This item has the following downloads:
1 SEXUAL RISK TAKING IN TOURISM By YELYZAVETA BERDYCHEVSKY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSO PHY UNIVERSITY OF FLORIDA 2013
2 2013 Y elyzaveta Berdychevsky
3 To my family
4 ACKNOWLEDGMENTS This dissertation came to fruition due to the help of many talented scholars and great people whom I would l ike to thank for being there for me when I needed them the most. I am especially grateful to my advisor and committee chair, Prof. Heather Gibson, who has been a true inspiration and an invaluable source of guidance, support, and advice throughout the exci ting journey of conducting this study. I would also like to thank my committee members, Prof. Mirka Koro Ljungberg, Dr. Kelly Semrad, Dr. Charles Gattone and Prof. Michael Sagas, who have provided helpful and constructive feedback as well as shared their expertise to make this dissertation better. Additionally, I would like to thank Dr. Galit Nimrod, Dr. Yaniv Poria, and Prof. Douglas Kleiber for their support, encouragement, and advice on multiple aspects of my research endeavors and development as a res earcher. Likewise, I would like to thank Prof. James Algina for his prompt and ample help with the implementation of the exploratory structural equation modeling in this study. I would like to express my appreciation to the UF Registrar for providing the s ampling frame for this study. I am also grateful to Eric Friedheim Tourism Institute for the Bill Simms Endowment Doctoral Student Research Award that partially funded this study. Likewise, I would like to thank the women who voluntarily participated in this study and shared their valuable perspectives on the researched phenomenon. Finally, I would have never achieved anything without the love and support of my family. My husband and my parents have always believed in me and motivated me. I am very gratef ul to my family for their limitless support and I dedicate this dissertation to them.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 11 Problem Statement ................................ ................................ ................................ .................. 14 Theoretical Framework ................................ ................................ ................................ ........... 18 Purpose Statement ................................ ................................ ................................ .................. 22 Research Questions ................................ ................................ ................................ ................. 23 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 25 Sensation Seeking and Sexual Risk Taking ................................ ................................ ........... 25 Gender, Sexual Double Standards, and Sexual Risk Taking ................................ ................. 29 Gender and Sexual Risk Taking in Adolescence and Young Adulthood ............................... 35 Sexual Behavior in Tourism ................................ ................................ ................................ ... 40 S ex Tourism ................................ ................................ ................................ ..................... 40 Sex in Tourism ................................ ................................ ................................ ................ 44 ................................ ................................ ........... 49 Summary ................................ ................................ ................................ ................................ 54 3 METHODS ................................ ................................ ................................ ............................. 56 Mixed Methods Research ................................ ................................ ................................ ....... 56 Legitima tion ................................ ................................ ................................ .................... 59 Phase 1 Qualitative Method Transcendental Phenomenology ................................ ......... 62 Data Collection ................................ ................................ ................................ ................ 66 Participants ................................ ................................ ................................ ...................... 69 Data Analysis ................................ ................................ ................................ ................... 70 Epoche and phenomenological reduction ................................ ................................ 71 Imaginative variation ................................ ................................ ................................ 73 Trustworthiness ................................ ................................ ................................ ............... 76 Phase 2 Quantitative Method ................................ ................................ ............................... 77 Instrument Construction ................................ ................................ ................................ .. 77 Instrument Validation ................................ ................................ ................................ ...... 79 Measurement ................................ ................................ ................................ ................... 80 Data Collection ................................ ................................ ................................ ................ 81 Sampling ................................ ................................ ................................ .......................... 82 Sample ................................ ................................ ................................ ............................. 83
6 Data Analysis ................................ ................................ ................................ ................... 85 4 FINDINGS ................................ ................................ ................................ .............................. 92 Findings from the Qualitative Phase ................................ ................................ ....................... 92 Composite Textural Description ................................ ................................ ...................... 92 taking in tourism ................................ ............ 93 Dim ensions of sexual risk taking in tourism ................................ ............................ 94 Physical/sexual health dimension of sexual risk taking in tourism ......................... 96 Social dimensio n of sexual risk taking in tourism ................................ ................... 98 Emotional dimension of sexual risk taking in tourism ................................ .......... 102 Mental dimension of sexual risk t aking in tourism ................................ ................ 104 Cultural dimension of sexual risk taking in tourism ................................ .............. 107 Composite Structural Description ................................ ................................ ................. 110 Sex definitions and attitudes underlying perceptions of sexual risk taking ........... 111 Social control of sexual double standards underlying perception s of sexual risk taking ................................ ................................ ................................ ........... 113 Young age underlying perceptions of sexual risk taking ................................ ....... 115 Facilitating and/or inhibiting factors of sexual risk taking in tourism ................... 116 Anonymity underlying sexual risk taking in tourism ................................ ............ 117 Leaving experiences behind as underlyi ng sexual risk taking in tourism .............. 118 Atmosphere and fun oriented mentality underlying sexual risk taking in tourism ................................ ................................ ................................ ................ 119 Peer pre ssure support solidarity vs. judgment underlying sexual risk taking in tourism ................................ ................................ ................................ ................ 119 Alcohol and drugs consumption underlying sexual risk taking in tourism ........... 120 Length, destination, and type of the tourist experiences underlying sexual risk taking ................................ ................................ ................................ .................. 122 Essence Statement ................................ ................................ ................................ ......... 123 Findings from the Quantitative Phase ................................ ................................ ................... 124 Research Question 1a: What Sexual Activities Are Perceived as Risky in Tourism by the Female College Students? ................................ ................................ ............... 124 Research Question 1b: What Are the Perceived Dimensions of Sexual Risk Taking in Tourism by the Female College Students? ................................ ............................ 125 Research Question 1c: Do the Perceived Dimensions of Sexual Risk Taking in Tourism Differ by Sexual Sensation Seeking Propensity among the Female College Students? ................................ ................................ ................................ ....... 129 Research Question 2: What Are the Dimensions of Motivations for Sexual Risk Taking in Tourism Among the Female College Students? ................................ ........ 138 Research Question 3a: What Are the Clusters of Sexual Risk Takers in Tourism Among the Female College Stude nts Based on the Perceptions of and Motivations for Sexual Risk Taking in Tourism? ................................ ..................... 140 Research Question 3b: What Are the Personality, Sexual, and Socio Demographic Profiles of the Clusters of Sexual Risk Takers in Tourism Among the Female College Students? ................................ ................................ ................................ ....... 143
7 Research Question 4a: What Are the Characteristics of the Tourist Experiences Facilitating and/or Inhibiting Sexual Risk T aking in Tourism Among the Female College Students? ................................ ................................ ................................ ....... 148 Research Question 4b: What Are the Perceptions of the Risk Taking Clusters Among the Female College Students Regarding the Characteristi cs Facilitating and/or Inhibiting Sexual Risk Taking in Tourism? ................................ ................... 149 5 DISCUSSION ................................ ................................ ................................ ....................... 165 Taking in Tourism ................................ ................... 165 Perceived Dimensions of Sexual Risk Taking in Tourism across the SSS Levels .............. 172 sk Taking in Tourism ................................ ................. 176 Clusters and Profiles of Sexual Risk Takers in Tourism among Women ............................ 183 Facilitating and Inhibiting Char acteristics of Sexual Risk Taking in Tourism .................... 190 Limitations and Delimitations ................................ ................................ .............................. 197 Recommendations for Future Research and Pract ical Implications ................................ ..... 20 0 Conclusions ................................ ................................ ................................ ........................... 203 APPENDIX A INFORMED CONSENT FOR THE QUALITATIVE PHASE ................................ ........... 207 B INTERVIEW GUIDE FOR THE QUALITATIVE PHASE ................................ ................ 209 C EMAIL INVITATION TO PARTICIPATE IN THE QUANTITATIVE PHASE .............. 210 D INFORMED CONSENT FOR THE QUANTITATIVE PHASE ................................ ........ 211 E INSTRUMENT FOR THE QUANTITATIVE PHASE ................................ ....................... 213 LIST OF REFERENCE S ................................ ................................ ................................ ............. 222 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 239
8 LIST OF TABLES Table page 3 1 Socio demographic chara cteristics of the participants in the qualitative phase ................ 89 3 2 Socio demographic characteristics the participants in the quantitative phase ................... 90 3 3 Sexual characteristics of the participants in the quantitative phase ................................ ... 91 4 1 Perceptions and experiences of sexual risk taking on vacation ................................ ....... 153 4 2 EFA solution for the sexual risk perceptions model ................................ ........................ 154 4 3 Summary of goodness of fit statistics for all models with sexual sensation seeking (SSS) index a s the grouping variable ................................ ................................ ............... 155 4 4 Sexual sensation seeking (SSS) levels ................................ ................................ ............. 156 4 5 Perceived sexual risk dimensions for lower and hig her levels of Sexual Sensation Seeking (SSS) based on MGI1 ................................ ................................ ........................ 157 4 6 EFA solution for the motivations model ................................ ................................ .......... 158 4 7 Classification of sexual risk takers in tourism ................................ ................................ 159 4 8 Sensation seeking profile of five clusters ................................ ................................ ........ 160 4 9 Sexual profile of five clusters ................................ ................................ .......................... 161 4 10 Socio demographic profile of five clusters ................................ ................................ ...... 162 4 11 Facilitating and inhibiting touristic characteristics of sexual risk t aking in tourism ....... 163 4 12 Clusters' perceptions of facilitating and inhibiting touristic characteristics .................... 164
9 Abstract of Dissertation Presented to the Gradua te School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SEXUAL RISK TAKING IN TOURISM By Y elyzaveta Berdychevsky August 2013 Chair: Heather J. Gibson Major: Health and Human Performance Tourism research suggests that sexual risk taking tends to be more prevalent in tourism environments. Tourism has been described as a context condoning higher degrees of sexual freedom, allowing for situational disinhibition, and ren dering sexual double standards less pervasive than in everyday life. The purpose of this study was to explore the links between tourism, risk taking, and sexual behavior among female college students. This study used a sequential mixed methods design, star ting with a qualitative and proceeding to a quantitative phase. The qualitative data were analyzed using a phenomenological lens. Fifteen individual, 1.5 2.5 hours long interviews were conducted with 13 women aged 19 24 years. The quantitative phase drew u pon the findings from the qualitative phase and relevant literature. The data were collected using an online survey, resulting in a sample of 853 university women (mean age 23.5 years). Both qualitative and quantitative findings revealed that sexual risk taking in tourism is a multidimensional and complex phenomenon. The qualitative phase identified five universal and common dimensions of sexual risk taking in tourism for the women, including physical/sexual health, social, emotional, mental/self percepti on, and cultural. The quantitative phase facilitated statistical exploration of sexual risk taking in tourism with a larger sample and identified three
10 dimensions: physical/sexual health, mental/emotional, and socio cultural. Both qualitative and quantitat taking in tourism environments. The quantitative results suggested that this variety of motivations can be grouped into three dimensions: anonymity/detachment/experimentation; safe exploration/thrill/ empowerment; and fun/less inhibitions/opportunity. In the quantitative phase, the women were clustered based on their perceived dimensions of sexual risk taking and motivations for it. Five clusters were identified: high and diver se risk perceivers motivated by anonymity and exploration; moderate and diverse risk perceivers motivated by fun; low risk perceivers with diverse motivations; low risk perceivers motivated by exploration; and unmotivated, high and diverse risk perceivers. The importance of this study stems from the development of a deeper understanding of sexual risk taking in tourism. In turn, this increased understanding may provide directions for health programs and/or information campaigns aimed at addressing sexual ri sk behaviors among young women.
