<%BANNER%>

Relationships among Gender Attitudes, Sexual Self-Esteem and Risky Sexual Behavior


PAGE 1

RELATIONSHIPS AMONG GENDER ATTITUDES, SEXUAL SELF-ESTEEM, AND RISKY SEXUAL BEHAVIOR By AUTUMN LAMPINEN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN EDUCATION UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Autumn Lampinen

PAGE 3

ACKNOWLEDGMENTS I thank Dr. Sondra Smith-Adcock for being my thesis chairperson and helping me through the thesis development process. I thank Dr. Cirecie West-Olatunji for being a member of my defense board. I thank Dave Marshall for getting me through the statistics. Finally, I thank my family for their support and encouragement; I could not have done this without them. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi ABSTRACT......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 Statement of the Problem..............................................................................................5 Need for the Study........................................................................................................6 Purpose of the Study.....................................................................................................8 Definitions....................................................................................................................9 2 LITERATURE REVIEW...........................................................................................11 (Sexual) Self-Esteem and Sexual Behavior................................................................12 Gender Differences and Sexual Behavior...................................................................17 Gender and [Sexual] Self-Esteem...............................................................................23 Summary and Critique of the Literature.....................................................................24 3 METHODOLOGY.....................................................................................................26 Participants.................................................................................................................26 Instruments.................................................................................................................26 Procedures...................................................................................................................33 4 RESULTS...................................................................................................................34 Data Management.......................................................................................................34 Multiple Linear Regression Analysis.........................................................................35 Full Model Regression................................................................................................36 Forward Stepwise Regression.....................................................................................37 Gender Specific Models.............................................................................................38 Additional Findings....................................................................................................40 5 DISCUSSION.............................................................................................................42 iv

PAGE 5

APPENDIX : DEFINITIONS OF INDEPENDENT VARIABLES.................................48 LIST OF REFERENCE.....................................................................................................50 BIOGRAPHICAL SKETCH.............................................................................................53 v

PAGE 6

LIST OF TABLES Table page 1 Gender means for risky sexual behavior (RSB).......................................................34 2 Slopes of independent variables for full models:.....................................................36 3 Summary of final solution (4 Steps)........................................................................38 4 Summary of final solution female stepwise (5 Steps)...........................................39 5 Summary of final solution male stepwise (2 Steps)..............................................40 6 Means and t-test scores of subscales by gender.......................................................41 vi

PAGE 7

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Master of Arts in Education RELATIONSHIPS AMONG GENDER ATTITUDES, SEXUAL SELF-ESTEEM, AND RISKY SEXUAL BEHAVIOR By Autumn Lampinen December 2005 Chair: Sondra Smith-Adcock Major Department: Counselor Education Many young adults are engaging in risky sexual behaviors that create negative consequences. Little research has been done regarding the reasons for engagement in such behaviors. Previous research shows a difference between men and womens sexual behavior. Gender, and an individuals attitudes about gender, may influence behavior within the context of an environment. It is possible that this influence extends to individual engagement in risky sexual behavior. Sexual self-esteem, as a part of the construction of global self-esteem, may also have an impact in what types of sexual behavior one engages in. The purpose of this study is to identify variables that possibly affect a persons likelihood of engaging in risky sexual behaviors. In particular, this study examines the possible relationship among gender, gender attitudes, sexual self-esteem, and risky sexual behavior. The study also examines the variance explained by gender attitudes and sexual self-esteem as they impact risky sexual behavior. The data for the current study came from a pre-existing database. The data include information vii

PAGE 8

from 175 participants from undergraduate courses at a large southern university that were given four questionnaires to assess sexual history, sexual risk-taking, gender attitudes, and sexual self-esteem. The questionnaires used included a demographic information questionnaire, the Sexual History Questionnaire, the Gender Attitude Inventory, and the Multidimensional Sexual Self-Concept Questionnaire. Five items from the SHQ made up the dependent variable of risky sexual behavior. The thirteen independent variables included gender and six subscales from both the GAI and the MSSCQ. Findings showed that while an overall multiple regression was statistically significant in predicting risky sexual behavior, the regression indicated the only independent variable that was significant in predicting risky sexual behavior was gender. That is, females are slightly less likely to engage in risky sexual behavior. Gender attitudes failed to show significance in predicting or explaining risky sexual behavior. However, sexual self-esteem had stronger findings, although only for males. Specifically, only internal sexual control was found to have statistical significance in predicting risky sexual behavior in males. The research findings imply that more research regarding risky sexual behavior would be beneficial. viii

PAGE 9

CHAPTER 1 INTRODUCTION In modern society, religion and politics place great emphasis on the impact sexuality and sexual activity have on people. For instance, about 20,000 members of conservative Christian churches in Washington State attended a Mayday for Marriage rally in opposition to homosexual marriage (Norris, 2004). In a State of the Union Address, President George Bush proposed to double the money spent on abstinence-only sex education for 2005, despite much research in favor of comprehensive sex education conducted by associations such as the Sexuality Information and Education Council of the United States [SIECUS] (Bowman, 2004). Decades ago, Alfred Kinsey created a research institute based on advancing the knowledge of human sexuality (Kinsey, 2005). However, as evidenced by these recent events, though advances and changes have been made, this knowledge has yet to create significant change in how society reacts to peoples sexuality and policies regarding sexual issues. With this renewed interest in traditional views of marriage and sexuality, many people recognize that our society also continues to place double standards on people regarding sexual behavior based on their gender. The Feminist Movement started out of womens realization of inequality based on gender and dissatisfaction with their lack of life choices (Beck, 1998). The double standard was no longer acceptable. The feminist ideology that both created and continues the feminist movement is very complex, more so than many individuals realize. One study suggested the general definition of what a typical feminist ideology and self-labeled feminist is: a woman that holds strong radical, 1

PAGE 10

2 socialist, and cultural feminist beliefs (Liss, Hoffner, & Crawford, 2000). One of these typical feminist groups is liberal feminists, which holds a basic belief that women and men are virtually the same and should be able to make decisions free from the governmental and social constraints that are based on gender. Liss et al. (2000) found that the majority of women in their study supported liberal feminist goals and beliefs, yet a majority of the participants (63%) in the study did not self-identify or label themselves as a feminist. Despite the Womens Movement and other advances toward equality that our culture has made, traditional gender scripts are being passed on to younger generations, such as the feeling that a wifes place is still in the home and it is the husbands responsibility to provide for his family (Vanwesenbeeck, Bekker, & van Lenning, 2001). In fact, it seems a negative stigma is still attached to the term feminist, which creates hesitance in women, even those who endorse pro-feminist goals and beliefs, to label themselves as such (Liss et al., 2000; Beck, 1998). The media, being one of the most influential entities in how culture and normalcy is defined, has oft promoted the idea of feminists as crazy Amazon lesbian man-haters (Beck, 1998). Given the current wave of anti-gay politics (Norris, 2004), the stigma of having a feminist label is often exacerbated by the assumption that one must me a lesbian to be a feminist (Beck, 1998). Beck (1998) goes on to note how media in Western culture shapes all things into dichotomies and opposition, meaning things are in black and white, good and bad terms with no room for potential grey. With this general media portrayal, masculine is opposite to feminine, and Western culture categorizes good as male/white/middle class/Christian. Feminism thus goes against the foundation on which the Western patriarchal society stands (Beck,

PAGE 11

3 1998). Third-wave feminism, the name for the current feminist era, tries to negotiate these dualistic presentations of being both feminist and feminine, and particularly discusses the importance of sexuality in this negotiation (Kinser, 2004). One final important note about traditional stereotypes is that it still remains a prominent way of how education regarding sexuality is presented. In particular, Vanwesenbeeck et al. (2001) acknowledge that the continuing trend of defining sexual behaviors in masculine terms helps maintain male power and control in heterosexual encounters. Additionally, traditional gender roles and assumptions in our society finds more casual and less cautious sexual behavior less acceptable when practiced by women, yet sexual behavior in general acceptable for men (Seal & Agostinelli, 1996). As a result of this double standard, women are more likely than men to experience negative consequences from risky sexual behavior such as social rejection due to unwanted pregnancies and STDs (Seal & Agostinelli, 1996; Milhausen & Herold, 1999). In addition to attitudes regarding gender roles, people have attitudes about ones own sexuality and the sexuality of others. Sexual self-esteem is one part of our self-concept that makes up the global definition of self-esteem (Gaynor & Underwood, 1995). How one feels about his or herself as a capable sexual being affects our mental wellbeing and how we deal with our relationships in sexual situations (Rosenthal, Moore, & Flynn, 1991). In fact, one study found high levels of [global] self-esteem are indicative of the likelihood that one will approach sexual behaviors and their consequences more carefully and with more forethought (Holmbeck, Crossman, Wandrei, & Gasiewski, 1994). However, little research seems to have taken up an interest in individual feelings of self-worth in the context of sexual self-esteem and how it impacts sexual behavior,

PAGE 12

4 rather than in terms of global self-esteem. Of the studies that consider self-esteem, most find a relationship between sexual or global self-esteem and risky sexual behavior. Some research shows that boys with higher self-esteem had higher amounts of sexual activity, whereas girls with higher self-esteem had lower amounts of sexual activity (Spencer, Zimet, Aalsma, & Orr, 2002). One possible reason for this discrepancy is that sexual self-esteem is formed in different ways based on ones gender (Breakwell & Millward, 1997). Research has also noticed different attitudes toward risky sexual behavior based on gender and gender attitudes. Young males have a tendency to be less concerned and less cautious regarding the consequences of risky sexual, believing that contraceptives and pregnancy concerns are the females responsibility, or problem (Lenskyj, 1990). Females appear to have a better grasp on some of the realistic problems that can accompany risky sexual behavior, particularly seeing unwanted pregnancy as detrimental to future goal attainment (De Gaston & Weed, 1996, p. 217). Yet, many females still participate in risky sexual behavior. Vanwesenbeeck et al. (1998) are among the researchers that show gender socialization in our masculine culture and phallocentric definitions of sexuality promote feelings of powerlessness in women, endorsing traditional gender roles. Researchers could continue to benefit from learning what about our gender socialization, and thus our gender attitudes, is making safe sexual behavior less consistent and less appealing to our young adults. What implications can be made based on social and cultural biases related to risky sexual behavior and their effect on the wellbeing and sexual behavior of young adults? If we uphold more traditional gender and life assumptions in our minds, how does this same mindset effect our beliefs and values when young adults are in the bedroom? In this

PAGE 13

5 study, some factors proposed to influence young peoples risky sexual choices are [sexual] self-esteem, and gender attitudes. Statement of the Problem What are the risks or negative outcomes when participating in risky sexual behavior? Pregnancy and sexually transmitted diseases [STDs] are two serious consequences of risky sexual behavior (Lenskyj, 1990). Approximately 34% of girls will become pregnant at least once by age 20, meaning there are approximately 820,000 teen pregnancies every year (National Campaign to Prevent Teen Pregnancy, 2004). One study noted only 57.9% of high school students reported using a condom during their last heterosexual intercourse encounter with females reporting less use (Halpern-Felsher, Kropp, Boyer, Tschann, & Ellen, 2004). The study did not include condom-use information regarding anal or oral sex. Statistics from the American Social Health Association [ASHA] (2005) report there are approximately 65 million people in the United States with a viral STD, and approximately 15 million new cases of STDs every year. There are an estimated 1,039,000 to 1,185,000 individuals in the United States living with HIV or AIDS with about 40,000 individuals being infected with HIV each year (Centers for Disease Control and Prevention [CDC], 2005). The CDC estimates that more than 50% of HIV/AIDS cases in the United States are acquired before the age of 25, and generally due to sexual intercourse (Florida Department of Health, 2004). The Department of Health (2004) also reports that two-thirds of the 12 million annual cases of new STD infections are in young people under the age of 25. Approximately 24-27% of the estimated population with HIV/AIDS is undiagnosed and unaware that they are even infected. Research also shows that a majority of college students rate other students chances of contracting the human

PAGE 14

6 immunodeficiency virus or other sexually transmitted diseases [HIV/STDs] higher than their own and do not identify themselves as a type of person who is at risk for HIV/STDs despite their own risky sexual behavior practices (Seal & Agostinelli, 1996). Are there gender differences when it comes to risky sexual behavior? If so, what possible social regulations are imposed on each gender? According to Vanwesenbeeck et al. (1998), All sexual behavior is mediated, facilitated, and/or justified by cognitions and beliefs, (p. 326). Gender attitudes are cognitions and beliefs about gender within a cultural context. There are clear gender double standards in assumptions about acceptable sexual behavior and contraceptive responsibility (Seal & Agostinelli, 1996). Sexually permissive women tend to be ostracized and stigmatized whereas men who engage in the same sexual behaviors are lauded as admirable and popular (Milhausen & Herold, 1999). Birth control, whether via oral contraceptives or having and using condoms tends to be primarily placed on the female, yet open-discussion or acknowledgement of preparedness for sexual behavior results in members of society calling these women derogatory names, such as sluts (Lenskyj, 1990). This indicates that there are definite societal impacts on human sexuality, gender socialization, and the differences in the expression of sexuality between the genders. What are the consequences of maintaining traditional gender roles and attitudes in relation to risky sexual behavior? How do these traditional attitudes impact our emotional and mental health, particularly sexual self-esteem? Need for the Study The dominant culture of the United States presents with discomfort on taboo topics such as sexual behavior, sex education, and premarital sex (Gudelunas, 2005). The factors that contribute to risky sexual behavior are unclear. Most studies focus on

PAGE 15

7 knowledge of risk in sexual activities, but fail to find a large impact on why young adults engage in risky sexual behavior (Hollar & Snizek, 1996). There exists a need to better understand the correlates of risky sexual behavior in order to make changes in social programming and socialization, to educate the population, and prevent negative outcomes due to risky sexual behavior. The current research will attempt to look at factors that potentially encourage risky sexual behavior. Furthermore, young adults are participating in premarital sexual behaviors despite abstinence education (Bowman, 2004; Starkman & Rajani, 2002; Caldas, 1994). Abstinence-only education is one outcome of societal denial regarding premarital sexual behavior (Starkman & Rajani, 2002). One thing that comprehensive sex education teaches is how to practice safe sex. It is suggested that positive attitudes toward using condoms increases the likelihood of using condoms (Halpern-Felsher et al., 2004). This could help prevent negative outcomes of risky sexual behavior, as well as reduce the likelihood of individuals engaging in risky sexual behavior. Without getting the knowledge to practice safe sex, how can we expect young adults and adolescents to reduce the risk involved in their sexual behavior? In addition to the criticism that [abstinence-only] sex education does not change or reduce adolescent or young adult sexual behavior, there are concerns about failing to bring emotional and moral issues of sexuality and sexual decision-making into the sex education process (Lenskyj, 1990). In other words, the positive aspects of sexuality, and how to build healthy sexual attitudes, are virtually ignored in sex education courses, as well as the providing of information about homosexuality and issues regarding gender identity (Lenskyj, 1990). When discussing teenage pregnancy and the prospects of

PAGE 16

8 unmarried parenthood as a result of combining negative sexual attitudes and low sexual self-esteem with risky sexual behavior, there are hidden costs that many people fail to consider. Only one-third of teenage mothers graduate from high school and 80% of unwed teen mothers are on welfare (Rinaldo, 2002). Many unwed teen mothers feel a loss of freedom (Rinaldo, 2002). Also, depression in young unmarried mothers is not uncommon (Lesser, Koniak-Griffin, & Anderson, 1999). The current study can help assess the potential costs of low sexual self-esteem, which includes negative attitudes about sexuality, and young adult risky sexual behavior, including unwanted pregnancies, STDs, depression, lowered self-esteem, and unreached future goals. The current study can also allow researchers and sex educators to consider more alternatives for the prevention of the negative consequences of risky sexual behavior. Purpose of the Study The purpose of this study is to determine if there are relationships between gender, gender attitudes, sexual self-esteem, and risky sexual behavior and to better define the nature of those relationships. This study will view the importance of gender when discussing the sensitive topic of sexual behavior. It is necessary to discover how attitudes about gender and the boundaries a culture imposes on each gender effects their risky sexual behavior. There has been a minimal amount of research that looks at peoples perceptions and beliefs about gender and sexuality and how these factors influence what one does or does not do with a sexual partner. This study will assess whether or not these gender attitudes are related to how one behaves sexually. It will also assess if gender attitudes affect sexual self-esteem. Another purpose this study is to determine how sexual self-esteem affects ones sexual behavior or the level of risk one

PAGE 17

9 takes in sexual activity. This study is concerned with how gender attitudes relate to individual sexual self-esteem and how this affects risky sexual behavior. The research questions addressed in this study are as follows: 1. What are the relationships between gender, gender attitudes, sexual self-esteem, and risky sexual behavior? 2. Does sexual self-esteem explain variance in risky sexual behavior above and beyond that explained by gender attitudes? 3. Which gender attitudes and sexual self-esteem factors explain the most variance in risky sexual behavior? Definitions Following are some definitions of factors for the purposes of this study. Self-esteem can be defined as the, self-evaluation of ones worth as a person (New Websters Dictionary and Thesaurus, 1993). Using Rogers concept of self, Gaynor and Underwood (1995) define sexual self-esteem as the tendency to value, versus devalue, ones own sexuality, thereby being able to approach rather than avoid sexual experiences both with self and others (p. 334). According to Andersen and Cyranowski, sexual self-esteem is defined as Sexual aspects of oneself that are derived from past experience, manifest in current experience, influential in the processing of sexually relevant social information, and guide sexual behavior (Mayers, Heller, & Heller, 2003, p. 270). The term gender attitudes refers to the beliefs one holds about others based on their gender. This includes such ideas as endorsement of traditional stereotypes, endorsement of female superiority, disapproval of female casual sex or sexual initiative, and acceptance of male heterosexual violence (Ashmore, Del Boca, & Bilder, 1995). Risky sexual behavior refers to sexual activity that places one at risk for physically negative outcomes such as unwanted and/or teen pregnancy, and the contraction of an STD or HIV/AIDS

PAGE 18

10 (Gil, 2005; Nzioka, 2004). Examples of risky sexual activities include multiple partners, one-time partners (a.k.a. one-night stands), and sex without use of a condom (Gil, 2005; Rosenthal, Moore, & Flynn, 1991).

