of sexual harassment across settings. Specifically, the study examined whether gender role stereotype beliefs, self- construals, and organizational contextual characteristics (i.e., participation in decision- making and perceptions of sexual harassment as a violations of organizational policy) differ between U.S. and Thai policewoman. It was hypothesized that female officers in the U.S. would adhere less to traditional gender role stereotype beliefs, have a higher interdependent view of self, have lower independent view of self, report higher 161 participation in decision-making and more often indicate that their organizational policy prohibits sexual harassment. These factors were then expected to influence perceptions of the severity of sexual harassment and concern about negative social reactions. These two factors were then anticipated to influence reactions to sexual harassment. Although stereotypes about Asian culture suggest that U.S. officers would be more assertive in responding to harassment, in the present research, this was not found to be the case, at least when women responded to questions about their worst-case incidents. In fact, female officers in Thailand were more likely to view sexual harassment as more severe and were more likely to adopt assertive reactions than American officers. As expected, Thai women show higher concern about social reactions, but concern about social reactions was not predictive of women’s use of assertive reactions. However, perceived severity was predictive of assertive reactions. The results suggested that Thai women reacted more assertively than American officers because they view the sexual harassment they experienced as severe. Even though Thai women had higher concern about social reactions, but they did not seem to be discouraged from adopting assertive reactions. Kauppinen—Toropainen & Gruber (1993) argued that when interpreting cross- national differences in women-unfiiendly experiences, one should consider differences in sensitization to harassment. The limited sensitization to sexual harassment among Thai employees may have led to their identification of only incidents that are more blatant and directed at specific targets, and therefore most of these worst case behaviors were evaluated by victims as severe. Alternatively, Thai women may come to the situations with different sets of expectations of male coworkers’s behaviors. For 162 example, the view that men should behave as gentlemen towards women, and treat them like ladies, combined with men’s behavior that did not conform to this norm may have led Thai women to find the men’s behaviors particularly offensive. American officers may include less serious incidents among those that they consider to be severe. The qualitative data showed that police work activities that promote a sense of camaraderie in the work group may affect how harassing behaviors are interpreted and handled. In the U.S., perhaps due to the both the greater sensitization among male coworkers to harassment, and the sense of cohesion among both female and male officers, male officers are more cautious and are receptive to women’s views. Both of these factors may have reduced the types of harassment that bother American women and therefore could have led to less need to react assertively. For both the Thai and the U.S. women, perceived severity, but not concern about social reactions to the response to harassment, was strongly and significantly associated with increased assertive reactions. Prior research also has found that, in the university setting, perceived severity is the most important factor in determining assertive reactions, such as reporting sexual harassment to an organizational authority (Brooks & Perot, 1991; Welsh, 1997). However, unlike prior research, the present study formd no support for the connection between concern about social reactions and assertive reactions (Wong, 1984; Adams et a1, 1983). This may be due to the population of women studied. Because these women were relatively more mature, well educated and in their mid thirties, they may base their reactions on perceived severity of harassment, rather than on concern about social reactions. Although measurement problems made it impossible to examine the association of 163 concern about social reactions to passive reactions for Thai participants, it was shown that concern about negative social reactions was associated with increased use of paSsive strategies in the U.S. group. At least in the U.S., those women who have higher concern about social reactions may be more willing to modify their own behaviors to improve the situations. For example, they more often ignore the behavior, avoid the person, and deflect the attention of the person to another issue. Thus, those who perceived harassment as severe and at the same time felt concerned about social reactions may favor the use of both passive and assertive responses to harassment. A question for future research is whether those who use both assertive and passive strategies develop low work motivations, as trying a variety of strategies might lead them to deplete their energy. In this study, measurement problems that existed for reactions to sexual harassment suggested that it might be to simplistic to view reactions to sexual harassment as consisting of two dimensions (Passive and Assertive). Low internal consistency of these scales, especially passive reactions for both groups, suggested that a finer distinction of passive reactions to sexual harassment is needed. For example, passive reactions such as going along may be different from passive reactions that convey the attempt to remove oneself from the behaviors such as avoiding the person, or deflecting the person’s attention. In terms of assertive reactions, as well, there needs to be a reclassification of assertive reactions that showed a higher degree of resistance (arguing) and those that connote milder forms of resistance (e. g., hinting dissatisfaction). Overall, the results of this study provided mixed support for the hypothesized relationships among variables. For beliefs about gender role stereotypes, the result tend 164 to confirm the expectations. Although there was a problem of internal consistency for the scale in the American sample, it does appear that most American female officers have high scores on this scale. Compared to Thai woman officers, American female officers less often endorse traditional gender role stereotypes beliefs. This result is consistent with prior literature which has described Asian society as patriarchal and therefore characterized by members’ socialization, to accept traditional sex role norms (Matsui et al, 1995; Ganguly, 1997). Also as expected, in the Thai sample, as acceptance of gender role stereotypes increases, sensitization to sexual harassment decreases (Malovich & Stake, 1990; Rubin, 1992). This finding was robust and independent of the effect of frequency of harassment or gender composition of the work group. This is consistent with the explanation that women with traditional gender role beliefs and liberal women use different sexual scripts in interpreting sexual harassment (Krahe et al., 2000). In addition, consistent with the explanation that traditional women are reluctant to sacrifice the loss of social relationships for social equality (Smith & Self, 1981), the acceptance of traditional gender role stereotypes was associated with increased concern about negative social reactions that might result from calling attention to sexual harassment. The effects of national differences on self-construal were in the direction hypothesized and therefore consistent with other findings that interdependent view of self is more of a characteristics of people in Asian culture, while independent view of self is more of a characteristics of people in Western culture (Singelis et al, 1999; Kim & Sharkey, 1995). Thai women had higher scores on interdependent view of self and lower scores on independent view of self, compared to the American counterparts. The 165 differences may be linked not only to the cultures of the two countries, but also to differences in the nature of work performed by women in the two samples. Prior work has explained that an interdependent view of self characterizes women, but independent view of self characterizes men (Cross & Madson, 1997). However, the U.S. women were used to performing job tasks associated with maleness, but the Thai women were not. The self-selection process in which individuals select jobs that match their traits, along with the police occupational socialization process, may contribute to high U.S. scores on independent view of self. However, the relationship of self-construals and perceived severity of sexual harassment was not significant. In both groups, women with an interdependent view of self, who placed strong emphasis on relationships did not find sexual harassment any less offensive than those with a low interdependent view of self. Consistently, people with an independent view of self, that is those who emphasized their personal freedom, were not more offended by harassment than people with a low independent view of self. The lack of connection between interdependent view of self and perceived severity of harassment may occur because although women who are high on interdependent view of self may accept harassment in order to maintain the social relationship; they may be at the same time more negatively affected by sexual harassment because it signifies a problematic social relationship. Also, in a society characterized by an interdependent view of self, high value is placed on the ability of people to guess what is on each other’s mind (Markus & Kitayarna, 1991). The failure of the harassers to conform to women’s expectations may result in higher perceptions of severity of harassment. This may have cancelled out the tendencies not to blame the harasser’s for fear of losing the 166 relationship, resulting in the lack of the relationship between interdependent self and perceived severity. Thus, the results suggested that interdependent self may not be such a bad thing. While it may lead women to modify themselves to fit with others’ needs, it may at the same times, lead them to expect others to be receptive to their needs and feelings, and result in sensitivity to the harmfirlness of sexual harassment. People with an independent view of self may not be offended by harassment any more than people with low independent view of self because they might perceive themselves to be “one of the boys”. Qualitative data highlights that many American female officers perceived themselves and male officers as brothers and sisters. Such perceptions, in turn, can lead them to become less critical of male officer’s behaviors and cancel out the result of independent view of self, which is being defensive when personal violations occur. Also, unlike unwanted sexual attention and sexual coercion, many forms of gender harassment did not involve a specific victim. Particularly gender harassment, which was involved in the majority of cases analyzed in the present study, may be irrelevant for women with independent self, as they do not feel it is directed at them. The connection of self-construals and concern about social reactions showed different patterns in the two samples, but both patterns were consistent with the predictions. For the Thai sample, interdependent view of self was positively associated with concern about social reactions. This provided support that people with an interdependent self are sensitive to others’views and constantly feel concerned about negative self-evaluations (Kim & Sharkey, 1995; Singelis et a1, 1999). For the American sample, independent view of self was negatively associated with concern 167 about social reaction, which supports the idea that people with an independent view of self focus on their uniqueness and positive qualities (Markus & Kitayarna,1991), which enhances their sense of self-determination in influencing existing circumstances (Chang et al., 1997). However, the fact that Thai women with high independent view of self were not more likely than Thai women with low independent view of self to have concern about social reactions perhaps reflects the existing realities in Thai society. In a collectivistic culture, where the attempts to minimize conflicts between in group members is emphasized and social interactions are guided by rules, obligations, and status, people may not benefit from an independent view of self. Their self- determination may not help them, since the reactions of others will still affect their well- being. On the contrary, in the American society, where personal freedom is emphasized and confrontation is an acceptable ways of handling conflicts, women with an interdependent view of self may have little reason for fearing negative reactions to their calling attention to the issue of sexual harassment. ‘ Inconsistent with initial expectations, the current study did not find that there were significant differences in participation in decision-making reported by female officers in the two countries. Although Thailand has been characterized as a high power-distance culture, which would be consistent with little decision-latitude being given to subordinates (Chen & Ditomoso, 1996), Thai women reported similar level of participation in decision-making as American female officers. The current study cannot rule out the possibility that the lower expectations of the Thai participants, due to being socialized into high power distance culture, may have led them to evaluate superior’s behaviors more leniently. However, this is unlikely, as previous research found that the 168 behaviors of supervisors as a result of power distance culture was so salient and accurately perceived by employees (Offermann & Hellmann, 1997). Given that the average rating of participation in decision-making