11 CHAPTER 1 INTRODUCTION Sex is an important part of human life and the link between sex and tourism has increasingly received scholarly attention Sex and tourism historically go hand in hand ( Hart & Hawkes, 2000; Littlewoo d, 200 2 ) and Bauer and McKercher (2003) claim ed tourism, considering that the sun, sea, sand, and sex became hallmarks signifying the essence of to urist experiences in the 1980s and 1990s both in popular culture and in the academic literature ( e.g., Crick, 1989; Hobson & Dietrich, 1994), sex cannot be ignored as a component of tourist experiences. Additionally, tourism has been conceptualized as a l iminoid sphere characterized by a sense of anonymity and transition, absence of everyday status distinctions and communitas, as well as freer behavior in general, and sexual behavior in particular (Lett, 1983; Turner, 1974). According to Littlewood (200 2 ), tourism is characterized by the supremacy of the which might be related to different patterns of sexual behavior in tourism vis vis everyday life Yet, f or many years researchers neglected the link between sexual activity and soci o psychological contexts in which this activity occurs. This fact is surprising considering sought but within the social context and discourses surrounding those acts. Similarly, Black (2000) suggest ed that the sexual behavior of an individual might alter in tourism, as one of the social influences on behavior (i.e., location) is shifted. Finally, some authors provide empirical evidence suggesting that tourist contexts constitute unique social arenas influencing sexual behavior in general and encouraging sexual risk taking in particular (Andriotis, 2010; Apostolopoulos, Snmez, & Yu, 2002; Eiser & Ford, 1995; Thomas, 2005).
12 I n the tourism literature commercial sex became the dominant focus in the 1990s, with Ryan, 2000; Ryan & Hall, 2001 the mid 1990s ( Pruitt & LaFont, 1995 ). These foci, whi le contributing important insights, have downplayed the incidence and importance of the sexual behavior among consent ing adults engag ed in sexual activity devoid of any commercial transaction (Josiam, Hobson, Dietrich, & Smeaton, 1998; McKercher & Bauer, 2 003) O ne exception to this trend was the study of casual sex and condom use among students on spring break (Apostolopoulos et al. 2002; Litvin, 2010; Maticka Tyndale & Herold, 1999 ; Mewhinney, Herold, & Maticka Tyndale, 1995). Yet, there are multiple typ es of tourist experiences that may exert different influences on sexual behavior in general and sexual risk taking in particular. Indeed, the existing literature on sex tourism provides a number of parameters that could be equally important in clarifyin g sexual risk taking in tourism. For example, the role of sex as a motiv e to get involved in tourist experience s (McKercher & Bauer, 2003; Oppermann, 1999) the length of time spent together and the relationship between sexual partners (Oppermann, 1999) the nature of the sexual experience (McKercher & Bauer, 2003; Ryan, 2000; Ryan & Hall, 2001) and the role of the tourism industry as a facilitator of sexual encounter s (McKercher & Bauer, 2003) However, t his tendency to focus on sex tourism does not do justice to the complexity of the meanings, perceptions, motivations, and potential risks involved in non commercial sexual behavior in tourism (Berdychevsky, Poria, & Uriely, 201 3a ). Moreover, McKercher and Bauer ( 2003) have argued that the relationship between touri sm and sex cannot be narrowly confined to the scope of the commercial trade. Likewise, Carter and Clift (2000) called to distinguish being a broader term including sexual behavior not involving a commercial transaction. Understanding tourists non commercial
13 sexual activity offers a different perspective on the interpersonal sexual dynamic in tourism compared to commercial sex tourism (Berdychevsky et al., 201 3a ), which is typically bound with a pow er differential between tourists and locals (Cohen, 1982; Cabezas, 2004; Snchez Taylor, 2001, 2006). In addition, the combination of sexual behavior between tourists themselves and the influence of gender relations and stereotypes ha s received little atte ntion (Berdychevsky, Poria, & Uriely, 2010), even in one of the most recent edited collections on sexual issues in the fields of tourism and leisure (Carr & Poria, 2010). Yet, gender relations are pivotal to understanding tourist experiences (Gibson, 2001; Swain, 1995) and their influence becomes even more crucial a whole host of taboos, prejudices, silence, and gendered double standards (Foucault, 1976). Alt hough, attempts have been made to distinguish between sex tourism and the of romance tourism ( Herold, Garcia, & DeMoya, 2001; Pruitt & LaFont, 1995 ), as well as to taking in tourism ( Ragsdale, Difrancheksco, & Pinkerton, 2006; Thomas, 2000, 2005; Wickens, 1997 ), there is still a need for an investigation of non commercial sexual risk taking in tourism and its potential links to gender relations Furthermore, Elsrud (2001) cal led for gender sensitive research of risk in tourism as it is related to narrating identity, and women have to negotiate conflicting ideals being caught between opportunity and tradition. Namely, on the one hand, risk taking in tourism can be an emancipato ry practice associated with feeling a strong, powerful, self reliant, and an exciting person. On the other hand, the risk and adventure narrative of travel has been historically considered as a masculine domain welcoming men and excluding women (Elsrud, 20 01).
14 Therefore, gender is an important variable for consideration in any investigation of sexual risk sexual risk taking in tourism, considering the effects of s exual double standards and anonymous tourist environments ( Eiser & Ford, 1995 ; Maticka Tyndale, Herold, & Oppermann, 2003 ; Mewhinney et al. 1995 ) perspectives are crucial to the understanding of thrill seeking and risk taking in tourism. She found that men are more prone to risk taking in tourism compared to women and that the early adulthood stage (17 45 years), especially the novice phase (17 33 years), are the pe ak eras for thrill seeking and risk taking in tourism. Following ; Levinson, Darrow, Klein, Levinson, & McKee, 1978 ) adult life cycle approach, Gibson as well as Gibson and Yiannakis (2002) suggested that adventurousness, avoidanc e of commitments, experimentation, and sense of immortality characterizing the transition into early adulthood complement and facilitate risk taking in tourism. Additionally, the lack of commitments associated with freedom to take risks and feelings of imm ortality contribute to a perceptual downplaying of the risks involved. These assumptions have informed the choice to focus in this dissertation on the particular life stages of late adolescence and young adulthood i.e., the college age. This is a decisio n that has precedence in the tourism literature on perceived risk and tourism (e.g., Lepp & Gibson, 2003, 2008). Problem Statement Tourism is associated with the geographical expansion of sexually transmitted diseases STDs (Clift & Forrest, 2000), as sex ual risk taking tends to be more prevalent in tourism environments compared to everyday life (Black, 1997). Therefore, one of the critiques of the various national surveys of sexual behavior and STDs is their failure to collect separate data on
15 respondents behavior in the context of international travel has drawn the attention of tourism researchers (Apostolopoulos et al., 2002; Black, 1997; Bloor et al., 2000; Thomas, 2000). Ho wever, the tendency to focus on commercial sex tourism has rendered non commercial sex in tourism and its potential health consequences with very limited research attention. A nuanced understanding of sexual risk taking is crucial considering sexually tra nsmitted diseases (STDs) in general, and HIV/AIDS in particular, which became pandemic in the 20th century and are of continued relevance today. According to AVERT (an international HIV and AIDS charity), about 33.3 million people worldwide were living wit h HIV/AIDS in 2009, with 2.6 million new infections per year. Furthermore, HIV/AIDS is a cause of 1.8 million annual deaths (AVERT, 2009). Based on the Centers for Disease Control and Prevention (CDC, 2007), around 19 million new cases of STDs occur in the US every year, with about half of infections among younger people ( 15 to 24 years ). Padian, Shiboski, Glass, and Vittinghoff (1997) argued that women are at particular risk, since the male to female HIV transmission was assessed to be approximately eight times greater than female to male transmission. In their earlier study, Padian, Shibowski, and Jewell (1991) found the incidence of female to male transmission was 1% compared to 20% for male to female transmission. Broaddus, Morris, and Bryan (2010), desc infections in women are caused by heterosexual unprotected sexual contact. Strikingly, a nationally representative CDC (2008) study revealed that one in four adolescent women (i.e., 26% = 3.2 million girls, 14 to 19 years) was infected at least once with some STD. The majority of the existing body of knowledge about HIV/AIDS has taken either medical and/or largely social perspectives. While the value of this literature should not be
16 underestimated, the dearth of research from interpersonal and/or personal perspectives, involving the concepts of age and life stage, gender, and sexuality, is problematic (Emmers Sommer & Allen, 2005). Understanding sexual risk taking requir es a consideration of gender and life stage that shape sexual behavior (Carpenter & DeLamater, 2012; CDC, 2007, 2008). This assumption, taking in adolescence constitutes a priority for public h ealth in the United States, has guided the focus of this dissertation on taking. Moreover, discussions of the consequences of sexual risk taking in tourism typically revolve mainly around STDs (Bloor et al., 2000; Cli ft & Forrest, 2000; Hart & Hawkes, 2000), while potential physical, mental, social, and emotional consequences can be more diverse. For instance, Zuckerman (2007) opened his chapter on sex and sensation seeking with the statement that safe sex outside a co mmitted relationship is an oxymoron, and suggested considering not only physical, but also psychological risks such as fear of losing reputation and/or feeling used among women and fear of inadequate performance and/or commitment among men. Therefore, a br oader and context specific understanding of sexual risk taking is warranted, leading to the adoption of a broad definition of sexual risk taking in this study. While most of the literature on sexual risk taking has focused on the behavioral aspects (e.g., sex with a stranger or sex without a condom) and their physical and epidemiological health consequences, this study considered psychological and sociological risks associated with sex as well, both in steady and casual relationships. Specifically, the gap with respect to sexual risk taking in tourism is threefold. First, taking in tourism and its consequences are in need of investigation. Considering the effect of the tourist social atmosphere that makes peopl e feel
17 perceptions of risk and its consequences might become distorted during their tourist experiences. Thus, the success of any endeavor to address sexual risk taking in tourism is conditioned on taking in tourism are in need of investigation. In this respect, considering the double standards of gender appropriate sexual behavior that are much more restrictive for women than for men (Jonason & Fisher, 2009; Maticka Tyndale et al., 2003; Rathus, Nevid, & Fichner Rathus, 1993), the impact of anonymous tourism environments (with lessened social contr ol encouraging sexual risk behaviors) might be particularly influential for women (Eiser & Ford, 1995; Mewhinney et al., 1995). Indeed, double standards of gender appropriate sexual behavior become somewhat subdued in the sexual scripts on vacation allowin g women a certain degree of experimentation (Maticka Tyndale & Herold, 1997) that might be related to risk taking. Likewise, Selnniemi (2002) found that women on vacation are more likely than men to report transgressive behaviors because such behaviors ar e more novel to them. Additionally, Ragsdale et al. (2006) claimed that there is an urgent need to opportunities for self exploration and self discovery denied to them in the ir home environment. Finally, research on solo female travelers illustrates the potential of negotiating gender stereotypes and constraints during tourist experiences as a source of empowerment, self making, independence, and emancipation (Jordan & Gibson, 2005; Obenour, 2005). Third, the characteristics of touristic social environments as contexts for sexual risk taking are in need of investigation. Researchers claim that tourist experiences provide a unique social atmosphere characterized by anonymity (Sh ields, 1990), liminality and liminoidity (Lett,