PAGE 19

CHAPTER 2 LITERATURE REVIEW Self-esteem and individual feelings of self-worth have a way of determining how we behave in many avenues of our lives. For as many people that exist in the world, there are just as many ways our self-esteem manifests itself in our daily routines. Many people seek therapy to improve their level of self-esteem. Some people lack feelings of self-worth, and thus self-esteem, when it comes to being in a classroom, others when it comes to getting a deserved promotion, and yet still others lack feelings of self-worth when it comes to making friends. Young adults sexual activities that they engage in are not immune to individual feelings of self-esteem and worth. Sexuality is just one facet of our self-esteem, but it is important to an individuals overall well-being. Of the studies that have been done in the area of sexual self-esteem and sexual behavior, most of the research involves high school age adolescents, and secondly, middle school age adolescents. Gender differences have been found in how levels of self-esteem affect risky sexual behavior (Spencer, Zimet, Aalsma, & Orr, 2002). There have also been findings that low self-esteem, both general and sexual self-esteem specific, affects sexual behavior (Young, Denny, & Spear, 1999; Hollar & Snizek, 1996). Gender differences in risky sexual behavior have also been studied. However, studies regarding gender attitudes, or beliefs about traditional gender roles, and how they relate to risky sexual behavior, have not been researched as much. One finding from the research is that males have a tendency to believe in traditional gender roles more often than women do (Vanwesenbeeck, Bekker, & van Lenning, 1998). Males also tend to be 11

PAGE 20

12 more sexually active (DeGaston & Weed, 1996), have more permissive attitudes toward sexual behaviors (1996), and be less concerned with the outcomes of risky sexual behavior (Breakwell & Millward, 1997). This chapter reviews more current literature about research that shows a relationship, or lack thereof, between sexual self-esteem, gender attitudes, and risky sexual behavior. The following literature review is segmented into relationships between two of each of these three variables. First, you will find a subsection on the relationship between sexual self-esteem and risky sexual behaviors. Second, there is a section regarding gender differences and risky sexual behavior. Third, you will find only one piece of literature discussing research about the relationship between sexual self-esteem and gender attitudes. This is due to a lack of research about that particular relationship in virtually any form. Finally, studies that research the relationship between all three variables of interest are virtually nonexistent. A subsection containing all three variables is thus absent from this chapter. (Sexual) Self-Esteem and Sexual Behavior Spencer, Zimet, Aalsma, and Orr (2002) conducted a study about adolescent gender differences in self-esteem and its potential impact on sexual behavior. In their longitudinal study, they used a sample of 188 adolescents, testing them first in seventh grade and later in ninth grade. Forty-three percent of the 188 participants were boys and fifty-seven percent were girls. All of the 188 participants were virgins during the seventh grade testing period. The researchers had the students complete a questionnaire that included the Rosenberg Self-Esteem Scale (RSES) and a single question concerning coital status, whether or not the participant had ever had sexual intercourse. The focus of the study is relevant to the current study and research in that it looked at self-esteem

PAGE 21

13 based on gender and how self-esteem based on gender affects the change in coital status from virginal to non-virginal. An analysis of covariance was used with gender and coital status effects on self-esteem, in conjunction with pubertal timing. Regarding the interests of the current research, the study showed a significant gender by coital status interaction for self-esteem results indicated that boys with high self-esteem at Time 1 [seventh grade] were 2.4 times more likely to initiate intercourse than boys with low self-esteem at Time 1. Conversely, girls with high self-esteem at Time 1 were three times more likely to remain virgins than those girls with low self-esteem. (p. 582) This study also examined if the change in intercourse status had an effect on self-esteem, taking gender into consideration. The study found that self-esteem remained stable. The study suggested some reasons for the difference in male versus female self-esteem in regards to sexual behavior, such as differing societal influences based on gender, and attempts at improving emotional needs. This study reinforced the need for more research regarding self-esteem, gender, and sexual behavior. Another study by Young, Denny, and Spear (1999) examined the relationship between adolescent self-esteem and sexual behavior of 1,659 junior and senior high school students ranging from seventh to twelfth grade. This study asserted that self-esteem is divided into three parts: peer self-esteem, family self-esteem, and school self-esteem. Peer self-esteem is based on relationships with friends and perceived acceptance or lack thereof. Family self-esteem refers to relationships with family members and perceived confidence from parents. School self-esteem is related to academic performance and perceived teacher response to students. The researchers used the Hare Self-Esteem Scale to conduct their study, which breaks down questions of self-esteem,

PAGE 22

14 sexual behavior, and sexual knowledge into the three parts of self esteem (peer, family, and school). The study found that peer self-esteem was positively correlated with both past and future sexual behavior. On the other hand, there was a negative correlation between sexual behavior and both family and school self-esteem, meaning the higher home and school self-esteem was in the participants, the less likely they were to engage in sexual behavior. This study provided information on individual factors that affect self-esteem and thus sexual behavior. Although it is important to realize there are multiple things that factor into our self-esteem development, the study came across a problem when combining all three types of self-esteem into an integrated model of overall self-esteem. It did not discuss the likelihood of sexual behavior given different combinations of the three areas of self-esteem. There was no information regarding the breakdown or possible differences based on gender, which is important to the current research. Breakwell and Millward (1997) also conducted a study on older adolescent sexual self-concept and sexual risk-taking. More specifically, the study examined the relationship between sexual self-concept and sexual activities based on gender. There were 474 participants ranging in age from 16 to 19, with the majority being female. The study was conducted via mailing questionnaires to participants. These questionnaires contained the Sexual Self-Concept Checklist, which has questions with choices of characteristics that participants felt described their sexual self-concept. They also contained questions on whether they felt they held traditional male/female roles in sexual relationships, the importance placed on sex and sexual attractiveness, and questions about their sexual activity.

PAGE 23

15 Breakwell and Millwards (1997) study suggested that sexual self-concept is formed in different ways based on gender and thus sexual risk-taking is expressed in different ways. For instance, responsibility for contraceptive use was found to be relatively irrelevant to males, but it was a central part of a sexual self-concept in females. A seemingly opposite finding was that there were no significant correlations between sexual self-concept and traditional male/female roles in sexual relationships. However, the study did not explain this finding or how the researchers came to this conclusion. They also did not define what they were seeking out when looking at traditional gender roles in sexual relationships. A study by Rosenthal, Moore, and Flynn (1991) investigates the sexual self-efficacy and sexual self-esteem of males and females and the relation between sexual risk-taking and these self-perceptions (p. 77). This study evaluated 1,008 predominantly female post-secondary students ranging in age from 17 to 20. All respondents had engaged in sexual activity, meaning they had engaged in oral, anal, or vaginal sex, or withdrawal. Participants filled out questionnaires on sexual self-efficacy, sexual self-esteem, and sexual behavior and risk. Sexual self-efficacy measured participant confidence and willingness in doing multiple kinds of sexual activities. Sexual self-esteem measured individuals perceptions and attitudes regarding their sexual adequacy and relationships. Sexual behavior and risk looked at whether sexual activities were engaged in with a casual versus regular partner and whether or not condoms were used in each scenario. Risk in sexual situations was evaluated based on partner type. One important aspect touched on in Rosenthal et al.s (1991) study is that they considered gender differences in sexual self-efficacy. In general, males are more

PAGE 24

16 confident in their condom use, masturbation, seeking out potential partners (dates), watching pornographic movies without embarrassment, and being able to get sexual needs met. On the other hand, women were more confident in saying no to sexual activity. The rest of the studys findings were made in relation to partner type, casual or regular. Overall, the study found there was little variance in risky behavior based on sexual self-efficacy and self-esteem measures. In regards to casual partners, the study found that the self-efficacy factor of the ability to say no was the only factor that correlated with lower levels of sexual risk-taking. When discussing regular partners, there were several factors increasing sexual risk-taking. Among these factors were the ability to say no and confidence in sexual need assertion. Regular partners were found to increase participant sexual self-esteem. The significance of the correlation of these findings was greater among the female participants than the males. Hollar and Snizek (1996) conducted a study regarding the connections between levels of self-esteem, sexual behavior, and knowledge of HIV/AIDS. The study used 353 university students as participants, consisting of 49.7% women and 50.3% men. The majority of students were in their first or second year of college. Instruments used in the study included a 14-item True/False HIV questionnaire with 2 additional HIV/AIDS multiple choice questions, the Rosenberg Self-Esteem Scale (RSES), and a multiple choice questionnaire regarding forms of sexual behavior they have participated in. The study focused on two different categories of risky sexual behavior which the researchers called conventional and non-conventional. Conventional risky sexual behavior included unprotected heterosexual sex, having multiple sex partners or having sex with promiscuous sex partners. Non-conventional risky sexual behavior included unprotected

PAGE 25

17 anal intercourse, syringe sharing or sex with someone who shares syringes, having sex with a prostitute, or having sex with someone who is knowingly HIV positive. Hollar and Snizeks (1996) study found that individuals of both genders with high self-esteem and more knowledge about HIV/AIDS were more likely to engage in conventional risky sexual behavior than others with lower self-esteem and less knowledge. However, the second most likely group to engage in conventional risky sexual behavior were the individuals classified as having low self-esteem and low levels of HIV/AIDS knowledge. As for non-conventional risky sexual behavior, the students with low self-esteem and low levels of HIV/AIDS knowledge were the most likely to engage in this type of behavior. In regards to their study, Hollar and Snizek (1996) expressed concern as to whether the cause of participants risky sexual behavior was due to high self-esteem or vice versa and they suggested a longitudinal study could help answer whether the behavior or the high self-esteem came first. This particular study pointed out that there were no gender differences in risky sexual behavior that pertains to HIV/AIDS. This goes against other research studies that have found gender differences in self-esteem and sexual behavior. More research on this topic can give clarity regarding this discrepancy. Gender Differences and Sexual Behavior DeGaston and Weed (1996) conducted research to better understand gender differences in the sexuality of adolescents. The research was done on the premise that emotional differences exist between the sexes regarding sexuality as well as distinct gender differences in sexual attitudes, values, motivation, expression, and reasons for avoiding sex (p. 218). The study enlisted more than 1,800 seventh and eight grade students from non-urban schools along the Pacific coast who were entering sex education

PAGE 26

18 courses. A list of 14 questions about sexuality was given to the participants. DeGaston and Weed (1996) predicted there would be gender differences in the level of permissiveness of sexual attitudes and activities as well as negativity toward outcomes of sexual behaviors. DeGaston and Weeds (1996) findings showed males were twice as sexually active as females. The study indicated females were more committed to abstinence and had less permissive attitudes regarding premarital sex. This commitment to abstinence included female beliefs that birth control and being in love are not reasons to justify having sex, despite the finding that females generally frame love and sex together more than males. The findings also noted that females saw adolescent sexual activity as an impediment to the attainment of future goals. In regards to family and social parameters of adolescent sexuality, females reported discussing sex with parents more often yet also found parents less approving of sexual behavior, to the point of placing more restrictions on females than on males. Among peers, females reported receiving less pressure to be sexually active. An additional interesting finding of the study was a higher likelihood of adolescent males believing sexual urges could not be controlled. Seal and Agostelli (1996) conducted a study about perceptions of risky sexual behavior prevalence among peers of college students. The study enlisted 217 participants, 96 of whom were men and 121 were women. The subjects ranged in age from 18 to 25 years old. All participants completed a questionnaire that assessed their individual sexual behaviors and safe sex practices. Upon completion, they were asked to give separate estimates, or percentages, of the prevalence of male and female peers sexual behaviors that they just rated themselves on. Specifically, there were five risky

PAGE 27

19 sexual behaviors that the study assessed: having at least 2 sex partners in the past year, having unprotected sex with anyone in the past year, having more than one sex partner in their life, having had any casual sexual encounters, and having casual sex without a condom. The study found a consistent overestimation of the prevalence of peer sexual behaviors among all participants, especially among those who engaged in the particular estimated behavior. Gender differences were found in both participant overestimation and the assumed gender or the peers that engaged in sexual behavior. Specifically, female participants made higher estimates than the male participants, and both male and female participants estimated sexual behaviors as being more frequent among male peers. Seal and Agostelli (1996) went on to discuss possible causes for gender differences both in the participants overestimations as well as the assumed peer sexual behaviors. Shearer, Hosterman, Gillen, and Lefkowitz (2005) conducted a study regarding the possible associations between traditional gender role attitudes and risky sexual behavior. Condom-related behaviors were a particular focus for risky sexual behaviors. Shearer et al. (2005) collected data from 154 participants between the ages of 18 and 25, of which 96 were women and 58 were men. Participants were asked to complete multiple scales in addition to answering questions about their sexual history. Sexual history questions were most important to the study in regards to condom use, engagement in casual sex, and participant relationship status. Instruments used in the study included the Condom Use Self-Efficacy Scale, the Outcome Expectancies for Condom Use (OECU), the Attitudes Toward Family Roles Scale (ATFRS) which looked at traditional marital

PAGE 28

20 and childrearing beliefs, and the Male Role Norms Scale (MRNS) which looked at traditional gender role beliefs in terms of anti-femininity and social status achievement. Shearer et al.s (2005) study found a moderate level of risky sexual behavior among the participants. Despite, overall positive beliefs about condom use, 82% of the participants reported not using a condom during sexual intercourse on at least one occasion. The study differentiated norms for different gender role beliefs, particularly male role norms and marital and childrearing role norms. Overall, women were found to have less traditional gender role beliefs than men. Women with more traditional beliefs regarding marital roles were more likely to believe there were greater barriers to (reasons not to) use a condom and believed sex would be less enjoyable when using a condom. Males with traditional beliefs about childrearing were more likely to believe they would enjoy sex less if a condom was used. Both male and female individuals with more traditional views of male role norms were more likely to use condoms. However, men who asserted anti-feminine beliefs about male roles were less likely to use condoms during intercourse and believe that sex without condoms would be less enjoyable. Also, males who attempted to achieve high social status were more likely to use condoms but were more likely to have casual sex as well. Lucke (1998) studied gender roles and sexual behavior of young women. In particular the study assesses the relationship between gender roles and unsafe sexual practices. These three sexual behaviors that may be considered unsafe are: having multiple partners, not using condoms, and using drugs or alcohol before or during sexual intercourse. Gender roles had three dimensions for the purposes of the study: personality traits, attitudes, and dating behavior. Women who were more egalitarian in their gender

PAGE 29

21 attitudes were said to have more masculine personality traits, which indicated the likelihood to engage in more risky sexual behavior. There were 400 participants from the Queensland, Australia area. Their ages ranged from 16 to 24 years old. Instruments used included the Personal Description Questionnaire (PDQ) to examine gender role personality traits, the Women in Society Questionnaire (WSQ) to assess gender role attitudes, and the DeLucia Dating Behavior Indices (DDBI) to assess gender role dating behavior. The participants were also assessed regarding their sexual behavior. Luckes (1998) findings differentiated between having masculine personality traits versus having masculine dating behaviors. The study showed the more that a woman endorsed masculine dating behaviors, the more she would also endorse feminine dating behaviors, meaning traditional gender roles. Women with egalitarian attitudes were more likely to have masculine traits. The study reported that these women with more masculine traits generally had more than one sexual partner were more liberal in beliefs about womens roles in society, and had less feminine dating styles. The study noted that the earlier a woman engaged in sexual intercourse, the less likely she was to use condoms later. An interesting finding the study made was that older participants were less likely to use condoms than the younger participants. Finally, alcohol and drug use was associated most with single women who had masculine personality traits, more egalitarian attitudes toward women, and non-steady or multiple sexual partners. Vanwesenbeeck, Bekker, and van Lenning (1998) looked at the connection between interactional competence in sexual encounters and the meaning of sexuality and attitudes regarding gender and gender relations. Interactional competence is the set of behaviors and attitudes one engages in to settle on an outcome of a sexual encounter with

PAGE 30

22 ones partner. This study used 386 university graduate students in the Netherlands who had experience with heterosexual intercourse. The participants consisted of 275 females and 111 males. The demographic characteristics between the male and female participants were similar allowing for the comparison to be possible. The questionnaire consisted of interactional behavior, sexual meaning, and gender relation attitude questions. The 55 interactional behavior questions assessed emotions and behaviors during heterosexual experiences. The 31 sexual meaning questions asked about sensation-seeking sexual behaviors, sexual compulsion, anxiety, and beliefs about the level of connection between love and sex. Attitudes about gender were measured with 41 questions regarding gender conservatism, acceptance of interpersonal violence and rape myths, and adverse beliefs and hostility toward each gender. Vanwesenbeeck et al.s (1998) study also found gender differences in interactional behavior and competence. In particular, females were found to be more defensively controlling of sexual situations and more anxious in terms of a lack of control of the situation. Males were found to be more pro-active in their controlling what they want and get in a sexual situation. In regards to sexual meaning, males scored higher for being sexually compulsive and sensation-seeking. Gender differences were also found in attitudes about sexuality. Male students were significantly more traditional in considering gender roles, were more accepting of rape myths, and engaged in more hostile feelings and beliefs toward either or both genders. Some of the variables were connected in the questionnaire based on gender as well. For women, the more the individual adhered to traditional gender roles, the less likely she was to be defensive in interactional behaviors. Women were also likely to be more hostile toward either gender

PAGE 31

23 if they had a higher degree of sexual anxiety. For men, the more hostile beliefs they engaged in, the more likely they were to be pro-actively controlling in sexual situations. Also, men with more traditional gender attitudes correlated with higher acceptance of rape myths and pro-active control. Gender and [Sexual] Self-Esteem Burnett, Anderson, and Heppner (1995) conducted a study that looked at environmental effects on self-esteem and gender roles. The study was not directly assessing sexual self-esteem. There were 236 participants in the study, 90 male and 146 female. All participants were in their first two years of college at a large Midwest university. The instruments used in the study were the Personal Attributes Questionnaire (PAQ), the Personal Attributes Questionnaire-Environmental form (PAQ-env), and the Coopersmith Self-Esteem Inventory (SEI). The PAQ assessed sex-role orientation, meaning whether a person identified themselves with characteristics of masculinity or femininity. The PAQ-env assessed the participants perceptions of environmental cues that encouraged masculinity or femininity. The SEI measured individual feelings regarding self-esteem. Burnett et al. (1995) found a significant correlation between having masculine traits and high self-esteem for both genders, but an insignificant correlation for feminine traits and self-esteem. The study also reported that the participants felt more encouragement to have masculine traits than feminine ones. This finding was significant for both genders. Finally, the study found that the individuals with the lowest levels of masculinity also had the lowest levels of self-esteem, which was particularly significant in women.

PAGE 32

24 Summary and Critique of the Literature There are some patterns that are emerging from the research. Studies that have looked at relationships between sexual self-esteem, risky sexual behavior, and gender differences are generally done with adolescent and young adult populations. Gender attitudes, the beliefs one holds about others based on their gender, including such ideas as endorsement of traditional stereotypes, endorsement of female superiority, disapproval of female casual sex or sexual initiative, and acceptance of male heterosexual violence (Ashmore, Del Boca, & Bilder, 1995), tend to be studied in the scope of traditional versus non-traditional, or individual feminine versus masculine, traits (Lucke, 1998; Burnett, Anderson, & Heppner, 1995). There are consistent findings that gender differences exist in sexual behaviors and levels of permissiveness for risky sexual behaviors. There are also consistent findings that many males and individuals having masculine traits of either gender are associated with higher levels of [risky] sexual behavior (Rosenthal et al., 1991; DeGaston & Weed, 1996; Lucke, 1998; Vanwesenbeeck et al., 1998). Traditional gender role beliefs are also being indicated as exacerbating levels of risky sexual behavior (Shearer et al., 2005). However, there are also inconsistent findings in this research area, such as how [sexual] self-esteem affects sexual behavior based on gender. Zimet et al. (2002) noted that high levels of self-esteem for boys, yet low levels of self-esteem for girls, increased amounts of risky sexual behaviors. On the other hand, Rosenthal et al. (1991) found little variance in sexual behavior when looking at sexual self-esteem. Yet another study (Young et al., 1999) noted that different types of self-esteem produced different likelihoods of engaging in sexual behavior. Perhaps the differences among the studies

PAGE 33

25 are based on whether they narrowed the scope of self-esteem to sexual self-esteem or other aspects of self-esteem, or studied self-esteem in a general sense. There are several areas of possible future topics to research in the area of sexual self-esteem, sexual behavior, and gender attitudes or gender roles. Studies that assess the direction of the relationships between these variables may increase our knowledge about sexual risk and self-esteem levels and allow us to come up with ways to effectively deal with these types of issues. Studies that are interested in socio-cultural impacts on these variables and how they are correlated are also an area that would benefit the field of counseling. Another area of interest may be how these variables interact to affect adults, especially older populations, such as individuals considered to be middle-aged or in old age. Only one-third of the studies which had self-esteem as a variable that were reviewed discussed sexual self-esteem rather than more general types of self-esteem (Breakwell & Millward, 1997; Rosenthal et al., 1991). One study did discuss that there are different elements to self-esteem (Young et al., 1999), but sexual self-esteem was not studied as one of the elements. It could be beneficial to the field of counseling and research to conduct more studies about the different facets of self-esteem and about the conceptualization of sexual self-esteem.