in both groups was moderate, it seems to be the case that despite the rhetoric of police organizations in the United States about their efforts to integrate women into the police force, the reality may be that in many departments surveyed, policewomen still feel that they have been offered limited opportunity in making decisions about their jobs. This is consistent with qualitative data, which found that many U.S. women feel that their managers are overprotective of women and seclude them from challenging jobs. They felt that police administrators have not done enough in offering policewomen equal opportunities as policemen. Therefore, it is not surprising that U.S. women did not report higher participation in decision-making. Participation in decision-making is not particularly useful in explaining variation in women’s evaluation of sexual harassment as severe. Participation in decision making may simply lead to stronger feeling, either positive and negative. It influences individuals to evaluate sexual harassment as more severe. However, it may simultaneously function as a form of social support in the organizational context, which helps to buffer any negative emotions that may be generated by sexual harassment behaviors. There is some empirical support that low participation in decision-making is connected to concern about social reactions. This connection is consistent with the idea that participation in decision-making leads to a sense of empowerment (Spreitzer, 1996), perceptions of organizational justice (Witt & Myers, 1997) and felt obligation to the organization (Salancik,l977). American officers who had high participation in 169 decision-making felt less concerned about negative social reactions that might result from calling attention to sexual harassment. However, this result was not found for the Thai sample. Different from the U.S., in Thailand higher-level police superiors are far removed from organizational realities and are subjected to change in assignment and location at anytime. Therefore the participation in decision-making with a particular superior may have little bearing on the general comfort of female officers. As expected, American female officers were more likely to believe that their organizational policy prohibits sexual harassment than Thai female officers. Perceptions of policy reflect the extent to which organizations have put efforts into specifying what kinds of behaviors are prohibited. As a result, female officers in the U.S., where policy exists, were more likely than Thais to indicate that their organization policy prohibits the type of behavior they described in their three worst cases of harassment. ' In the ethical decision making fiamework (Jone, 1991), social cues provided by organizational policy should influence individual’s judgment of seriousness of harassment. However, the data only partially support this prediction. Perceptions of organizational policy is positively related to perceived severity of harassment only among Americans, but was not related to perceived severity of harassment among Thais. The lack of common definitions of harassment in the Thai police department is likely to result in women’s excusing the harassers to some extent, for they carmot be expected to be aware of unacceptable behavior. In the U.S., women know that men are aware of what behavior is acceptable according to policy, and thus they consider harassment as more serious. 170 For both Thailand and the U.S., perception of policy is unrelated to concern about social reactions. This suggests that a list of prohibited conduct in policy may be not be sufficient to ease victims’ concern about social reactions to their responses to harassment. It makes sense that perceptions of an organization’s willingness to stand behind sexual harassment victims rather than the mere existence of a policy would be more important in influencing concern about negative social reactions and victims’ response to victimization. Study Limitations The research presented here is an initial attempt to investigate reactions to harassment by female officers in Thailand and the United States. As is often true for studies which are done at the rudimentary stages of theory development, there are some weaknesses that must be acknowledged. First of all, several measures have lower reliability than expected. For example, the internal reliability of gender role stereotypes in the American group and passive reactions in the Thai group was very low. As a result, the study could not make use of these variables and could not determine the hypothesized effects, as originally planned. In some cases, scale variables had quite low reliability, between .60 and .70. One way to solve the problem is to focus on the data fiom only one group and select only items that have high loading from that group and then conduct the analysis. However, this approach would undermine the purpose of the study, which is to compare reactions to sexual harassment between female officers in Thailand and the U.S.. Given that all of the scales used have reliability above .60, the data were viewed as sufficient to follow the original plan of the study. Second, female officers in Thailand and the United States may not be the idea] 171 sample for tests of the theory. There are some key structural differences in policing in the two countries, including gender compositions of the work group, the nature of jobs women performed, and organizational tenure. These factors may influence their perceptions and reactions to harassment. ‘ Third, as in prior quantitative field research on sexual harassment and reactions to it, the present study cannot control or measure the detailed characteristics of the worst case of sexual harassment situations that each female officer encountered. For example, the degree of intrusiveness of the behaviors, and the characteristics and the personalities of the harassers may vary from case to case. Thus, these variables may be responsible for the results. However, this appears to be a common problem for research on sexual harassment in field settings, as the objective characteristics of the events can never be quantified, unless the research adopts a scenario-based approach, which has its own drawbacks. This limitation does need to be noted so that there is awareness of alternative explanations of findings. Forth, another potential limitation of the present study involves the generalizability of the results. This study was conducted by using female officers who are in law enforcement agencies in Thailand and in the U.S. Although the organizational structure may appear similar to other organizations, finding cannot be easily generalized beyond police departments. Moreover, the data were obtained fi'orn women who were willing to participate in the study, and in departments where the police administrators agreed to the study. The samples may not be an accurate representation of policewomen, for example, the women who took part may be especially interested in women’s issues. 172 Organization Implications Several characteristics of individuals may lead to differential interpretations of sexual harassment. In turn, these differing perceptions influence reactions to sexual harassment. The results of the study provided some information that is important in developing effective policy, grievance procedures, and intervention programs. In the Thai sample, findings are consistent with the proposition that beliefs about gender role stereotypes play a role in decreasing perceived severity of harassment and increasing concern about social reactions in responding to harassment. In lights of this finding, organizational interventions aimed at increasing sensitivity to harassment and providing support among traditional women may be effective in increasing the willingness of victims to confront harassment. For example organizational training could allow women and men to discuss the effect of sex role socialization on how a person interprets and copes with harassment, and how at the same time it can contribute to the persistence of harassment. This could heighten awareness of the problem of sexual harassment and toleration of it in organizations, and also benefit victims, who would better understand their own interpretations of and reactions to harassment. Moreover, exchanging views with other people in the training could provide the opportunity for traditional women to be aware of social support from other women. Although gender role stereotypes result fiom a long-term sex role socialization process and may be difficult to change, it might be important to try to change sexist stereotypes. Organizational practices that reinforce differences in women’s and men’s job tasks should be eliminated. For example, the use of women to serve coffee, wash dishes, and perform only secretarial work should be prohibited. Promotion based on 173 merit rather than being sexually cooperative should be a focus. Under the conditions that women’s role and men’s role in the workplace become less discrepant, employees in organizations may appreciate the worth of women and view sexual harassment as serious. It might also eliminate myths about harassment, such as the women use their sexual advantages in exchange of j ob benefits. Female officers who encounter harassment may be more able to elicit understanding and support from other organizational members. These changes may take many years, but they would address the root of sexual harassment problems. Moreover, for Thai women, having an interdependent view of self was negatively associated with concern about social reactions. Although the findings are optimistic in a sense that concern about social reactions did not appear to reduce assertive reactions, while perceived severity was positively associated with assertive reactions, the fact that several women were willing to respond to harassment that bothered them, despite the fact that they were not comfortable using those assertive strategies suggests that many women, especially those with an interdependent view of self, may experience some psychological discomfort while adopting assertive strategies. When victims engage in assertive actions, despite the fact that they themselves felt uncomfortable with those actions, the strategies used may not be equally effective as when victim feel more confident about what they are doing. As such, top management as well as middle managements should attempt to encourage and reward victims who come forward and confront their harassers by minimizing any negative consequences they may encountered. Women and male coworkers should be educated that it is not wrong to assert themselves or feel bad about themselves, when they were violated, regardless of 174 who the harassers are. The complaint handling process of organizations itself needs to be structured to place a high emphasis on justice, rather than the consideration of seniority or coalitions within the organization. Assertiveness training should also be provided so that victims can learn how to communicate their view and feel comfortable with it. Given that people with interdependent view of self has been viewed as placing high emphasis on significant others (Cross & Madson, 1997), organizations that also encourage third-parties, especially those who are well-respected in the organizations to intervene when they witnessed potential sexually harassing situations on the behalf of the victims may also provide victim’s with more comfort in dealing with harassment themselves in the future. Moreover, informal procedures where victims can turn to someone they can trust, and report harassment should be provided so that women who feel concerned about social reactions can have more options to deal with harassment. In the United States, where an independent self was negatively related to concern about social reactions, interventions aimed at increasing women’s positive self- evaluations or self-worth might reduce their concern about social reactions, and allow them to respond assertively when they are harassed. Clinical counseling that emphasized positive self-concepts and the sense of agency should be provided for interested employees. Additionally, given that participation in decision making was associated with low concern about social reactions, organizational efforts to provide autonomy for female employees in their work roles may be particularly beneficial in helping women feel that other employees will respond favorable if they try to stop harassment assertively. The qualitative data suggested that many female officers in the U.S. felt dissatisfied with the way they were treated at work, particularly the lacks of 175 trust in their competence. They felt like they were exempted from difficult jobs and were assigned to perform the jobs that no one wants. Thus, organizations that wish to increase perceptions of work autonomy of women need to reduce gender disparities in job assignments and for example, supervisors, should be encouraged to and rewarded for allowing women to perform the same job as men. In addition, informal and formal grievance procedures for woman who feel they are treated unfairly would be useful. The fact that U.S. women who were concerned about social reactions also experienced the most harassment suggests that they are in organizations that tolerate harassment. Individuals’ understanding of sexual harassment events as a clear violation of policy was the only variable associated with perceived severity of sexual harassment in the U.S. group. Perception of organizational policy was the most important predictor of the view that harassment is serious, and the view of harassment as serious is connected to assertive reactions. Therefore, it is important that organizations have policies that are clear and that these are communicated to all employees. Training, including role- playing and the discussion of cases and films