18 facilitating a nd inhibiting factors of sexual risk taking in tourism is of crucial importance. To conclude, the perception of sex as not serious or indecent, and the difficulties in researching such a sensitive topic, may explain the lack of research on sexual behavior in tourism (Poria & Carr, 2010; Turner & Rubinson, 1993). Perhaps, the sentiments of the Victorian and Puritan ethics still overshadow sex related matters rendering them as a taboo topic. These pervasive attitudes may have contributed to the situation wher e sex is considered to be marginal and unworthy of scholarly attention. Indeed, the topic of sexual risk taking in tourism should not be trivialized as sexual risk taking on vacation may have serious mental and physical health consequences that may follow people to their everyday lives. Theoretical Framework Any focus on sexual behavior and risk in tourism warrants careful attention to the conceptualization s of tourist experience s, sexual behavior and risk. For this dissertation, broad and pluralistic fram eworks are deemed to be appropriate due to the implementation of transcendental phenomenology in the qualitative phase. Transcendental phenomenology implies subjective exp eriences. As such, it is suggested that a pluralistic framework with technical elements proposed by Poria, Butler, and Airey (2003) for conceptualizing tourist experiences may allow adequate breadth and flexibility to capture the nderstandings. Poria et al. suggest capturing tourist experiences in terms of the following aspects: (1) activity i.e., whether tourism is perceived as recreation or a non recreation al activity; (2) space i.e., whether the experience is taking place i n the home context or at the tourist destination, regardless of the geographical distance between them; and (3) time i.e., whether time during
19 the experience is perceived as leisure and free time vs. non leisure and non free time. The interaction between experiences, activities, as well as the space and time frames they act in. Understanding s exual behavior in tourism also requires a broad perspective in this dissertation. Oppermann ( 199 8, 1999) argued that the narrow definition of sex as a penile vaginal intercourse is inadequate for approaching sexual behavior in tourism, since it omits a wide variety of relevant sexual behaviors. Thus, he suggested understanding sexual behavior in tour ism as any activity with sexual overtone, including activities with and without penetration. In line with this approach, a be accommodated, such as oral sex, anal sex, masturbation, ac tivities associated with voyeurism and exhibitionism (Oppermann, 1998, 1999; Ryan & Martin, 2001), petting (sexual touching), Tyndale & Herold, 1997, p. 319). As for ca sual sex, while the specifics will be defined by the participants, it will be generally approached as a sexual encounter between either strangers or brief acquaintances with or without sexual intercourse, typically, occurring once and not involving expecta tions of a further relationship (Paul, McManus, & Hayes, 2000). Risk in tourism contexts will be understood in this dissertation according to the flexible framework suggested by Elms (1998) and adapted to tourism by Ryan (2003) involving three major compon ents of context, likelihood, and consequence. First, Ryan emphasized the uniqueness of tourism environments (that will be further elaborated in the section Sexual be comes psychologically predisposed toward risk taking, hedonistic behaviors, and loss of routine responsibilities (p. 56). The second factor of likelihood refers to the technical probability
20 component and even more importantly to the subjective perceptual c omponent, as a social psychological construct, where tourists exercise choice and decide to take risks to gain benefits if the perceived rewards of risk taking outweigh the costs i.e., and the likelihood and magnitude of loss. Lastly, the consequence com ponent refers not only to loss, but also to reward whereas, for tourist risk takers rewards are typically psychological in nature, including excitement, adrenalin, and potential immediate and/or long term contribution to sense of well being. Additionally, sensation seeking theory will guide the quantitative phase of the dissertation. Personality characteristics, in addition to age and gender, are essential factors to the understanding of adolescent risk taking, as well as significant predictors of it (Gullo ne & Moore, 2000). Additionally, personality characteristics are frequently explored with respect to sexual risk taking and the trait of sensation seeking is one of the most common (Turchik, 2007). by the seeking of varied, novel, complex, and intense sensations and experiences, and the willingness to take physical, social, legal, and financial risks for the sake of such experience (1994, p. 27; 2007, p. 49). Zuckerman (2007) self identified with the group of theorists who argue that personality traits have their roots in biological mechanisms and genes as well as their interactions with the environment. Zuckerman started his involvement with the idea of a sensation seeking trait in the early 1960s wi th the development of his first sensation seeking scale SSS (Zuckerman, Kolin, Price, & Zoob, 1964). Zuckerman et al. decided to use the optimal level of arousal (OLA) and optimal level of stimulation (OLS) theories as a basis for the first version of th e questionnaire, since they hypothesized that high sensation seekers would: (1) have higher OLA OLS, (2) underestimate the risks or judge benefits as outweighing risks, and (3) seek novel situations providing intense
21 sensations to reach higher OLA OLS. Zuc kerman (1994, 2007) argued that sensation seeking is a correlate of risk taking, but not a primary motive for it. In other words, sensation seekers accept risk as a possible price for achieving arousal, but they are not attracted to risk for its own sake. The sensation seeking scale has been shown by numerous researchers to be valid in understanding various phenomena bound with risk taking, including sexual behavior, heavy drinking, smoking, drug use, criminal activities, gambling, reckless driving, etc. ( Greene, Krcmar, Walters, Rubin, & Hale, 2000; Roberti, 2004; Zuckerman, 1994, 2007; Zuckerman & Kuhlman, 2000). The initial version of the sensation seeking scale has been revised several times. Zuckerman (2007) described five reincarnations of the scale a nd the development of multiple additional versions for specific uses. One of the key reincarnations SSS IV and SSS V happened in the late 1970s, when Zuckerman, Eysenck, and Eysenck (1978) factor analyzed the revised items and found four dimensions/sub scales of (1) thrill and adventure seeking, (2) experience seeking, (3) disinhibition, and (4) boredom susceptibility. According to Zuckerman et al. (1978), thrill and adventure seeking refers to the desire to partake in physical activities offering unusu al and strong sensations, such us skydiving, scuba diving mountain climbing etc Experience seeking indicates pursuing new experiences and sensations though the senses, mind, and general nonconforming lifestyle. Stimulating experiences for the mind and s ense include art, music, travel, etc. Disinhibition represents seeking sensations through a hedonistic lifestyle and other people, including involvement in wild parties, drinking, sexual variety, etc. Finally, boredom susceptibility refers to aversion to m onotonous conditions and sense of restlessness when subjected to such conditions. Another key transformation was the development of the broader five factor personality questionnaire Zuckerman and Kuhlman Personality Questionnaire ZKPQ by Zuckerman,
22 Kuhlman, Joireman, Teta, and Kraft (1993), which includes a scale for impulsive sensation seeking. This scale was successfully tested with respect to risk taking in general and sexual risk taking in particular (Smith 2007; Turchik, 2007; Zuckerman & Kuhlma n, 2000). Interestingly, gender and age were found to be the two most significant demographic characteristics affecting sensation seeking (Greene et al., 2000; Roberti, 2004; Zuckerman, 1994, 2007; Zuckerman et al., 1978), which is important for this study considering its focus. Men typically scored higher than women on all the subscales (with exception of experience seeking) with the biggest gender differences in disinhibition and thrill and adventure seeking (Zuckerman, 2007). Additionally, sensation seek ing scores peaked in late adolescence and then steadily declined with age, with the most remarkable descents in thrill and adventure seeking and disinhibition (Roberti, 2004; Zuckerman, 2007). Purpose Statement In order to address the gaps described in th e problem statement, a sequential mixed methods design was implemented, starting with qualitative and proceeding to quantitative approaches to investigating th e depth and meanings of sexual risk behaviors in tourism. The qualitative phase of this dissertation provided rich data with a diversity of ideas and perspectives on the researched topic. This approach inductively established the breadth of the construct a nd contributed to the development of foundational knowledge about sexual risk taking in tourism. The second phase was quantitative drawing upon the findings from the first phase. The goal was, first, to develop and to validate a measurement scale of sexual risk taking in tourism that may serve as an instrument to both academics and professionals in the field. Second, perceptions of, and motivations for, sexual risk taking in tourism were explored with a larger sample contributing to the generalizability of the findings.
23 The purpose of the study was to investigate the link between tourism, risk taking, and sexual behavior, while looking at the role of gender relations and life course stage. The overall ptions of and motivations for sexual risk taking in tourism, as well as the characteristics of the tourist social environment associated with sexual risk taking? Specifically, the purpose of this study was threefold: 1. subjective perspectives of what constitutes sexual risk taking in tourism and its potential consequences. 2. meanings attached to them. 3. To explore the charac teristics of the tourist social environments (i.e., facilitating and/or inhibiting factors) as the contexts for sexual risk taking among female college students. Research Questions The first taki ng in tourism perspectives on what constitutes sexual risk taking in tourism and its potential consequences? In other words, what is the phenomenon of sexual ri sk taking in tourism as perceived by the female college students? The draft of the instrument for the second quantitative phase was developed based on the qualitative findings and relevant insights from the literature. The following research questions we re examined in the second phase: 1a. What sexual activities are perceived as risky in tourism by the female college students? 1b. What are the perceived dimensions of sexual risk taking in tourism by the female college students? 1c. Do the perceived dim ensions of sexual risk taking in tourism (from RQ 1b) differ by sexual sensation seeking propensity among the female college students?
24 2. What are the dimensions of motivations for sexual risk taking in tourism among the female college students? 3a. Wha t are the clusters of sexual risk takers in tourism among the female college students based on the perceptions of (from RQ 1b) and motivations for (from RQ 2) sexual risk taking in tourism? 3b. What are the personality (in terms of sensation seeking) sexu al and socio demographic profiles of the clusters of sexual risk takers in tourism among the female college students (from RQ 3a)? 4a. What are the characteristics of the tourist experiences facilitating and/or inhibiting sexual risk taking in tourism am ong the female college students? 4b. What are the perceptions of the risk taking clusters among the female college students (from RQ 3a) regarding the characteristics facilitating and/or inhibiting sexual risk taking in tourism (from RQ 4a)?