PAGE 34

CHAPTER 3 METHODOLOGY Participants Initially, there were 175 participants in the study (98 females and 77 males) receiving education from a large southeastern public university. However, due to the rate of attrition, the final number of response sets used in the study was 126 (76 females and 50 males). All students were at the undergraduate level and were currently enrolled in either Stress Management or Interpersonal Communications Skills courses. The mean age was 21.3 years old, with ages ranging from 18 to 37 years. The majority of the participants were unmarried. The mean age for participants losing their virginity was 16.8 years with a standard deviation of 2.9 years. All participation was voluntary and they could leave the study at any time they requested without concern of penalty. Participants were notified of the potentially sensitive and personal material that was going to be assessed. Procedures for informed consent were followed, and participants confidentiality was also protected. Instruments Demographic information. Participants initially filled out a questionnaire regarding demographic data. Demographic data collected for this study included participant age, gender, and marital status. Sexual History Questionnaire (SHQ; Cupitt, 1998). The dependent variable in this study is risky sexual behavior. To define risky sexual behavior, we took 5 items from Cupitts (1998) Sexual History Questionnaire. This questionnaire is a self-report 26

PAGE 35

27 measure that was created to assess the level of risk of contracting HIV based on an individuals sexual behavior. The questions on the Sexual History Questionnaire are an array of multiple choice, yes/no, and Likert-type 5-point scales. The questionnaire should only take approximately 5 to 10 minutes to complete. These 5 questions sought information regarding how many sexual partners the participant had in the last month, whether the last sexual encounter was unprotected vaginal or anal intercourse with a regular sexual partner, how able the participant felt in expressing wants and needs regarding sex, and self-perception of risk from HIV/AIDS. This questionnaire is not scored in the traditional sense, but does provide information about a participants risk for the contraction of HIV. The scores, upon being coded for the purposes of this study, of these five items were added together in such a way that higher scores indicated a higher level of risky behavior. The minimum possible score is 2. A range of scores is technically not possible due to the open-ended number of sexual partners had in the last month. In this particular sample, the possible range is 2 to 34. The questionnaire begins with collecting demographic data, and then continues with questions placed into four different sections. First in the SHQ (1998), participants are asked if they have sex with men or women in order to distinguish between sexual identity and sexual behavior. Follow-up questions ask whether or not participants have ever had penetrative sex, and if so, whether or not a condom was used. Second, the questionnaire asked about a participants sexual activity in the last month. In particular, it asks how many sexual partners they have had in the past month, how many of those were regular vs. casual partners, how many sexual encounters were had with each partner, and how many of the occasions included the use of a condom. Third, the

PAGE 36

28 questions are related to the participants last sexual encounter. Questions related to how long ago the last sexual encounter was, partners gender, discussion about condom use and other concerns about sex, contraception, and what type of intercourse was engaged in. Types of intercourse included vaginal, anal, oral, or masturbatory and whether or not a condom was used. Finally, personal perceptions of risk and knowledge of HIV was assessed (e.g. How much at risk do you consider yourself from HIV/AIDS?). Cupitt (1998) advises the information the Sexual History Questionnaire seeks to collect is not stable, since the sexual activity of all individuals changes over time. The third section, regarding ones last sexual encounter was added in an attempt to strengthen the questionnaires reliability. However, this assessment has a high level of reliability (correlation < .80, with p < .001) in regards to the fourth section that asks about perceptions of risk and knowledge about HIV. A list of definitions of the sexual behaviors was included to ensure participants understood the questions in similar ways, creating a high level of face validity (1998). Gender Attitude Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995). There are 13 independent variables, one of which is gender. Six of the remaining twelve independent variables come from Ashmore et al.s (1995) Gender Attitude Inventory (GAI). This inventory is structured to assess topical attitudes regarding gender. The inventory is specifically created for the assessment of college students in the United States. The Gender Attitude Inventory consists of 109 questions, broken into 14 different gender attitude scales that are content-specific, or 3 second-order factor scales, which assess patterns of social behaviors based on gender. The questions were either framed in an attitudinally traditional (42 items) or modern (67 items) style. All responses to the

PAGE 37

29 questions are formed in Likert-type 7-point scores, ranging from strongly disagree to strongly agree. A higher score on any scale indicates a more traditional attitude towards gender. This questionnaire can be scored utilizing the 14 separate scales or the 3 second-order factor scales. The Gender Attitude Inventory (1995) has 14 different scales in assessing specific gendered behaviors and attitudes. The 109 items of this inventory were placed under one of these scales. The 14 scales and example items of each follow: Endorsement of Female Superiority (e.g. Compared to women, men have many undesirable qualities), Acceptance of Traditional Stereotypes (e.g. Men are more independent than women), Condemnation of Homosexuality (e.g. Male homosexuality is merely a different kind of lifestyle which should not be condemned), Disapproval of Female Sexual Initiative (e.g. The man should always be the one to initiate sex with a woman), Disapproval of Female Casual Sex (e.g. It is acceptable for a woman to have sex with a casual acquaintance), Endorsement of Chivalry (e.g. Chivalry is generally demeaning to women), Acceptance of Male Heterosexual Violence (e.g. Women provoke rape by their appearance or behavior), Endorsement of Family Roles (e.g. Care of children should be shared equally by both spouses), Belief in Differential Work Roles (e.g. All occupations should be equally accessible to both men and women), Rejection of Female Political Leadership (e.g. I would vote for a politically qualified woman to be President of the U.S.), Opposition to Womens Rights (e.g. Passage of legislation to further womens rights is necessary), Endorsement of Individual Action (e.g. Most women have only themselves to blame for not doing better in life), Opposition to Funded Day Care (e.g. Government-sponsored child care centers would be a big waste of money),

PAGE 38

30 and Opposition to Abortion (Abortion should be treated legally as a form of murder). Also, the information from 67 items gathered from this inventory allow us to view 3 second-order factors that assist in predicting patterns of social behaviors based on gender attitudes. These second-order factors are: Societal Organization which includes gender and societal role issues, Sexual Relationships which considers sexual behavior acceptance, and Stereotypes which evaluates the overall endorsement of traditional gender roles. The six subscales that were used in this study from the GAI are: acceptance of traditional stereotypes (ATS), disapproval of female sexual initiative (FSI), disapproval of female casual sex (FCS), endorsement of family roles (EFR), belief in differential work roles (DWR), and opposition to womens rights (OWR). The remaining seven GAI subscales did not have questions specifically relevant to the interests of the study and were thus removed from analysis. The scores for each item in a given subscale were summed up to create a total subscale score. The subscale Acceptance of Traditional Stereotypes consisted of 10 items with a score range of 10 to 70. Disapproval of Female Sexual Initiative consisted of 11 items with a score range of 11 to 77. Disapproval of Female Casual Sex consisted of 3 items with a score range of 3 to 21. Endorsement of Family Roles consisted of 11 items with a score range of 11 to 77. Belief in Differential Work Roles consisted of 9 items with a score range of 9 to 63. Finally, Opposition to Womens Rights consisted of 10 items with a score range of 10 to 70. Once each subscale has a score, all of the subscale scores are added up to create a single measure of the respondents gender attitudes. After several items were reverse coded, higher scores on each of the subscales indicates a more traditional gender attitude.

PAGE 39

31 Ashmore et al. (1995) found strong reliability in both internal consistency and temporal stability for all 14 scales and the 3 second-order factor scales. The Alphas and the test-retest correlations for all three second-order factors were higher than those for the 14 primary scales. In regards to internal consistency, the mean of the Alpha levels for males was .83, which was a little higher than the .79 mean Alpha levels for females. When assessing the temporal stability of the GAI, the correlations for the test-retest mean of males was .81 and .78 for females. The GAI is a valid instrument and attained statistical significance for 9 of the scales that men scored as holding more traditional attitudes than women. The GAI was compared to other standardized measures of sex-role ideology. In particular, when compared to the Attitudes Toward Women Scale (AWS), there were statistically significant correlations for 10 of the 14 scales for men, and 12 of the 14 scales for women. The Multidimensional Sexual Self-Concept Questionnaire (MSSCQ; Snell, Jr., 2001). The remaining 6 independent variables come from subscales from the MSSCQ. This questionnaire is a self-report measure that was created to assess 20 aspects of the self-concept of human sexuality. There are 20 subscales, each having 5 questions, alternating in ascending order throughout the questionnaire for a total of 100 items. Responses to all of the sexual self-esteem items were in Likert-type format ranging from 1 to 5, where 1 represents not at all characteristic of me and 5 represents very characteristic of me. The MSSCQ (2001) provides 20 different scores for each of the 20 subscales. The scores for each item in a given subscale were summed up to create a total subscale score. The range of scores per subscale is 5 to 25. Once each subscale has a score, all of the subscale scores are added up to create a single measure of the

PAGE 40

32 respondents sexual self-esteem. The range of the sum of subscale scores is 30 to 150. Several items were reverse coded in the MSSCQ, and a higher score on each of the subscales indicates a higher tendency of correspondence to the aspect of sexual self-concept that the subscale measures. The questionnaire should take approximately 45 minutes to one hour to complete. The 20 subscales of the Snells (2001) questionnaire are as follows: sexual anxiety, sexual self-efficacy, sexual consciousness, motivation to avoid risky sex, chance/luck sexual control, sexual preoccupation, sexual assertiveness, sexual optimism, sexual problem self-blame, sexual monitoring, sexual motivation, sexual problem management, sexual esteem, sexual satisfaction, power-other sexual control, sexual self-schemata, fear of sex, sexual problem prevention, sexual depression, and internal sexual control. Some example items include I am very aware of my sexual feelings and needs, I am satisfied with the status of my own sexual fulfillment, I have a fear of sexual relationships, My sexual behavior is mostly determined by people who have influence and control over me, and The sexual aspects of my life are a matter of fate (destiny). However, only six MSSCQ subscales were utilized as independent variables in the current study are: sexual anxiety, sexual self-efficacy, sexual esteem, sexual satisfaction, power-other sexual control, and internal sexual control. The remaining 14 subscales of the MSSCQ were not used due to their lack of fit in relation to the focus of the current study. Snell (2001) found high internal consistency for all 20 of the subscales, with Cronbach alphas ranging from .72 to .94 for all participants. Validity was established through recent research regarding the MSSCQ. The subscales were able to predict the

PAGE 41

33 likelihood of university-age male and female contraceptive behavior based on different scores of different subscales. Procedures Prior to the distribution of the assessments, participants signed informed consent forms and given a copy of the signed form. Participants were also given assurance of their confidentiality of participation in the study. Participants were also informed of the sensitive and personal nature of the assessment questions. Subjects were asked to complete four questionnaires. Participants first filled out the demographic information questionnaire. Second, they filled out Cupitts (1998) Sexual History Questionnaire. Third, the participants filled out Ashmore et al.s (1995) Gender Attitude Inventory. Finally, the participants filled out Snells (2001) Multidimensional Sexual Self-Concept Questionnaire. All questionnaires were filled out in their entirety, despite the current studys use of only certain subscales from each measurement. The questionnaires took approximately two hours to complete.

PAGE 42

CHAPTER 4 RESULTS The purpose of the study was to assess if the extent of risky sexual behavior in young adults is predicted by gender, gender attitudes, and sexual self-esteem. Gender attitudes and sexual self-esteem were each defined by 6 subscales. Data Management Initially, the sample contained 175 participants. However, several participants were eliminated due to extensive missing data in the response sets. Participants were eliminated from the sample if any of the subscales contained no data. Of the respondents who may have missed an item or two within a subscale, missing data was replaced by using the mean of the items responded to within that subscale. This method has the advantage of producing more accurate estimated means for each of the subscales. However, the tradeoff of using mean replacement is a reduced overall variance. The final number of response sets used in the study was 126. There were 76 females and 50 males included in the response set used for the analysis. The gender means for risky sexual behavior (RSB) are provided in the following: Table 1: Gender means for risky sexual behavior (RSB). Gender N Mean Std. Dev. -95% CI +95% CI Male 50 6.20 3.31 5.25 7.14 Female 76 4.19 1.54 3.84 4.55 34

PAGE 43

35 Multiple Linear Regression Analysis To determine which of the 13 independent variables were significant in predicting risky sexual behavior, a multiple linear regression equation was calculated. One categorical variable, gender (GEN), was used. The remaining variables were continuous. The six subscales that were used from the Gender Attitude Inventory are: acceptance of traditional stereotypes (ATS), disapproval of female sexual initiative (FSI), disapproval of female casual sex (FCS), endorsement of family roles (EFR), belief in differential work roles (DWR), and opposition to womens rights (OWR). The six MSSCQ subscales utilized in the current study are: sexual anxiety (SA), sexual self-efficacy (SSE), sexual esteem (SE), sexual satisfaction (SS), power-other sexual control (PSC), and internal sexual control (ISC). The resulting linear regression equation is expressed as: y = a + 1(GEN) + 2(ATS) + 3(FSI) + 4(FCS) + 5(EFR) + 6(DWR) + 7OWR)8(SA) + 9SSE) + 10(SE) + 11(SS) + 12(PSC) + 13(ISC) + e Where: y = predicted value of Total Score a = y intercept GEN = gender (M=0, F=1) 1 = slope of gender ATS = acceptance of traditional stereotypes 2 = slope of acceptance of traditional stereotypes FSI = disapproval of female sexual initiative 3 = slope of disapproval of female sexual initiative FCS = disapproval of female casual sex 4 = slope of disapproval of female casual sex EFR = endorsement of family roles 5 = slope of endorsement of family roles DWR = belief in differential work roles 6 = slope of belief in differential work roles OWR = opposition to womens rights 7 = slope of opposition to womens rights SA = sexual anxiety 8 = slope of sexual anxiety SSE = sexual self-efficacy

PAGE 44

36 9 = slope of sexual self-efficacy SE = sexual esteem 10 = slope of sexual esteem SS = sexual satisfaction 11 = slope of sexual satisfaction PSC = power-other sexual control 12 = slope of power-other sexual control ISC = internal sexual control 13 = slope of internal sexual control e = residual error term Full Model Regression Risky sexual behaviors were analyzed using a standard linear regression model. The full model which includes all variables was shown to be significant: F(13,112) = 3.01, p < .0008. Overall, the model produced an R2 = .259, accounting for 25.9% of the overall variance. Although the overall model was significant in predicting risky sexual behavior (RSB), the only independent variable that proved to be significant was gender (GEN). A summary of the results for the full model is provided in Table 2. Table 2: Slopes of independent variables for full models: Variable Value Std Err t(112) p -value a 10.38 2.53 4.09 < .00001* 1 -1.40 0.56 -2.52 0.013* 2 0.01 0.03 -0.56 0.575 3 0.02 0.03 0.65 0.520 4 0.04 0.06 0.70 0.484 5 0.03 0.03 -1.28 0.203 6 -0.02 0.03 -0.52 0.605 7 0.02 0.03 0.69 0.491 8 -0.04 0.07 -0.58 0.561 9 0.08 0.09 0.95 0.343

PAGE 45

37 Table 2: Continued Variable Value Std Err t(112) p -value 10 -0.07 0.09 -0.80 0.42 11 -0.07 0.08 -0.98 0.331 12 -0.01 0.07 -0.14 0.888 13 -0.13 0.09 -1.45 0.15 Where: indicates significance The resulting linear equation from using all the variables in the full model: RSB = 10.38 1.40(GEN) + 0.01(ATS) + 0.02(FSI) + 0.04(FCS) + 0.03(EFR) 0.02(DWR) + 0.02OWR)(SA) + 0.08SSE) 0.07(SE) 0.07(SS) 0.01(PSC) 0.13(ISC) + e Forward Stepwise Regression A forward stepwise regression method was used to determine if a reduced model would reveal any significant independent variables which would be useful in predicting RSB. A forward stepwise regression technique adds one variable at a time to the model and determines which individual variable explains the most variance. Subsequent variables are added one at a time in an order which accounts for the most variance in the model. Final solutions for a forward stepwise regression method usually have less than the full number of original variables selected. For the final solution using the forward stepwise technique, the model was significant F(4,121) = 8.92, p < 0.0001, and produced a Multiple R2 value of 0.227, accounting for 22.7% of the overall variance. The only variable which was significant in this model was gender (GEN).

PAGE 46

38 Table 3: Summary of final solution (4 Steps) Variable R2 increase Value Std Err t(121) p -value Step 0 Intercept 10.57 1.28 8.25 <.0001* Step 1 Gender 0.144 -1.57 0.46 -3.45 <.0007* Step 2 Internal Sexual Control 0.056 -0.07 0.07 -1.03 0.30 Step 3 Endorsement of Family Roles 0.013 -0.03 0.02 -1.56 0.12 Step 4 Sexual Esteem 0.014 -0.10 0.07 -1.47 0.14 Where: indicates significance A practical interpretation of the results suggests that female respondents scored an average of 1.57 points lower than male respondents on the score of risky sexual behavior (RSB) scale. While internal sexual control (ISC), endorsement of family roles (EFR), and sexual esteem (SE) did help to explain more of the total overall variance and would help produce a more accurate estimate of expected risky sexual behavior (RSB) scores, they were not statistically significant. Gender Specific Models Since no variables other than gender were found to be significant in either of the models tested, separate analyses were conducted for both males (N = 50) and females (N = 76) to determine if any of the remaining independent variables were significant in predicting risky sexual behavior (RSB) in regard to gender. Risky sexual behavior (RSB) for females was analyzed using a standard linear regression model. The full model, including all variables was not found to be significant: F(12,63) = 1.1137, p < .365. Overall, the model produced an R2 = .175, accounting for 17.5% of the overall variance. Alternatively, a forward stepwise regression was found to

PAGE 47

39 be significant: F(5,70) = 2.55, p < .035. The forward stepwise regression model produced an R2 = .154, accounting for 15.4% of the overall variance in the reduced model. Of the variables included in this model, only the intercept was found to be significant. Table 4: Summary of final solution female stepwise (5 Steps) Variable R2 Increase Value Std Err t(70) p -value Step 0 Intercept 5.16 1.46 3.53 <.0001* Step 1 Sexual Anxiety 0.056 0.07 0.05 1.58 0.11 Step 2 Accept Traditional Stereotypes 0.034 -0.02 0.02 -1.10 0.27 Step 3 Sexual Esteem 0.014 -0.06 0.04 -1.51 0.13 Step 4 Disapprove Female Sexual Initiative 0.013 0.03 0.02 1.80 0.07 Step 5 Endorsement of Family Roles 0.034 -0.03 0.02 -1.69 0.09 Where: indicates significance A practical interpretation of the results suggests that for female respondents, the predicted value of risky sexual behavior (RSB) is 5.16. While sexual anxiety (SA), acceptance of traditional stereotypes (ATS), sexual esteem (SE), disapproval of female sexual initiative (FSI), and endorsement of family roles (EFR) did help to explain more of the total overall variance and would help produce a more accurate estimate of expected risky sexual behavior (RSB) scores, they were not statistically significant. Risky sexual behavior (RSB) for males was analyzed using a standard linear regression model. The full model, including all variables was not found to be significant:

PAGE 48

40 F(12,37) = 1.05, p < .425. Overall, the model produced an R2 = .254, accounting for 25.4% of the overall variance. Alternatively, a forward stepwise regression was found to be significant: F(2,47) = 5.44, p < .007. The forward stepwise regression model produced an R2 = .188, accounting for 18.8% of the overall variance in the reduced model. Of the variables included in this model, the intercept and internal sexual control (ISC) were found to be significant. Table 5: Summary of final solution male stepwise (2 Steps) Variable R2 increase Value Std Err t(47) p -value Step 0 Intercept 9.56 2.70 3.54 <.0001* Step 1 Internal Sexual Control 0.156 -0.35 0.11 -3.18 <.002* Step 2 Disapprove of Female Casual Sex 0.031 0.20 0.15 1.36 0.18 Where: indicates significance A practical interpretation of the results suggests that for male respondents internal sexual control (ISC) is significant in helping to predict risky sexual behavior (RSB). While disapproval of female casual sex (FCS) did help to explain more of the total overall variance and would help produce a more accurate estimate of expected risky sexual behavior (RSB) scores, this was not statistically significant. Additional Findings Although the multivariate analyses performed did not reveal a significant explanation for the reasons behind the gender effects using the factors examined in this study, an alternative and less powerful way of looking at the data is to view the univariate means for each subscale based upon gender. Performing a series of t-tests on the means

PAGE 49

41 of each subscale may begin to give an idea of where to focus future research into this area: Table 6: Means and t-test scores of subscales by gender Variable Male (N = 50) Female (N = 76) t(124) p -value Accept Traditional Stereotypes 33.51 39.18 3.14 .002* Disapprove Female Sexual Initiative 60.27 59.99 -0.14 .890 Disapprove of Female Casual Sex 14.49 13.25 -1.77 .791 Endorsement of Family Roles 48.10 57.29 4.50 <.001* Belief in Differential Work Roles 36.04 44.42 5.67 <.001* Oppose Womens Rights 40.88 47.97 5.27 <.001* Sexual Anxiety 12.61 10.91 -2.35 .202* Sexual Self Efficacy 17.54 17.27 -0.36 .722 Sexual Esteem 16.73 17.55 1.09 .278 Sexual Satisfaction 16.83 17.56 0.90 .371 Power-other Sexual Control 12.38 9.60 -3.68 <.001* Internal Sexual Control 17.75 18.78 1.46 .148 Where: indicates significance

PAGE 50

CHAPTER 5 DISCUSSION Although the overall multiple regression model was statistically significant in predicting risky sexual behavior, the most significant finding is that there are gender differences in risky sexual behavior as it is defined in the study. Specifically, females are slightly less likely to engage in risky sexual behavior. Other than gender, no other independent variables were considered to be statistically significant. Since the strongest test, a full multiple regression, only found gender to be statistically significant in helping to predict risky sexual behavior, a reduced model multiple regression was used to assess whether any of the independent variables would have statistical significance in predicting risky sexual behavior based on gender. This reduced model found that internal sexual control (ISC) is the only independent variable that is significant in predicting risky sexual behavior for males. The reduced model for females found no independent variables to be significant. The finding that there are gender differences, particularly that women are slightly less likely to engage in risky sexual behavior, is supported by previous research. However, the studys lack of further significant findings is both surprising and disappointing. The study failed to indicate what other relationships existed among gender, gender attitudes, sexual self-esteem, and risky sexual behavior. In particular, the study was unable to isolate which of the independent variables that defined gender attitudes and sexual self esteem were significant in predicting risky sexual behavior. The independent variables that represented gender attitudes within the study were not found to 42

PAGE 51

43 be significant in any model used and only one of the six sexual self-esteem variables, internal sexual control, came up as significant for males only. Internal sexual control, defined as a belief that ones sexuality is within ones own control, may have come up as significant for predicting risky sexual behavior in males because of gender socialization and cultural reinforcement that men are able to maintain control in situations requiring interaction and negotiation with others. In the realm of sexuality, this supports findings in previous research that men are more confident and able to get sexual needs met. Internal sexual control in males may also lead to sexual risk-taking to get sexual needs met in a way that indicates assumptions of invincibility, which is more characteristic of younger individuals. Much of the previous research relating self-esteem to risky sexual behavior is in terms of global self-esteem, rather than a specific construct of sexual self-esteem, which may explain the lack of findings in the current study. Also, the questionnaire utilized to define sexual self-esteem, was meant to define sexual self-concept, with sexual esteem being only one factor of sexual self-concept. Despite the lack of significant findings in this study, the model itself was not problematic. The scales that the independent variables were taken from were reliable. The model did what it was supposed to, which was to assess the independent variables to see if they were significant in predicting risky sexual behavior. The results that were expected were simply not found, except for the fact that gender affected the likelihood of engagement in risky sexual behavior. However, the series of t-tests that were performed based on gender means showed some very interesting patterns. Womens t-test scores for several of the independent variables were contrary to what previous research suggests. The women in the current study were more likely to accept traditional stereotypes,

PAGE 52

44 endorse traditional family roles, believe in the differentiation of work roles, and oppose womens rights. In fact, the t-test scores for these four variables were found to have statistical significance. Thus, a practical interpretation of these results suggests that the women in this study were overall more traditional in their gender attitudes and beliefs. Given the findings of the t-tests, the lack of findings from the overall model used in the study and what previous research has found, it is important to look at the social and cultural underpinnings of the results. The study found less traditional gender attitudes among the men rather than the women. Previous research suggests that women generally have more liberal gender attitudes than men (Vanwesenbeeck et al., 1998; Shearer et al., 2005) even though they are less likely to engage in risky sexual behavior (DeGaston & Weed, 1996). In fact, previous research has not only shown that men, who purportedly have more traditional gender attitudes, engage in more risky sexual behavior, but individuals who claim to uphold traditional and/or conservative gender beliefs also engage in higher rates of risky sexual behavior (Shearer et al., 2005). This finding from previous research goes against many sociocultural assumptions that being liberal in gender attitudes and lifestyle choices is more indicative of risky sexual behavior. What implications can be made about this research? When considering the traditional nuclear family and who was given a voice in the family, traditional gender attitudes supported womens deference to their husbands, including the topic of safe sexual practices. Rosenthal et al. (1991) saw a correlating pattern in regular partners, where the inability to say no to sexual activities and lack of confidence in sexual need assertion increased the likelihood of risky sexual behavior. Womens belief systems have fluctuated over the years. In present day, womens roles, sociocultural beliefs, and gender attitudes matter

PAGE 53

45 and are relevant to the current findings. What once defined liberal feminism as radical beliefs is now seen as basic to the rights of women in present day (Liss et al., 2000). A possible, if not likely, interpretation of feminisms importance on gender attitudes and how it may affect sexuality and sexual behavior is that it both makes women more independent in their decisions regarding gender and sexuality yet may encourage a backlash to liberal feminist beliefs due to negative stigma endorsed by the media. There are several limitations to the current study. First, the respondents in the study were obtained from a college population. This may make it difficult to generalize to other populations. Second, it is possible that a sample bias exists in that the sample was taken from two different courses in which the students elected to take the course. The participants were not asked what their major was, which could create a bias and limit the generalizability of the findings. Third, the participants came from a convenient sample. The two subject classes that participants were recruited from were Stress Management and Interpersonal Communication Skills, which are both held in the College of Education. Fourth, other demographic information that could greatly impact study findings, particularly hometown or regional culture and religiosity, were not used in the study. Fifth, the four surveys that were utilized in the study were lengthy and took an average of two hours to complete. Due to the time necessary for a participant to invest in the study, there was a significant amount of data lost by participants not finishing the surveys. Sixth, in relation to lost data, mean replacement for missing data also had an impact on the findings. Specifically, mean replacement decreases the variance of the sample. Seventh, the number of surveys completed and utilized for this study may be too small for the number of variables examined. When the data was separated based on

PAGE 54

46 gender, the study was especially small. When the sample was broken down by gender, there were 50 males and 76 females. One rule of thumb for sample size is to have ten times the number of variables assessed in the study. Using this standard, given that there are 12 independent variables aside from gender, it would have been preferable to have a minimum of 120 males and 120 females. Eighth, the reliability and validity of the definition of risky sexual behavior for the purposes of this study are uncertain. Additionally, some items within the Sexual History Questionnaire (SHQ) that were used to define risky sexual behavior were open-ended and subject to multiple interpretations. Ninth, items within the subscales may have been interpreted differently per each respondent. Finally, the various subscales used to define gender attitudes and sexual self-esteem were ineffective in uncovering gender differences in risky sexual behavior. Additional valid measures may be needed for risky sexual behavior, as well as for the independent variables that purportedly measure gender attitudes and sexual self-esteem. Future research may be beneficial in several different areas. The t-tests on the gender means, while overall a weak test in this application, still indicate potentially significant differences on six of the subscales used which may be followed up on in later studies: acceptance of traditional stereotypes, endorsement of family roles, belief in differential work roles, opposition to womens rights, sexual anxiety, and power-other sexual control. Also, future studies may benefit from shorter surveys to decrease survey incompletion rates. If future studies regarding gender attitudes, sexual self-esteem, and risky sexual behavior are done again, it may be beneficial to find additional reliable and valid assessments for the particular constructs being assessed and/or increase the number of students surveyed.

PAGE 55

47 In conclusion, this study investigates the possible relationships between gender attitudes, sexual self-esteem, and risky sexual behavior and the potential differences of those relationships based on gender. Of the original 175 participants, only 126 participants data sets were included due to survey incompletion. Participants completed four questionnaires: a demographic information questionnaire, the Sexual History Questionnaire (SHQ; Cupitt, 1998), the Gender Attitude Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995), and the Multidimensional Sexual Self-Concept Questionnaire (MSSCQ; Snell, 2001). Findings showed that while an overall multiple regression was statistically significant in predicting risky sexual behavior, the regression indicated the only independent variable that was significant in predicting risky sexual behavior was gender. Additional findings are helpful in consideration of future directions of research.

PAGE 56

APPENDIX DEFINITIONS OF INDEPENDENT VARIABLES ATS: Acceptance of Traditional Stereotypes refers to the belief that men and women are different, such that women in some ways are considered inferior to men. An example of this is the idea of men being more independent than women, or that women are weak in comparison to men. DWR: belief in Differential Work Roles refers to the separation of men and women in careers based on gender due to the assumption that one gender is less capable of engaging in certain tasks, such as restricting a woman from getting a job that is considered too physically difficult or labeling an occupation as womens work or mens work. EFR: Endorsement of Family Roles refers to the belief that women are better equipped to stay at home to raise children than men are and that men should take the lead in family decision making. FCS: disapproval of Female Casual Sex refers to disapproving of a womans sexual relations outside of a [monogamous] loving relationship. FSI: disapproval of Female Sexual Initiative refers to the belief that it is not alright for a woman to take the first step in dating or engaging in sexual activities. ISC: Internal Sexual Control is the belief that the sexual aspects of ones life are determined by ones own personal control. OWR: Opposition to Womens Rights refers to belief that the womens liberation movement and equal rights laws are unnecessary. 48

PAGE 57

49 PSC: Power-other Sexual Control is the belief that the sexual aspects of ones life are controlled by others who are more powerful and influential than oneself. SA: Sexual Anxiety is the tendency to feel tension, discomfort, and anxiety about the sexual aspects of ones life. SE: Sexual Esteem is the general tendency to positively evaluate ones own capacity to engage in healthy sexual behaviors and to experience ones sexuality in a satisfying and enjoyable way. SS: Sexual Satisfaction is the tendency to be highly satisfied with the sexual aspects of ones life. SSE: Sexual Self-Efficacy is the belief that one has the ability to deal effectively with the sexual aspects of oneself.

PAGE 58

LIST OF REFERENCE American Social Health Association. (August 2005). Learn about STIs/STDs. Available from http://www.ashastd.org/learn/ learn_statistics.cfm Ashmore, R. D., Del Boca, F. K., & Bilder, S. M. (1995). Construction and validation of the Gender Attitude Inventory, a structured inventory to assess multiple dimensions of gender attitudes. Sex Roles, 32(11/12), 753-785. Beck, D. B. (1998). The f word: How the media frame feminism. NWSA Journal, 10(1), 139-153. Bowman, D. H. (2004). Abstinence-only debate heating up. Education Week, 23(22), 1-10. Breakwell, G. M., & Millward, L. J. (1997). Sexual self-concept and sexual risk-taking. Journal of Adolescence, 20, 29-41. Burnett, J. W., Anderson, W. P., & Heppner, P. P. (1995). Gender roles and self-esteem: A consideration of environmental factors. Journal of Counseling & Development, 73, 323-326. Caldas, S. J. (1994). Teen pregnancy. Phi Delta Kappan, 75(5), 402-406. Centers for Disease Control & Prevention. (August 2005). Divisions of HIV/AIDS Prevention Basic Statistics. Available from http://www.cdc.gov/hiv/stats.htm Cupitt, C. (1998). Sexual History Questionnaire. In C. M. Davis (Ed.), Handbook of sexuality-related measures (pp. 106-108). Thousand Oaks, CA: Sage Publications. DeGaston, J. F., & Weed, S. (1996). Understanding gender differences in adolescent sexuality. Adolescence, 31(121), 217-230. Florida Department of Health. (August 2005). HIV/AIDS 2004 Fact Sheets. Available from http://www.doh.state.fl.us/Disease_ctrl/aids/updates/facts/facts.html Gaynor, P. A., & Underwood, J. K. (1995). Conceptualizing and measuring sexual self-esteem. In P. E. Shrout, & S. T. Fiske (Eds.), Personality research, methods, & theory: A festschrift honoring Donald W. Fiske (pp. 333-347). Hillsdale, NJ: Erlbaum Associates. 50

PAGE 59

51 Gil, S. (2005). Personality traits and coping styles as mediators in risky sexual behavior: A comparison of male and female undergraduate students. Social Behavior and Personality, 33(2), 149-158. Gudelunas, D. (2005). Talking taboo: Newspaper advice columns and sexual discourse. Sexuality & Culture, 9(1), 62-87. Halpern-Felsher, B. L., Kropp, R. Y., Boyer, C. B., Tschann, J. M., & Ellen, J. M. (2004). Adolescents self-efficacy to communicate about sex: Its role in condom attitudes, commitment, and use. Adolescence, 39(155), 443-456. Hollar, D. S., & Snizek, W. E. (1996). The influences of knowledge of HIV/AIDS and self-esteem on the sexual practices of college students. Social Behavior and Personality, 24(1), 75-86. Holmbeck, G. N., Crossman, R. E., Wandrei, M. L., & Gasiewski, E. (1994). Cognitive development, egocentrism, self-esteem, and adolescent contraceptive knowledge, attitudes and behavior. Journal of Youth & Adolescence, 23(2), 169-194. Kinser, A. E. (2004). Negotiating spaces for/through third-wave feminism. NWSA Journal, 16(3), 124-153. Kinsey Institute, The. (August 2005). Mission Statement. Available from http://www.indiana.edu/~kinsey/ Lenskyj, H. (1990). Beyond plumbing and prevention: Feminist approaches to sex education. Gender & Education, 2(2), 217-231. Liss, M., Hoffner, C., & Crawford, M. (2000). What do feminists believe? Psychology of Women Quarterly, 24, 279-284. Lesser, J., Koniak-Griffin, D., & Anderson, N. L. R. (1999). Depressed adolescent mothers perceptions of their own maternal role. Issues in Mental Health Nursing, 20, 131-149. Lucke, J. C. (1998). Gender roles and sexual behavior among young women. Sex Roles, 39(3/4), 273-297. Mayers, K. S., Heller, D. K., & Heller, J. A. (2003). Damaged sexual self-esteem: A kind of disability. Sexuality and Disability, 21(4), 269-282. Milhausen, R. R., & Herold, E. S. (1999). Does the sexual double standard still exist? Perceptions of university women. The Journal of Sex Research, 36(4), 361-368.

PAGE 60

52 National Campaign to Prevent Teen Pregnancy. (August 2005). Teen pregnancy So what? Available from http://www.teenpregnancy.org/ whycare/sowhat.asp New Websters Dictionary and Thesaurus of the English Language. (1993). Danbury, CT: Lexicon Publications, Inc. Norris, M. (2004). A crowd of nearly 20,000 rally against marriage equality. Lesbian News, 29(11), 16. Nzioka, C. (2004). Unwanted pregnancy and sexual transmitted infection among young women in rural Kenya. Culture, Health, and Sexuality, 6(1), 31-34. Rinaldo, D. (2002). The tough life of a teen mom. Scholastic Choices, 17(5). Retrieved August 25, 2005, from Academic Search Premier database. Rosenthal, D., Moore, S., & Flynn, I. (1991). Adolescent self-efficacy, self-esteem and sexual risk-taking. Journal of Community & Applied Social Psychology, 1, 77-88. Seal, D. W., & Agostinelli, G. (1996). College students perceptions of the prevalence of risky sexual behavior. AIDS Care, 8(4), 453-467. Shearer, C.L., Hosterman, S. J., Gillen, M. M., & Lefkowitz, E. S. (2005). Are traditional gender role attitudes associated with risky sexual behavior and condom-related beliefs? Sex Roles, 52(5/6), 311-324. Snell, Jr., W. E. (1998). The Multidimensional Sexual Self-Concept Questionnaire. In C. M. Davis (Ed.), Handbook of sexuality-related measures (pp. 521-524). Thousand Oaks, CA: Sage Publications. Spencer, J. M., Zimet, G. D., Aalsma, M. C., & Orr, D. P. (2002). Self-esteem as a predictor of initiation of coitus in early adolescents. Pediatrics, 109(4), 581-584. Starkman, N., & Rajani, N. (2002). The case for comprehensive sex education. AIDS Patient Care and STDs, 16(7), 313-318. Vanwesenbeeck, I., Bekker, M., & van Lenning, A. (1998). Gender attitudes, sexual meanings, and interactional patterns in heterosexual encounters among college students in the Netherlands. The Journal of Sex Research, 35(4), 317-327. Young, M., Denny, G., & Spear, C. (1999). Area specific self-esteem and adolescent sexual behavior. American Journal of Health Studies, 15(4), 181-188.