in relations to what is acceptable and not acceptable should occur regularly in order to promote common definitions of harassment. In Thailand, individual’s perception of policy has no relationship with perceived severity of harassment. Establishing sexual harassment policy could produce a shared understanding, and reduce the variation that results from individual differences, such as gender role stereotype beliefs. As a result, formal sexual harassment policy may increase organizational members’ awareness of the harmfulness of harassment. Organizations need to recognize that a woman sometimes uses both assertive and passive strategies for responding to harassment. Even if victims do report an incident, 176 they may also modify their own behaviors to improve the situation. In other words, some women who confronted their harassers, also avoid the person and limit their own opportunities to enjoy freedom of movement in the organizations. Therefore, if organizations wish to evaluate the effectiveness of the interventions such as sexual harassment training program, victim’s passive strategies should be considered as an important index of success. The use of multiple indicators of success would be the best measure of organizational success in promoting a women-fiiendly workplace environment. Recommendation for Future Research In order to expand on and confirm the findings, several areas remain open for firture research investigations. First of all, although it was hypothesized that women in Thailand would be less assertive in reacting to sexual harassment, it was found that they were not. Indeed, Thai women were more likely to perceive sexual harassment as more severe and also reported more concern about social reactions. The Situational factors surrounding sexual harassment (e. g., how others in the context view harassment and how they viewed assertive reactions and passive reactions) may be important as the attitudes of the victims themselves. Future research both in Asian and Western countries should also examine perceptions of harassment and perceptions of victims’ reactions to sexual harassment fi'om the perspective of others in women’s work context, including higher police administrators, male officers, women who are non-victims, and victims of harassment. Other parties’ evaluations of the events may be important in determining what happened and how it was resolved and may better explain variance in victim’s reactions and help us identify barriers victims are facing. Studies that examine 177 reactions to harassment may also use victim’s perceptions of normative expectations of seriousness of the events or about the use of assertive behaviors as additional predictors of reactions to harassment. It might be beneficial to investigate whether reactions of women with strong beliefs about gender roles, for example, are influenced more by normative expectations, rather than by their own evaluations of harassment. Second, future research on reactions to sexual harassment need to investigate women’s perceptions of the effectiveness of the strategies that were employed. This would provide more direction for organizations as well as researcher to focus on the most effective reactions for the victims, which at the same time are strategies that do not make victims suffer fiom the lack of opportunities to work in their work place. It may be that pe0ple in different countries see the same strategy differently. For example, do Thai feel less satisfied with the outcomes of interactions when they use assertive reactions, compared to people in the U.S.? and do women with traditional role attitudes feel more satisfied with the outcomes, when passive reactions were employed? This type of information will be helpful for organizations to understand women’s need and at the same times explore what types of interventions would be needed. Third, the current study is a cross-sectional study, therefore the nature of causality could not be determined specifically. For example, assertive reactions can lead to perceived severity of harassment because it can provide comfort to the victims to think that how they reacted was justified after the fact. In the firture, longitudinal studies should also be conducted by examining panel data beginning at the point when individuals enter the organizations, tracking sexual harassment events, their perceptions of individuals and their coping patterns. This could be done by examining how 178 individual differences and organizational characteristics might play a role in this process. The combination of both quantitative analysis and in-depth qualitative accounts would be particularly useful in giving details about how this process has developed and would provide useful insights into the process. Forth, the less antagonistic feelings among American participants, their low concern about social reactions, and their limited use of assertive reactions suggests the importance of occupational identity in structuring gender relations within the work place. This may have helped policemen and policewomen to perceived their similarity, which in turn, lead them to be able to communicate much more easily and resolve sexual harassment on their own. It would be fruitful for future research to investigate how occupational socialization influences perceptions of harassment severity, victim reactions to harassment and reactions of other parties to victims’ behaviors. This might be done through an examination of perceptions of solidarity in the work group, and of organizational and occupational identification. Such research could be the basis for recommending additional interventions. On the other hand, factors that lead Thai women to become sensitive to harassment should be explored and identified. For example, whether women’s standards of what is appropriate men’s behaviors during the interactions with women might be responsible for sensitivity to harassment found among Thais remains an interesting research question. Fifth, future cross-national quantitative research on reactions sexual harassment should employ only scales that have been validated across samples. If it is necessary to shorten the questionnaire, the researcher may want to select items that have high loadings. In addition, a pilot study should be conducted prior to the data collection 179 process and open discussion with the subjects about the instruments used should be done so that the measurement can be modified. Sixth, the study focused on reactions to sexual harassment based on the operational definitions of sexual harassment events identified by prior empirical work in the U.S. As a result, the current study considers only reactions of harassment that include Americans’ definition of harassment. To develop a more comprehensive list of harassment events and understand people’s reactions to those events, future research should also use interviews about things that women view as sexually harassing. Then, researchers could use that information to develop different scenarios which include events depicted by various groups. The use of a scenario approach would also allow the researcher to determine the impact of various factors, while being able to control or manipulate sexual harassment event characteristics. Despite all of these limitations and future research that is needed to provide a clearer understanding of reactions to sexual harassment in police organizations, this research is significant in the sense that it is the first study that examines reactions to sexual harassment of policewomen in Thailand. The comparison to the U.S. is also unique. The findings raise awareness that existing factors that differ between the West and the East, that have been found in one setting to be predictive of perceptions and reactions to interpersonal conflicts or sexual harassment situations, may not always operate in a consistent manner. The study highlights the complexities inherent in the nature of reactions to sexual harassment in real life settings. While previous research that used scenarios found that Americans were more critical of harassment than Asians (Neale, 1991), this doesn’t mean that Asian female victims would be less critical of harassment 180 that happened in their own organizations. Instead, the lack of sensitivity of the issue to the issue by others in their surroundings can lead the Thai women to perceive more threats fi'om harassment and have more reasons to behave more assertively. This finding is new and demonstrates the need to be cautious about generalizing from research that presents artificial scenarios abstracted from real world settings. It also provides some ground for future work to consider cultural values, which influence victims’ reactions via the behaviors of others in the situations. It is important that research considers cultural context and not concentrate on victim attributes as the sole determinants of reactions of sexual harassment. 181 APPENDICES 182 APPENDIX A Informed Consent Statement (English Version) 183 Informed Consent Statement You are invited to participate in a study concerning interactions with people of the opposite sex at work. The purpose of the present study is to gathering the information about social interactions between policewomen and policemen in the workplace. The information from your perspective will provide valuable insights on informing your administrators to improve working conditions for female police officers. Your participation involves reading and filling out a questionnaire, which should take about 15-20 minutes. Some of the questions may require you to reflect on past events that may cause some distress. However, the main benefit of your disclosure is the attainment of information to better understand those events. Any information obtained in connection with this study that can be identified with you will remain confidential. To protect your privacy, the obtained data will be stored on a password protected computer in a locked room. Only the responsible project investigator and the secondary investigator will have access to the data. The data will be kept for no more than two years. In any reports or publications, no one will be identified and only group data will be presented. Your privacy will be protected to the maximum extent allowable by law. Your participation is voluntary and you may choose not to participate at all. If you agree to participate in the study and then wish to withdraw, you are flee to do so at anytime without any penalty/loss of benefits. If you would like a summary of the results of this study, please fill in your address so that it will be mailed to you at the conclusion of the study. If you have any questions about this study, please contact me or the responsible 184 project investigator: Merry Morash, Ph.D., Professor, 514 Baker Hall, School of Criminal Justice, Michigan State University, East Lansing, Mi 48824-1118, phone: (517) 353-0765, E-mail: morashm@msu.edu. If you have any questions or concerns regarding your role and rights as a subject of this research or dissatisfied at any time with any aspect of the study, you may contact—anonymously, if you wish-Ashir Kumar, M. D., Chair of the University Committee, on research involving human subjects (U CRIHS) by phone: (517) 355-2180, fax: (517) 432-4503, E-mail: uchrihs@msu.edu or regular mail: 202 Olds Hall, Michigan State University, East Lansing, Mi 48824. Thank you for your cooperation Sincerely Yours, Somvadee Chaiyavej, Secondary Investigator The School of Criminal Justice, 560 Baker Hall Michigan State University, Mi 48824 E-mail: chaiyave@pilot.msu.edu Phone: (517) 355-9860 You indicate your voluntary agreement to participate by completing and returning the questionnaire. (Please fill in your address only if you would like a summary of the results of the study) Your address 185 APPENDD( B Informed Consent Statement (Thai Version) 186 luuanenmufiuaau I v A r r A v . A V (v d. o muléisuunrylfifimuflulumsfinmoaafiarmrilrlgtiuwufinumnnssiwluuwmu t Av til :5. 22 al v v a" l v 2; Al t1nurummmmmeuutfluhlmamm’maagammnumsflsusziuwufissm'medumu o A 9.1 v A O _ Al 0 v 4i. are! t 1' IV nqmaumtuastedumunamaadalunwmu ragaulnsumnuuuawasmurflumdugau 1 ti. 0 t 9/ A I I d. u it 0 uraqmmmsmlrltauamagurmrluumaasuaaamumarlendgaamwmnaaulumsu'mu v 11 dd. 0 A rastequu’mmmeuma d I I I A, i! I 4’ 91 msumummmmuuesflssnaunanmsmuuasnranunusaumu maximum rlrsmtu 15-20 mil 5'1mumritrmeesvi'tlfiu'méimssfinfrqmrjmrtu'luafimisadaeznalfi A d A! v I 6 6'0 v A I ad 11v v Lnnmdmnrunuuln ausaisnmu firsluaumntymnmrulntumarmunne msinrura 4' al. 0 v v r I :1 dQI c; 21 a {J b v of 114 IV gammauesmlmmlemamsmmmuunm‘nu ragalnnmuulnmaummrarnums finmfifimmsnvii/t1uluslrlfiavhuipieslpii’umsrlnilalitflumetuti’u Liainmmyutiluaqu v 1 v tr :4 2; vv a «V d. A (1141/ 2/ v A d. :3 nsraamu raunmuunulnmeslnrumrmurnmlutnsatnaummarnmarluwaulnmnr l I IV v a! v at I 9; AlA A at A 2A A! A 9A I :1 all agmuluumuumn NL'WENLLEI ”’3 Ell/1 ll‘Vl’a‘lJClfil‘ll’fllJifl‘a‘Gfi’lfi‘LLfitv’l El‘a‘flx‘lL‘I'lElleVl’lltllfll’QS v d mmsnfi’ii’agafifiagimli‘lé’ 1781381141sitiiumstr‘iui'nml'itflur: smmlrltfiuaasil 11.1 d 6' 1d A d A, 11 d. O :1 (J‘qflilflulllffluvlflqulllm'lllunqzluuani‘znlt‘llfiuifilfllllUiflmfimqau Llfiz‘llfluvfimufllflufluuqx y . I v I w v d u u- Li'lulrllursaunqu emutilumumuarmuas'lnsumsqunear munnguma rytgn lilriiflulrl'lsi’aehsgtqn d r I I I: V I Ki ‘3 I 4 Al I n’lrumumuuaamuutfluirlmum’maumister/twitter inautuuwa'reestaanuezlu 1’ d I 0‘ I 3’ 4 I t i! d I I v I 1" U all t‘n’itltl'm‘i‘qtlttitmtlflulfl “8‘9 “qanlyl'lufilnflxllrflfll'lnaqumllLLfirJVl'lUQ’lN'l‘fllVl’qznflumq13.13191 187 finnulneirlflluuu‘iemufimlrslu‘mflnq mnvhuriiasmsuufindenarnufid’engmslfifiaguaa 1 a; A. d. I I 91v 0 d. A mulemaumesmmslmumIufitanmrtuamrinmqnm I d 9.1 (d. v Au «:1, A I 91 ‘ ‘11 d 9A 0! U Adv A mnmuuflrymumlelnqtnmnmnuqeuungmwnnau'mtnnusagoeeuanumnnuau lartmrioifi Merry Morash, PhD., Professor, 514 Baker Hall, School of Criminal Justice, Michigan State University, East Lansing, Mi 48824—1118 lurd’wrflsfl) 353-0765 51.3421 t d 0 cl or ad. or 4::F morashm@msu.edu mnmuummulnq usenonunmalalnqtnmnuuumuunshared 1 CS 9.! 