25 CHAPTER 2 LI TERATURE REVIEW The literature suggests links between personality characteristics, gender, life course stage, and sexual behavior. Therefore, the following three sections focus on (1) sensation seeking and sexual risk taking, (2) gender, sexual double stan dards, and sexual risk taking in general, and (3) gender and sexual risk taking in adolescence and young adulthood in particular. Adopting the principle of intersectionality, understanding sexual behavior and sexual risk taking, through the kaleidoscope of intersecting axes of power (e.g., g ender, age, and personality characteristics) has the potential to unveil more richness and complexity associated with sexual behavior and risk taking Further, the review proceeds to the literature on sexual behavior in tourism and the behavior in tourism specifically. Sensation Seeking and Sexual Risk Taking The link between sensation seeking and sexual risk taking can be summarized in sensations in casual sexual encounters. Furthermore, impu lsive sensation seekers can be at particular risk due to their characteristic lack of restraint in potentially pleasurable situations (Zuckerman, 2007). Some of the most implemented versions of the sensation seeking scale to the understanding and/or predic ting of sexual risk taking are: (1) the SSS V and its subscales (Zuckerman et al., 1978); (2) the ZKPQ, particularly the impulsive sensation seeking subscale (Zuckerman et al., 1993); and (3) the sexual sensation seeking scale SSSS (Kalichman, Johnson, A dair, Rompa, Multhauf, & Kelly, 1994; Kalichman & Rompa, 1995).
26 Drawing upon the SSS V or its subscales (Zuckerman et al., 1978), the following studies investigated the link between sensation seeking and sexual risk taking. Bancroft et al. (2003) studied p ersonality traits, as measured by the SSS V, and sexual risk taking among gay men (focusing on unprotected anal intercourse, oral sex, number of casual partners, and cruising behavior) and found that disinhibition was related to all of these risk behaviors and boredom susceptibility was related to casual sex. Likewise, Bancroft et al. (2004) conducted equivalent research with straight men and found that the disinhibition subscale predicted the number of sexual partners during the past year and whether sexua l intercourse took place within the last six months. Donohew et al. (2000) found strong associations between the sensation seeking trait and sexual risk taking among the US adolescents, where sexual risk taking was measured by the intentions to have sex an d actual experience, number of lifetime sexual partners, using drugs and/or alcohol before sex, using a condom, having been pregnant or caused a pregnancy, among others. Greene et al. (2000) explored adolescent risk taking behaviors and found that both sen sation seeking and egocentrism are associated with risky sexual behaviors (measured as the number of sexual partners in the past two years and six months, and the patterns of condom use). Finally, Wagner (2001) found that sensation seeking predicts risky s exual behavior among the US undergraduate students. The following studies investigated the link between sensation seeking and sexual risk taking based on the ZKPQ and its impulsive sensation seeking subscale (Zuckerman et al., sensation seeking is associated with risky sex, but does not predict it. Cohen and Fromme (2002) found that sensation seeking correlates with substance use and high risk sex, both with regul ar and new sexual partners, among young adults. Zuckerman and Kuhlman (2000) investigated
27 college students and found impulsive sensation seeking to be significantly related to sexual risk taking among the other risk taking areas such as drinking, smoking, and taking drugs. In this case, sexual risk taking was measured by the number of sexual partners throughout a and how often some method of birth control and /or condom is used. Finally, Hoyle, Fejfar, and Miller (2000) reviewed 53 studies investigating the relationships between personality traits and sexual risk taking (defined by the number of partners, unprotected sex, and high risk sexual encounters e.g. having sex with a stranger). Out of 53 studies, 38 used sensation seeking scales (mainly SSS V and ZKPQ instruments) and found this trait to be the primary predictor of sexual risk taking, correlated with all three aforementioned categories and accountin g for 64% of the effect sizes. Considering the evidence of the relationship between sensation seeking and sexual risk taking, Kalichman et al. (1994) argued for the development of a specific scale for sexual propensity to attain optimal levels of sexual well as a measure of nonsexual experience seeking, through sampling and revising the items from the instruments of Zuckerman et al. (1964, 1978). Kalichman et al. tested the scales with gay men demonstrating the internal consistency, time stability, as well as convergent, discriminant, and divergent validity of the measures. Sexual sensation seeking was consistently r elated to various measures of sexual risk taking. Kalichman and Rompa (1995) slightly revised the scales and further tested the measures with men and women, confirming their reliability, validity, and usefulness in predicting sexual risk taking.
28 In furthe r implementations, Chng and Gliga Vargas (2000) found sexual sensation seeking to be one of the significant predictors of unprotected anal sex among gay men of color. Likewise, Parsons, Bimbi, and Halkitis (2001) found sexual sensation seeking to be assoc iated with sexual compulsivity, which in turn was associated with HIV sexual risk behaviors among gay/bisexual male escorts. However, McCoul and Haslam (2001) found that sexual sensation seeking was associated with a higher frequency of unprotected sex and a higher number of sex partners among heterosexual men, but not among gay men. The sexual sensation seeking scale was also tested and validated with college student samples. For instance, Gaither and Sellbom (2003) investigated the link between sexual se nsation seeking and various patterns of sexual risk taking among college students with a particular focus on gender differences. They found that men scored significantly higher on sexual sensation seeking than women. Yet, the correlations between sexual se nsation seeking and various sexually permissive behaviors were stronger for women compared to men (Gaither & Sellbom, 2003). Wiederman and Hurd (1999) found that increased sexual sensation seeking was related to extra dyadic dating and sexual activity amon g college students (both men and women). Wiederman (1999) also explored volunteer bias (self selection bias) in sexuality research and found that, among other characteristics, the volunteers scored hirer on sexual sensation seeking. Therefore, a substanti al body of knowledge reveals that general nonsexual sensation seeking and specific sexual sensation seeking are related to various kinds of risky sexual behaviors. Furthermore, sensation seeking theory has demonstrated its relevance with respect to the tou rism related topics as well, such as tourist roles, risk perceptions, preferences in terms of traveling styles, destinations, and chosen vacation activities (Lepp & Gibson, 2008; Pizam, Reichel, & Uriely, 2002; Pizam et al., 2004). Higher sensation seekers are typically more likely
29 to be independent travellers, to prefer more active and spontaneous vacations, to have travelled internationally, to have traveled to the destinations perceived as riskier, and to be more likely to engage in risky activities on v acation. As such, sensation seeking theory was deemed to be appropriate to guide the quantitative phase of this dissertation. Gender, Sexual Double Standards, and Sexual Risk Taking Gender relations and sexual double standards are the foundational issues t o an understanding of sexual behavior and risk taking. Gender has been defined as socially constructed roles associated with biological sex, as cultural and historical expectations with respect to performing femininity and masculinity, and as a system of p ower where men have more power compared to women (Muehlenhard, Peterson, Karwoski, Bryan, & Lee, 2003). Butler (1990) stated that gender roles are culturally and discursively constructed where the socially expected roles with respect to sexuality and sexua l behavior are perpetuated via repetitive performative acts. Gender organizes most aspects of sexual behavior since men and women may adopt various socially sanctioned patterns of sexual behavior, and may experience and understand the same sexual activity differently (McCabe, Tanner, & Heiman, 2010). Social double standards of appropriate sexual behavior sexual double standards are particularly restrictive for women compared to men, setting gender specific standards of allowable sexual permissiveness ( Crawford & Popp, 2003; Eaton & Rose, 2011; Jonason & Fisher, 2009; McCabe et al., 2010; Muehlenhard et al., 2003). Based on these double standards, a single, sexually active woman runs the risk of jeopardizing her reputation and being branded a an unmarried, sexually active man is unlikely to deal with the similar negative consequences (Rathus, Nevid, & Fichner e who do not subscribe to the double standards of sexual behavior (Attwood, 2007). Therefore, sexual
30 double standards are differential based on gender, with a considerably harsher and restrictive version for women (Muehlenhard et al., 2003). Crawford and Popp (2003) argued that women are typically faced with conflicting ideals i.e., the dichotomy of Madonna whore, where the former encourages purity and virginity and the latter refers to sexual permissiveness. Yet, they also emphasized that gendered doubl e standards in general, and sexual double standards in particular, are local constructions shaped by double standards (i.e., forbidding premarital intercourse f or women) and transitional double standards (i.e., allowing for premarital intercourse for women in a committed relationship), Crawford and Popp indicated the declining relevance of the orthodox, but not of the transitional double standards. The discussio n of sexual double standards invites reference to the sexual scripting sexual lives are governed by social norms with respect to sexual desires and behaviors, rath er than by biological factors. Gagnon and Simon introduced three interrelated levels at which sexual scripting takes place: cultural, interpersonal, and intrapsychic levels. Cultural scripts refer to the social scenarios of what is appropriate in general ( e.g., gender specific double standards of appropriate sexual behavior). Interpersonal scripts refer to the level of social interaction in e.g., a particular sexual encounter. Th fantasies, which are of course influenced by cultural and interpersonal scripts. Drawing upon sexual double standards, Eaton and Rose (2011) reviewed three recent decades of s cholarship on dating in the journal Sex Roles and argued that gender typed dating
31 practices and etiquette continue to be perpetuated and institutionalized both at the level of cultural scripts (e.g., general expectations and beliefs) and interpersonal scri pts (e.g., actual interpersonal behaviors and emotions). They explained that from a feminist perspective, dating is a crucial scene for investigating the progress toward gender equality because it is one of the stage as well as establish the trajectory for future relationships that will either disrupt or perpetuate gender double standards, both at the individual and collective levels. Eaton and Rose suggested that the persistence of gender stereotypes embedded in dating scripts can be explained by the following: (1) perhaps, gender equality is not achieved yet; (2) adhering to established scripts may reduce the anxiety associated with the uncertainty of dating; and (3) following a cultural script may demonstrate s ocial awareness and savvy. However, investigating the impact of gender expectations on the meanings assigned to the cultural level are aligned with traditional gender stereotypes and sexual scripts, yet, their personal experiences elicited multiple contradictions and complexities resulting in a wide range were inex tricably permeated with gender and cultural beliefs about appropriate sexual scripts, arguing that women and men are fundamentally different, that sex is physical for men and re is rarely typically objectified. Furthermore, McCabe et al. (2010) found that more women than men adopted a broad definition of sex, where sex is not confined to g enital contact.