PAGE 61

BIOGRAPHICAL SKETCH Autumn Lampinen graduated cum laude in 2002 from the University of Florida with a Bachelor of Science in psychology and a Bachelor of Arts in criminology. She is currently completing requirements for a Master of Arts in Education and Specialist in Education in mental health counseling at the University of Florida. Autumns research interests include sexuality, self-esteem, feminism, substance abuse, and criminal rehabilitation. She has been a member of Chi Sigma Iota, Beta Chapter, and the Counselor Education Student Association. She also holds membership in the American Counseling Association. Autumn is currently awaiting publication of an article in which she has secondary authorship, Teaching Family Systems Theory through Service-Learning. 53


Permanent Link: http://ufdc.ufl.edu/UFE0013339/00001

Material Information

Title: Relationships among Gender Attitudes, Sexual Self-Esteem and Risky Sexual Behavior
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013339:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013339/00001

Material Information

Title: Relationships among Gender Attitudes, Sexual Self-Esteem and Risky Sexual Behavior
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013339:00001


This item has the following downloads:


Full Text












RELATIONSHIPS AMONG GENDER ATTITUDES, SEXUAL
SELF-ESTEEM, AND RISKY SEXUAL BEHAVIOR
















By

AUTUMN LAMPINEN


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN EDUCATION

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Autumn Lampinen















ACKNOWLEDGMENTS

I thank Dr. Sondra Smith-Adcock for being my thesis chairperson and helping me

through the thesis development process. I thank Dr. Cirecie West-Olatunji for being a

member of my defense board. I thank Dave Marshall for getting me through the

statistics. Finally, I thank my family for their support and encouragement; I could not

have done this without them.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES .............. ................................................ ........ vi

ABSTRACT .............. ..................... .......... .............. vii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

State ent of the P problem ............................................................................. ........ 5
N eed for the Study ........................................................ ................ .6
Purpose of the Study ............... ............... .................................. .8
D definition s ......................................................................... . 9

2 LITERATURE REVIEW ........................................................... ....... ........11

(Sexual) Self-Esteem and Sexual Behavior...................................... ............... 12
Gender Differences and Sexual Behavior ............... .......................................17
G ender and [Sexual] Self-Esteem ..................................................... ...... ......... 23
Summary and Critique of the Literature ...................................... ...............24

3 M E T H O D O L O G Y ............................................................................ ................... 26

P a rtic ip a n ts ........................................................................................................... 2 6
In stru m e n ts ........................................................................................................... 2 6
P ro c e d u re s .......................................................................................3 3

4 R E S U L T S .............................................................................3 4

D ata M anagem ent ................................................................................................ 34
M multiple Linear R egression A analysis ....................................................... 35
Full M odel Regression............................................. 36
Forward Stepwise Regression............................................................37
G en der Sp ecific M models ....................................................................................... 3 8
A additional F findings ..............................................................40

5 D ISC U S SIO N ............................................................................... 42



iv










APPENDIX: DEFINITIONS OF INDEPENDENT VARIABLES .............................48

L IST O F R E F E R E N C E ............................................................................ ....................50

B IO G R A PH IC A L SK E T C H ...................................................................... ..................53






















































v
















LIST OF TABLES


Table p

1 Gender means for risky sexual behavior (RSB)............... ....... ............... 34

2 Slopes of independent variables for full models: ...............................................36

3 Sum m ary of final solution (4 Steps) ...................................................................... 38

4 Summary of final solution female stepwise (5 Steps)................ ..................39

5 Summary of final solution male stepwise (2 Steps).................... ..................40

6 Means and t-test scores of subscales by gender....................... ...............41















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Master of Arts in Education

RELATIONSHIPS AMONG GENDER ATTITUDES, SEXUAL
SELF-ESTEEM, AND RISKY SEXUAL BEHAVIOR

By

Autumn Lampinen

December 2005

Chair: Sondra Smith-Adcock
Major Department: Counselor Education

Many young adults are engaging in risky sexual behaviors that create negative

consequences. Little research has been done regarding the reasons for engagement in

such behaviors. Previous research shows a difference between men and women's sexual

behavior. Gender, and an individual's attitudes about gender, may influence behavior

within the context of an environment. It is possible that this influence extends to

individual engagement in risky sexual behavior. Sexual self-esteem, as a part of the

construction of global self-esteem, may also have an impact in what types of sexual

behavior one engages in. The purpose of this study is to identify variables that possibly

affect a person's likelihood of engaging in risky sexual behaviors. In particular, this

study examines the possible relationship among gender, gender attitudes, sexual self-

esteem, and risky sexual behavior. The study also examines the variance explained by

gender attitudes and sexual self-esteem as they impact risky sexual behavior. The data

for the current study came from a pre-existing database. The data include information









from 175 participants from undergraduate courses at a large southern university that were

given four questionnaires to assess sexual history, sexual risk-taking, gender attitudes,

and sexual self-esteem. The questionnaires used included a demographic information

questionnaire, the Sexual History Questionnaire, the Gender Attitude Inventory, and the

Multidimensional Sexual Self-Concept Questionnaire. Five items from the SHQ made up

the dependent variable of risky sexual behavior. The thirteen independent variables

included gender and six subscales from both the GAI and the MSSCQ. Findings showed

that while an overall multiple regression was statistically significant in predicting risky

sexual behavior, the regression indicated the only independent variable that was

significant in predicting risky sexual behavior was gender. That is, females are slightly

less likely to engage in risky sexual behavior. Gender attitudes failed to show

significance in predicting or explaining risky sexual behavior. However, sexual self-

esteem had stronger findings, although only for males. Specifically, only internal sexual

control was found to have statistical significance in predicting risky sexual behavior in

males. The research findings imply that more research regarding risky sexual behavior

would be beneficial.














CHAPTER 1
INTRODUCTION

In modern society, religion and politics place great emphasis on the impact

sexuality and sexual activity have on people. For instance, about 20,000 members of

conservative Christian churches in Washington State attended a "Mayday for Marriage"

rally in opposition to homosexual marriage (Norris, 2004). In a State of the Union

Address, President George Bush proposed to double the money spent on abstinence-only

sex education for 2005, despite much research in favor of comprehensive sex education

conducted by associations such as the Sexuality Information and Education Council of

the United States [SIECUS] (Bowman, 2004). Decades ago, Alfred Kinsey created a

research institute based on advancing the knowledge of human sexuality (Kinsey, 2005).

However, as evidenced by these recent events, though advances and changes have been

made, this knowledge has yet to create significant change in how society reacts to

people's sexuality and policies regarding sexual issues.

With this renewed interest in traditional views of marriage and sexuality, many

people recognize that our society also continues to place double standards on people

regarding sexual behavior based on their gender. The Feminist Movement started out of

women's realization of inequality based on gender and dissatisfaction with their lack of

life choices (Beck, 1998). The double standard was no longer acceptable. The feminist

ideology that both created and continues the feminist movement is very complex, more so

than many individuals realize. One study suggested the general definition of what a

typical feminist ideology and self-labeled feminist is: a woman that holds strong radical,









socialist, and cultural feminist beliefs (Liss, Hoffner, & Crawford, 2000). One of these

typical feminist groups is liberal feminists, which holds a basic belief that women and

men are virtually the same and should be able to make decisions free from the

governmental and social constraints that are based on gender. Liss et al. (2000) found

that the majority of women in their study supported liberal feminist goals and beliefs, yet

a majority of the participants (63%) in the study did not self-identify or label themselves

as a feminist.

Despite the Women's Movement and other advances toward equality that our

culture has made, traditional gender scripts are being passed on to younger generations,

such as the feeling that a wife's place is still in the home and it is the husband's

responsibility to provide for his family (Vanwesenbeeck, Bekker, & van Lenning, 2001).

In fact, it seems a negative stigma is still attached to the term "feminist", which creates

hesitance in women, even those who endorse pro-feminist goals and beliefs, to label

themselves as such (Liss et al., 2000; Beck, 1998). The media, being one of the most

influential entities in how culture and normalcy is defined, has oft promoted the idea of

feminists as crazy Amazon lesbian man-haters (Beck, 1998). Given the current wave of

anti-gay politics (Norris, 2004), the stigma of having a feminist label is often exacerbated

by the assumption that one must me a lesbian to be a feminist (Beck, 1998). Beck (1998)

goes on to note how media in Western culture shapes all things into dichotomies and

opposition, meaning things are in black and white, good and bad terms with no room for

potential grey. With this general media portrayal, masculine is opposite to feminine, and

Western culture categorizes "good" as male/white/middle class/Christian. Feminism thus

goes against the foundation on which the Western patriarchal society stands (Beck,









1998). Third-wave feminism, the name for the current feminist era, tries to negotiate

these dualistic presentations of being both feminist and feminine, and particularly

discusses the importance of sexuality in this negotiation (Kinser, 2004).

One final important note about traditional stereotypes is that it still remains a

prominent way of how education regarding sexuality is presented. In particular,

Vanwesenbeeck et al. (2001) acknowledge that the continuing trend of defining sexual

behaviors in masculine terms helps maintain male power and control in heterosexual

encounters. Additionally, traditional gender roles and assumptions in our society finds

more casual and less cautious sexual behavior less acceptable when practiced by women,

yet sexual behavior in general acceptable for men (Seal & Agostinelli, 1996). As a result

of this double standard, women are more likely than men to experience negative

consequences from risky sexual behavior such as social rejection due to unwanted

pregnancies and STDs (Seal & Agostinelli, 1996; Milhausen & Herold, 1999).

In addition to attitudes regarding gender roles, people have attitudes about one's

own sexuality and the sexuality of others. Sexual self-esteem is one part of our self-

concept that makes up the global definition of self-esteem (Gaynor & Underwood, 1995).

How one feels about his or herself as a capable sexual being affects our mental wellbeing

and how we deal with our relationships in sexual situations (Rosenthal, Moore, & Flynn,

1991). In fact, one study found high levels of [global] self-esteem are indicative of the

likelihood that one will approach sexual behaviors and their consequences more carefully

and with more forethought (Holmbeck, Crossman, Wandrei, & Gasiewski, 1994).

However, little research seems to have taken up an interest in individual feelings

of self-worth in the context of sexual self-esteem and how it impacts sexual behavior,









rather than in terms of global self-esteem. Of the studies that consider self-esteem, most

find a relationship between sexual or global self-esteem and risky sexual behavior. Some

research shows that boys with higher self-esteem had higher amounts of sexual activity,

whereas girls with higher self-esteem had lower amounts of sexual activity (Spencer,

Zimet, Aalsma, & Orr, 2002). One possible reason for this discrepancy is that sexual self-

esteem is formed in different ways based on one's gender (Breakwell & Millward, 1997).

Research has also noticed different attitudes toward risky sexual behavior based

on gender and gender attitudes. Young males have a tendency to be less concerned and

less cautious regarding the consequences of risky sexual, believing that contraceptives

and pregnancy concerns are the female's responsibility, or "problem" (Lenskyj, 1990).

Females appear to have a better grasp on some of the realistic problems that can

accompany risky sexual behavior, particularly seeing unwanted pregnancy as

"detrimental to future goal attainment" (De Gaston & Weed, 1996, p. 217). Yet, many

females still participate in risky sexual behavior. Vanwesenbeeck et al. (1998) are among

the researchers that show gender socialization in our masculine culture and phallocentric

definitions of sexuality promote feelings of powerlessness in women, endorsing

traditional gender roles. Researchers could continue to benefit from learning what about

our gender socialization, and thus our gender attitudes, is making safe sexual behavior

less consistent and less appealing to our young adults.

What implications can be made based on social and cultural biases related to risky

sexual behavior and their effect on the wellbeing and sexual behavior of young adults? If

we uphold more traditional gender and life assumptions in our minds, how does this same

mindset effect our beliefs and values when young adults are in the bedroom? In this









study, some factors proposed to influence young people's risky sexual choices are

[sexual] self-esteem, and gender attitudes.

Statement of the Problem

What are the risks or negative outcomes when participating in risky sexual

behavior? Pregnancy and sexually transmitted diseases [STDs] are two serious

consequences of risky sexual behavior (Lenskyj, 1990). Approximately 34% of girls will

become pregnant at least once by age 20, meaning there are approximately 820,000 teen

pregnancies every year (National Campaign to Prevent Teen Pregnancy, 2004). One

study noted only 57.9% of high school students reported using a condom during their last

heterosexual intercourse encounter with females reporting less use (Halpern-Felsher,

Kropp, Boyer, Tschann, & Ellen, 2004). The study did not include condom-use

information regarding anal or oral sex.

Statistics from the American Social Health Association [ASHA] (2005) report

there are approximately 65 million people in the United States with a viral STD, and

approximately 15 million new cases of STD's every year. There are an estimated

1,039,000 to 1,185,000 individuals in the United States living with HIV or AIDS with

about 40,000 individuals being infected with HIV each year (Centers for Disease Control

and Prevention [CDC], 2005). The CDC estimates that more than 50% of HIV/AIDS

cases in the United States are acquired before the age of 25, and generally due to sexual

intercourse (Florida Department of Health, 2004). The Department of Health (2004) also

reports that two-thirds of the 12 million annual cases of new STD infections are in young

people under the age of 25. Approximately 24-27% of the estimated population with

HIV/AIDS is undiagnosed and unaware that they are even infected. Research also shows

that a majority of college students rate other students' chances of contracting the human









immunodeficiency virus or other sexually transmitted diseases [HIV/STDs] higher than

their own and do not identify themselves as a type of person who is at risk for HIV/STDs

despite their own risky sexual behavior practices (Seal & Agostinelli, 1996).

Are there gender differences when it comes to risky sexual behavior? If so, what

possible social regulations are imposed on each gender? According to Vanwesenbeeck et

al. (1998), "All sexual behavior is mediated, facilitated, and/or justified by cognitions and

beliefs," (p. 326). Gender attitudes are cognitions and beliefs about gender within a

cultural context. There are clear gender double standards in assumptions about

acceptable sexual behavior and contraceptive responsibility (Seal & Agostinelli, 1996).

Sexually permissive women tend to be ostracized and stigmatized whereas men who

engage in the same sexual behaviors are lauded as admirable and popular (Milhausen &

Herold, 1999). Birth control, whether via oral contraceptives or having and using

condoms tends to be primarily placed on the female, yet open-discussion or

acknowledgement of preparedness for sexual behavior results in members of society

calling these women derogatory names, such as "sluts" (Lenskyj, 1990). This indicates

that there are definite societal impacts on human sexuality, gender socialization, and the

differences in the expression of sexuality between the genders. What are the

consequences of maintaining traditional gender roles and attitudes in relation to risky

sexual behavior? How do these traditional attitudes impact our emotional and mental

health, particularly sexual self-esteem?

Need for the Study

The dominant culture of the United States presents with discomfort on taboo

topics such as sexual behavior, sex education, and premarital sex (Gudelunas, 2005). The

factors that contribute to risky sexual behavior are unclear. Most studies focus on









knowledge of risk in sexual activities, but fail to find a large impact on why young adults

engage in risky sexual behavior (Hollar & Snizek, 1996). There exists a need to better

understand the correlates of risky sexual behavior in order to make changes in social

programming and socialization, to educate the population, and prevent negative outcomes

due to risky sexual behavior. The current research will attempt to look at factors that

potentially encourage risky sexual behavior.

Furthermore, young adults are participating in premarital sexual behaviors despite

abstinence education (Bowman, 2004; Starkman & Rajani, 2002; Caldas, 1994).

Abstinence-only education is one outcome of societal denial regarding premarital sexual

behavior (Starkman & Rajani, 2002). One thing that comprehensive sex education

teaches is how to practice safe sex. It is suggested that positive attitudes toward using

condoms increases the likelihood of using condoms (Halpern-Felsher et al., 2004). This

could help prevent negative outcomes of risky sexual behavior, as well as reduce the

likelihood of individuals engaging in risky sexual behavior. Without getting the

knowledge to practice safe sex, how can we expect young adults and adolescents to

reduce the risk involved in their sexual behavior?

In addition to the criticism that [abstinence-only] sex education does not change

or reduce adolescent or young adult sexual behavior, there are concerns about failing to

bring emotional and moral issues of sexuality and sexual decision-making into the sex

education process (Lenskyj, 1990). In other words, the positive aspects of sexuality, and

how to build healthy sexual attitudes, are virtually ignored in sex education courses, as

well as the providing of information about homosexuality and issues regarding gender

identity (Lenskyj, 1990). When discussing teenage pregnancy and the prospects of









unmarried parenthood as a result of combining negative sexual attitudes and low sexual

self-esteem with risky sexual behavior, there are hidden costs that many people fail to

consider. Only one-third of teenage mothers graduate from high school and 80% of

unwed teen mothers are on welfare (Rinaldo, 2002). Many unwed teen mothers feel a

loss of freedom (Rinaldo, 2002). Also, depression in young unmarried mothers is not

uncommon (Lesser, Koniak-Griffin, & Anderson, 1999). The current study can help

assess the potential costs of low sexual self-esteem, which includes negative attitudes

about sexuality, and young adult risky sexual behavior, including unwanted pregnancies,

STD's, depression, lowered self-esteem, and unreached future goals. The current study

can also allow researchers and sex educators to consider more alternatives for the

prevention of the negative consequences of risky sexual behavior.

Purpose of the Study

The purpose of this study is to determine if there are relationships between

gender, gender attitudes, sexual self-esteem, and risky sexual behavior and to better

define the nature of those relationships. This study will view the importance of gender

when discussing the sensitive topic of sexual behavior. It is necessary to discover how

attitudes about gender and the boundaries a culture imposes on each gender effects their

risky sexual behavior. There has been a minimal amount of research that looks at

people's perceptions and beliefs about gender and sexuality and how these factors

influence what one does or does not do with a sexual partner. This study will assess

whether or not these gender attitudes are related to how one behaves sexually. It will also

assess if gender attitudes affect sexual self-esteem. Another purpose this study is to

determine how sexual self-esteem affects one's sexual behavior or the level of risk one









takes in sexual activity. This study is concerned with how gender attitudes relate to

individual sexual self-esteem and how this affects risky sexual behavior.

The research questions addressed in this study are as follows:

1. What are the relationships between gender, gender attitudes, sexual self-
esteem, and risky sexual behavior?

2. Does sexual self-esteem explain variance in risky sexual behavior above and
beyond that explained by gender attitudes?

3. Which gender attitudes and sexual self-esteem factors explain the most
variance in risky sexual behavior?

Definitions

Following are some definitions of factors for the purposes of this study. Self-

esteem can be defined as the, "self-evaluation of one's worth as a person" (New

Webster's Dictionary and Thesaurus, 1993). Using Rogers' concept of self, Gaynor and

Underwood (1995) define sexual self-esteem as "the tendency to value, versus devalue,

one's own sexuality, thereby being able to approach rather than avoid sexual experiences

both with self and others" (p. 334). According to Andersen and Cyranowski, sexual self-

esteem is defined as "Sexual aspects of oneself that are derived from past experience,

manifest in current experience, influential in the processing of sexually relevant social

information, and guide sexual behavior" (Mayers, Heller, & Heller, 2003, p. 270). The

term "gender attitudes" refers to the beliefs one holds about others based on their gender.

This includes such ideas as endorsement of traditional stereotypes, endorsement of

female superiority, disapproval of female casual sex or sexual initiative, and acceptance

of male heterosexual violence (Ashmore, Del Boca, & Bilder, 1995). Risky sexual

behavior refers to sexual activity that places one at risk for physically negative outcomes

such as unwanted and/or teen pregnancy, and the contraction of an STD or HIV/AIDS






10


(Gil, 2005; Nzioka, 2004). Examples of risky sexual activities include multiple partners,

one-time partners (a.k.a. one-night stands), and sex without use of a condom (Gil, 2005;

Rosenthal, Moore, & Flynn, 1991).














CHAPTER 2
LITERATURE REVIEW

Self-esteem and individual feelings of self-worth have a way of determining how

we behave in many avenues of our lives. For as many people that exist in the world,

there are just as many ways our self-esteem manifests itself in our daily routines. Many

people seek therapy to improve their level of self-esteem. Some people lack feelings of

self-worth, and thus self-esteem, when it comes to being in a classroom, others when it

comes to getting a deserved promotion, and yet still others lack feelings of self-worth

when it comes to making friends. Young adults' sexual activities that they engage in are

not immune to individual feelings of self-esteem and worth.

Sexuality is just one facet of our self-esteem, but it is important to an individual's

overall well-being. Of the studies that have been done in the area of sexual self-esteem

and sexual behavior, most of the research involves high school age adolescents, and

secondly, middle school age adolescents. Gender differences have been found in how

levels of self-esteem affect risky sexual behavior (Spencer, Zimet, Aalsma, & Orr, 2002).

There have also been findings that low self-esteem, both general and sexual self-esteem

specific, affects sexual behavior (Young, Denny, & Spear, 1999; Hollar & Snizek, 1996).

Gender differences in risky sexual behavior have also been studied. However,

studies regarding gender attitudes, or beliefs about traditional gender roles, and how they

relate to risky sexual behavior, have not been researched as much. One finding from the

research is that males have a tendency to believe in traditional gender roles more often

than women do (Vanwesenbeeck, Bekker, & van Lenning, 1998). Males also tend to be









more sexually active (DeGaston & Weed, 1996), have more permissive attitudes toward

sexual behaviors (1996), and be less concerned with the outcomes of risky sexual

behavior (Breakwell & Millward, 1997).