9.1 V A V d I 9.24 I U A“ El. CI I I I mulumsulugmsumssee use mnmuganluwalanumrfinmutualnnnnulmnasualn a I A 1 t A v I 91 Ad. . nmu mummsnnnnalnemnnslurzuuamtar (mnmunarms) lrlu Ashtr, Kumar, M.D., Chair of the University Committee. on research involving human subject (UCRIHS) rm lnsdmi : (517) 355-21801mms : (517) 432-4503 511121: ucrihs@msu.edu u‘ialrlrnri'lel : 202 Olds Hall, Michigan State University, East Lansing, Mi, 48824 rauauwrsnmlumwidufia n’qemnufiufin Somvadee Chaiyavej, rpm (mils) amfi 11181911 Secondary Investigator, faintest The School of Criminal Justice, mnt‘rmnrzmun'rrefifirruwmmty'r 560 Baker Hall, 560771r1111mt18r Michigan State University, umfinmfinuuti’gfiiunu East Lansing, Mi 48824, Lilatfimiuau’tir igii‘iunu 48824 USA. rlssmn anijami‘m 51:15: chaiyave@pilot.msu.edu lusrlnri: (517) 335-9860 188 A 4’ v - I I A A I I v u martenuasnuuuuaaumuunaumnm’rmuauueumi'numusoumam'maunr'ln (neutdnsaniiatjLannsnstiifivi'run’atmruufindonamu’rti’e) d l I naguaqmu 189 APPENDIX C Instruction of the Sexual Harassment Experiences Questionnaire (EnglishVersion) 190 Instruction I : On the following pages are questions about your interactions with men at work. You will be asked to indicate how often you have been in certain situations, where any male officers at this agency engaged in some forms of behaviors towards you within the past two years. These situations are divided into three sections. At the end of each section, please stop and think about the worst event that has happened to you. Then, use that experience to assess how much you agree that the statements listed accurately described your reactions to the situation. Your responses may range from 5 (Strongly Agree) to l (Strongly Disagree). In completing the questionnaire, if there are many male officers involved in the situation, please focus on the person who had the greatest impact on you. If you have never been in any of the situations within the past two years, please think of the worst event that has happened to your female coworkers in this agency instead. This situation may have taken place at any time, not limited to within the past two years. This situation may be based on her disclosure to you, your direct observation of her experience, or through the accounts of her story from other coworkers. We would like to know hbw you would react to that situation, if it was happening to you, right now. You are simply using her experience as a hypothetical situation. Ifthe person involved was your coworker’s direct supervisor, check that this person is a direct supervisor. Then, imagine that you have been recently assigned to work with a new supervisor who initiates this act towards you. What would be your reaction to the situation? If the person involved was her subordinate, check that the person is a 191 subordinate. Then, imagine that your new subordinate initiates this act towards you. What would be your reaction to the situation? Answer the questions in terms of how much you agree that the statements listed fit with your “probable reactions”. Please take into account the reality of your circumstances at work when deciding how you would behave. You and the female officer may differ in the way that you interpret and handle the situations. In responding to the questions, please focus on your own reactions. Please understand that there are no right or wrong answers. Individuals may differ in the ways they react to these situations. 192 APPENDIX D Instruction of the Sexual Harassment Experiences Questionnaire (Thai Version) 193 o A I A'- o I A u A A v I Iv A md’au 1 z'lumihna‘lrluummumaq Innanumsmlgd’uwuéuasmunuIj'rna'lun . I w l I 4 I 0 menu muae‘lasumsaaummurmurednnag‘luamumsmmsamtfisms’IamaIfln G A I fl v o A I I I 3 A nmuuuuqmnuu anaequnnssumegrluuunemuuaanssmm‘ln malurias A I 0' I A, I I A I seasna'raes‘ilumum amumstuma'ruumeantilumumu quauunasdau A A (A v A A A A’ u I v 5 v nennuqnuasunmrlseaumstunIaqmunqnmnutnnuunumu uaqannuu'lrt v I o I I a A I A A uszaumstfimnmoneummm'r mumuei'qammlmnriamaumrlau'l'i’usaa'ta A A I I (15 II I I! g I 1111mmnauauaanasmunaamumsmuu'lnams9nnot maeuuesmummsmfiu & I as I A A I G3 I A' 1111919141191 5 (muti'nuamqm)'lrleun~1 1 ('lumuri'naamtm) A a AA A, A o A A II 'lumsnauuuuaaumudaum urlgnsmnauaueeu mnummamanmmum A A AA I I A umunu neonIaanau'lauunnanuuanflnunamumnuqn I I I A U ti‘Imm'rmulumannag‘luamumsm‘lmaa maluszaznmaaqilnmum nqtm A A 0‘41. 1! A A ”A a: v A I i! A 1 I a: reantmlssaumrtumms'muqnulntnnuunumamaumugurywarmuluvmamuuuuu a; A 1 A C I I 0 II 0 V I Ci. I amumsmumnetstnmutualnnn'm lumenLiluesnarmnnagmaluse sromsarilnmum winrl’u amumsulfimnesmmnmsti‘lnInterimti‘iauvhutmltivhuus’m mnmstirtnmm I or (AA A, y A l d A :3 I ”V 0 mutationnernumnmsruuanunumaurarmu usauazuumumeazlnsufirmnmuan 1 A U A A I A. I d. 21 I I 1mtnmnmrassmmmaumumnmaumumunuaw} mnaamsm'mmmuesnauaum I (I: I I (:0 U A I: v I A" I A I I) naamumstuuamals mnmamumsmumaaInmunumutarlmtusu mummunla ureaunsstfiraufiauvhumLfluan'mn'lsnlrinaamhri’u UV" v val :5 v A t . o 4' aquflnaQLnUQTIflfluLflu El. \lfilJ'U[U‘ll’lififilmi'qmflflL'Wflufluqflumflfl'n’lu qufifflfluuqflfifllfl ”UN 1 21: d I] 2/ A A I I A 210’ 9! Q'lliflflflau fl'fl % QfllJlJCy‘ll'lifiElm‘tN uflqnumu’lmrunn’mm Vl'lllLWQlC’lS‘Un'li‘NfllJliu'lEl‘L“ 194 My” unwitting a 11 munnuluu' irnssfianfinssun’rnam whitetailmsnauauaanatuqmstfi A" l A A 21 A" 91 v V V A I 0 A I II utuuls? mnunnamnmrmuti‘lutflnmnuumunrarmaumu mrnsasumearlrlmunnag we! 91 or 4:1"! 920’ A A I 9.1 910! IV (V O A U I a“ una @nmnuumm Ltmnunmmsunmmnmcg’flnmnuuty'mnesqunnssumnmou u I I A AAA I 6': I 0 IA I I I: v mu mussuflgnrmnauauarnaamumrtuut‘nu‘lr? naummuluurmnmumumn A I D A A Q! U “ AAA A I 91” I medians ransnuutaeulanrnu rignrennaunuarnmestflulrlln ‘lltlx‘l‘ll’mt'fld O A A A A O 1 A. v A I: I I A .A I nemammnramwmamfluarfluummuuarmu Luannaulammnmunzrlgun111111? I 0 A 9.! ti A V V f I I I V muttasmsqeutyst‘l‘uumnnsnnewuasenmsnuamumstumsqamrunnn'mnu lunar neufinmumdnfingmnaulafi mjnauguawaavhgtm 9 I! I IA. A A All A AA I ng‘nmmlomm'r luumnaulngnmaun uauunnameeemlgnsmnauauarna amumrrfimdnfilé‘lfimfleufiu 195 APPENDIX B Table 12 Sexual Harassment Experiences Questionnaire (English Version) 196 Table 12. Sexual Harassment Experiences Questionnaire (English Version) Part I : During the past 2 years, have you Never Once Some Often ever been in a situation where any male or times officers who are members of this agency did Twice the following? (Gender Harassment) Most Of The l. Habitually told suggestive stories or ofi‘ensive jokes? 2. Made crude and offensive sexual remarks, either publicly (eg., in the office, meeting), or to you privately? 3. Treated you differently because of your sex (eg mistreated, slighted or ignored you) ? 4. Displayed, used, or distributed sexist or suggestive materials(e. g. pictures, stories, pornography) to you! or in your presence ? 5. Frequently made sexist remarks to you (suggesting that women are too emotional to assume a leadership role.)? 6.”Put you down” or was condescending toward you due to your sex? Reactions to the situation 1. The person who initiated the act is a Ddirect supervisor Dhigher superior coworker Dcoworker Usubordinate Bother (specify) 2. Was this experience your own? D Yes D No 197 Table 12 (cont’d). Please indicate to what extent you agree that Strongly Not Strongly the following statements accurately describe Agree Agree Sure Disagree Disagree your reactions to the situation. 1. Going along with the behavior 2. Informing an authority to take action against the person 3. Ignoring the behavior. 4. Comrmmicating your displeasure to the person indirectly (e.g. becoming distant). 5. Changing your habitual ways of doing things to avoid this type of situation from recurring (e.g. seeking a transfer, bringing friends along). 6. Telling the person about the negative consequence (3) he may face from this behavior. 7. Attempting to deflect the person’s attention to another issue. 8. Protesting the person explicitly (e. g. arguing, telling him to stop). 9. Finding this act offensive. 10. Being concerned that by calling attention to the issue, coworkers might think worse of you. ll. Believing the official polices/rules of this agency clearly prohibit this act. 198 Table 12 (cont’d). Part II : During the past 2 years, have you ever been in a situation where any male officers who are members of this agency did the following? (Unwanted Sexual Attention) Never Once 01' Twice Some times Often Most Of The time 7. Made unwanted attempts to draw you into a discussion of personal or sexual matters (e.g. attempted to discuss or corrnnent on your sex life)? 8. Gave you unwanted sexual attention? 9. Attempted to establish a romantic sexual relationship with you despite your efforts to discourage him? 10. Has continued to ask you for dates, drinks, dinner etc. even though you have said “No”? ll. Touched you (e.g., laid a hand on your bare arm, put an arm around your shoulders) in a way that made you feel Imcomfortable? 12. Made unwanted attempts to stroke or fondle you (e.g., stroking your leg or neck, touching your breast, etc.)? 13. Made unwanted attempts to have sex with you that resulted in you pleading, crying or physically struggling? 199 Table 12 (cont’d). Reactions to the situation 1. The person who initiated the act is a Ddirect supervisor Dhigher superior coworker Dcoworker Usubordinate Bother (specify) 2. Was this experience your own? D Yes D No Please indicate to what extent you agree that Strongly Not Strongly the following statements accurately describe Agree Agree Sure Disagree Disagree your reactions to the situation. 1. Going along with the behavior 2. Informing an authority to take action against the person 3. Ignoring the behavior. 4. Communicating your displeasure to the person indirectly (e.g., becoming distant). 5. Changing your habitual ways of doing things to avoid this type of situation from recurring (e.g. seeking a transfer, bringing friends along). 6. Telling the person about the negative consequence (5) he my face from this behavior. 7. Attempting to deflect the person’s attention to another issue. 8. Protesting the person explicitly (e. g. arguing, telling him to stop). 9. Finding this act offensive. 200 Table 12 (cont’d). Please indicate to what extent you agree that the following statements accurately describe your reactions to the situation. Strongly Agree Agree Not Strongly Sure Disagree Disagree 10. Being concerned that by calling attention to the issue, coworkers might think worse of you. 11. Believing the official polices/rules of this agency clearly prohibit this act. Part III : During the past 2 years, have you ever been in a situation where any male officers who are members of this agency did the following? (Sexual Coercion) Never Once or Twice Some Often Most times of the time 14. Made you feel like you were being subtly bribed with some sort of reward or special treatment, to engage in sexual behavior? 15. Made you feel subtly threatened with some sort of retaliation for not being sexually cooperative (e.g., the mention of upcoming evaluation, review etc.)? l6. Implied faster promotions or better treatment if you were sexually cooperative? 17. Made it necessary for you to respond positively to sexual or social invitations in order to be well treated? 18. Made you afraid that you would be treated poorly if you didn’t cooperate sexually? 201 Table 12 (cont’d). your reactions to the situation. Part III : During the past 2 years, have you Never Once Some Often Most ever been in a situation where any male or times of the officers who are members of this agency did Twice time the following? 19. Treated you badly for refusing to have sex ? Reactions to the situation 1. The person who initiated the act is a Ddirect supervisor Dhigher superior coworker Dcoworker Usubordinate Bother (specify) 2. Was this experience your own? D Yes D No Please indicate to what extent you agree that Strongly Not Strongly the following statements accurately describe Agree Agree Sure Disagree Disagree 1. Going along with the behavior 2. Informing an authority to take action against the person 3. Ignoring the behavior. 4. Communicating your displeasure to the person indirectly (e.g., becoming distant). 5. Changing your habitual ways of doing things to avoid this type of situation from recurring (e.g. seeking a transfer, bringing friends along). 6. Telling the person about the negative consequence (5) he may face from this behavior. 202 Table 12 (cont’d). Please indicate to what extent you agree that Strongly Not Strongly the following statements accurately describe Agree Agree Sure Disagree Disagree your reactions to the situation. 7. Attempting to deflect the person’s attention to another issue. 8. Protesting the person explicitly (e. g. arguing, telling him to stop). 9. Finding this act offensive. 10. Being concerned that by calling attention to the issue, coworkers might think worse of you. ll. Believing the official polices/rules of this agency clearly prohibit this act. 203 APPENDIX F Sexual Harassment Experiences Questionnaire (Thai Version) 204 Table 13. Sexual Harassment Experiences Questionnaire (Thai Version) I d d I mun 1: mn'lurioeaeaznm 2 3mm: 4 I I (A ““0 fl mu muraang’lumqmamnfiafia ... '- 14 . . ma man man a . J. . hunt] ma I; maulanmu 'luuuonmuunwvhae am ,3 ma mm 5 am pm Iin'lih‘lrfialzi? I 4' 4 III. I 1. rmLiniriilnunronrnannnmnmnng Lflurlrzr-i'i ? I!!! 0 al. Al al. v 2. l‘nnnnmnguusa uranmnmn biliary I I ‘H. I O U mo lumanflunmmmx (mu lumun d 4 cl I I mu mommies-in) neannmulilumr doufiqfimn ? 3. flfirfinanaundwunnIiaq'lfla'imfiu mr'izrinuli'lummrfiamiu flfififiinn’iu arias‘ln'nnma 11111315835 niolfimansio Ii'iu)? do! II 4 I v A. 4. “$914111 usausna'imaq in mm: a gnmnurfir u‘ia munnmw (11in $me 4' 4 4 «J I Lamina ammunition) lurnlznmu flmngsi’oagifion? 