32 The notion of power or asymmetry of power, associated with sexual double standards has also gained some research attention. For instance, Browning, Kessler, Hatfield, and Choo (1999) investigated the links between gender, power, and sexual behavior, and found that the power of a dyadic relationship (i.e., the power of both partners) and the experience of power as a sexual motive (i.e., dominance and submission) were important predictors of sexual behavior. Furthermore, dominance was identifi ed as a sexual motive in both genders, but submission was associated with sexual behavior only among women, while among men it was associated with argued that the common male female asymmetry in compliant sexual behavior puts woman in a position of greater compliance rendering her vulnerable to sexual risk taking, sexual violence, and traumatic emotional reactions. Among the potential reasons for differential sexu al compliance across gender, Impett and Peplau addressed: (1) differential opportunities men are expected to initiate sexual relationships while women are expected to either comply or to turn down the initiation, (2) ease of influence according to trad itional gender roles men are perceived as active agents and women as influenced and submissive targets, (3) beliefs about male sexuality an idea that men have stronger sexual desire compared to women, (4) gender roles a woman is supposed to be responsi recreational orientation focusing on bodily needs, and (6) power and dependence the extent to which a woman has less power (economic, social, physical, etc.) in the relationship makes her
33 Additionally, Higgins and Browne (2008) argued that compliance and sexual risk taking are bound not only with of middle class people to the opinions of poor and working class people, and found that while c ompared to those of women, middle class participants saw this difference as socially constructed, while poor and working class participants saw it as a biological mandate (which in turn was used as a justification and legitimation argument for infidelity, conflict, and coercion). yper sexual needs are perceived as biological and irrepressible urges. Thus, poor and working class women are particularly vulnerable in this arrangement since they report not only situations when sex is desirable, but also situations when sex is performed to avoid conflict and/or violence, as well as to prevent infidelity. Therefore, sexual double standards persist at the cultural level, as women are expected to be sexually available, but not active. This asymmetry grants men with more opportunities for au tonomy, pleasure, and sexual self determination compared to women (Blanc, 2001). However, egalitarian sexual attitudes may be associated with increased sexual satisfaction (Muehlenhard et al., 2003). Moreover, Crawford and Popp (2003) discussed the detrime ntal effects of the being, emphasizing both physical and psychological consequences. Specifically, sexual double standards discourage women from negotiating safer sex and from being prepared in terms of contraception (e.g., bringing their own condoms), at least during the first sexual encounter, leading women to face an ambivalent choice between the risks associated with unprotected sex and risks associated with potential negative
34 judgments from their se xual partners (Crawford & Popp, 2003; Higgins & Browne, 2008; Tolman, Striepe, & Harmon, 2003). Furthermore, sexual double standards with the aforementioned conflicting ideals may be expressing sexuality may be socially sanctioned, but repressing sexuality may lead to women growing disconnected from their desires and identities. This dissonance may even lead to developing a negative sexual identity when women start perceiving themselv es as objects and/or victims of sexual encounters instead of agents and active participants (Crawford & Popp, 2003). Likewise, Tolman et al. to construct authen feelings. Indeed, a sexual body as a target of power and sexual double standards can serve as mechanisms of control, discipline, normalization, dividing and regulatory practice s for the purpose of reification of gender relations, and the prevention of sexual behaviors that may threaten the social status quo (Butler, 1990, 1996; Deveaux, 1996; Grimshaw, 1993; Sawicki, 1996). In this respect, Grimshaw argued that women are histori cally subject to the rules of stated that female sexuality is socially constructed as passive and subordinate to male desires and needs because a sexually active wo women can resist sexual double standards and construct novel patterns for self understanding, creating room for contestation and subversion (Butler, 1990, 1996; Sawicki, 1996). Moreover, by gaining co ntrol over their bodies (e.g., in terms of sexual pleasure, contraception, and reproduction) women ca n empower themselves and transform local gender relationships
35 have its costs, since Grello, Welsh, and Harper (2006) found that men with a history of casual sex repo rt the least symptoms of depression, while women engaging in casual sex report the most depressive symptoms. Gender, sexual double standards, and available sexual scripts are indeed crucial to understanding sexual behavior, but several other factors, such as life course stage and transitions are also essential to understanding sexual behavior and risk taking (Carpenter & DeLamater, 2010, 2012). Thus, the next section in this review looks at the studies focusing on the particular age span relevant to this st udy adolescence and young adulthood with or without the reference to the role of gender. Gender and Sexual Risk Taking in Adolescence and Young Adulthood The literature on adolescence and young adulthood is marked by a lack of consistency in defining t hese life stages in terms of chronological age. For example, in some studies, college students are referred to as young adults, while in others as adolescents. This inconsistency, as well as the fact that recent conceptions of adolescence encompass the fir st stages of early adulthood (Arnett, 2010), prompted the decision to discuss the phases of adolescence and young adulthood together. Moreover, preventative efforts of CDC (2007, 2008), focusing on sexually transmitted diseases, also target these life stag es as one group (i.e., young people 15 to 24 years of age). In their conceptual paper, Hovell et al. (1994) discussed evidence of high sexual behavior and experimentation rates, frequently without a condom, among adolescents putting them at increased risk of unplanned pregnancy and/or acquiring various STDs, including HIV/AIDS. According to the CDC (2008), adolescent sexual risk taking includes inconsistent condom use, first intercourse at an early age, and the combination of sex with alcohol and/or drugs. As such,
36 the CDC has identified sexual risk taking among adolescents as a priority for public health since it has devastating effects in the United States. Drawing upon the National Longitudinal Survey of Youth (ages 15 to 24; assessed in 1998, 2000, 2002 and 2004), Brookmeyer and Henrich (2009) found that sexual risk taking was prevalent, and in the majority of the cases, it occurred in conjunction with consuming alcohol. Likewise, Broman (2007) used the data from the National Longitudinal Survey of Adol escent Health (grades 7 through 12; assessed in 1994/1995, 1996, and 2001) and focused on sexual risk taking among black adolescents. He found that sexual risk taking was associated with age, gender, substance use, and family factors. In other words, older male adolescents were more likely to partake in sexual risk taking. Use of substances was positively associated with sexual risk taking, while maternal warmth was associated with lower engagement in sexual risk taking among adolescents. Turchik and Garske (2009), drawing upon a sample of US college students, found that men express greater intentions for sexual risk taking as well as report more actual involvement in sexual risk taking compared to women. Sylwester and Pawlowski (2011) also found among under graduate students in the UK that risk takers (not necessarily sexual risk takers, but in the broad sense of the word) are more alluring to both genders in the context of short term sexual relationships, which has a potential to make these relations even ri skier. Investigating AIDS related sexual risk taking behavior among US college students, Baldwin and Baldwin (1988) found that women were more cautious in general in comparison to men, but the average number of sexual partners per year and the age at first sexual intercourse higher number of sexual partners per year were at higher risk of contracting HIV). Interestingly, Baldwin and Baldwin also found that havin g accurate knowledge regarding the transmission of
37 HIV/AIDS was not associated with cautious sexual practices, concluding that the public health programs should consider not only providing knowledge, but also generating some fear (yet, not panic or hysteri a). While some of the aforementioned studies in this section have alluded to gender as one of the variables in the context of sexual behavior at particular life stages, the following studies l, Feng, and Schoenrock (1999) investigated the link between gender and sexual risk in US college students, focusing on perceptions of risk were higher compared to sexual risk taking. Additionally, Bell et al. found that men engage in sexual risk taking for different reasons compared to women, in that sensation seeking was associated with sexual risk for men, but not fo r women, while women emphasized the relational and emotional aspects of sex. Milhausen and Herold (1999) also focused on gender and perceptions of sexual double standards among college students. Drawing upon a sample of Canadian female students, the author s revealed a discrepancy where women believed in the existence of a sexual double standard at the societal level, but rejected it on the personal level. Relevant to the discussion of campuses. Wade and Heldman (2012) explored the hooking up culture among US college students referring to non relational sex with strangers or friends, typically under the influence of alcohol. They found that two thirds of the participants were involved in at least one hook up and some even acquired up to 13 sexual partners during the first year of college. Alcohol consumption was essential to the hook up script, since it offered not only a disinhibiting effect,
38 but also preserved the illus ion of meaningless and carelessness. Wade and Heldman (2012) stated that students were seeking pleasure, empowerment, and meaningfulness in sex, yet, most of them, and particularly women, failed to achieve these aspirations in hook up sex. Likewise, Bradsh aw, Kahn, and Saville (2010) argued that the hook up script is more frequent than the dating script among college students. They analyzed risks and benefits associated with dating and hooking up for women and men, and concluded that due to sexual double st andards women benefit more from the dating script while men from the hooking up script. Bradshaw et al. also illustrated that the participants indicated similar risks and benefits across gender with respect to dating and hooking up, but to a different degr ee. Namely, in the hooking up script, women were more concerned than men about becoming emotionally attached to the partner and wanting a relationship. Men, in the hooking up script, were also concerned that women would get attached to them, but they viewe d sexual gratification as more beneficial compared to women. As for the dating script, women were more concerned than men about losing friendship and/or being more interested in the relationship than the partner. Men, in the dating script, were more concer ned than women about losing independence and freedom, and saw the benefits of physical intimacy as more important compared to women. Drawing upon sexual double standards, Broaddus et al. (2010) investigated US college ation strategies (i.e., explanation of the consequences, refusal of sex in the absence of condom, and eroticization of the condom use where condom is introduced as fun and erotic activity) and condom proposers based on gender (i.e., whether man or woman su ggests using condom). A female proposer was generally judged no more harshly, and sometimes even more positively, than a male proposer. Yet, women, specifically, evaluated
39 the female proposer who adopted the eroticization strategy more harshly than men, ju dging her as promiscuous and even less mature. Cho and Span (2010) investigated the effects of gender and alcohol on US college taking. They found that intoxicated and placebo treated men reported more vigilan ce and lower intentions to engage in sex than men in the control group (i.e., no alcohol or placebo). Conversely, intoxicated and placebo treated women reported greater intentions to engage in sex compared to their sober counterparts. Explaining these find ings, Cho and Span drew upon the distinction between two alternative models of alcohol effect impairment model (emphasizing the pharmacological effect) and expectancy a rguing belief that she has consumed alcohol may serve as an excuse and as a factor partially alleviating responsibility for socially unacceptable behavior for wome n. Men, however, are not subjected to the same amount of social scrutiny and do not need excuses for sex to the same extent as women do. Another gender related factor associated with sexual risk taking is sexual coercion. Struckman Johnson, Struckman John son, and Anderson (2003) investigated tactics of sexual coercion (e.g., physical restraint and threats, emotional manipulation and lies, sexual arousal, and intoxication) among US college students and found that 78% of women and 58% of men were subjected t o such tactics, while 26% of women and 40% of men have employed such tactics. These gender differences in coercion are related to sexual risk taking and cannot be ignored. Thus, there is a need for a comprehensive and gender sensitive model of adolescent