This chapter reviews more current literature about research that shows a

relationship, or lack thereof, between sexual self-esteem, gender attitudes, and risky

sexual behavior. The following literature review is segmented into relationships between

two of each of these three variables. First, you will find a subsection on the relationship

between sexual self-esteem and risky sexual behaviors. Second, there is a section

regarding gender differences and risky sexual behavior. Third, you will find only one

piece of literature discussing research about the relationship between sexual self-esteem

and gender attitudes. This is due to a lack of research about that particular relationship in

virtually any form. Finally, studies that research the relationship between all three

variables of interest are virtually nonexistent. A subsection containing all three variables

is thus absent from this chapter.

(Sexual) Self-Esteem and Sexual Behavior

Spencer, Zimet, Aalsma, and Orr (2002) conducted a study about adolescent

gender differences in self-esteem and its potential impact on sexual behavior. In their

longitudinal study, they used a sample of 188 adolescents, testing them first in seventh

grade and later in ninth grade. Forty-three percent of the 188 participants were boys and

fifty-seven percent were girls. All of the 188 participants were virgins during the seventh

grade testing period. The researchers had the students complete a questionnaire that

included the Rosenberg Self-Esteem Scale (RSES) and a single question concerning

coital status, whether or not the participant had ever had sexual intercourse. The focus of

the study is relevant to the current study and research in that it looked at self-esteem









based on gender and how self-esteem based on gender affects the change in coital status

from virginal to non-virginal. An analysis of covariance was used with gender and coital

status effects on self-esteem, in conjunction with pubertal timing. Regarding the interests

of the current research, the study showed

a significant gender by coital status interaction for self-esteem...
results indicated that boys with high self-esteem at Time 1 [seventh grade]
were 2.4 times more likely to initiate intercourse than boys with low self-
esteem at Time 1. Conversely, girls with high self-esteem at Time 1 were
three times more likely to remain virgins than those girls with low self-
esteem. (p. 582)

This study also examined if the change in intercourse status had an effect on self-esteem,

taking gender into consideration. The study found that self-esteem remained stable. The

study suggested some reasons for the difference in male versus female self-esteem in

regards to sexual behavior, such as differing societal influences based on gender, and

attempts at improving emotional needs. This study reinforced the need for more research

regarding self-esteem, gender, and sexual behavior.

Another study by Young, Denny, and Spear (1999) examined the relationship

between adolescent self-esteem and sexual behavior of 1,659 junior and senior high

school students ranging from seventh to twelfth grade. This study asserted that self-

esteem is divided into three parts: peer self-esteem, family self-esteem, and school self-

esteem. Peer self-esteem is based on relationships with friends and perceived acceptance

or lack thereof. Family self-esteem refers to relationships with family members and

perceived confidence from parents. School self-esteem is related to academic

performance and perceived teacher response to students. The researchers used the Hare

Self-Esteem Scale to conduct their study, which breaks down questions of self-esteem,









sexual behavior, and sexual knowledge into the three parts of self esteem (peer, family,

and school).

The study found that peer self-esteem was positively correlated with both past and

future sexual behavior. On the other hand, there was a negative correlation between

sexual behavior and both family and school self-esteem, meaning the higher home and

school self-esteem was in the participants, the less likely they were to engage in sexual

behavior. This study provided information on individual factors that affect self-esteem

and thus sexual behavior. Although it is important to realize there are multiple things that

factor into our self-esteem development, the study came across a problem when

combining all three types of self-esteem into an integrated model of overall self-esteem.

It did not discuss the likelihood of sexual behavior given different combinations of the

three areas of self-esteem. There was no information regarding the breakdown or

possible differences based on gender, which is important to the current research.

Breakwell and Millward (1997) also conducted a study on older adolescent sexual

self-concept and sexual risk-taking. More specifically, the study examined the

relationship between sexual self-concept and sexual activities based on gender. There

were 474 participants ranging in age from 16 to 19, with the majority being female. The

study was conducted via mailing questionnaires to participants. These questionnaires

contained the Sexual Self-Concept Checklist, which has questions with choices of

characteristics that participant's felt described their sexual self-concept. They also

contained questions on whether they felt they held traditional male/female roles in sexual

relationships, the importance placed on sex and sexual attractiveness, and questions about

their sexual activity.









Breakwell and Millward's (1997) study suggested that sexual self-concept is

formed in different ways based on gender and thus sexual risk-taking is expressed in

different ways. For instance, responsibility for contraceptive use was found to be

relatively irrelevant to males, but it was a central part of a sexual self-concept in females.

A seemingly opposite finding was that there were no significant correlations between

sexual self-concept and traditional male/female roles in sexual relationships. However,

the study did not explain this finding or how the researchers came to this conclusion.

They also did not define what they were seeking out when looking at traditional gender

roles in sexual relationships.

A study by Rosenthal, Moore, and Flynn (1991) "investigates the sexual self-

efficacy and sexual self-esteem of males and females and the relation between sexual

risk-taking and these self-perceptions" (p. 77). This study evaluated 1,008 predominantly

female post-secondary students ranging in age from 17 to 20. All respondents had

engaged in sexual activity, meaning they had engaged in oral, anal, or vaginal sex, or

withdrawal. Participants filled out questionnaires on sexual self-efficacy, sexual self-

esteem, and sexual behavior and risk. Sexual self-efficacy measured participant

confidence and willingness in doing multiple kinds of sexual activities. Sexual self-

esteem measured individual's perceptions and attitudes regarding their sexual adequacy

and relationships. Sexual behavior and risk looked at whether sexual activities were

engaged in with a casual versus regular partner and whether or not condoms were used in

each scenario. Risk in sexual situations was evaluated based on partner type.

One important aspect touched on in Rosenthal et al.'s (1991) study is that they

considered gender differences in sexual self-efficacy. In general, males are more









confident in their condom use, masturbation, seeking out potential partners (dates),

watching pornographic movies without embarrassment, and being able to get sexual

needs met. On the other hand, women were more confident in saying no to sexual

activity. The rest of the study's findings were made in relation to partner type, casual or

regular. Overall, the study found there was little variance in risky behavior based on

sexual self-efficacy and self-esteem measures. In regards to casual partners, the study

found that the self-efficacy factor of the ability to say no was the only factor that

correlated with lower levels of sexual risk-taking. When discussing regular partners,

there were several factors increasing sexual risk-taking. Among these factors were the

ability to say no and confidence in sexual need assertion. Regular partners were found to

increase participant sexual self-esteem. The significance of the correlation of these

findings was greater among the female participants than the males.

Hollar and Snizek (1996) conducted a study regarding the connections between

levels of self-esteem, sexual behavior, and knowledge of HIV/AIDS. The study used 353

university students as participants, consisting of 49.7% women and 50.3% men. The

majority of students were in their first or second year of college. Instruments used in the

study included a 14-item True/False HIV questionnaire with 2 additional HIV/AIDS

multiple choice questions, the Rosenberg Self-Esteem Scale (RSES), and a multiple

choice questionnaire regarding forms of sexual behavior they have participated in. The

study focused on two different categories of risky sexual behavior which the researchers

called conventional and non-conventional. Conventional risky sexual behavior included

unprotected heterosexual sex, having multiple sex partners or having sex with

promiscuous sex partners. Non-conventional risky sexual behavior included unprotected









anal intercourse, syringe sharing or sex with someone who shares syringes, having sex

with a prostitute, or having sex with someone who is knowingly HIV positive.

Hollar and Snizek's (1996) study found that individuals of both genders with high

self-esteem and more knowledge about HIV/AIDS were more likely to engage in

conventional risky sexual behavior than others with lower self-esteem and less

knowledge. However, the second most likely group to engage in conventional risky

sexual behavior were the individuals classified as having low self-esteem and low levels

of HIV/AIDS knowledge. As for non-conventional risky sexual behavior, the students

with low self-esteem and low levels of HIV/AIDS knowledge were the most likely to

engage in this type of behavior. In regards to their study, Hollar and Snizek (1996)

expressed concern as to whether the cause of participants' risky sexual behavior was due

to high self-esteem or vice versa and they suggested a longitudinal study could help

answer whether the behavior or the high self-esteem came first. This particular study

pointed out that there were no gender differences in risky sexual behavior that pertains to

HIV/AIDS. This goes against other research studies that have found gender differences

in self-esteem and sexual behavior. More research on this topic can give clarity

regarding this discrepancy.

Gender Differences and Sexual Behavior

DeGaston and Weed (1996) conducted research to better understand gender

differences in the sexuality of adolescents. The research was done on the premise that

emotional differences exist between the sexes regarding sexuality as well as "distinct

gender differences in sexual attitudes, values, motivation, expression, and reasons for

avoiding sex" (p. 218). The study enlisted more than 1,800 seventh and eight grade

students from non-urban schools along the Pacific coast who were entering sex education









courses. A list of 14 questions about sexuality was given to the participants. DeGaston

and Weed (1996) predicted there would be gender differences in the level of

permissiveness of sexual attitudes and activities as well as negativity toward outcomes of

sexual behaviors.

DeGaston and Weed's (1996) findings showed males were twice as sexually

active as females. The study indicated females were more committed to abstinence and

had less permissive attitudes regarding premarital sex. This commitment to abstinence

included female beliefs that birth control and being in love are not reasons to justify

having sex, despite the finding that females generally frame love and sex together more

than males. The findings also noted that females saw adolescent sexual activity as an

impediment to the attainment of future goals. In regards to family and social parameters

of adolescent sexuality, females reported discussing sex with parents more often yet also

found parents less approving of sexual behavior, to the point of placing more restrictions

on females than on males. Among peers, females reported receiving less pressure to be

sexually active. An additional interesting finding of the study was a higher likelihood of

adolescent males believing sexual urges could not be controlled.

Seal and Agostelli (1996) conducted a study about perceptions of risky sexual

behavior prevalence among peers of college students. The study enlisted 217

participants, 96 of whom were men and 121 were women. The subjects ranged in age

from 18 to 25 years old. All participants completed a questionnaire that assessed their

individual sexual behaviors and safe sex practices. Upon completion, they were asked to

give separate estimates, or percentages, of the prevalence of male and female peers'

sexual behaviors that they just rated themselves on. Specifically, there were five risky









sexual behaviors that the study assessed: having at least 2 sex partners in the past year,

having unprotected sex with anyone in the past year, having more than one sex partner in

their life, having had any casual sexual encounters, and having casual sex without a

condom.

The study found a consistent overestimation of the prevalence of peer sexual

behaviors among all participants, especially among those who engaged in the particular

estimated behavior. Gender differences were found in both participant overestimation

and the assumed gender or the peers that engaged in sexual behavior. Specifically,

female participants made higher estimates than the male participants, and both male and

female participants estimated sexual behaviors as being more frequent among male peers.

Seal and Agostelli (1996) went on to discuss possible causes for gender differences both

in the participants' overestimations as well as the assumed peer sexual behaviors.

Shearer, Hosterman, Gillen, and Lefkowitz (2005) conducted a study regarding

the possible associations between traditional gender role attitudes and risky sexual

behavior. Condom-related behaviors were a particular focus for risky sexual behaviors.

Shearer et al. (2005) collected data from 154 participants between the ages of 18 and 25,

of which 96 were women and 58 were men. Participants were asked to complete multiple

scales in addition to answering questions about their sexual history. Sexual history

questions were most important to the study in regards to condom use, engagement in

casual sex, and participant relationship status. Instruments used in the study included the

Condom Use Self-Efficacy Scale, the Outcome Expectancies for Condom Use (OECU),

the Attitudes Toward Family Roles Scale (ATFRS) which looked at traditional marital









and childrearing beliefs, and the Male Role Norms Scale (MRNS) which looked at

traditional gender role beliefs in terms of anti-femininity and social status achievement.

Shearer et al.'s (2005) study found a moderate level of risky sexual behavior

among the participants. Despite, overall positive beliefs about condom use, 82% of the

participants reported not using a condom during sexual intercourse on at least one

occasion. The study differentiated norms for different gender role beliefs, particularly

male role norms and marital and childrearing role norms. Overall, women were found to

have less traditional gender role beliefs than men. Women with more traditional beliefs

regarding marital roles were more likely to believe there were greater barriers to (reasons

not to) use a condom and believed sex would be less enjoyable when using a condom.

Males with traditional beliefs about childrearing were more likely to believe they would

enjoy sex less if a condom was used. Both male and female individuals with more

traditional views of male role norms were more likely to use condoms. However, men

who asserted anti-feminine beliefs about male roles were less likely to use condoms

during intercourse and believe that sex without condoms would be less enjoyable. Also,

males who attempted to achieve high social status were more likely to use condoms but

were more likely to have casual sex as well.

Lucke (1998) studied gender roles and sexual behavior of young women. In

particular the study assesses the relationship between gender roles and unsafe sexual

practices. These three sexual behaviors that may be considered unsafe are: having

multiple partners, not using condoms, and using drugs or alcohol before or during sexual

intercourse. Gender roles had three dimensions for the purposes of the study: personality

traits, attitudes, and dating behavior. Women who were more egalitarian in their gender









attitudes were said to have more masculine personality traits, which indicated the

likelihood to engage in more risky sexual behavior. There were 400 participants from the

Queensland, Australia area. Their ages ranged from 16 to 24 years old. Instruments used

included the Personal Description Questionnaire (PDQ) to examine gender role

personality traits, the Women in Society Questionnaire (WSQ) to assess gender role

attitudes, and the DeLucia Dating Behavior Indices (DDBI) to assess gender role dating

behavior. The participants were also assessed regarding their sexual behavior.

Lucke's (1998) findings differentiated between having masculine personality

traits versus having masculine dating behaviors. The study showed the more that a

woman endorsed masculine dating behaviors, the more she would also endorse feminine

dating behaviors, meaning traditional gender roles. Women with egalitarian attitudes

were more likely to have masculine traits. The study reported that these women with

more masculine traits generally had more than one sexual partner were more liberal in

beliefs about women's roles in society, and had less feminine dating styles. The study

noted that the earlier a woman engaged in sexual intercourse, the less likely she was to

use condoms later. An interesting finding the study made was that older participants

were less likely to use condoms than the younger participants. Finally, alcohol and drug

use was associated most with single women who had masculine personality traits, more

egalitarian attitudes toward women, and non-steady or multiple sexual partners.

Vanwesenbeeck, Bekker, and van Lenning (1998) looked at the connection

between interactional competence in sexual encounters and the meaning of sexuality and

attitudes regarding gender and gender relations. Interactional competence is the set of

behaviors and attitudes one engages in to settle on an outcome of a sexual encounter with









one's partner. This study used 386 university graduate students in the Netherlands who

had experience with heterosexual intercourse. The participants consisted of 275 females

and 111 males. The demographic characteristics between the male and female

participants were similar allowing for the comparison to be possible. The questionnaire

consisted of interactional behavior, sexual meaning, and gender relation attitude

questions. The 55 interactional behavior questions assessed emotions and behaviors

during heterosexual experiences. The 31 sexual meaning questions asked about

sensation-seeking sexual behaviors, sexual compulsion, anxiety, and beliefs about the

level of connection between love and sex. Attitudes about gender were measured with 41

questions regarding gender conservatism, acceptance of interpersonal violence and rape

myths, and adverse beliefs and hostility toward each gender.

Vanwesenbeeck et al.'s (1998) study also found gender differences in

interactional behavior and competence. In particular, females were found to be more

defensively controlling of sexual situations and more anxious in terms of a lack of control

of the situation. Males were found to be more pro-active in their controlling what they

want and get in a sexual situation. In regards to sexual meaning, males scored higher for

being sexually compulsive and sensation-seeking. Gender differences were also found in

attitudes about sexuality. Male students were significantly more traditional in

considering gender roles, were more accepting of rape myths, and engaged in more

hostile feelings and beliefs toward either or both genders. Some of the variables were

connected in the questionnaire based on gender as well. For women, the more the

individual adhered to traditional gender roles, the less likely she was to be defensive in

interactional behaviors. Women were also likely to be more hostile toward either gender









if they had a higher degree of sexual anxiety. For men, the more hostile beliefs they

engaged in, the more likely they were to be pro-actively controlling in sexual situations.

Also, men with more traditional gender attitudes correlated with higher acceptance of

rape myths and pro-active control.

Gender and [Sexual] Self-Esteem

Burnett, Anderson, and Heppner (1995) conducted a study that looked at

environmental effects on self-esteem and gender roles. The study was not directly

assessing sexual self-esteem. There were 236 participants in the study, 90 male and 146

female. All participants were in their first two years of college at a large Midwest

university. The instruments used in the study were the Personal Attributes Questionnaire

(PAQ), the Personal Attributes Questionnaire-Environmental form (PAQ-env), and the

Coopersmith Self-Esteem Inventory (SEI). The PAQ assessed sex-role orientation,

meaning whether a person identified themselves with characteristics of masculinity or

femininity. The PAQ-env assessed the participants' perceptions of environmental cues

that encouraged masculinity or femininity. The SEI measured individual feelings

regarding self-esteem.

Burnett et al. (1995) found a significant correlation between having masculine

traits and high self-esteem for both genders, but an insignificant correlation for feminine

traits and self-esteem. The study also reported that the participants felt more

encouragement to have masculine traits than feminine ones. This finding was significant

for both genders. Finally, the study found that the individuals with the lowest levels of

masculinity also had the lowest levels of self-esteem, which was particularly significant

in women.









Summary and Critique of the Literature

There are some patterns that are emerging from the research. Studies that have

looked at relationships between sexual self-esteem, risky sexual behavior, and gender

differences are generally done with adolescent and young adult populations. Gender

attitudes, the beliefs one holds about others based on their gender, including such ideas as

endorsement of traditional stereotypes, endorsement of female superiority, disapproval of

female casual sex or sexual initiative, and acceptance of male heterosexual violence

(Ashmore, Del Boca, & Bilder, 1995), tend to be studied in the scope of traditional versus

non-traditional, or individual feminine versus masculine, traits (Lucke, 1998; Burnett,

Anderson, & Heppner, 1995).

There are consistent findings that gender differences exist in sexual behaviors and

levels of permissiveness for risky sexual behaviors. There are also consistent findings

that many males and individuals having masculine traits of either gender are associated

with higher levels of [risky] sexual behavior (Rosenthal et al., 1991; DeGaston & Weed,

1996; Lucke, 1998; Vanwesenbeeck et al., 1998). Traditional gender role beliefs are also

being indicated as exacerbating levels of risky sexual behavior (Shearer et al., 2005).

However, there are also inconsistent findings in this research area, such as how

[sexual] self-esteem affects sexual behavior based on gender. Zimet et al. (2002) noted

that high levels of self-esteem for boys, yet low levels of self-esteem for girls, increased

amounts of risky sexual behaviors. On the other hand, Rosenthal et al. (1991) found little

variance in sexual behavior when looking at sexual self-esteem. Yet another study

(Young et al., 1999) noted that different types of self-esteem produced different

likelihood of engaging in sexual behavior. Perhaps the differences among the studies









are based on whether they narrowed the scope of self-esteem to sexual self-esteem or

other aspects of self-esteem, or studied self-esteem in a general sense.

There are several areas of possible future topics to research in the area of sexual

self-esteem, sexual behavior, and gender attitudes or gender roles. Studies that assess the

direction of the relationships between these variables may increase our knowledge about

sexual risk and self-esteem levels and allow us to come up with ways to effectively deal

with these types of issues. Studies that are interested in socio-cultural impacts on these

variables and how they are correlated are also an area that would benefit the field of

counseling. Another area of interest may be how these variables interact to affect adults,

especially older populations, such as individuals considered to be middle-aged or in old

age. Only one-third of the studies which had self-esteem as a variable that were reviewed

discussed sexual self-esteem rather than more general types of self-esteem (Breakwell &

Millward, 1997; Rosenthal et al., 1991). One study did discuss that there are different

elements to self-esteem (Young et al., 1999), but sexual self-esteem was not studied as

one of the elements. It could be beneficial to the field of counseling and research to

conduct more studies about the different facets of self-esteem and about the

conceptualization of sexual self-esteem.