205 Table I3 (cont’d). I A A A I mun 1: mn'luri'neenenm 2 ilmnu I I (A mu mummg‘lumqmemmhjca I al’ ' mmflafimn luummauunsevhaa Iiahh‘i’u'ia'lai? 4' mm a 4 Mn . . ma man man lama ma 2;, and 3 ma mm x. ma ms 5 linemanqgnmnnnjarinvimoqnannia (trunnnoda finrfiafiu msnrfiuileilmulfiu ndqazv‘iwfi’ifilflucfifiwlé’)? 6. fialfinauifinfian’on n‘i‘a unmvhfianri’q asmmrnr’izfinau mmnfiuri'whulflu nfla? I22: UVV D cflfifiafinfimm El 51.11 I! U mmnuaumdnamum'a‘tfi 1. nnnaé’nszfiawqfinrsnn’anm Li‘lu v War a v v I Iv A I D QUGPIUUEQ‘DWIVIEJWN D gmnnnty‘n'i‘numnmuu D maummu 9.1 (3:14) 2. flexanmrrfilfluflsmnmsrfirmnmm Ulii Uln'lri I I a II A. I nqnnseua'imumummwm’lmn v 0 I A, A A nomwman‘lfluanmnm flfifi‘imaauauawmvhueia amumam’lfi’adnegnfi’m mu Lfiu 1:qu lu‘n‘iu tmfiu 9/ ll 91 91 man man In man man and: said as as 81 v A K! a; 1. naanmnnannnssnuu‘lflmn 206 Table I3 (cont’d). I I G A I a I: I I I a l I: nqmaaeua'imumuvi’anmm‘lma mu mu 1mm lnmu lumu 2! Q! Q! 91 I. e. .r- .. ma ma la ma ma nanonnmaalilunnmum . fltl’N arias flfifi‘imannaunwmmuna ad in anaunaem‘lfi'nziwgnfi'm 2. ué’alfipfifiéamsfimfiandainns'mlfin ‘1 mmnsmfinmrfi’nnnnafifi 3. Lfintanfianfinrinn’anaaman it 4. uamirialriwalannnnnafilfllnné’nm I 0 v I A A! (Lam mmmnnuru) 5. iJi’mflfinu‘éfimamsnsznwainwj “not naufimnnszfi’nfluflnfilfl Lfiofiszln’nfin d. . a, (o AVA 1 Dd Laminimmumsmmuaaummu’lnnn (Lin realm noriiaulfln’on) I! U 6. nannaaofinnnnafinalist-martini ‘. v I 91 o A" Lnansaut‘nulnmi mnmsna‘ww 207 Table 13 (cont’d). I I C5 91 A I newnesnmmumumnmadam y IV I 4’ A :- nonmamaalflua nmnm ilfifi'i‘mnauauawmvhusia amumsm'ln'hemgnfi'm 91 1! ma ma la mu Lfiu laiuu' lu’n‘iu lailiiu D 11 man man odor EN 11 7. rimmmhnmumwaularomnnagu tug? adn‘u 8. flexfiqanannnaéfilflaflw‘i’nLau (min ln’ué’i nonlfinqn) C: I 0 t: A d. all I 9. mummsneemmflumnnqulnmru finals 10. iiinfi’madamnfianmnnulsméfi‘mfi d. I a; A 4" I I twaumumuaznmnnmmuluudaum al. I A I ll. L‘D‘B’J’luifil‘U’lfJ / 33L'lJleJflfil'NLljul’l’Nn’l? I t: 212/ 0 U I D nndnmamuu llfilrl’lun'Wllfi‘zl’l'lanfl'l’fl’J adaaimau 208 Table I3 (cont’d). d A I claim 2: mn’ludaaeenenm 2 i‘lnmu nil: " ' “ Lflu mu muraann’lumamemnfiama .. ~ . " ' | use ‘U’rJEl man a , x . . lmnn ma :3 mmflnnnw 'luumm'iuuneemaa am :2 mm mm b are me violih‘iu’a‘o‘lzi? D V I V A I i! i! 7. wmmmnfiamnmulnnnrmrlu d. I 0’ d A I nunmu‘namumnsnammoniamn (L‘D‘u cl :1 a; d. v wmmmmunmnrnmnnzfamnlnmnu 4' i/ I Linslunwmmu? 8. unmmwaulanonaué’unfimflanfl naulnn’mmr ? 9. wmmnfiazni’mnownfiufiuflulfiaimo V I :1 d. I V :1 nnmumq nmunmmnnumué’q ? .4 .1 . .1 .. 10. mmem:Linnmnazmumulilmmnu sitar/inflame lilmnzlsfinn’onri’un‘i‘alrliu flsznaunamsfiq new Llfi'i'wi'iuazuandfi "134'"? 209 Table I3 (cont’d). A I A A I dorm annulumaaznenm 2 rlmnu d. nus mii vhumaon'lumnmerfifishfia A. . “J“ . “ ‘ . 185‘?) lJ‘BEJ 95mm] 6 . 4, . . lama ma 5 mmflnnmn 'luumm'iuunsemae am 5 am nm 5 am am dnlflflui‘olfi? 11. zinfiaéaammhu (win mafianuuru vim lnnlnzivi'm) luzi’nnmrfifinlfinaui Enfjné’mls lilamnla ? 12. wmmnaequlz’ifinfia 11??) main I I 91 d 9.! 71'1“ (L‘Ilu WflqflqNQ3§U1flqu39mufiflmflfl 9 d I naun‘innucl’nnmaanranim Innnmuln' gimme)? 13. 71mmnazfimmiufi'ufirimhulnafi I I‘ll A v 0 VI v v mulmmmr sumnnmlnmunmraa 9.1 91 91 A : I swam eaalnneanurunozfi ? minaugymnagmumirfi 1.nnnafinsxfiqwqfinswfianmo Li’lu 1.71611 ”mnalnnnei U fifiaé’ufirym‘i’umndnfi’u El Linuiqnmu ‘11 :21: El El {55 man "rpm El 3111 (can) 2.1Jsmnmacfli‘luflrmnmanlrnovhulna D111 D'lu'lri 210 Table 13 (cont’d). I I a V d l a C I I I C! I a nqmeeummumumamadam um um 1mm lnmu lama I. u. .r- .. 5021 1:12:21 ls cii'm firm nemwmnn‘lrluanmam . odor sans flfifi‘imnauaumnmnnuna as in amumsm‘lfi’ndnagnfi’aa 1. né’nnnnnfianfinsrnfiu‘lfln’on I ”9A0 d 2. uri’alnnnmmsnmfinndni’nnmmfia Q! mmnsmfinmrfi’nnnnnfifi 3. Lfimanfi’qufinrsnfirndmlfin 4. uanifivi'ilrlnolanonnnafilfllnné’nnq I 0 v I A 1 (Lou mmmamuru) 5. 11inLilii‘tiufifimqn'iensrvi'ifiwm'l 11m I A O C A A A mumnnnsanufluflnnlfl manaz’lé’nfin A I v r0 :A A] and Lnminimmuma‘mmumnmmulnnn . ..1 (ml ration mmnulrlfian) F 11°! 6. unnna’mr‘i’nnanaguuriqlraszn’mrfion D I II o A anulnulmma mnmanarmu 7. wmmnmnmumnnnularomnnaé’fi liltjfisaiiu 211 Table 13 (cont’d). I I I: v A I q a I I I a I a nqrmeena'm'iumumnmm’lmi um um 1mm lnmu lnmu I. I... .r- .. coins: visa la (31°38 51°28 nonmamaa‘lfluanmnm . anon sans ilfifi’imaounumnmvhuna 3'3 an amumem‘ln’bziwgnfi'ae 8. Harriesnannnafifilflnn’mimau (L‘ll‘Ll 1.8m nonlfinnqn) 9. Lfiudnma‘nrefinfifluaafimulfiuwju finals 91“ v I A IA 5 10. Efinnfiqaqquqnmfiflqquflulqu'lgw’flflu Li‘inui’mmmzn’mfinfianauluudanm d. I d I a 11. L‘nm’mlnmn / 33LUE11J‘BEJ’NL1JNW’NH’15‘ I A" 211/ 0 v I 9/ “11156131836401.8141 lnn’mmrnrzmmnmfl’) ariasrinmu 212 Table I3 (cont’d) I A A A I mun 3: mu'luri'neenenm 2 ilmnu A, I (A mu mulauag'lumqmsmnshiga a I A’ II ' mmfinnmu lunu'am'iuunsemaa rialfli‘lfin'lii? 1.311918 1.1111 U’N min not] aid filfi‘flfi L’Jfl’l 14. fialfinauifindnfi’ié’agnhnaunu nmm lonmrlfisw'i’an'ifian'irilfifin naunnaifitflufiLnnlugilnnnlngilunnnfia A 9/ I A A v LWQIVWWUNWQWn5‘5‘3J71'NLWfifi’JEI 7 15. fiwlfinnui’findoqnqnmnnmnq '11s: A v v I I A I I nn’ia‘unununmalnnmanua mm'm’m SI zlnlfi’mnnfmnonlumrmanaunalul’ian M i! I A A A 1W2 ma (1111.1 nnnamsfla‘ununmnun 2511an mannmunamu ‘12‘1‘11 7 v A A t I I: 16. 1.1.21nufluunmmnanununmamnm A A 0A ”A: I A nemeflgnnanmulnmumnmunms naunumlul’iqfi’mon’qn ? 213 Table 13 (cont’d). A A A I alarm 3: mn'luri'neenenm 2 ilmnu A flux} 11’“? vhumaan'lumnmerfifisheos .. . “J“ . “ ' . nave mm man a , .r . . lmnn ma 5 mafianmn'luumnnuunsemae 21m 3 ma mm x no: WIN Iia'liliiu'ia'lii 0 II I 0 21 v 0 1 A 17. mlnmumnmnnnmmmnmrylu d A 91 A A 210’ marianunsnlumgmomnnaz’lnsums ilfiilfifi‘n'lumamsmu ? 18. fialfinaulnsrdnnauarlfii’unasrlfiim =1 .d I I . I a 711an mmamulnnonsennnlumsnau A II V aumlul‘na‘gmnmn ? 19. flfiifisi’wimimaemlmian man: I I A U V 0’!) mulnnannmnauwufimn ? mannugumsingmummi 1.nnnacfinrzfiqunsrnfiinnw 11:11.1 Irv v Iv D vv w v :1 I t! D A I QUQfiuumfiqbflm?Q fiUGHUUKQ‘Dq‘Duflfll-quluu LW’fl'LWQNQ'lU U cflfifian‘nfimm U 611'! (am) 2.rJemumanilijurlseannastfiroari'mm C1111 [1111111 214 Table I3 (cont’d). I I a V d l a u I I I a I d nqmezmfimumumumm’lmfi um um 1mm 1mm 1mm ,, u. .I'- .. 5021 (309.1 13 first] 0306 mamwmm'lfluafimum I mm mm mm flfifi‘émnauauawmfiwfia 4 4 .' m m m amumetfi'lfimiwgnfi’m 1. flé’mmufiqufinsmfiu‘lflfiw 2. fialfififiéwwfimfiandfiwmu Lfirammmmfimmsfiuufimé’fi 3. Lfinmflfianfins‘a‘ufi’anmmfia 4. uamfivhlfiwahsimmafiiflimaé’flm I 9 6’ I A 3 (mu II'ImII'NLIaqu) 5. 1J§IJLflfiaufifimqmsnszfiwfiamaq “um fifiu‘VILmanixfiflLfluflnfilfl Lfiefiqzlé’vfin A I I! 0'0 Ala AX pd Laminimmunwmmumummu’tmn (L‘li‘u mafi'm mfiflulflé’oa) D 0’ 6. uanndwfi’uumaéuudqqué’mLam v I 9; o .5 'a‘fluvnulmma mnmmsxmu 215 Table 13 (cont’d). I I a a I nqmfiezqmmumufi’wmeII'lm'I v v I a" A - mflnafiumna'lfluaamflm flfifi’iflfimauauawmmufia amumsmflfi’adwgnfi’m Lfiu TII'LLIJI 1.341131; ’IIJLfiu ma 1% firm my @dfia @uIa EN Ea {I 7. wmmuu’wLuummaulqmmmmfifi lfléémfiu 8. Mafiwfiiemfiafifilflafifiqfimu (wiu IG‘I’LLE‘I’I nanlfivqm) C: I 0 1:, Al Id. i/I I 9. mummmszmmflumwmulumqu Lamb 1!“ V I ‘ IiiI A] 10. 3annmm’m’mmmwaulfimegwmu Lfiauéwmuqzfimfimaafiwluudaum dl I d I 11. L‘namuiama / Immuaa'mflumanqs‘ I I: 9/ 81 ° N I 91 IImIImm'IIIu Immumamzmmnmqh 'aziqq‘fi’mfi-Iu 216 APPENDIX G Table 14 Items Assessing Beliefs about Gender Role Stereotypes (English Version) 217 Table 14 Items Assessing Beliefs about Gender Role Stereotypes (English Version) (Strongly Agree) to l (Strongly Disagree). Instruction 11: Please indicate if you agree or disagree with the following statements. (Check the appropriate box). Your responses may range from 5 Strongly Not Strongly Agree Sure Disagree 5 4 3 2 l 1. Women should take care of running their homes and leave running the country to men. 2. Most men are better suited emotionally for politics than are most women. 3. A working mother can establish as warm and secure a relationship with her children as a mother who does not work 4. A preschool child is likely to suffer if his or her mother works. 5. It is better for all involved if the man is the achiever outside the home and the women takes care of the home and the family 218 APPENDIX H Table 15 Items Assessing Beliefs about Gender Role Stereotypes (Thai Version) 219 Table 15 Items Assessing Beliefs about Gender Role Stereotypes (Thai Version) " d I I G I a V I p 0 ' I n’IaIVIZ: ngmanummumufiom'ia'lnmué’onnuiomaum‘hji'i (mm‘i‘owma'lu'nm I ' 0 I :3 I G I A. d I 3 I owimmmu) mnommmummflu‘lfllfimum 5 (mufionnmam) rm (Inmué’otmma EN) tfiué’oa 1.31mi IIJLfiu I! ma nemfia 1a orhafia 5 4 3 2 l 9 A u A A. 11 I VA 11 1. mamanormamau'luts‘mmrqunmu‘noflmmmm uazfldanlfimrfislumrquafiautfimtfluufiwfi‘um 9! N‘il’IEl Q 9 d II (A I 2. Q‘n’ltld’mmnumw L“N’135N1flfi'lflfl'1?flmnqzlflu mrtfimmnndqé’mfia . Id. 0 3 J v 0’ iii. I 3. uunmmunmmsnmzamanonud'uwufinouquua: fiunqlfifiuqnq 1aImulé’tdutfimIi'uuzifi‘lfilé’fiaq'm 4. tfin'lmioa'iufiaufiaufiumifinfiszfimti‘luvgn'i mnuu‘ aan‘lflfiamuunnfiau I 0 v | J 5. fiztflumrfinmmumnndanfimmi’oa mmj‘mmflu Q 9 6 D J SI SI IJ'IIIIerIJm'IIImquIImIImamuunnmuuazguma .I .I ., . Ifludqanquamqmuum’nmunm APPENDIX 1 Table 16 Items Assessing Self-Construals (English Version) 221 Table 16 Items Assessing Self-Construals (English Version) (Strongly Agree) to 1 (Strongly Disagree). Instruction 11: Please indicate if you agree or disagree with the following statements. (Check the appropriate box). Your responses may range from 5 Strongly Not Strongly Agree Sure Disagree (Interdependent Self 6-1 5, Independent Self 1 5-23) 5 4 3 2 1 6. I will sacrifice my self-interest for the benefit of the group I am in 7. I act as fellow group members would prefer. 8. I stick with my group even through difficulties 9. It is important for me to maintain harmony within my grow 10. It is important to me to respect decisions made by the group 11. I will stay in a group if they need me, even when I am not happy with the group 12. Even when I strongly disagree with group members, I avoid an argument. 13. I respect people who are modest about themselves. 14. I often have the feeling that my relationship with others is more important than my own accomplishments. 222 Table 16 (cont’d). (Strongly Agree) to 1 (Strongly Disagree). Instruction II: Please indicate if you agree or disagree with the following statements. (Check the appropriate box). Your responses may range from 5 Strongly Not Strongly Agree Sure Disagg 5 3 2 l 15. My happiness depends on the happiness of those around me. 16. I don’t change my opinions in conformity with those of the majority 17. I don’t support my group when they are wrong. 18. I assert my opposition when I disagree strongly with the members of my group 19. I act the same way no matter who I am with. 20. I enjoy being unique and different from others in many respects. 21. I am comfortable with being singled out for praise or rewards. 22. Speaking up in a work/task group is not a problem for me 23. I value being in good health above everything. 223 APPENDIX I Table 17 Items Assessing Self-Construals (Thai Version) 224 Table 17 Items Assessing Self-Contruals (Thai Version) I ‘1 A I I C it A I G v V v I d" 0 4" I n'IINVIZ: nemarzummumumnnralumumnnmnmwmlflu (mmrawmn'lunm I l 0 I :1 I C I A. A I C mafimmmu) mnnmnamummflu'hflfimun 5 (mufionamam) m 1 (Iumua’lon adnqfia) tfiué’on IIJLtti Iiitfiu v mu minim In minim 5 4 3 2 1 6. ism-inseamLannnzflrzla'nfidousi'oLionéufii’nwté’a defiant} 7. ii'mLé’nfinqxnrzfinluaafitfiauamin'luntjnsimme'lfi nszfia II II 4 v III I I! v d. 8. IWWLQ’IEJ‘LWEJfiflg‘ll’Nleil‘llflQTl'MQ’I uflumunmn énuan 9. mrinmnoaudufiufifiundulfitflu‘lflndfiamifiufin .1. do v 0 v v v Lflutrmnmnrymnmmmm 10. n'IerrnIumaVTnaulr-nmnéufintflufi‘mfiénfiry VII 9 down ’ML‘I I! 11. innvi’nqwonngfluntiufiusin'lfl mnnufin'luna’n I! III II II Id 0 I :I a mam? LtII'I'III'IIILs'Ir-Izluumaumnunqnuunmn IQ Ix.) VI Table 17(cont’d) o ‘1 A I I G v d I a 1! v ii I ‘ I: 0 4" I mMVIZ: nrmnsznmmumumnnmhmumnnmomnnmolflu (mmsmnmnlwnm I ' 0 I :1 I a I A. A I C 'J'Nfimmzan) mnommmummflu‘lfllfimun 5 (mué’onaqu) m 1 (Iumué’on odoafia) a v I I I a mumn 134L114 lumu I! ma minim In nei'IIEN 5 3 2 l 12. ufidfiawLfi’asz‘lmfiun’qandaafiar‘i’unmfinlunzju D ‘l/ a d 1i. 91 D “I’mwannwamamms‘lmum A 13. I'mtr-‘I’Ié’éinfiufinnnmafiumwo’oufiounounu i! D V d 224 I V (dd I d 14. “11'1"Lr'I'lN1'11]WQ'INEZIUQ'IWQ'INZI’NWUBWNWBIIflnflfiu {incityalaihwvinmnn'inmnmhiwmmu D :3 I I v ‘11 15. manamnw’mfi’mmungnumamgm‘nmgnu 91 i‘fliJ‘Il’N 16. iiwwvi'IIIJLIJ‘z'itIumwfimfimmnutmtfialfiannna’m ri'umnmfiudwlnng 9! i! I V I i! 21 4" G 17. II'IIIInnlnauuauunqmmmwLsntnnmnI'n'ILIJuIl'ItI an 18.1’IWL'infiunmufinfi’nu Lfialfitfiué’qmi‘lundwfiafiu amifinlunzju 19 10 0‘ Table 17 (cont’d) ‘1 A I I 6 i! G. I 3 2/ V v I a‘: 0 r I 'IINVIZ: nqmnszummumumnnro’lumumnnumnmwmnlflu (mma‘mnmnlwnm I 0 I :I I a I A A I G mafimmmn) mmommmummifilu’lflln’mun 5 (mué’oaamam) m 1 (Immun’on ndaafia) c: 21 I I I <3 mumn 131 w human 9! non nemm 1s semen 91 II a 0‘ v d w I I 21 II I V G 19. mmmflgun ouumnmnu’lmnmmmnzag ‘lJ "Inn ‘II [51111 20. iamihfimanqmri’umsfitonfinuniuazunnn'wnnnlfl “I I, nongoulwmn'l mu 21. iinwfiahliannfinnzmals findé’i’uéwumnn‘in V I d 91 1‘ I’Imaoqmmgmm I II I I 0 IV I 22. mennmuqnlumrflrmunqnfimmunulu’lélti‘lu fltymziwi’uifim'ia d v v - aI «J an a «I 23. ‘Il’IWLQ'IL‘IWI‘y‘n'I‘J‘NQ‘anlWVlflkuu’flafiflulfi 227 APPENDIX K Table 18 Items Assessing Participation in Decision-Making (English Version) 228 Table 18 Items Assessing Participation in Decision-Making (English Version) Instruction 111: Overall, since you have been working in this agency, to what extent do your superiors (your direct supervisors and higher ups) do the following? (Check the apprOpriate box). Your responses may range from 5 (A great deal) to 1 (Not at all). A great Not at deal all 5 4 3 2 l 1. Allow you to decide how you do your job 2. Help you to have a say or influence on what goes on in your department. 