40 Peterson called for the need to conceptualize such sexual empowerment as a multidimensional agency and control to some more objective measure of institutional and cultural power of women. Likewise, Russel, Van Campen, and Muraco (2012) suggested that healthy sexual development in adolescence should be constructed based on the positive development of sexual agency, rather than on historical discourses of restriction and risk. Finally, various social contexts ought to be considered in this respect, since social environments condoning higher degrees of sexual freedom may constitute arenas for sexual self exploration and experimentation, substantiating a rationale for exploring sexual behavior in tourism. Sexual Behavior in Tourism The tourism literature uses confusing and overlapping terminology with respect to sexual behavior in tourism. Namely, sex tourism (Ryan, 2000; Ryan & Hall, 2001; Ryan & Kinder, 1996) prostitut ion tourism (Jeffreys, 1999) romance tourism (Belliveau, 2006; Dahles & Bras, 1999; Pruitt & LaFont, 1995) and sex in tourism (Carter & Clift, 2000; McKercher & Bauer, 2003) Each term implies different connotations based on the various scripts associat ed with sexual behavior in tourism (i.e., sexual actors involved, motivations, power relations, gender, race, and presence or absence of any commercial transaction). Yet, the inevitable overlap between these scripts causes confusion. Sex Tourism T he major ity of literature on commercial sex tourism creates an illusion that all the rest of tourist activity is sex free. For instance, Ryan (2000) define d sexual intercourse while away from home an all inclusive term, but one which permits a di scussion of different experience to sex tourism may prove inadequate in understanding the complexity of this
41 phenomenon. In fact, sex tourism is a term that is very di fficult to define ( Snchez Taylor, 2001). The l ast decade contributed a variety of studies on sex tourism from different geographical areas (e.g. Aston, 2008; Bauer & McKercher 2003; Brennan, 2004; Cabezas, 2004; Clift & Carter, 2000; Jeffreys, 1999; Kemp adoo, 1999; Piscitelli, 2007; Ryan & Hall, 2001; Snchez Taylor, 2001), emphasizing that sex tourism is a complex phenomenon. The literature provides examples of a narrow definition of sex tourism (i.e., sex for money) and of broad approaches that conside 004) approach is an example of the former stating that sex tourism involves male tourists expecting that their vacation s will involve paid sexual experiences, where sexual experience can range from engaging in sex with prostitutes to visiting strippers and/or using escorts. Conversely, other researchers disapprove of such narrow definitions of sex tourism as an exchange of money for sex (e.g., Cabezas, 2004; Herold, Garcia & De M oya, 2001; McKercher & Bauer, 2003; Oppermann, 1999; Ryan & Hall, 2001). For instance, Cabezas claimed that sex tourism is more than an illicit commercial exchange of sex for money, as it involves ambiguous elements such as different kinds of opportunities for recreation, romance, consumptio n, travel, migration and marriage. Furthermore, a narrow definition omits attention to the variety of sex workers, their self perceptions, and gendered double standards that affect the outcomes of the sexual encounter for the sex worker. For example, Cab ezas, based on her research of sex tourism in Cuba and the Dominican Republic, and Piscitelli (2007), based on her research of sex tourism in Fortaleza, in north east Brazil, claim ed that there are different types of sex workers and not all of them recogni ze themselves as such even if they receive some kind of economic help and/or present. Additionally, Cabezas illustrate d how double standards of appropriate sexual behavior for men
42 onquering the (p. 1008). Thus, Snchez Taylor (2001) call ed for the develop ment of a conceptual model for sex tourism that could reflect the diversity of tou rist related sexual economic exchange and the complexity of the power relations that corroborate them. There are also debates regarding whether the term sex tourism is adequate or it should substituted with another term prostitution tourism. Jeffreys (19 99), for instance, based on her study of sexual exploitation of women in the Philippines, argued objectification and the abuse of women (p. 180). She claim ed that male sex tourism is promoted by the government in the Philippines because of its profitability and call ed to understand it as prostitution tourism nnot be tolerated in the leisure industry as it is detrimental to women and children gender and economic interests in South east Asia, labeled this phenomenon as t ourism a form of organized risk behavior for spreading the HIV/AIDS virus (p. 177). De Albuquerque (1998) cautioned about not being misled by the illusion of romance relationship and to understand both male and female sex tourism relationship s as However, Oppermann (1999), Ryan (2000) and Herold et. al. (2001) criticized such studies for essentializing all forms of sex tourism as a form of victimization. In turn, the broader perspective s employed by tourism researchers to define sex tourism encompass multiple parameters. For instance, Ryan and Hall (2001) suggested that in order to
43 define sex tourism s exual encounters should be located along two axes: (1) a continuum between voluntary participation on behalf of the sex worker on the one end and a position of total exploitation on the other; and (2) the level of commerciality that underpin s the sexual en counter. Ryan (2000) added a third axis where on the one extreme a sexual encounter that enhances feeling s of self integrity is located while on the other extreme a sexual encounter that degrad es a person s sense of integrity is placed Likewise McKerch er and Bauer ( 2003) suggested examining the following three dimensions: (1) the role of sex as a motivation for travel; (2) the nature of the sexual encounter positive and mutually rewarding for all the participants vs. negative and exploitative for at l east one of the partners; and (3) the role of the tourism industry in facilitating sexual encounter s Further, Oppermann ( 1999) proposed understanding sex tourism as more than just a traditional stereotype of monetary exchange and offered a holistic model with six parameters in order to grasp the complexity of sex tourism: (1) the importance of intention to engage in sex with strangers as a purpose of travel; (2) the nature of the financial exchange; (3) amount of time spent with a sex worker; (4) prostitut e tourist relationship; (5) the nature of the sexual encounter; and (6) consideration of who travels (sex seeker, sex provider, or both). However, even list of parameters might not be comprehensive enough to explain the complexity of sexual be havior in tourism and the diversity implied by gender differences. Sexual behavior is related to identity formation, building relationships, conforming to social stereotypes or, conversely, negotiating them via sexual behavior. Moreover, using the idea of sex tourism as a v a ntage point to approaching all sexual behavior s in tourism and/or juxtaposing all sexual activities vis vis aforementioned parameters for defining sex tourism might not do justice to the complexity and heterogeneity of sexual behaviors in tourism.
44 Moreover, analytically, it may serve only a quite limited range of the research questions in this area and may not be particularly applicable to certain populations. For instance, Pritchard and Morgan (2006) claimed that couples constitute a d ominant tourist unit in most hotels and tourist attractions and it would make sense to suggest that a substantial amount of sexual activity in tourism occurs among those couples. Lumping their sexual activity under the category of sex tourism might not be adequate. Likewise, the variety of sexual encounters in tourism without a commercial transaction might not fit conceptually under the category of sex tourism. Sex in Tourism Sex in tourism, as opposed to sex tourism, is a broader term including sexual beh avior without a commercial transaction ( Carter & Clift, 2000) Additionally, it seems to be less loaded with negative connotations, leaving more room for interpretations and, perhaps, also a consideration of the social atmosphere in tourism contexts. Touri sm scholars argue that the distinctive features of tourism time and space may influence a since people leave behind some of their social baggage when they travel (McKercher & Bauer, 2003). Therefore, tourism offers a special socia Eiser & Ford, 1995; Shields, 1990; Wickens, 1997). Clarifying the uniqueness of the social context in tourism is vital to an understanding of sexual behavior in tourism and it has often been conce ptualized using the concepts of an understanding of liminality: separation, transition, and incorporation. Transition is a relevant phase for understanding tourist behavior as it signifies the time place of between and betwixt, characterized by ambiguity, where a person feels detached from everyday life. The tourist leaves the familiar environment of everyday life (i.e., separation) for an experience of temporary ch ange in a different environment (i.e., transition) in order to eventually return to everyday life (i.e.,
45 quasiliminal, to refer to the stage in leisure or tourism that resembles the limin al phase. Tourism can be understood as a liminoid realm associated with transition, absence of status distinctions and equality, anonymity, communitas, less constrained behavior in general and sexual behavior in particular. By entering a liminoid state, o r in simple terms, by travelling, people are provided with an opportunity to express things and to fulfill fantasies that they would otherwise suppress (Ryan & Kinder, 1996). Due to the anonymity and perceived tolerance offered by a tourist destination, in dividual s may explore more radical sexual behavior compared to their home environments either in choice of partners, or in frequency, or in nature of the sexual experience (Bauer & e charter yacht a kind of reli ef of tension through the expression of repressed thoughts and feelings (p. 54) T ourists reported play ing with the rules of everyday life, especially those applicable to social relationships, personal indulgence, and sexual behavior. Lett stated that the 51). Additionally those charter y acht tourists who described sexual activity with a steady partner bound with vacation sex. In turn, Selnniemi (2003) offer ed an understanding of liminoid sun lust tourism as a fourfold transition/tran sgression incorporating the spatial transition from place (w hich is home )
46 enable a control and reveal usually hidden characteristics (p. 27). Finally, a sensual transition occurs as individuals may become more aware of their heightened bodily senses that mi ght be triggered differently, as well as may respond differently to the same stimuli, in liminoid tourism environments. Selnniemi also state d that liminality explains why tourism and sex are so closely linked and Black (2000) argue d that it is often assum ed that travel per se is liminal and liberatory. a counter site for transgression, inversion, and contestation of everyday social order has been applied to understand sexual behavior in leisure and to urism experiences. For instance, Andriotis (2010) conceptualized a gay nude beach as a heterotopia, suggesting that in this setting people can experiment with inversions of morals and practices, as well as with their bodies, sexuality, and pleasure. ConFes t an alternative lifestyle event held in Australia that became one of the contemporary pilgrimage centers was also conceptualized by St John (2001) as a heterotopia offering opportunities for fulfilling various fantasies, disrupting gender identities, and resisting everyday social norms and expectations. This uniqueness of the social atmosphere in liminoid tourist experiences is associated by some scholars with existential authenticity (Kim & Jamal, 2007; Wang, 1999, 2000). E xistential authenticity refe rs to the quest for the true self, individuality, self identity and self making, meaning making, and anxiety if the aforementioned are not achieved (Steiner & Reisinger, 2006). Wang theorized that the liminoid tourist experience allows people to transcend everyday roles because of its freer, spontaneous, simpler, and romantic nature. He suggested distinguishing between intra and inter personal aspects of existential authenticity. conceptualization
47 was empirically supported by study of tourist experiences at a Renaissance festival where public nudity and casual sex were found to be expressions of intra personal existential authenticity atmosphere of the festival Researchers cite feeling s of situationa l disinhibition, sense of anonymity, and liminality/liminoidity to explain sexual promiscuity and increased rates of sexual risk taking in tourism, both in terms of the accelerated temporal progression of the sexual relations and choice of sexual partners (Apostolopoulos et al., 2002; Maticka Tyndale et al. 2003). For instance, Shields (1990) conceptualized seaside resorts as characterized with liminality, temporary loss of social bearings, and the primary aim of gaining pleasu re. Likewise, Pritchard and Morgan (2006) dirty weekend as contributing to the transgression of everyday norms out of time, out of place, and out of mind Similarly Apostolopoulos et al. (2002 ) define d various types of leisure tourist experiences and drugs encourag ing the suspension and rejection of personal inhibitions and everyday social norms ( p. 733 ) Ford and Eiser ( 19 95 ) exploring the sexual activity of young tourists (16 29 years) suggested the concept the sense of being a different person on vacation and fe e ling less sual sex in tourism (p. 323 ). Furthermore, certain destinations such as Ibiza, Faliraki and Rhodes have bec a me the Gomorrah of & Laustsen, 2004, p. 99). The general attitude of disinhibition and tolerance toward sexual permissiveness and transgressions of social norms in some tourist destinations is reflected in the social myths and marketing slogans