CHAPTER 3
METHODOLOGY

Participants

Initially, there were 175 participants in the study (98 females and 77 males) receiving

education from a large southeastern public university. However, due to the rate of

attrition, the final number of response sets used in the study was 126 (76 females and 50

males). All students were at the undergraduate level and were currently enrolled in either

Stress Management or Interpersonal Communications Skills courses. The mean age was

21.3 years old, with ages ranging from 18 to 37 years. The majority of the participants

were unmarried. The mean age for participants losing their virginity was 16.8 years with

a standard deviation of 2.9 years. All participation was voluntary and they could leave

the study at any time they requested without concern of penalty. Participants were

notified of the potentially sensitive and personal material that was going to be assessed.

Procedures for informed consent were followed, and participants' confidentiality was

also protected.

Instruments

Demographic information. Participants initially filled out a questionnaire

regarding demographic data. Demographic data collected for this study included

participant age, gender, and marital status.

Sexual History Questionnaire (SHQ; Cupitt, 1998). The dependent variable in

this study is risky sexual behavior. To define risky sexual behavior, we took 5 items

from Cupitt's (1998) Sexual History Questionnaire. This questionnaire is a self-report









measure that was created to assess the level of risk of contracting HIV based on an

individual's sexual behavior. The questions on the Sexual History Questionnaire are an

array of multiple choice, yes/no, and Likert-type 5-point scales. The questionnaire

should only take approximately 5 to 10 minutes to complete. These 5 questions sought

information regarding how many sexual partners the participant had in the last month,

whether the last sexual encounter was unprotected vaginal or anal intercourse with a

regular sexual partner, how able the participant felt in expressing wants and needs

regarding sex, and self-perception of risk from HIV/AIDS. This questionnaire is not

scored in the traditional sense, but does provide information about a participants' risk for

the contraction of HIV. The scores, upon being coded for the purposes of this study, of

these five items were added together in such a way that higher scores indicated a higher

level of risky behavior. The minimum possible score is 2. A range of scores is

technically not possible due to the open-ended number of sexual partners had in the last

month. In this particular sample, the possible range is 2 to 34.

The questionnaire begins with collecting demographic data, and then continues

with questions placed into four different sections. First in the SHQ (1998), participants

are asked if they have sex with men or women in order to distinguish between sexual

identity and sexual behavior. Follow-up questions ask whether or not participants have

ever had penetrative sex, and if so, whether or not a condom was used. Second, the

questionnaire asked about a participants' sexual activity in the last month. In particular,

it asks how many sexual partners they have had in the past month, how many of those

were regular vs. casual partners, how many sexual encounters were had with each

partner, and how many of the occasions included the use of a condom. Third, the









questions are related to the participants' last sexual encounter. Questions related to how

long ago the last sexual encounter was, partner's gender, discussion about condom use

and other concerns about sex, contraception, and what type of intercourse was engaged

in. Types of intercourse included vaginal, anal, oral, or masturbatory and whether or not

a condom was used. Finally, personal perceptions of risk and knowledge of HIV was

assessed (e.g. How much at risk do you consider yourself from HIV/AIDS?).

Cupitt (1998) advises the information the Sexual History Questionnaire seeks to

collect is not stable, since the sexual activity of all individuals changes over time. The

third section, regarding one's last sexual encounter was added in an attempt to strengthen

the questionnaire's reliability. However, this assessment has a high level of reliability

(correlation < .80, with p < .001) in regards to the fourth section that asks about

perceptions of risk and knowledge about HIV. A list of definitions of the sexual

behaviors was included to ensure participants understood the questions in similar ways,

creating a high level of face validity (1998).

Gender Attitude Inventory (GAI; Ashmore, Del Boca, & Bilder, 1995). There

are 13 independent variables, one of which is gender. Six of the remaining twelve

independent variables come from Ashmore et al.'s (1995) Gender Attitude Inventory

(GAI). This inventory is structured to assess topical attitudes regarding gender. The

inventory is specifically created for the assessment of college students in the United

States. The Gender Attitude Inventory consists of 109 questions, broken into 14 different

gender attitude scales that are content-specific, or 3 second-order factor scales, which

assess patterns of social behaviors based on gender. The questions were either framed in

an attitudinally traditional (42 items) or modern (67 items) style. All responses to the









questions are formed in Likert-type 7-point scores, ranging from strongly disagree to

strongly agree. A higher score on any scale indicates a more traditional attitude towards

gender. This questionnaire can be scored utilizing the 14 separate scales or the 3 second-

order factor scales.

The Gender Attitude Inventory (1995) has 14 different scales in assessing specific

gendered behaviors and attitudes. The 109 items of this inventory were placed under one

of these scales. The 14 scales and example items of each follow: Endorsement of

Female Superiority (e.g. "Compared to women, men have many undesirable qualities"),

Acceptance of Traditional Stereotypes (e.g. "Men are more independent than women"),

Condemnation of Homosexuality (e.g. "Male homosexuality is merely a different kind of

lifestyle which should not be condemned"), Disapproval of Female Sexual Initiative (e.g.

"The man should always be the one to initiate sex with a woman"), Disapproval of

Female Casual Sex (e.g. "It is acceptable for a woman to have sex with a casual

acquaintance"), Endorsement of Chivalry (e.g. "Chivalry is generally demeaning to

women"), Acceptance of Male Heterosexual Violence (e.g. "Women provoke rape by

their appearance or behavior"), Endorsement of Family Roles (e.g. "Care of children

should be shared equally by both spouses"), Belief in Differential Work Roles (e.g. "All

occupations should be equally accessible to both men and women"), Rejection of Female

Political Leadership (e.g. "I would vote for a politically qualified woman to be President

of the U.S."), Opposition to Women's Rights (e.g. "Passage of legislation to further

women's rights is necessary"), Endorsement of Individual Action (e.g. "Most women

have only themselves to blame for not doing better in life"), Opposition to Funded Day

Care (e.g. "Government-sponsored child care centers would be a big waste of money"),









and Opposition to Abortion ("Abortion should be treated legally as a form of murder").

Also, the information from 67 items gathered from this inventory allow us to view 3

second-order factors that assist in predicting patterns of social behaviors based on gender

attitudes. These second-order factors are: Societal Organization which includes gender

and societal role issues, Sexual Relationships which considers sexual behavior

acceptance, and Stereotypes which evaluates the overall endorsement of traditional

gender roles.

The six subscales that were used in this study from the GAI are: acceptance of

traditional stereotypes (ATS), disapproval of female sexual initiative (FSI), disapproval

of female casual sex (FCS), endorsement of family roles (EFR), belief in differential

work roles (DWR), and opposition to women's rights (OWR). The remaining seven GAI

subscales did not have questions specifically relevant to the interests of the study and

were thus removed from analysis. The scores for each item in a given subscale were

summed up to create a total subscale score. The subscale Acceptance of Traditional

Stereotypes consisted of 10 items with a score range of 10 to 70. Disapproval of Female

Sexual Initiative consisted of 11 items with a score range of 11 to 77. Disapproval of

Female Casual Sex consisted of 3 items with a score range of 3 to 21. Endorsement of

Family Roles consisted of 11 items with a score range of 11 to 77. Belief in Differential

Work Roles consisted of 9 items with a score range of 9 to 63. Finally, Opposition to

Women's Rights consisted of 10 items with a score range of 10 to 70. Once each

subscale has a score, all of the subscale scores are added up to create a single measure of

the respondent's gender attitudes. After several items were reverse coded, higher scores

on each of the subscales indicates a more traditional gender attitude.









Ashmore et al. (1995) found strong reliability in both internal consistency and

temporal stability for all 14 scales and the 3 second-order factor scales. The Alphas and

the test-retest correlations for all three second-order factors were higher than those for the

14 primary scales. In regards to internal consistency, the mean of the Alpha levels for

males was .83, which was a little higher than the .79 mean Alpha levels for females.

When assessing the temporal stability of the GAI, the correlations for the test-retest mean

of males was .81 and .78 for females. The GAI is a valid instrument and attained

statistical significance for 9 of the scales that men scored as holding more traditional

attitudes than women. The GAI was compared to other standardized measures of sex-

role ideology. In particular, when compared to the Attitudes Toward Women Scale

(AWS), there were statistically significant correlations for 10 of the 14 scales for men,

and 12 of the 14 scales for women.

The Multidimensional Sexual Self-Concept Questionnaire (MSSCQ; Snell,

Jr., 2001). The remaining 6 independent variables come from subscales from the

MSSCQ. This questionnaire is a self-report measure that was created to assess 20 aspects

of the self-concept of human sexuality. There are 20 subscales, each having 5 questions,

alternating in ascending order throughout the questionnaire for a total of 100 items.

Responses to all of the sexual self-esteem items were in Likert-type format ranging from

1 to 5, where 1 represents not at all characteristic of me and 5 represents very

characteristic of me. The MSSCQ (2001) provides 20 different scores for each of the 20

subscales. The scores for each item in a given subscale were summed up to create a total

subscale score. The range of scores per subscale is 5 to 25. Once each subscale has a

score, all of the subscale scores are added up to create a single measure of the









respondent's sexual self-esteem. The range of the sum of subscale scores is 30 to 150.

Several items were reverse coded in the MSSCQ, and a higher score on each of the

subscales indicates a higher tendency of correspondence to the aspect of sexual self-

concept that the subscale measures. The questionnaire should take approximately 45

minutes to one hour to complete.

The 20 subscales of the Snell's (2001) questionnaire are as follows: sexual

anxiety, sexual self-efficacy, sexual consciousness, motivation to avoid risky sex,

chance/luck sexual control, sexual preoccupation, sexual assertiveness, sexual optimism,

sexual problem self-blame, sexual monitoring, sexual motivation, sexual problem

management, sexual esteem, sexual satisfaction, power-other sexual control, sexual self-

schemata, fear of sex, sexual problem prevention, sexual depression, and internal sexual

control. Some example items include "I am very aware of my sexual feelings and

needs", "I am satisfied with the status of my own sexual fulfillment", "I have a fear of

sexual relationships", My sexual behavior is mostly determined by people who have

influence and control over me", and "The sexual aspects of my life are a matter of fate

(destiny)". However, only six MSSCQ subscales were utilized as independent variables

in the current study are: sexual anxiety, sexual self-efficacy, sexual esteem, sexual

satisfaction, power-other sexual control, and internal sexual control. The remaining 14

subscales of the MSSCQ were not used due to their lack of fit in relation to the focus of

the current study.

Snell (2001) found high internal consistency for all 20 of the subscales, with

Cronbach alphas ranging from .72 to .94 for all participants. Validity was established

through recent research regarding the MSSCQ. The subscales were able to predict the









likelihood of university-age male and female contraceptive behavior based on different

scores of different subscales.

Procedures

Prior to the distribution of the assessments, participants signed informed consent

forms and given a copy of the signed form. Participants were also given assurance of

their confidentiality of participation in the study. Participants were also informed of the

sensitive and personal nature of the assessment questions. Subjects were asked to

complete four questionnaires. Participants first filled out the demographic information

questionnaire. Second, they filled out Cupitt's (1998) Sexual History Questionnaire.

Third, the participants filled out Ashmore et al.'s (1995) Gender Attitude Inventory.

Finally, the participants filled out Snell's (2001) Multidimensional Sexual Self-Concept

Questionnaire. All questionnaires were filled out in their entirety, despite the current

study's use of only certain subscales from each measurement. The questionnaires took

approximately two hours to complete.














CHAPTER 4
RESULTS

The purpose of the study was to assess if the extent of risky sexual behavior in

young adults is predicted by gender, gender attitudes, and sexual self-esteem. Gender

attitudes and sexual self-esteem were each defined by 6 subscales.

Data Management

Initially, the sample contained 175 participants. However, several participants

were eliminated due to extensive missing data in the response sets. Participants were

eliminated from the sample if any of the subscales contained no data. Of the respondents

who may have missed an item or two within a subscale, missing data was replaced by

using the mean of the items responded to within that subscale. This method has the

advantage of producing more accurate estimated means for each of the subscales.

However, the tradeoff of using mean replacement is a reduced overall variance. The final

number of response sets used in the study was 126. There were 76 females and 50 males

included in the response set used for the analysis.

The gender means for risky sexual behavior (RSB) are provided in the following:

Table 1: Gender means for risky sexual behavior (RSB).
Gender N Mean Std. Dev. -95% CI +95% CI

Male 50 6.20 3.31 5.25 7.14

Female 76 4.19 1.54 3.84 4.55









Multiple Linear Regression Analysis

To determine which of the 13 independent variables were significant in predicting

risky sexual behavior, a multiple linear regression equation was calculated. One

categorical variable, gender (GEN), was used. The remaining variables were continuous.

The six subscales that were used from the Gender Attitude Inventory are: acceptance of

traditional stereotypes (ATS), disapproval of female sexual initiative (FSI), disapproval

of female casual sex (FCS), endorsement of family roles (EFR), belief in differential

work roles (DWR), and opposition to women's rights (OWR). The six MSSCQ

subscales utilized in the current study are: sexual anxiety (SA), sexual self-efficacy

(SSE), sexual esteem (SE), sexual satisfaction (SS), power-other sexual control (PSC),

and internal sexual control (ISC).

The resulting linear regression equation is expressed as:

y = a + Pi(GEN) + 32(ATS) + P3(FSI) + P4(FCS) + 35(EFR) + 36(DWR) +
P7(OWR) + Ps(SA)+ 39(SSE) + Pio(SE) + P11(SS) + P12(PSC) + 313(ISC) +e

Where: y = predicted value of Total Score
a = y intercept
GEN= gender (M=0, F=1)
1i = slope of gender
ATS = acceptance of traditional stereotypes
32 = slope of acceptance of traditional stereotypes
FSI= disapproval of female sexual initiative
33 = slope of disapproval of female sexual initiative
FCS = disapproval of female casual sex
N4 = slope of disapproval of female casual sex
EFR = endorsement of family roles
35 = slope of endorsement of family roles
DWR = belief in differential work roles
16 = slope of belief in differential work roles
OWR = opposition to women's rights
P7 = slope of opposition to women's rights
SA = sexual anxiety
P8 = slope of sexual anxiety
SSE = sexual self-efficacy









P9 = slope of sexual self-efficacy
SE = sexual esteem
o10 = slope of sexual esteem
SS = sexual satisfaction
P11 = slope of sexual satisfaction
PSC = power-other sexual control
P12 = slope of power-other sexual control
ISC = internal sexual control
113 = slope of internal sexual control
e = residual error term

Full Model Regression


Risky sexual behaviors were analyzed using a standard linear regression model.

The full model which includes all variables was shown to be significant: F(13,112) =

3.01, p < .0008. Overall, the model produced an R2 = .259, accounting for 25.9% of the

overall variance.

Although the overall model was significant in predicting risky sexual behavior

(RSB), the only independent variable that proved to be significant was gender (GEN). A

summary of the results for the full model is provided in Table 2.

Table 2: Slopes of independent variables for full models:
Variable Value Std Err t(112) p-value
a 10.38 2.53 4.09 < .00001*

1P -1.40 0.56 -2.52 0.013*
32 0.01 0.03 -0.56 0.575
P3 0.02 0.03 0.65 0.520

P4 0.04 0.06 0.70 0.484
P5 0.03 0.03 -1.28 0.203
P6 -0.02 0.03 -0.52 0.605
P7 0.02 0.03 0.69 0.491
p1 -0.04 0.07 -0.58 0.561
39 0.08 0.09 0.95 0.343









Table 2: Continued
Variable Value Std Err t(112) p-value

Pio -0.07 0.09 -0.80 0.42
P11 -0.07 0.08 -0.98 0.331
P12 -0.01 0.07 -0.14 0.888
P13 -0.13 0.09 -1.45 0.15
Where: indicates significance
The resulting linear equation from using all the variables in the full model:

RSB = 10.38 1.40(GEN) + 0.01(ATS) + 0.02(FSI) + 0.04(FCS) + 0.03(EFR) -

0.02(DWR) + 0.02(OWR) 0.04(SA)+ 0.08(SSE) 0.07(SE) 0.07(SS) 0.01(PSC) -

0.13(ISC) + e

Forward Stepwise Regression

A forward stepwise regression method was used to determine if a reduced model

would reveal any significant independent variables which would be useful in predicting

RSB. A forward stepwise regression technique adds one variable at a time to the model

and determines which individual variable explains the most variance. Subsequent

variables are added one at a time in an order which accounts for the most variance in the

model. Final solutions for a forward stepwise regression method usually have less than

the full number of original variables selected.

For the final solution using the forward stepwise technique, the model was

significant F(4,121) = 8.92, p < 0.0001, and produced a Multiple R2 value of 0.227,

accounting for 22.7% of the overall variance. The only variable which was significant in

this model was gender (GEN).









Table 3: Summary of final solution (4 Steps)
Variable R p Val Std Err t(12 2-
increase ue 3 1) value
Step 0 Intercept 10.57 1.28 8.2 <.000
5 1*
Step 1 Gender 0.144 -1.57 0.46 <.000
3.45 7*
Step 2 Internal 0.056 -0.07 0.07 0.30
Sexual Control 1.03
Step 3 Endorsement 0.013 -0.03 0.02 0.12
of Family Roles 1.56
Step 4 Sexual Esteem 0.014 -0.10 0.07 0.14
1.47
Where: indicates significance

A practical interpretation of the results suggests that female respondents scored an

average of 1.57 points lower than male respondents on the score of risky sexual behavior

(RSB) scale. While internal sexual control (ISC), endorsement of family roles (EFR),

and sexual esteem (SE) did help to explain more of the total overall variance and would

help produce a more accurate estimate of expected risky sexual behavior (RSB) scores,

they were not statistically significant.

Gender Specific Models

Since no variables other than gender were found to be significant in either of the

models tested, separate analyses were conducted for both males (N = 50) and females (N

= 76) to determine if any of the remaining independent variables were significant in

predicting risky sexual behavior (RSB) in regard to gender.

Risky sexual behavior (RSB) for females was analyzed using a standard linear

regression model. The full model, including all variables was not found to be significant:

F(12,63) = 1.1137, p < .365. Overall, the model produced an R2 = .175, accounting for

17.5% of the overall variance. Alternatively, a forward stepwise regression was found to









be significant: F(5,70) = 2.55, p < .035. The forward stepwise regression model

produced an R2 = .154, accounting for 15.4% of the overall variance in the reduced

model. Of the variables included in this model, only the intercept was found to be

significant.

Table 4: Summary of final solution female stepwise (5 Steps)
Variable R2 p Val Std Err t(70) p-
Increase ue 3 value
Step Intercept 5.16 1.46 3.53 <.000
0 1*
Step Sexual Anxiety 0.056 0.07 0.05 1.58 0.11
1
Step Accept 0.034 -0.02 0.02 0.27
2 Traditional 1.10
Stereotypes
Step Sexual Esteem 0.014 -0.06 0.04 0.13
3 1.51
Step Disapprove 0.013 0.03 0.02 1.80 0.07
4 Female Sexual
Initiative
Step Endorsement of 0.034 -0.03 0.02 0.09
5 Family Roles 1.69
Where: indicates significance

A practical interpretation of the results suggests that for female respondents, the

predicted value of risky sexual behavior (RSB) is 5.16. While sexual anxiety (SA),

acceptance of traditional stereotypes (ATS), sexual esteem (SE), disapproval of female

sexual initiative (FSI), and endorsement of family roles (EFR) did help to explain more

of the total overall variance and would help produce a more accurate estimate of expected

risky sexual behavior (RSB) scores, they were not statistically significant.