3. Encourage you to have a say or influence on decisions that affect your job. 4. Are receptive to your ideas 5. Encourage you to participate in decision-making regarding matters that relate to your job. 229 APPENDIX L Table 19 Items Assessing Participation in Decision—Making (Thai Version) 230 Table 19 Items Assessing Participation in Decision-Making (Thai Version) UV 0 d 91 t! I I II 0 I. 0 I v d. V v mi’an3:Inamnummunmuwnmmmunglumunuwmnflqaguu mama I II c! v v v u v v u I: . I v 0 . I v 1.1mm (£11 anuurymllmnmquazfi Innntu‘mrznugmu'lfl),finuonaaufinrznaaanalflfi mm’iamfimln (Ii'ILh‘imnmnlmimdwfimmzau) finnammvhummifluhl'lé’ei’qun 5 (azi'mnn) an 1 (Minna) ari'II 1min N'Il'l LflEl 1. numnn'lfivhummrnfinau'lalé’tmdaamlfififimurm Iinuaeinqlr 2. ‘IIQUIII‘VI’IIALLEWNHQ’INfifitfiuu‘i‘flfififlfiwflfilfln’nmflu lfllunuomnu'nmnnu 3. nfiunuu‘lfinmuammnufimfiun‘inlfififinfiwaname V I A. I V II A I 91913141951361 mmmnnunmnmrmurmmu 4. %flamnnfintfiunmnw 5. quanu‘lfivinufidouianlunnrfinaula qu‘i‘mfitfim 1’ v II A I ‘nflonunu’mmrmmmmu 231 APPENDIX M An opened ended question assessing sexual harassment experiences 232 Instruction IV: Based on your experiences with male officers in this agency, please describe things that any male officers have done or said, which made you, as being a woman feel uncomfortable (ambivalent, tongue-tied, embarrassed, annoyed, feel _’ treated inappropriately etc.) List as many situations as you can. In describing each of these situations, please be specific about what was said and done. Also briefly explain how you reacted to it and what made you behave the way that you did but please do not enter any names. 0 ‘l A 6’ I 0! 0 I I! I I in. '- mam 4 : mnflsmnmsmnmmunummsmnlunmnmuiifimum ngmamsmnfiam I d. 0 a 91 0 d y al. 0 91 IV I 1‘ sm'] nmmsmmfianmu tnnlmnrzmnrn'qnluannmznmlnmmutm Lug'tyzm‘i‘ly A $24 I I A v A d. I Iv I 0 91:9 920’ 18qu ganlnazmnln (mu ananamao wmmlnann nszannszmuls mmty ganlnsun'I-I flfiufianwlfimmmn nan) 121'amumrniaa‘liflt‘fln’mnfiqnLnnfinnummrnfialn’ Iums 6‘ I I: I d tfl . O A A“ nssmnamumsmmmu ngmnsxntmmalflmummnoel? mamsmwalnmmu n‘r‘imnzi’u'] é’ondnvhulé’muaumlflL‘Iiulsua:aflrfinlfi’vhunsxfiflflL'IiuIIu usinqrmath Inlaiimmcflnmlflfi’aau APPENDIX N Demographic Questionnaire 234 InstructionV: Demographic Information Please answer each of the following questions (All of your personal information will remain confidential and will be used for data analysis purpose only). 1. Age Years 2. Marital Status: Cl Single E] Married [:1 Separated/Divorced/Widowed 3. Ethnic group [JWhite (non-Hispanic) DAfn'can-American “a: E] Hispanic E1 Others (specify) 4. Highest level of education: El High School [:1 Associate’s Degree 1:1 Some college E1 Bachelor’s Degree El Master’s Degree [:1 PhD. 5. Rank (Please Write in) 6. Have you ever been in a supervisory position? UYes E] No 7. How long have you been working in this agency ? Years 8. How likely is it that you will remain in police work until retirement? (Circle one) Very Likely 5 4 3 2 1 Very Unlikely 235 9. To what extent does your current work routine require you to interact primarily with male coworkers in the absence of female coworkers? (Circle one) A great deal 5 4 3 2 1 Not at all 10. Please describe the types of job responsibilities you are currently assigned. If what you are doing is different from what you used to do within the past 2 years at this agency, please also describe the nature of those former job responsibilities and indicate when the changes took place (e. g. 6 months ago, 12 months ago) Current responsibilities Former responsibilities 236 U 0 o d I I 21 I v 91 I Iv I mam 5 nagamum nrmnnanmmnunawanalflfi (Ianamummaamu t’ G' v 9 9/ 45' A rv I d I :1 mnumzqnani'lumnuanuazaxm‘hfl‘nmama‘omflzmagaunmmmauu) 2. ammnwmsama 1:11am Dam/attain Uunnri’uaal/nehi’w/Lflunfim 3. renumsfinmqaqa [1175614112qu Uflrznqnfiafimmfi‘nfiw Dclaumrfinmluiiuamufinmmfiw Dilitutmn'i Utfitytynl‘n Dtl‘rrutyntan 4. EM 5. vinumnéhrarwiaumiaé’ "Mn "tymfia’lzi ? Elma [1134'th ' 6. Ii'Iuvi'IInuiuutionaquiim‘lai’tflus: :L'Imiii] ? ............................. U 7. fim'nuflu'hllé’mnfinaLfi'mlnfiviauazi'nmamafinmalflaunrsv’faLnfiammgramme? (Qanawfiwmmm) Liiulillé’unn 5 4 3 2 1 Lilu'lilloi’fiaamn 8. gIJLLIJIImafiamumaavimlummzflaqfi’u finlfinaufiaafiamufi’uLfiauiwmu‘mamu o y 4i. Idcl. I A II; p d mm Iu'nmznlnnmaumnmunmqflmngagmnmnummmln ? (canawi‘tanmmam) Li'luatmmn 5 4 3 2 1 1mm 4‘ v a: y A 4" I 91w ’1 0 I v 9. namnuramananmnPmnrnIImammmunmu'laanuauumnlnnazmagluflsquu I I 0 I finfiqanfinmfia ”Inqagluflaqfiufiumnnadmnaafinwtnanszfiqmnaulumiomwfi mn'luii'nr: :L'Imaaqi‘lfichum nammrsmné’nnmwfifiman-int?mantrinma'niu I. . .I . r. .. .r .I . .I ..I ., .. ..I .. ma uawzmnmnflaauuflaammuummutualn (mu 6 Laaunum 12 maunum) ’1 d. v A v nunnmnumnmanflaqnu nfinfimwi’nfimamfiu Appendix 0 Debriefing Statements 239 The questionnaires that you have just filled out are intended to examine the differences in perceptions of and reactions to sexual harassment between Thai and American policewomen. The situations depicted are those that could be interpreted as sexual harassment in the U.S. The purpose of the study is to assess if American policewomen would fine these situations more inappropriate and react to those situations in a more assertive manner than Thai policewomen. This is because sexual harassment in the workplace is an issue that has been viewed as a problem in the U.S. much longer than in Thailand. It is very likely that this might differentiate the ways that Thai and American policewomen perceive and react to these situations. If you have been exposed to any of those situations mentioned in the survey and feel that you need to talk to someone about these experiences, or need advice about how to deal with the situations, please contact: Michigan Self-Help Clearing House 1- 800-777-5566 or (517) 484-7373 or at http://www.mpas.org. 240 d. I 1:. 11 G A” I: :5. 0 I v 91 unuaaumnnmumalnmantarsaulflutflulfltnaaasoanmnunnmalumssng an: manauaumnamsqnmumqmmwdqq énmanflalnn uass‘inrowtfiaam‘sfiu (A. O 6’ I d. d 91 I amumrtunmmtiluamumrmmaqnmaszmnonnlaqatfiunnrqnmumsmn‘luilrzmn y A a' A u A“ d. A I O A A v a I anagamsm amissmmmmmaautflu’lflmaflrunmqmmanmaaLnrnuazmmn amumrnimziafilnmmsaIJII'Inm'n uammauwasviamnnqstfitnanfilfltmui’nman‘fi I 1‘ I ’0” d” d. 0 II. (J ”V I nutaamnmwralu mutflumrnzmrqnmnmamnIunnnqnuLflqunanlarunnrumon Lilufltymluanigam‘imminmundnluflszmale/121mm fiafi’mi‘lu‘ltflé’daqnfiaaaazv’inlfi A I AA A. O A A w I tnnmanumnmqluonnnannnmsnmalnauazaturnu liflumrum uaznnuaumna amumrcfima’qi’l mnvhulé’tnanufi’uma}macflafimannln’namfifluunnaaumu unzifindqnau it- A v 0 ' d I I v v r I y fi‘immsfiaznnnaflrmumrtfimmu manonmamrl‘inmmazanmannamumimmmfi Iumalnlfifina nrmqfimnagafi‘fitfiaumfiq 386/61 inmfitnn 44 (mantafineg‘n) nuni’amfimn mama nay-In? nsatnn 10900 171??va1 (02)513-2780, 513-1929 REFERENCES Adams, J ., Kottke, J. & Padgit, J. S. (1983). Sexual harassment of university students. Journal of College Student Personnel. November, 484-490. Adams-Roy, J. & Barling, J. (1998). Predicting the decision to confront or report sexual harassment. Journal of Organizational Behavior. 19, 329-336. Anderson, R., Brown, J. & Campbell, E. A, (1993). Aspects of sex discrimination within the police service in England and Wales. Home Office Police Research Group: London. . Appier, J. (1998). Policing Women: The Sexual Politics of Law Enforcement and the LAPD. Philadelphia: Temple University. Anneli, S. E. R. Fasolo, P. & Lynch, P. (1998). Perceived Organizational support and police performance: The Moderating Influence of Socioemotional Needs. Journal of Applied Psychology. 83. 2. 288-297. Backhouse, C., & Cohen, L. (1981). Sexual harassment on the iob. Englewood Cliffs, NJ: Prentice-Hall. Bails, C. R. (1994). Females Reactions to Sexual Harassment in the Workplace and the Impact of Race. Unpublished Doctoral Dissertation. Temple University. Baker, D. D. , Terpstra , D. E. & Larntz , K. (1990). The influence of individual characteristics and severity of harassing behavior on reactions to sexual harassment. Sex Roles. 22. 305-326. Baker, N. L. (1989). Sexual lgrassment and iob satisfaction in trafiditional and nongditional industrial occupations. Unpublished Doctoral Dissertation. California School of Professional Psychology. Baroni, B. J. & Petrini C. M. (1992). Unwelcome Advances-Legal Liability: Power Struggles. Training & Developmeng, 46, 5, 19-21. Bartol, C. R., Bergen, G. T., Volckens, J. S., & Knoras, K. M. (1992). Women in small- town policing. Criminal Justice and Behavior 19,240-260. Beirene, P. & Messerschmidt, J. (1995). Criminologl Orlando, Florida: Harcourt Brace. Belk, S. S., Snell, W. E., Holzman,W. H. Hemandez-Sanchez, J. E., & Garcia,- Falconi, R. (1989). The impact of Ethnicity, Nationality, Counseling Orientation and 243 Mental Health Standards on Stereotypic Beliefs About Women: A Pilot Study. Sex Roles. 21. 9-10, 671-695. Benson, D. J., & Thompson, G. E. (1982). Sexual harassment on a university campus: The confluence of authority relations, sexual interest and gender stratification. Social ProblemsJ9L236-251. Bentler, P. M. (1985). Theory and implementation of EQS: A structural equations program. Lost Angeles: BMDP Statistical Software. Bhatiasevi, A. (1998). Law will protect women against sexual harassment. [Online] Available http://scoop.bangkokpost.co.th/bangkopostnes/bp199801 15/150198news05.htm1, February, 12, 2001. Bingharn, S. G. & Scherer, L. L. (1993). Factors Associated with Responses to Sexual Harassment and Satisfaction with Outcome. Sex roles. 29,3-4, 239-269. Brooks, L. & Perot, A. R. (1991). Reporting Sexual Harassment: Exploring a Predictive Model. Psychology of Women Quarterly. 15. 31-47. Brown , J. & Heidensohn, F. (2000). Gender apd Policing: Cormaarative Perspectives. New York: St. Martin’s Press. Brown, J. M. (1998). Aspects of discriminatory treatment of women police officers serving in forces in England and Wales. Britis_h Journal of Criminology, 38, 2, 165-282. Bularzik, M. (1978). Sexual harassment at the workplace: historical notes. Radical American. 12. 4. 25-44. Byers, S. E. (1996). How well does the traditional sexual script explain coercion?. Review of a program of research. In S. E. Byers & L. F. O’Sullivan (Eds), Sexual coercion in dating relationships (pp.7-25L Binghamton, NY: Haworth Press. Chang, W. C., Chua, W. L. & Toh, Y. (1997). The concept of Psychological Control in the Asian Context. In L. Kwok., U. Kim, Yarnguchi, S. & Kashima, Y. (Eds). Progress in Asian Social PsvchologYWoll .. pp.95-117). New York : Wiley & Sons. Chamey, D. A., & Russell, R. C. (1994). An overview of sexual harassment. American JoumaiofPsvchiatrv. ISILI. 10-17. Chen, C. C. & Ditomaso, N. (1996) Performance Appraisal and Demographic Diversity : Issues Regarding Appraisals, Appraisers, and Appraising. In Kossek, E. E. & L. 244 A. Sharon (Eds)., Managing Diversity : Human resource strategies for transforming the workplace (pp.137-163). Cambridge: Blackwell. Chu, J. (1987). Asian Pacific American women in mainstream politics. Asian American Psychological Association Journal. 12, 1. 19-33. Cochran, C. C., Frazier, P., A. & Olson, A. M. (1997). Predictors of responses to unwanted sexual attention. Psychology of Women Quarterly. 2L 2, 207-226. Coles, F. S. (1986). Forced to quit: Sexual harassment complaints and agency response. Sex RolesL14, 81-95. Collings, S. J. (1987). Baniers to rape reporting among White Afiican Women, South Afiican. Journal of Psychology. 17. 20.-24. Collins, E. G. C., & Blodgett, T. B. (1981). Sexual harassment..some see it...some won’t. Harvard Business Review. 59. 76—95. Connell, D. S. (1991). Effective Sexual Harassment Policies: Unexpected Lessons from Jacksonville Shipyards. Employee Relations 17 2 191-206. Conte, A. (1997). Legal Theories of Sexual Harassment In W. O’Donohue (Ed).,Sexual Harassment : Theory Research and Treatment. (pp.50-83) Needharn Heights, MA: Allyn & Bacon. Conway, M. R. (1998). Sexual harassment aboard. Wtflcforce. 34. 8-9. Crittenden, K. S. (1991). Asian self-effacement or feminine modesty? Attributional patterns of women university students in Taiwan. Gender and society. 5; l. 98- 117. Cross, S. E. & Madson, L. (1997). Models of the Self : Self Contstruals and Gender. PsychologigBulletin, 122. 1. 5-37 Crull, P. (1982). Stress effects of sexual harassment on the job: Implications for counseling. Amegan Journal of Onhopsychiam, 52. 539-544. De Coster, S., Estes, S. B. & Mueller, C. W. (1999). Routine Atcivities and Sexual Harassment in the Workplace. Worl_<_and Occupations. 26.1. 21-49. Denton, M. & Zeytinoglu, I. U. (1993). Perceived participation in decision-making in a university setting: The impact of gender. Industrial & Labor Relzmnp Review 46. 2, 320-333. DeSouza, E. R., Pryor, J. B., & Hutz, C. S. (1998). Reactions to Sexual Harassment 245 Charges Between North Americans and Brazilians. Sex Roles. 39. 11-12, 913- 928. Detroit News. (1990). Sexually Harassed, Female Cops Say. Detroit News. Dec 9, A11- A15, Bl. Dubois, C. L. 2., Faley, R. H., Kustis, G. A., Knapp, D. E. (1999). Perceptions of organizational responses to formal sexual harassment complaints. Journal of Managerial Issues. 1L 2. 198-212. East Asian Executive Reports. (1998). Thailand: Labor and business differ on new Thai labor law. East Asian executive Reports. 20. 4. 9, -22. Equal Employment Opportunity Commission. (1992). Guidelines on discrimination based on sex. 29 Code of Federal_ReguIations (Labor). 1604.11. Equal Employment Opportunity Commission: Guidelines on discrimination because of sex , 29 C. F. R. 1604.11 (1980). Erickson, K. T. (1966). Wayward Puritans: A study in the Sociology of Deviance. New York: Wiley. Faley, L. (1978). Sexual Shakedown: The sexual harassment of women on the m. New York: McGraw-Hill. Faley, R. H. (1982). Sexual Harassment: Critical review of legal c-ases with general principles and preventive measures. Personnel Psychology. 35. 