48 happens in Vegas, Maticka Tyndale, Herold, & Mewhinney, 1998, p. 262 Tenerife, stays in Tenerife ( Thomas, 2005, p. 571). Focusing specifically on spring break vac ation s and college students Maticka Tyndale et al. (1998) found that 21% of men and 17% of women had casual sex on spring break because of the social atmosphere being conducive to sexual experimentation and promiscuity. I n studies o f sexual risk taking am ong students on spring break, opportunities for casual sex (in combination with binge drinking and taking drugs) were found to be the major motivating factors for getting involved in such vacation s ( Apostolopoulos et al., 2002; Snmez et al., 2006). As for the potential gender differences in sexual behavior on spring break, the literature provides contradictory findings. While some studies indicate that men engage in more casual sex on spring break than women ( Josiam et al., 1998; Litvin, 2010; Maticka Tynd ale et al. 2003 ), other studies reveal no gender differences in casual sex on spring break (Maticka Tyndale & Herold, 1997; Maticka Tyndale et al., 1998). In line with the former studies men were found to report higher levels of approval of casual sex co mpared to women on vacation in general and on spring break in particular (Eiser & Ford, 1995; Maticka Tyndale et al., 1998). However Maticka Tyndale and Herold (1997) found that women appear to have a greater license for being sexually active on spring br eak vis vis everyday life due to the spring break sexual cultural script being comprised of partying alcohol consumption sexual arousal, sexual pressuring, and participation in sexual environment. Likewise, Mewhinney et al. (1995) used Gagnon and Simon (1973) scripting theory i.e., cultural, interpersonal and intrapsychic levels d the feeling s of freedom from everyday norms and responsibilities, as well the sense of anonymit y, as
49 contributing to the approval of casual sex on spring break particularly for women. They argued that the unique sub cultural context of s pring b reak combine d with more permissive norms and vacation related sexual attitudes render sexual double stan dards less relevant and encourage sexual permissiveness in the scripts for both genders Namely, at the cultural level, the scripts reflect more permissive norms of behavior when the person is on vacation. At the interpersonal level, spring break scripts o f interaction make it easier to become involved in sexual behavior more quickly as illustrated by the rapid physical intimacy during dance i.e., grinding. Yet, at the intrapsychic scripting, women reported internal conflict i.e., despite the greater a cceptance of casual sex on spring break many women indicated that they would feel guilty for engag ing in casual sex. Thus, cultural and interpersonal scripts favoring casual sexual activity on spring break were not always congruent with intrapsychic level scripts (Mewhinney et al., 1995). Gender is an essential issue in the study of sexual behavior in tourism, highlighting the continued need for gender sensitive research (Berdychevsky et al., 2013a) Considering sexual do uble standards, the significance of anonymous, liminoid, tourist environment s might stem from their potential to offer settings for exploring sex ual behaviors by women without jeopardizing their reputations. Reflecting the general trend in the field, resea sexual behavior in tourism has also primarily focused on the commercial aspect ( Dahles & Bras 1999 ; Jeffreys, 2003; Pruitt & LaFont 1995 ) In this respect, t he literature reveals disagreement as to whether female tourists who engage in sex ual activity with locals should be included among the ranks of sex tourists or whether they constitute a separate category called romance tourism. Pruitt and LaFont (1995), in their study of female tourists in Jamaica, coined the term romance tourism, wh ich they use instead of sex tourism, in order to distinguish and to emphasize
50 term r difference is that both female tourists and their local lovers engage in negotiation of gender identity through experimentation with new gender roles, and neither female tourists nor local men perceive their relationships as prostitution. In turn, Dahles and Bras (1999) based on their anthropological fieldwork in Indonesia, suggest ed conceptualizing local males engaging in sexual relationships with female tourists as romantic entrepreneurs since they perceive white girlfriends as vehicles to fulfill dream s, make a living and secure their future. However, these romantic entrepreneurs do not like to be called gigolos as they do not perceive the money they get from female tourists as payment for sex. The concept of romance tourism is loaded with controversi es that have generated a debate regarding the similarities and differences between sex tourism and romance tourism. For in stance, Jeffreys (2003) explained that the main similarity proposed by those researchers who seek to include women in sex tourism is t he economic, class and racial privileged status of Western tourists compared to their local sexual partners. In addition, de Albuquerque (1998) wa s critical of the concept of romance tourism as he saw the relationships between female tourists and beach bo ys as money for sex and thus as sex tourism. Similarly, Snchez Taylor (2001) Republic and Jamaica, describe d four problems with the distinction between male sex tourism and female romance tourism. First, sexual economic exchanges are not always straightforward and male tourists, as well as local women, might not necessarily perceive themselves as prostitute users and prostitutes. Second, both female and male tourists can be sexually predatory and hostile. Third, this distinction reproduces essentialist understandings of female and male
51 sexuality where m e n benefit from the sexual access to wom e ies Finally, the idea that female tourist s having sex with l ocal men are not sex tourists tends to downplay the significance of racialized and economic power. However, Jeffreys (2003) criticize d a recent tendency among researchers to include women within the ranks of sex tourists and argue d on to the power relations, context, meanings and effects of the behaviors of male and female tourists who engage 2003, p. 223), while an ungendered approach to sexual behavior in tourism fails to demonstrate those differences and obscures the power relations. She argue d that t he numbers of women involved in pursuing sexual relationships with local men while on holiday are tiny compared to that of men, and wom e s with local m e n might be considered similar to a holiday romance with a tourist mate. In turn, Herold et. al. (2001), based on their study in the Dominican Republic suggest ed that the differences between romance tourism and sex tou rism can be seen in the following elements: the role of romance and courtship vs. sex and money, characteristics of beach boys and female sex workers, motivations for involvement, degree of freedom in selecting the target on behalf of the beach boy vs. sex worker, the seduction process, the sexual relationships (i.e., emotional involvement, fe e ling of control and condom usage), obtainment of payment, and the feasibility of longer term relationships. Herold et al. conceptualize d sex tourism and romance tour ism as two ends of a continuum of motivations rather than as distinct categories and their findings reveal that more of the female tourists were located toward the romance end and more of the male tourists toward the sex end of the continuum. Thus, positio sexual behavior with local men as an adjunct of male sex tourism may not be the best approach
52 Moreover, the same question about the relevance (or the lack of it) of sex tourism or non commercial sexual behavior in tourism persists. T he tourism literature consists of only a few studies that have endeavored to commercial sexual behavior in tourism. For instance, Wickens (1997) in al behavior in Chalkidiki Northern Greece, found that women explained their sexual promiscuity with the idea that the social atmosphere in their tourist experiences grants them the (p. 151). Later Wickens and Snmez (2007) also desc In her study of the factors that may facilitat e or new sexual partner abroad Thomas (2000) found that many women experienced sexual permissiveness during the ir trip s and engaged in sexual intercourse significantly earlier th a n they would do at home because of the perceived anonymity and temporal changes that they experienced while on holiday. The women explained their decisions to have sex with a new partner they met while travelling based on the factors of attraction and pleasure, presence or absence of a steady relationship at home, alcohol consumption, notions of space and priva cy, as well as emotions and fear of vulnerability. In another study, Thomas (2005) found that the women she interviewed establish ed unqualified trust and ha d sex with relative stranger s on vacation in Tenerife, the Canary Islands, sooner than they would at home. Thomas suggested this was associated with a number of factors, including the faster progression of sexual relationships the temporary freedom from norms and constraints in a liminoid tourism environment, perceived anonymity, and the perception of t ime as being compressed.
53 Studying sexual behavior of female travelers in Costa Rica, Ragsdale et al. (2006) found that women travelling alone or with a female companion were more likely to engage in casual sex than women travelling in groups. Additionally, women having expectations for casual sex while travelling were more likely to experience it. However, these women were also more likely to procure and to initiate condom use with a new sexual partner they met on vacation compared to women without anticipa tions, putting the latter at a higher risk of unsafe sex. Poria (2006) also suggested that researchers should consider the importance of sexuality in understanding sexual behavior in tourism as he found that lesbian women might change their sexual comportm ent in tourism as part of the self expression and practice of existential authenticity due to a sense of perceived anonymity in tourism. tourism should also be consider ed. Berdychevsky et al. (2013a) examined a variety of tourist experiences (e.g., rest and relaxation, city break, backpacking, and business trips) in researching suggeste perceptual, and behavioral factors, where in leisure based tourist experiences women reported various changes in their sexual behavior vis vis everyday life and emphasized diffe rent degrees in the importance of sex in terms of their motivations for, and satisfaction with, their tourist experiences. Berdychevsky et al. (2010) found that sexual activities perceived by women as unconventional and adventurous such as casual sex comprise an integral part of the adventurous ethos of the backpacking experience. Likewise, Falconer (2011), investigating the narratives of female
54 backpackers, suggeste d that sexual behavior is at the core of the internal conflict between responsibility for personal safety and aspiration for exciting and potentially risky experiences. On the one hand, sex on the road is bound with negotiating gender constraints, vulnerab ility, and potential loss of control O n the other hand, however, for risky and intense sensual experiences is a way of establishing the identity of the liberated and empowered risk taker. Summary To conclude, pluralistic frameworks for unde rstanding tourism (Poria et al., 2003), sex (Oppermann, 1998, 1999), and risk (Ryan, 2003) were adopted in this dissertation to reflect its concepts. Following the sug gestion that late adolescence and young adulthood are the peak eras for exploration, experimentation, risk taking (Gibson, 1996) and sensation seeking (Zuckerman, taking is a pri ority for public health, these life stages were the target population for this dissertation. Additionally, since gender and sexual double standards are the pivotal concepts to investigating sexual behavior (Crawford & Popp, 2003; Eaton & Rose, 2011; Jonaso n & Fisher, 2009), gender was a crucial variable considered. Men and women may have different motivations for sexual risk taking (Bell et al., 1999) and various degrees of different risks and benefits associated with sexual activities (Bradshaw et al., 201 0; Wade & Heldman, 2012). In this respect, tourism should be explored as a context condoning higher degrees of sexual freedom with the liminoid characteristics potentially rendering sexual double standards less prevalent than in everyday life. Tourism has been conceptualized as a liminoid phenomenon (Let, 1983; Selnniemi, 2003) and a heterotopia (Andriotis, 2010; St John, 2001) allowing for situational disinhibition (Apostolopoulos et al., 2002; Ford & Eiser, 1995), as well as expression
55 of existential au thenticity in general (Wang, 1999, 2000), and through sexual behavior in particular (Kim & Jamal, 2007). The tourism social atmosphere provides women with opportunities for sexual self exploration and self discovery unavailable to them at home (Berdychevsk sexual risk taking behaviors in tourism might be a source of richer findings compared to men (however, gender based comparison may unveil multiple complexities). Moreover, t here is a efforts aimed at educating women regarding safer sex precautions in tourism (Ragsdale et al., 2006). Indeed, it is important to consider that sexual exploration in anonymous environments is related to sexual risk taking (Black, 1997; Thomas, 2005), which may lead to various physical and mental health consequences, including contracting and bringing home STDs, pregnancy, damaged reputation, internal co nflicts, and physical and/or emotional traumas. Yet, the topic of sexual risk taking in tourism remains under researched and the purpose of this study was to address this gap.