Risky sexual behavior (RSB) for males was analyzed using a standard linear

regression model. The full model, including all variables was not found to be significant:









F(12,37) = 1.05, 1 < .425. Overall, the model produced an R2 = .254, accounting for

25.4% of the overall variance. Alternatively, a forward stepwise regression was found to

be significant: F(2,47) = 5.44, p < .007. The forward stepwise regression model

produced an R2 = .188, accounting for 18.8% of the overall variance in the reduced

model. Of the variables included in this model, the intercept and internal sexual control

(ISC) were found to be significant.

Table 5: Summary of final solution male stepwise (2 Steps)
Variable R p Value Std Err t(47) p-value
increase P
Step Intercept 9.56 2.70 3.54 <.0001*
0
Step Internal Sexual 0.156 -0.35 0.11 <.002*
1 Control 3.18
Step Disapprove of 0.031 0.20 0.15 1.36 0.18
2 Female Casual
Sex
Where: indicates significance

A practical interpretation of the results suggests that for male respondents internal

sexual control (ISC) is significant in helping to predict risky sexual behavior (RSB).

While disapproval of female casual sex (FCS) did help to explain more of the total

overall variance and would help produce a more accurate estimate of expected risky

sexual behavior (RSB) scores, this was not statistically significant.

Additional Findings

Although the multivariate analyses performed did not reveal a significant

explanation for the reasons behind the gender effects using the factors examined in this

study, an alternative and less powerful way of looking at the data is to view the univariate

means for each subscale based upon gender. Performing a series oft-tests on the means










of each subscale may begin to give an idea of where to focus future research into this

area:

Table 6: Means and t-test scores of subscales by gender
Variable Male (N= Female (N = 76) t(124) p-value
50)
Accept Traditional 33.51 39.18 3.14 .002*
Stereotypes
Disapprove Female 60.27 59.99 -0.14 .890
Sexual Initiative
Disapprove of 14.49 13.25 -1.77 .791
Female Casual Sex
Endorsement of 48.10 57.29 4.50 <.001*
Family Roles
Belief in Differential 36.04 44.42 5.67 <.001*
Work Roles
Oppose Women's 40.88 47.97 5.27 <.001*
Rights
Sexual Anxiety 12.61 10.91 -2.35 .202*

Sexual Self Efficacy 17.54 17.27 -0.36 .722

Sexual Esteem 16.73 17.55 1.09 .278

Sexual Satisfaction 16.83 17.56 0.90 .371

Power-other Sexual 12.38 9.60 -3.68 <.001*
Control
Internal Sexual 17.75 18.78 1.46 .148
Control

Where: indicates significance














CHAPTER 5
DISCUSSION

Although the overall multiple regression model was statistically significant in

predicting risky sexual behavior, the most significant finding is that there are gender

differences in risky sexual behavior as it is defined in the study. Specifically, females are

slightly less likely to engage in risky sexual behavior. Other than gender, no other

independent variables were considered to be statistically significant.

Since the strongest test, a full multiple regression, only found gender to be

statistically significant in helping to predict risky sexual behavior, a reduced model

multiple regression was used to assess whether any of the independent variables would

have statistical significance in predicting risky sexual behavior based on gender. This

reduced model found that internal sexual control (ISC) is the only independent variable

that is significant in predicting risky sexual behavior for males. The reduced model for

females found no independent variables to be significant.

The finding that there are gender differences, particularly that women are slightly

less likely to engage in risky sexual behavior, is supported by previous research.

However, the study's lack of further significant findings is both surprising and

disappointing. The study failed to indicate what other relationships existed among

gender, gender attitudes, sexual self-esteem, and risky sexual behavior. In particular, the

study was unable to isolate which of the independent variables that defined gender

attitudes and sexual self esteem were significant in predicting risky sexual behavior. The

independent variables that represented gender attitudes within the study were not found to









be significant in any model used and only one of the six sexual self-esteem variables,

internal sexual control, came up as significant for males only. Internal sexual control,

defined as a belief that one's sexuality is within one's own control, may have come up as

significant for predicting risky sexual behavior in males because of gender socialization

and cultural reinforcement that men are able to maintain control in situations requiring

interaction and negotiation with others. In the realm of sexuality, this supports findings

in previous research that men are more confident and able to get sexual needs met.

Internal sexual control in males may also lead to sexual risk-taking to get sexual needs

met in a way that indicates assumptions of invincibility, which is more characteristic of

younger individuals. Much of the previous research relating self-esteem to risky sexual

behavior is in terms of global self-esteem, rather than a specific construct of sexual self-

esteem, which may explain the lack of findings in the current study. Also, the

questionnaire utilized to define sexual self-esteem, was meant to define sexual self-

concept, with sexual esteem being only one factor of sexual self-concept.

Despite the lack of significant findings in this study, the model itself was not

problematic. The scales that the independent variables were taken from were reliable.

The model did what it was supposed to, which was to assess the independent variables to

see if they were significant in predicting risky sexual behavior. The results that were

expected were simply not found, except for the fact that gender affected the likelihood of

engagement in risky sexual behavior. However, the series oft-tests that were performed

based on gender means showed some very interesting patterns. Women's t-test scores for

several of the independent variables were contrary to what previous research suggests.

The women in the current study were more likely to accept traditional stereotypes,









endorse traditional family roles, believe in the differentiation of work roles, and oppose

women's rights. In fact, the t-test scores for these four variables were found to have

statistical significance. Thus, a practical interpretation of these results suggests that the

women in this study were overall more traditional in their gender attitudes and beliefs.

Given the findings of the t-tests, the lack of findings from the overall model used

in the study and what previous research has found, it is important to look at the social and

cultural underpinnings of the results. The study found less traditional gender attitudes

among the men rather than the women. Previous research suggests that women generally

have more liberal gender attitudes than men (Vanwesenbeeck et al., 1998; Shearer et al.,

2005) even though they are less likely to engage in risky sexual behavior (DeGaston &

Weed, 1996). In fact, previous research has not only shown that men, who purportedly

have more traditional gender attitudes, engage in more risky sexual behavior, but

individuals who claim to uphold traditional and/or conservative gender beliefs also

engage in higher rates of risky sexual behavior (Shearer et al., 2005). This finding from

previous research goes against many sociocultural assumptions that being liberal in

gender attitudes and lifestyle choices is more indicative of risky sexual behavior. What

implications can be made about this research? When considering the traditional nuclear

family and who was given a voice in the family, traditional gender attitudes supported

women's deference to their husbands, including the topic of safe sexual practices.

Rosenthal et al. (1991) saw a correlating pattern in regular partners, where the inability to

say no to sexual activities and lack of confidence in sexual need assertion increased the

likelihood of risky sexual behavior. Women's belief systems have fluctuated over the

years. In present day, women's roles, sociocultural beliefs, and gender attitudes matter









and are relevant to the current findings. What once defined liberal feminism as radical

beliefs is now seen as basic to the rights of women in present day (Liss et al., 2000). A

possible, if not likely, interpretation of feminism's importance on gender attitudes and

how it may affect sexuality and sexual behavior is that it both makes women more

independent in their decisions regarding gender and sexuality yet may encourage a

backlash to liberal feminist beliefs due to negative stigma endorsed by the media.

There are several limitations to the current study. First, the respondents in the

study were obtained from a college population. This may make it difficult to generalize

to other populations. Second, it is possible that a sample bias exists in that the sample

was taken from two different courses in which the students elected to take the course.

The participants were not asked what their major was, which could create a bias and limit

the generalizability of the findings. Third, the participants came from a convenient

sample. The two subject classes that participants were recruited from were Stress

Management and Interpersonal Communication Skills, which are both held in the College

of Education. Fourth, other demographic information that could greatly impact study

findings, particularly hometown or regional culture and religiosity, were not used in the

study. Fifth, the four surveys that were utilized in the study were lengthy and took an

average of two hours to complete. Due to the time necessary for a participant to invest in

the study, there was a significant amount of data lost by participants not finishing the

surveys. Sixth, in relation to lost data, mean replacement for missing data also had an

impact on the findings. Specifically, mean replacement decreases the variance of the

sample. Seventh, the number of surveys completed and utilized for this study may be too

small for the number of variables examined. When the data was separated based on









gender, the study was especially small. When the sample was broken down by gender,

there were 50 males and 76 females. One rule of thumb for sample size is to have ten

times the number of variables assessed in the study. Using this standard, given that there

are 12 independent variables aside from gender, it would have been preferable to have a

minimum of 120 males and 120 females. Eighth, the reliability and validity of the

definition of risky sexual behavior for the purposes of this study are uncertain.

Additionally, some items within the Sexual History Questionnaire (SHQ) that were used

to define risky sexual behavior were open-ended and subject to multiple interpretations.

Ninth, items within the subscales may have been interpreted differently per each

respondent. Finally, the various subscales used to define gender attitudes and sexual self-

esteem were ineffective in uncovering gender differences in risky sexual behavior.

Additional valid measures may be needed for risky sexual behavior, as well as for the

independent variables that purportedly measure gender attitudes and sexual self-esteem.

Future research may be beneficial in several different areas. The t-tests on the

gender means, while overall a weak test in this application, still indicate potentially

significant differences on six of the subscales used which may be followed up on in later

studies: acceptance of traditional stereotypes, endorsement of family roles, belief in

differential work roles, opposition to women's rights, sexual anxiety, and power-other

sexual control. Also, future studies may benefit from shorter surveys to decrease survey

incompletion rates. If future studies regarding gender attitudes, sexual self-esteem, and

risky sexual behavior are done again, it may be beneficial to find additional reliable and

valid assessments for the particular constructs being assessed and/or increase the number

of students surveyed.









In conclusion, this study investigates the possible relationships between gender

attitudes, sexual self-esteem, and risky sexual behavior and the potential differences of

those relationships based on gender. Of the original 175 participants, only 126

participants' data sets were included due to survey incompletion. Participants completed

four questionnaires: a demographic information questionnaire, the Sexual History

Questionnaire (SHQ; Cupitt, 1998), the Gender Attitude Inventory (GAI; Ashmore, Del

Boca, & Bilder, 1995), and the Multidimensional Sexual Self-Concept Questionnaire

(MSSCQ; Snell, 2001). Findings showed that while an overall multiple regression was

statistically significant in predicting risky sexual behavior, the regression indicated the

only independent variable that was significant in predicting risky sexual behavior was

gender. Additional findings are helpful in consideration of future directions of research.














APPENDIX
DEFINITIONS OF INDEPENDENT VARIABLES

ATS: Acceptance of Traditional Stereotypes refers to the belief that men and

women are different, such that women in some ways are considered inferior to men. An

example of this is the idea of men being more independent than women, or that women

are weak in comparison to men.

DWR: belief in Differential Work Roles refers to the separation of men and women

in careers based on gender due to the assumption that one gender is less capable of

engaging in certain tasks, such as restricting a woman from getting a job that is

considered too physically difficult or labeling an occupation as "women's work" or

"men's work".

EFR: Endorsement of Family Roles refers to the belief that women are better

equipped to stay at home to raise children than men are and that men should take the lead

in family decision making.

FCS: disapproval of Female Casual Sex refers to disapproving of a woman's sexual

relations outside of a [monogamous] loving relationship.

FSI: disapproval of Female Sexual Initiative refers to the belief that it is not alright

for a woman to take the first step in dating or engaging in sexual activities.

ISC: Internal Sexual Control is the belief that the sexual aspects of one's life are

determined by one's own personal control.

OWR: Opposition to Women's Rights refers to belief that the women's liberation

movement and equal rights laws are unnecessary.









PSC: Power-other Sexual Control is the belief that the sexual aspects of one's life

are controlled by others who are more powerful and influential than oneself.

SA: Sexual Anxiety is the tendency to feel tension, discomfort, and anxiety about

the sexual aspects of one's life.

SE: Sexual Esteem is the general tendency to positively evaluate one's own

capacity to engage in healthy sexual behaviors and to experience one's sexuality in a

satisfying and enjoyable way.

SS: Sexual Satisfaction is the tendency to be highly satisfied with the sexual

aspects of one's life.

SSE: Sexual Self-Efficacy is the belief that one has the ability to deal effectively

with the sexual aspects of oneself.















LIST OF REFERENCE


American Social Health Association. (August 2005). Learn about STIs/STDs. Available
from http://www.ashastd.org/lear/ learn_statistics.cfm

Ashmore, R. D., Del Boca, F. K., & Bilder, S. M. (1995). Construction and validation of
the Gender Attitude Inventory, a structured inventory to assess multiple
dimensions of gender attitudes. Sex Roles, 32(11/12), 753-785.

Beck, D. B. (1998). The fword: How the media frame feminism. NWSA Journal, 10(1),
139-153.

Bowman, D. H. (2004). Abstinence-only debate heating up. Education Week, 23(22), 1-
10.

Breakwell, G. M., & Millward, L. J. (1997). Sexual self-concept and sexual risk-taking.
Journal ofAdolescence, 20, 29-41.

Burnett, J. W., Anderson, W. P., & Heppner, P. P. (1995). Gender roles and self-esteem:
A consideration of environmental factors. Journal of Counseling & Development,
73, 323-326.

Caldas, S. J. (1994). Teen pregnancy. Phi Delta Kappan, 75(5), 402-406.

Centers for Disease Control & Prevention. (August 2005). Divisions ofHIV/AIDS
Prevention Basic Statistics. Available from http://www.cdc.gov/hiv/stats.htm

Cupitt, C. (1998). Sexual History Questionnaire. In C. M. Davis (Ed.), Handbook of
sexuality-related measures (pp. 106-108). Thousand Oaks, CA: Sage
Publications.

DeGaston, J. F., & Weed, S. (1996). Understanding gender differences in adolescent
sexuality. Adolescence, 31(121), 217-230.

Florida Department of Health. (August 2005). HIV/AIDS 2004 Fact Sheets. Available
from http://www.doh.state.fl.us/Disease_ctrl/aids/updates/facts/facts.html

Gaynor, P. A., & Underwood, J. K. (1995). Conceptualizing and measuring sexual self-
esteem. In P. E. Shrout, & S. T. Fiske (Eds.), Personality research, methods, &
theory: Afestschrift honoring Donald W. Fiske (pp. 333-347). Hillsdale, NJ:
Erlbaum Associates.










Gil, S. (2005). Personality traits and coping styles as mediators in risky sexual behavior:
A comparison of male and female undergraduate students. Social Behavior and
Personality, 33(2), 149-158.

Gudelunas, D. (2005). Talking taboo: Newspaper advice columns and sexual discourse.
Sexuality & Culture, 9(1), 62-87.

Halpern-Felsher, B. L., Kropp, R. Y., Boyer, C. B., Tschann, J. M., & Ellen, J. M. (2004).
Adolescents' self-efficacy to communicate about sex: Its role in condom attitudes,
commitment, and use. Adolescence, 39(155), 443-456.

Hollar, D. S., & Snizek, W. E. (1996). The influences of knowledge of HIV/AIDS and
self-esteem on the sexual practices of college students. Social Behavior and
Personality, 24(1), 75-86.

Holmbeck, G. N., Crossman, R. E., Wandrei, M. L., & Gasiewski, E. (1994). Cognitive
development, egocentrism, self-esteem, and adolescent contraceptive knowledge,
attitudes and behavior. Journal of Youth & Adolescence, 23(2), 169-194.

Kinser, A. E. (2004). Negotiating spaces for/through third-wave feminism. NWSA
Journal, 16(3), 124-153.

Kinsey Institute, The. (August 2005). Mission Statement. Available from
http://www.indiana.edu/-kinsey/

Lenskyj, H. (1990). Beyond plumbing and prevention: Feminist approaches to sex
education. Gender & Education, 2(2), 217-231.

Liss, M., Hoffner, C., & Crawford, M. (2000). What do feminists believe? Psychology of
Women Quarterly, 24, 279-284.

Lesser, J., Koniak-Griffin, D., & Anderson, N. L. R. (1999). Depressed adolescent
mothers' perceptions of their own maternal role. Issues in Mental Health Nursing,
20, 131-149.

Lucke, J. C. (1998). Gender roles and sexual behavior among young women. Sex Roles,
39(3/4), 273-297.

Mayers, K. S., Heller, D. K., & Heller, J. A. (2003). Damaged sexual self-esteem: A kind
of disability. Sexuality and Disability, 21(4), 269-282.

Milhausen, R. R., & Herold, E. S. (1999). Does the sexual double standard still exist?
Perceptions of university women. The Journal of Sex Research, 36(4), 361-368.









National Campaign to Prevent Teen Pregnancy. (August 2005). Teen pregnancy -So
what? Available from http://www.teenpregnancy.org/ whycare/sowhat.asp

New Webster's Dictionary and Thesaurus of the English Language. (1993). Danbury,
CT: Lexicon Publications, Inc.

Norris, M. (2004). A crowd of nearly 20,000 rally against marriage equality. Lesbian
News, 29(11), 16.

Nzioka, C. (2004). Unwanted pregnancy and sexual transmitted infection among young
women in rural Kenya. Culture, Health, and Sexuality, 6(1), 31-34.

Rinaldo, D. (2002). The tough life of a teen mom. Scholastic Choices, 17(5). Retrieved
August 25, 2005, from Academic Search Premier database.

Rosenthal, D., Moore, S., & Flynn, I. (1991). Adolescent self-efficacy, self-esteem and
sexual risk-taking. Journal of Community & Applied Social Psychology, 1, 77-88.

Seal, D. W., & Agostinelli, G. (1996). College students' perceptions of the prevalence of
risky sexual behavior. AIDS Care, 8(4), 453-467.

Shearer, C.L., Hosterman, S. J., Gillen, M. M., & Lefkowitz, E. S. (2005). Are traditional
gender role attitudes associated with risky sexual behavior and condom-related
beliefs? Sex Roles, 52(5/6), 311-324.

Snell, Jr., W. E. (1998). The Multidimensional Sexual Self-Concept Questionnaire. In C.
M. Davis (Ed.), Handbook ofsexuality-related measures (pp. 521-524). Thousand
Oaks, CA: Sage Publications.

Spencer, J. M., Zimet, G. D., Aalsma, M. C., & Orr, D. P. (2002). Self-esteem as a
predictor of initiation of coitus in early adolescents. Pediatrics, 109(4), 581-584.

Starkman, N., & Rajani, N. (2002). The case for comprehensive sex education. AIDS
Patient Care and STDs, 16(7), 313-318.

Vanwesenbeeck, I., Bekker, M., & van Lenning, A. (1998). Gender attitudes, sexual
meanings, and interactional patterns in heterosexual encounters among college
students in the Netherlands. The Journal of Sex Research, 35(4), 317-327.

Young, M., Denny, G., & Spear, C. (1999). Area specific self-esteem and adolescent
sexual behavior. American Journal of Health Studies, 15(4), 181-188.















BIOGRAPHICAL SKETCH

Autumn Lampinen graduated cum laude in 2002 from the University of Florida

with a Bachelor of Science in psychology and a Bachelor of Arts in criminology. She is

currently completing requirements for a Master of Arts in Education and Specialist in

Education in mental health counseling at the University of Florida.

Autumn's research interests include sexuality, self-esteem, feminism, substance

abuse, and criminal rehabilitation. She has been a member of Chi Sigma Iota, Beta

Chapter, and the Counselor Education Student Association. She also holds membership

in the American Counseling Association. Autumn is currently awaiting publication of an

article in which she has secondary authorship, "Teaching Family Systems Theory through

Service-Learning."