583-600. Faley, R. H., Knapp, D. E., Kustis, G. A., & Dubois, C. L. Z. (1994). Organizational cost of sexual harassment in the workplace. Paper presented at the Ninth Annual Conference of the Society for Industrial and Organizational Psychology, Nashville, TN. Feldman-Summers, S. & Norris J. (1984). Differences between rape victims who report and those who do not report to a public agency. Journal of Applied Socia_1_ Psychology. 14. 562-573. Feldman-Summers, S. & Ashworth, C. (1981) Factors related to intentions to report a rape. Journal of Social Issues 37 53-70. Firestone, J. M., & Harris, R. J. (1994). Sexual harassment in the U.S. military. Individualized and environmental contexts. Armed Forcesfi and Society. 21. 1. 25- 43 F iske, S. T., & Glick, P. (1995). Ambivalence and stereotypes cause sexual harassment: 246 A theory with implications for organizational change. J ournal of socgl Issues. 51, _1_, 97-1 15. Fitzgerald L. F., & Shulhnan, S. L. (1993). Sexual Harassment: A research analysis and agenda for the 19908. Journal of Vocanional Behavior. 42. 5-27. Fitzgerald, L. F., Gelfand, M. J. & Drasgow, F. (1995). Measuring sexual harassment: Theoretical and psychometric advances. Basic and Applied Socfl Psychology. 17, 4, 425-445. Fitzgerald, L. F., Swan, S. & Magley, V. J. (1997). But was it really sexual harassment? Legal, Behavioral and Psychological Definitions of the Workplace Victimization of women. In O’Donohue W. (Ed). Sexual Harassment: Theory. Research and Treatment (pp.5-281. Needharn Heights: Allyn & Bacon. Fitzgerald, L. F. (1990). Sexual Harassment: The Definition and Measurement of a Construct. In M. Paludi (Ed). Ivory Power: Sexual harassment on campus (pp. 21-44) . New York: Suny. Fitzgerald, L. F., Shullman, S. L., Bailey, N., Richards, M., Swecker, J., Gold, Y. Ormerod, M. & Weitzman, L. (1988). The incidence and dimensions of sexual harassment and in academia and the workplace. Journal of Vocational Behavior 3_2_, 152-175. Ganguly, A. (1997). Predictors of attitudes towards wife-beating among Asian Indians in America. Unpublished Doctoral Dissertation: California Institute of Professional Psychology. Garson, D. (2003). Multiple Regression. [Online] Available http://www2.chass.ncsu.edu/garson/pa765/regress.htm#toleran_ce, March 1, 2003. Geare, A. J. (1998). Sexual Harassment: Modern issue—ancient problem. New Zealand Journal of Industrial Relations 22-23 3/1, 241-276. Gelfand, M. J., Fitzgerald, L. F. Drasgow, F. (1995). The structure of sexual harassment: A confirmatory analysis across cultures and settings. I oum_al of vocational behavior 47 2, 164-177. Gillogly, K. (2000). Forced sex endemic in Th_ai society. [Online] Available http/lscoop.bangkokpost.co.th/bangkokpostnews/pr 0001 01 5/ 1 5 1 000news26.htm 1, 12 February 12, 2001. Goldfarb, R. G. (1985). Effects of a workshop designed to promote effective coping with sexual harassment and its associated effects: A single case design. Unpublished Doctoral Dissertation. University of Maryland. 247 Gosselin, H.L. (1984). Sexual harassment on the job: Psychological, social and economic repercussions. Canada's Mengl Health. 32, 3, 21-24. Greenberger, D. B., & Strasser, S. (1991). The role of situational and dispositional factors in the enhancement of personal control in organizations. Research in Organizational Behavior. 13, 111-145. Gruber, J. E. (1990). Methodological problems and policy implications in sexual harassment research. Population Research and Policy Review, 9, 235-254. Gruber, J. E. (1997). An Epidemiology of Sexual Harassment: Evidence from North American and Europe. In O’Donohue W. (Ed). Sexual Harassment: Theoryg Research and Treatment (pp.84-98). Needharn Heights: Allyn & Bacon. Gruber, J. (1998). The Impact of male work environments and organizational policies on women’s experiences of sexual harassment. Gender and Society. 124;, 301-320. Gruber, J. E. & Bjorn, L. (1982). Blue-collar blues: The sexual harassment of women autoworkers. WorLand Occupation. 9. 3, 271-298. Gruber, J. E. & Bjorn, L. (1986). Women’s responses to sexual harassment: An analysis of sociocultural, organizational, and personal resource models. SociaLScience Quarterly, 67, 814-826. Gruber, J & Smith, M. (1995). Women’s responses to sexual harassment: A multivariate analysis. Basignd Applied Social Psychology. 17. 543-562. Gutek, B. A. (1985). Sex and the Worflace: The impact of sexual behavior 3&1. pgassment on women. men and organizations. San Francisco: J ossey-Bass. Gutek, B. A. & Koss, M. P. (1993). Changed women and changed organizations: Consequences and Coping with Sexual Harassment. Journal of Vocational Behavior, 42, 28-48. Gutek, B. A., & Morasch, B. (1982). Sex-ratios, sex role spillover, and sexual harassment of women at work. Journal of Social Issues. 38, 4. 55-74. Gutek, B. A., Cohen, A. G., & Konrad, A. M. (1990). Predicting social-sexual behavior at work: A contact hypothesis. Agdemy of Management Journal. 33, 560-577. Gutek, B. A., Kakumara, C. Y., Gahart, M. & Handschumacher, I. (1980). Sexuality and the workplace: Basic_and Applied 80% Psychology. 1. 3. 255-265. Herrera, R. S., DelCampo, R. L., & Ames, M. H. (1993). A serial approach for translating family science instrumentation. Family Relations, 42, 357-360. 248 Hofstede G. (1980). Culture’s corpsguences: International differgtnces in work; related values. Beverly Hills, CA: Sage Hofstede G. (1986). Cultural differences in teaching and Learning. International Journi of Intercultural Relations. 10, 301-320. Hong, S. M. & Giannakopoulos, E. (1994). The relationship of satisfaction with life to personality characteristics. Journal of Psychology, 128, 5, 547-558. Hsieh, T. , Shybut, J. & Lotsof, E. (1969). Internal vs. External control and ethnic group membership: a cross-cultural comparison. Journal of Consulting and Clinical Psychology, 33, 122-124. Hunt, J. (1990). The Logic of Sexism among Police. Women and Criminal justice,l, 3- 30. Jack, D. C. (1991). SilencinLthe self: Women ami depression. Cambridge, MA: Harvard University Press. Jacobi, L. L. (1999). The application of role theory to the sexual harassment paradigr_n. A policy capturing approach. Unpublished Doctoral Dissertation. Old Dominion University. Janoff-Bullman, R. (1979). Characterological versus behavioral self-blame: Inquiries into depression and rape. Journal of Personality and Social Psychology. 37, 10. 1 798-1809. Janus, S. S., Janus, O, Lord, L. K. & Power, T. (1988). Women in Police work-Annie Oakley or Little Orphan Annie? Police Studies. I 1. 3. 124-127. J eanquart-Barone, S. (1996). Examination of Supervisory Satisfaction in Traditional and Nontraditional Gender Based Reporting Relationships. Sex role4L9/10, 717- 728. Jenson, I. & Gutek, B. I. (1982). Attributions and assignment of responsibility in sexual harassment. JournaL of Social Issuesa38. 121-136. Jone, T. (1991) Ethical decision making by individuals in organizations: An issue- contingent model. Academy of Management Review, 16, 366-3 95. Jones, T. S. & Remland, M. S. (1992). Sources of variability in perceptions of and responses to sexual harassment. Sex Roles, 27, 121-141. Judge, T. & Ferris, G. (1993). Social Context of Performance Evaluation Decisions. Academy of Management Journal, 36, 1, 80-105. 249 Kauppinen-Toropainen, K. & Gruber, J. E. (1993). Antecedents and Outcomes of Women-Unfriendly Experiences: A study of Scandinavian, Former Soviet, and American Women. Psychology of Women Quarterly, 17, 431-456. Kenig, S., & Ryan J. (1986). Sex differences in levels of tolerance and attribution of blame for sexual harassment on a university campus. Sex Roles, 15, 535-549. Kim, M. & Sharkey, W. F. (1995). Independent and interdependent construals of self: Explaining cultural patterns of interpersonal communication in multi-cultural organizational settings. Communication Quarterly, 43, 1, 20. Kim, M., Hunter, J ., Miyahara, A. & Horvath, A. (1996). Individual- Vs. Cultural-level dimensions of individualism and collectivism: Effects on preferred conversation styles. Communication Monographsa63, 1. 28-29. Kitayarna, S., Markus, H. R, Matsurnoto, H. & Narasakkunkit, V. (1997). Individual and collective Processes in the construction of the self: Self-enhancement in the United States and self-criticism in Japan. Journal of Personalitxand Social_ Psychology. 72. 6. 1245-1267. Knapp, D. E. Paley, R. H., Ekeberg. S. E. & Dubois, C. L. Z. (1997). Determinants of target responses to sexual harassment: A conceptual framework. Academy of management Review. 22. 3. 687-729. Koss, M. P. (1990). Changed lives: The psychological impact of sexual harassment. In M. Paludi (Ed). Ivory Power: Sexual harassment ommpus. (pp. 73-92) New York: Suny. Krahe, B., Scheinberger-Olwig, R. & Kolpin, S. ( 2000). Ambiguous Communication of Sexual Intentions as a Risk Marker of Sexual Aggression. Sex Roles, 42. 5/6. 313-337. Kulik C. T., Perry, E. L. & Schmidtke, J. M. (1997). Responses to sexual harassment: The effect of perspective. J ouflal of Managerial Issues, 9a], 37-53. Lafontaine, E., & Tredeau, L. (1986). The fi'equency, sources and correlates of sexual harassment among women in traditional male occupations, sex roles. 15. 433-438. Lai, T. A. (1985). East Asian Women: Resisting the violence. Working Together: To Prevent Sexual_ and Domestic Violence, 5 (5). Landrine, H. & K'lonoff, E. A. (1997) Discrimflrtion against women : prevalence, consegences, remedies. Thousand Oaks, California: Sage Publications. Lazarus R. S. & Folkrnan, S. (1984). Stress.appraisal, and coping .NY: Springer 250 Publishing Company. Lee, A. Y., Gardner, W. L., & Aaker, J. L. (2000). The Pleasures and Pains of Distinct Self-Construals: The Role of Interdependence in Regulatory Focus. Journal of Personjiity and Social Psychology. 78. 6. 1122-1134. Lerner, M. J. (1971). Observer’s evaluation of a victim: Justice Guilt, and veridical perception. J oumal of Personflty am! Social Psychology. 20, 127-135. Lindsey, K. Sexual harassment on the job and how to stop it. (1997). Ms. Magazine, 47- 48. Littler -Bishop, 8., Seidler-Feller, D. , & Opaluch, R. E. (1982). Sexual harassment in the workplace as a fimction of initiator’s status: The case of airline personnel. Journal of Social Issues. 38. 4. 5-22. Livingston, J. A. (1982). Responses to sexual harassment on the job: Legal, organizational and individual actions. Journal of Social Issues. 38. 4. 5-22. Loy, P. H. & Stewart, L. P. (1984). The extent and effects of the sexual harassment of working women. Sociological Focus, 17, 31-43. Luzio, R. C. (1993). Effects of “bystander”versus “direct” sexual harassment on women’s emotional reactions, motivation and wLflr-related coping strategies. Unpublished Dissertation. California School of Professional Psychology. MacKinnon, C. A. (1979). Sexual Harassment of Working Women: A Case of Sex Discrimination. New Haven, CT: Yale University Press. Mahithirook, A. (2000). Flight Crews want legislation to stamp out rowdy travellers. [Online] Available http://scoop.bangkokpost.co.th/ bangkonkpostnews/pr0000705/050700 news05.htrnl, February 12, 2001. Malovich, N. J ., & Stake, J. E. (1990). Sexual harassment on campus: Individual differences in attitudes and beliefs. Psychology of Women Qaarterly. 14. 63-81. Mangione-Lambie, M. G. (1994). Sexual Harassment: The Effects of Perceiver Gender Race and Rank on Attitudes and Actiofins. Unpublished Doctoral Dissertation. California School of Professional Psychology. Mansfield, P. K., Koch, P. B., Henderson, J ., Vicary, J. R., Cohn, M., & Young, E. W. (1991). The job climate for women in traditionally male blue-collar occupations. Sex Roles, 25, 63-79. Marin, A. J. & Guadagno, R. E. (1999). Perceptions of sexual harassment victims as a 251 function of labeling and reporting. Sex-Roles. 4L 11-12, 921-940. Markert, J. (1999). Sexual Harassment and the Communication Conundrum. Gender Issues, 17, 3, 34-52. Markus, H. R. & Kitayarna, S. (1991). Culture and the Self: Implications for cognition, emotion and motivation. Psychological Review, 98, 2, 224-253. Martin, S. E. (1990). On the Move: The Status of Women in Policing. Washington: Police Foundation. Martin, S. E. (1978). Sexual Politics in the Workplace: The Interactional World of Policewomen. Symbolic Interaction. L 2, 44-60. Martin, S. E. (1980). Breaking aad Enter-rag: Policewomen on Patrol. Berkeley: University of California Press. Martin, S. E. (1993). Female Officers on the Move?: A Status Report on Women in Policing. In R. D. Dunham & G. P. Alpert (Eds), Critical Issues in Policing (pp. 327-347). Prospect Heights, IL: Waveland. Martin, S. E. (1994). Outsider within the station house: The impact of race and gender on black women police. Social problems, 41, 383-400. Matsui, T., Kakuyama, T., Mary-Lou, O. & Martin, O. (1995). Women’s perceptions of social-sexual behavior: A cross-cultural replication. Journal of Vocatiorgl_ Behavior. 46. 2, 203-215. Mattrnan Jr., G. L. (2000b). A global view of sexual harassment. HRMagazine. 45, 1, 151-158. Mattrnan, Jr., G. L. (2000a). Harassment, discrimination laws go global. National Underwriter, 104. 37. 34-35. Maypole, D. E. & Skaine, R. (1982). Sexual harassment of blue-collar workers. Journal of Sociologyand Social_Weljrre.9, 682-695. Mclnnis, M. & Fitzgerald, L. F. (1997). Sexual Harassment: A preliminary test of an integrative model. Jourpal of Applied SocialPsychology, 27a10, 877-901. Miller, J. G. (1984). Culture and the Development of everyday social explanation. _ Journal of Personality and Social Psychologya46, 961-978. Millham, J ., & Smith, L. E. (1981). Sex-role Differentiation among Black and White Americans: A comparative study. The Journal of Blaak Psychology. 7. 2. 77-90. 252 Mishra, A. K. & Spreitzer, G. M. (1998). Explaining how survivors respond to downsizing: The role of trust, empowerment, justice and work redesign. Academy of Management, 23, 3, 567-588. Moore, K. S. (1995). Perceptions of sexual harassment in organizatioaral settings: the _c_a_§e of subordinate-initiaged harassment. Unpublished Doctoral Dissertation. Ohio State University. Morash, M. , Hoffman, V., Lee, Y. H. & Shim, Y. H. (1999). Wife Abuse in South Korea. In R. Muraskin (Ed.), Women and Justice: Development of International Policy (pp. 47-85). Amsterdam: Gordon and Breach. Morash, M., & Haarr, R. (1995). Gender, workplace problems, and stress in policing. Justice Quarterly. 12. 113-140. Mori, L., Bemat, J. A., Glenn, P. A., Selle, L. L. & Zarate, M. G. (1995). Attitudes toward rape: Gender and Ethnic Differences across Asian and Caucasian College ""- Students. Sex Roles, 32, 7-8, 457-467. Morris, M. W. & Peng, K. (1994). American and Chinese attributions for social and physical events. Journal of Personality and Social PsychologY. 67, 6. 949-971. Morse, H. B. (1995). Factors affecting rpporting behavior and psychological effects on femalegraduate students who are sexually harassed. Unpublished Doctoral Dissertation. University of Kentucky. National Center for Women & Policing. (1999). Equflty Denied: The Status of Women in Policipg: 1998. The Feminist Majority Foundation. Neale, N. C. (1991). The role of culture and gender in perceptions of sexual harassment. Unpublished Doctoral Dissertation. University of Minnesota. Neumann, J .E. (1989). Why people don’t participate in organizational change?. In R. Woodman & W. Pasmore (Eds.), Research in organizational change and development, Vol. 3. (pp. 1 81-212). Greenwich, CT: JAI Press. Niebuhr, R. E., & Boyles, W. R. (1991) Sexual harassment of military personnel: An examination of power differentials. IntematiorarLJoumal of Intercultural Relations, 15-445-457. Nurius, P. S., Noriss, J ., Young, D. S., Gramham, T. L., Gaylord, J. (2000). Interpreting and Defensively Responding to Threat: Examining Appraisals and Coping with acquaintance Sexual Aggression. Violence and Victims. 1522. 187-208. O’Meara, D. J. (1989). Cross-Sex Friendship: Four Basic Challenges of an Ignored 253 Relationship. Sex Roles, 21L7-8. 525-543. Oetzel, J. G. (1998a ). The effects of self-construals and ethnicity on self-reported conflict styles. Communication Reports, 11, 2, 133-144. Oetzel, J. G. (1998b). Culturally homogeneous and heterogeneous groups: Explaining communication processes through individualism-collectivism and self-construal. International Journal of Intercultural Relations, 22a2, 135-161. Offerrnann, L. R. & Hellmann, P. S. (1997). Culture’s Consequences for leadership behavior: National values in action. Journal of Cross-cultural Psychology. 28. 3. 342-3 5 l . Osborne, J. W. & Waters, E. (2002). Four assumptions of Multiple Regression that Researchers Should Always Test. fiactigrl Assessment. Research & Evaluation 8, 2. [Online] Available Imp://ericae.net/pare/getvn.asp?m§p?v=8&n=2, March 1, 2003. Oshinsky, J. C. (1980). Sexual harassment of women students in higher educaaon (1980). Unpublished Doctoral Dissertation. University of Florida. Paetzold, R. L. & O’Leary-Kelly, A. M. (1993). Organizational communication and the legal dimensions of hostile environment sexual harassment. In G. L. Kreps (Ed.), Sexual harassment: Communication implications (pp.63-77). Cresshill, NJ: Hampton, Press. Pike, D. L. (1992). Women in police academy training: some aspects of organizational response. In M. L, Imogene (Ed), The Changing Roles of Women in the Criminal Justice System (pp251-270L Prospect Heights, IL: Waveland Press Poasa, K., H., Mallinckrodt, B., Suzuki, L. A. (2000). Causal attributions for problematic family interactions: A qualitative, cultural comparison of Western Samoa, American Samoa and the United States. Counseling Psychologist, 28, 1, 32-60. Popovich, P. M. (1988). Sexual Harassment in Organizations. Employee Responsibilities fld Riglats Journal. 1. 4. 273-282. Radelet, L. A. & Carter, D. L. (1994). The Policaand the Community. Ontario: Macmillan College Publishing. Ragins, B. R. & Scandura, T. A. (1995). Antecedents and work-related correlates of reported sexual harassment: An empirical investigation of competing hypothesis. Sex Roles. 32. 429-455. Raksiltham, U. (1984). Police DiscretignagyJudgment on The Use of Force: An Empirical 254 Study of the Bangkok Metromlitan Police, Thailand. Unpublished Doctoral Dissertation. Claremont Graduate School. Rospenda, K. M. , Richman, J. A. & Nawyne, S. J. (1998). Doing Power: The Confluence of Gender, Race, and Class in Contrapower Sexual Harassment. Gender & Society. 12. 1. Feb, 40-60. Rotter, J. B. (1990). Internal versus external control of reinforcement: A case history of a variable. American Psychologist. 45. 4, 489-493. Rowe, M. P. (1996). Dealing with Harassment: A systems Approach. In M. S. (Ed) , Sexual harassment in the Workplace Perspectives, Frontiers and Re§ponse Slategies (V. 5 I. (pp. 241-271). Thousand Oaks: Sage. Rubin, L ( 1992). Sexgl Harassment: Individrfl Differences in Rgporting Behaviors. Unpublished Doctoral Dissertation. University of Kansas. Salancik, G. R. (1977). Commitment and the control of organizational behavior and belief. In B. M. Staw & G. R. Salancik (Eds.), New directions in organizational behavior. (pp.1-54). Chicago: St. Clair Press. Salisbury, J ., Ginorio, A. B., Remick, H. & Stringer, D. M. (1986). Counseling victims of sexual harassment. Psychotherapy, 23, 316-323. Savestarran, N. (1991). Police calet attitudes toward police corruption in Thailand. Unpublished Doctoral Dissertation. Florida State University. Schneider, K. T. & Swan, S. (1994). Job-related, psychological and health-related outcomes of sexual harassment. fiper presented aghe Ninth Annual Conference of the Socieg for Industrial and Organizational Psychology, Nashville, TN. Schulz, D. M. (1995). From social worker to crimefigpter: Women in United States Municipal Policing. Westport, CT: Greenwood. Segrave, K. (1995) Police Women: A History. Jefferson, NC: McFarland. Shaver, K. (1970). Defensive attribution: Effects of severity and relevance on the responsibility assigned for an accident. Journal of Perso_nality anad Socia_l_ Psychology. 14. 101-113. Shelton, N. (1999). Black and White College Women’s Perceptions of Sexual Harassment. Sex roles. 404 7-/8, 593-615. Singelis, T. M. (1994). The measurement of independent and interdependent self- construals. Personality and Social Psychology Bulletin, 20, 580-591. 255 Singelis, T. M. , Bond, M., H., Sharkey, W. F., Lai, C. S. Y. (1999). Unpackaging culture’s influence on self-esteem and embarrassability: The role of self- construals. Journal of Cross Cultural Psychology. 30. 3. 315-341. Skrocki, M. R. (1978). Sexual pressures on the iob. McCalls, March, 1978. Smith, M. D., & Self, G. D. (1981). Feminists and traditionalists: An attitudinal comparison. Sex Roles. 7, 2, 183-188. Smith, P. A. , & Midlarsky, E. (1985). Empirically derived conceptions of femaleness and maleness: A current View. Sex roles, 12 ,3-4L 313-328. Somers, P. A. & Clementson-Mohr, J. (1979). Sexual extortion in the workplace. Personnel Administrator 23-28. Sorod, B. (1991). The influence of national and organizational cultures on managerial values, attitudes, and performance. Dissertation Abstacts International, 52, 4B- 2338. (IS No. 0419-4217). Sperry, L. B. & Powell, G. N. (1999). Observers’ reactions to social-sexual behavior at work: An ethical decision making perspective. Journal of Management, 25, 6, 779-802. Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions, measurement and validation. Academy of Management Journal, 38, 5, 1442-1465 Spreitzer, G. M. (1996). Social Structural Characteristics of Psychological empowerment. Academy of Management Journal. 39. 2. 483-504 Stockdale, M. S. (1996). What we Know and Need to Learn. In M. S. (Ed.) , Sexual harassment in the Workplace Perspectives, Frontiers and Resporise Strategies (V. 5, pp. 3-25.). Thousand Oaks: Sage. Sullivan, M. & Bybee, D. I. (1987). Female students and sexual harassment: What factors predict decisions and thoughts about reporting. J ournai of the National Aasociation for Women Dean Administrators & Couraelers. 50, 11-16. Summers, R. J. (1991). Determinants of Judgments of and Responses to a Complaint of Sexual Harassment. Sex Roles, 25, 7-8, 379-392. Swanson, C. R., Territo, L. & Taylor, R. W. (1993). Police Administration: Structures, Processes and Behavior ,(3rd ed.). Englewood Cliffs, NJ: Prentice-Hall. Tabachnick B. G., & Fidel], L. S. (1989). Using Multivariate Statistics. New York: HarperCollins. 256 Tangri, S. S. & Hayes, S. M. (1997). Theories of Sexual Harassment. In O’Donohue, W. (Ed.), Sexual Harassment: Theory. Research and TreagentNeedham Heights, Lpp.112-128). Allyn & Bacon. Tangri, S. S., Burt, M. R. and Johnson, L. B. (1982). Sexual Harassment at Work: Three Explanatory Models. Journal of Social Issues. 38. 4, 33-54). Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological perspective on mental health. Psychological Bulletin, 103, 193-210. Thacker, R. A. (1996). A Descriptive Study of Situational and Individual Influences upon Individual’s Responses to Sexual harassment. Human Relations 49 8 1 105-1122 The Commission on Women’s Affairs and Women’s Role Development . (1981). Current llaai Women’s Stag. The National Economic and Social Development Broad of Thailand. The Royal Thai Police Department, (2000) Annual Report on police performance, 1999 :Bangkok: The Royal Thai Police Printing office. Thailand. The Thai labour protection Act (1998). National Executive Council No 103. Thibaut, J. & Walker, L. (1975). Procedural Justice: A Psychological Analysis. Hissdale, N. J. Lawrence Earlbaum. Thongpao, r. (1998). In the name of human digriig. [Online] Available http://scoop.bangkokpost.co.th/bangkokpostnews/bpl 9980809/090898perspective 06.html, February, 12, 2001. Till, F. (1980). Sexual harassment: A report on the sexuaflarassment of smdeami Washington, DC.: National Advisory Council on Women’s Educational Programs. Tsui A. S., Egan, T. D., & O'Reilly III, C. A. (1992). Being different: Relational demography and organizational attachment. Administrative Science Quarterly, 31, 549-579. U.S. Merit Systems Protection Broad. (1988). Sexual harassment in the FederaL Government : An upd_a_ta Washington, DC: Office of Merit Systems Reviews and Studies. Government Printing Office. U.S. U.S. Merit Systems Protection Broad. (1981). Sergil harassment in the F edapaL Workplace: Is it aproblem? Washington, DC: Office of Merit Systems Reviews and Studies. Government Printing Office. Van Wormer, K. S. & Bartollas, C. (2000). Women and the Criminal Justice System. 257 Allyn and Bacon: Boston. Vaux, A. (1993). Paradigrnatic assumptions in sexual harassment research: Being guided without being misled. J ourna of Vocational Behavior. 42. 116-135. Victor, B., Trevino, L. K., & Shapiro, D. L. (1993). Peer reporting of unethical behavior: The influence of justice evaluations and social context factors. Journal of business ethics 12 253-263. Vroom, V. (1964). Work and motivation. New York: John Wiley. Wallace, S. H. & Brislin, R. W. (1973). Evaluating language translations: Experiments on three assessment methods. Journal of Applied Psychology, 57. 3, 328-334. Wasigasin, W. & Hemaprasit, S. (1996). Values and Attitudes of the Thai Society that Affect the Family Institute and the Society. Thammasat University: Bangkok Thailand. Watsi, A. S., Bergman, M. E., Glomb, T. M. & Drasgow, F. (2000). Test of the Cross- Cultural Generalizability of a Model of Sexual Harassment. Journal of Applied Psychology. 85. 5. 766-778. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 72, 548-573. Welsh, T. M. (1997) V_ariables influencing the reporting of sexual harassment. Unpublished Doctoral Dissertation. Indiana State University. Wexler, J. G. & Logan, D. D. (1983). Sources of Stress among Women Police Officers. Jouraal of Police Science and Administration 11 46-53. White, J. K., & Ruh, R. A. (1973). Effects of personal values on the relationship between participation and job attitudes. Administrative Science Quarterly, 18, _4 506-514. Wickliffe, V. P. (1998). A cross-cultural analysis of the relationship between decision making styles, con_sumer demographics, and product characteristics. Unpublished Doctoral Dissertation. Michigan State University. Williams, C. W., Brown, R. S., Lees-Harley, P. R. & Price R. J. (1995). An Attributional (Causal Dimensional) Analysis of Perceptions of Sexual Harassment. J om] of Applied Social Psychology. 25. 13. 1169-1183. Witt, A. L. & Myers, J. G. (1992). Perceived environment uncertainty and participation in decision making in the prediction of perceptions of the fairness of personnel decisions. Review of Public Personnel Administration, 12, 3, 49-56. 258 Witt, L A. Andrews, M. C. & Kackman, K. M. (2000). The role of participation in decision-making in the organizational politics-job satisfaction relationship. Human Relations. 53.3. 341-358. Wong, M. M. (1984). Sexual Harassment at Work: Female Police Officers. Unpublished Dissertation. University of Pittsburgh. Wong, S. S., Heiby, E. M., Karneoka, V. A., Dubanoski, J. P. (1999). Perceived control, self-reinforcement and depression among Asian American and Caucasian American Elders. Journal of Applied Gerontology. 18. 1. 46-62. Worchel, S. & Brehm, J. W. (1971). Direct and implied social restoration of freedom. Journal of Persanality an_d Social PSYChOlOELIS. 3. 294-304. Worell, J. & Remer, P. (1992). Feminist perapectives in therapy: An empowerment model for women. New York: Wiley. Wright, D. W. & Young, R. (1998). The effects of family structure and maternal employment on the development of gender-related attitudes among men and women. Journal of Family Issues. 19. 3. 300-314. Yoder, J. D. & Aniakudo, P. (1996). When pranks become harassment: The case of Afiican American Women Firefighters. Sex roles. 35. 5/6, 253-270 Zlotnick, C. L. (1994). The incidence of universitxasexual harassment complaints related to institutional factors. Unpublished Doctoral Dissertation: University of Rhode Island. 259