56 CHAPTER 3 METHODS Mixed Methods Research Mixed methods designs refer to resea interpreting quantitative and qualitative data in a single study or in a series of studies that methods research has been d approach has been presented as a new third research paradigm and a distinctive methodology (Greene, 20 08; Johnson, Onwuegbuzie, & Turner, 2007; Leech & Onwuegbuzie, 2009). Methodological eclecticism and pluralism associated with mixed methods designs provide a potential for superior research vis vis a monomethod approach (Johnson & Onwuegbuzie, 2004). A ny mixed methods study attempts some form of integration, but the debates around the incommensurability of the qualitative and quantitative paradigms have often created unhelpful boundaries impeding analytic integration (Bazeley, 2009). However, the dialec tical process of learning and growing through debate and interaction is an essential part of research philosophy and methodology (Johnson, 2008). Greene (2007, 2008) described various approaches to mixing methods/paradigms/mental models, including (1) a pu rist stance paradigms are incommensurable and cannot be mixed in the same study; (2) a complementary strength stance when mixing methods they should be kept separate from each other due to the incompatibility of the paradigms; (3) a dialectic stance dialogical engagement with paradigmatic differences; (4) an alternative paradigm stance reconciling incommensurabilities through new paradigms, such as pragmatism, (5) an a paradigmatic stance paradigmatic assumptions are logically
57 independent and, thu s, can de mixed; and (6) a substantive theory stance paradigms can be intertwined with substantive theories. Pragmatism offers a middle ground philosophical and methodological position between epistemological underpinnings of qualitative and quantitative traditions (naturalistic/interpretive/ constructivist vs. positivist), providing a practical, instrumental, and outcome oriented method of inquiry for producing socially useful knowledge (Cameron, 2009; Feilzer, 2010; Morgan, 2007). As such, pragmatism ha s been described as a pluralistic partner of and a philosophical foothold for the mixed methods research (Cameron, 2009; Feilzer, 2010; Johnson & Onwuegbuzie, 2004). Morgan explained that pragmatism does not ignore the importance of epistemology, but it do es deny the hypothesis of incommensurability between the paradigms and grants the emphasis to ecting traditional dualisms; recognizing both physical and socio psychological worlds; viewing knowledge as both constructed and based on the lived reality; endorsing eclecticism, pluralism, and practical empiricism as a way for determining what works; app roving fallibilism viewing research conclusions as imperfect, non absolute, and tentative (Johnson & Onwuegbuzie, 2004, p. 18). Conversely, Sale, Lohfeld, and Brazil (2002) advocated for incommensurability of the paradigms arguing that different views of reality dictate a different view of the studied phenomenon. As such, they claimed that quantitative and qualitative researchers do not study the same phenomena. However, Sale et al. stated that this should not discourage the implementation of multiple met hods in the same study for complementary purposes; yet, the distinction between the studied phenomena should be clarified. Applying their logic to this mixed methods study, the phenomenon for the qualitative phase was the lived experience of sexual risk ta king in tourism.
58 This, in turn, informed the development of a measure of sexual risk taking in tourism by women. taking in tourism may appear the same The literature reflects the plurality of mixed methods designs, calling for embracing the differences, as opposed to imposing h omogeneity (Harrits, 2011). For instance, a sequential mixed method design (as opposed to a concurrent/parallel design) refers to the situation where one type of data serves as a basis for constructing, collecting, and interpreting another type of data (Ca meron, 2009; Leech & Onwuegbuzie, 2009). In addition to this time aspect, mixed methods designs are typically classified based on the level of mixing (fully mixed vs. partially mixed) and the emphasis of approaches i.e., dominant status vs. equal status (Johnson et al., 2007; Leech & Onwuegbuzie, 2009). Johnson et al. even conceptualized a mixed methods design methods research (p. 124). Th is study was informed by the pragmatist approach and conducted following a sequential mixed method s design starting with a qualitative phase and proceeding to a quantitative phase Both qualitative and quantitative approaches in this study had an equal status and the aspiration was t o get as close as possible to a fully mixed methods design across research purposes, a consistent content core across the qualitative and quantitative types of data, analytical techniques, as well as interpretations and conclusions. An e xploratory, qualita tive strategy wa s essential for this study in order to approach the question inductively and to start creat ing an empirically rich base of knowledge for this topic. Th is strategy shed some light on
59 perceptions of sexual risk taking in tourism as w ell as associated motivations and meanings. Mason, Augustyn, and Seakhoa King (2010) argued that exploratory studies have been advocated for increasingly in the social sciences when dealing with new research topics. However, while advocating for the explo ratory stage as part of a sequential mixed methods design in tourism research, Mason et al. emphasized the importance of the methodological rigorousness of this phase for it to be a reliable basis for the next quantitative phase. To address this requiremen t, transcendental phenomenology was conducted as the first phase in this dissertation. Since t he theoretical perspective of phenom enology informed the qualitative phase in this study, a ed any sort of risk both with casual or steady sexual partners was understood as sexual risk taking Legitimation T he issue of validity, or its equivalent in mixed methods research, needs to be addressed. The notion of validity in mixed methods research has been addressed sparingly in the literature (Dellinger & Leech, 2007). Historically, validity in mi x ed methods studies was assessed separately for the qualitative and quantitative parts. However recent scholarship in the area has developed additional crit eria for assessing validity in mixed methods contexts (Dellinger & Leech, 2007; Onwuegbuzie & Johnson, 2006; Tashakkori & Teddlie, 2003). Since the concept of validity has connotations associated with quantitative research, scholars suggest using more neut ral terminology in the context of mixed methods studies, such as the concepts of inference quality (Tashakkori & Teddlie, 2003) and legitimation (Onwuegbuzie & Johnson, 2006). The former approach was chosen for this study due to its comprehensive set of re commendations including nine legitimation types: sample integration legitimation, inside outside legitimation, weakness minimization legitimation, sequential legitimation, conversion legitimation,
60 paradigmatic mixing legitimation, commensurability legitima tion, multiple validities legitimation, and political legitimation (Onwuegbuzie & Johnson, 2006) Sample integration legitimation refers to the potential of constructing meta inferences by making inferences across both the qualitative and quantitative sam ples. Such potential was maximized in this study by : (1) recruiting participants for both qualitative and quantitative phases from the same target population and (2) implementing a systematic random sampling with proportional representation according to c lass standing in the quantitative phase (more on this issue is presented in the s ampling section). Inside outside legitimation indicates the degree of accuracy in terms of presenting an participant) and an observer researcher) viewpoints. The former was fostered through member checking (more on this issue can be found under qualitative data analysis), while the latter was enhanced by soliciting peer reviews for various components of this study. Weakness mi nimization legitimation addresses the capacity of one methodological approach to compensate for the weaknesses of another. In this study, the qualitative phase provided richness and depth, while the quantitative phase addressed generalizability. Sequentia l legitimation refers to minimizing the effect of reversing the sequence of the qualitative and quantitative phases on meta inferences In this study, the qualitative start appeared to be the only way to approach inductively the under researched phenomenon of and to provide a solid foundation for quantitative exploration Conversion legitimation points to the extent to which quantitizing (e.g., counting frequencies in qualitative data) and qualitizing (e.g., constru cting narrative profiles such as the average profiles or comparative profiles) results can yield quality meta inferences. To enhance this type of legitimation, counting frequencies for the sake of establishing universality in the
61 qualitative phase was done cautious ly to avoid acontextual c ounting and over counting. Additionally, constructing profiles in the quantitative phase was done cautious ly so as not to over generalize them or present them as unrealistic. Paradigmatic mixing legitimation indicates t he extent to which the epistemological underpinnings of the qualitative and quantitative approaches are combined into a usable package. Adopting philosophical and methodological pragmatism in this study, as well as making explicit the philosophical assumpt ions in each research phase, helped in attaining this purpose. Commensurability legitimation refers to the degree to which the meta inferences reflect an iterative cognitive process of switching and integration between qualitative and quantitative lenses. Chapter 5 reflects the integration and the iterative back and force analysis between the qualitative and quantitative viewpoints in an attempt to create a holistic view of the studied phenomenon. Multiple validities legitimation refers to the validity ass essment of the qualitative and quantitative phases according to their respective standards of validity or its equivalent. Such an evaluation was conducted in this study and is presented in respective sections of Chapter 3. Finally, political legitimation p oints to the extent to which the meta inferences derived from both the qualitative and quantitative components can be valuable to consumers of mixed method research. In this study, such a potential was fostered through : (1) addressing the socially meaningf taking in tourism; (2) advocating pluralism of perspectives in addressing this knowledge gap; and (3) striving to generate practical results and an understanding that could be used in designing efficient health programs and/or information campaigns.
62 Phase 1 Qualitative Method Transcendental Phenomenology The goal of phenomenology is to understand the hidden meanings of the experience, to clarify the essence of the exper ience, as well as the ways through which an individual makes sense of it ( Moustakas 1994 ). Thr ee main traditions in phenomenological research are classical transcendental phenomenology Edmund Husserl existential phenomenology Maurice Merleau Ponty an d Jean Paul Sartre, and hermeneutic phenomenology Martin Heidegger, Hans George Gadamer, Paul Ricoeur, and Max van Manen (Grbich, 2007 ). Edmund Husserl is the father of the classical transcendental method of phenomenology H e introduced the approach in h is book Ideas (1913). His approach is descriptive in nature (i.e. the assumption is that the researcher finds the meaning and describes it) as opposed to interpretative hermeneutic phenomenology where the researcher interprets and, thus, inevitably gives m eaning to the studied phenomenon (Vagle, Hughs, & Durbin, 2009). Additionally, the essential components of transcendental phenomenology are phenomenological reduction and epoche bracketing (all will be explained later) while according to both existentia l and hermeneutic phenomenologies these are impossible (Grbich, 2007). based on studying material things, arguing that this approach failed to consider the consciously expe riencing person. Transcendental B y adopting a phenomenological attitude a researcher focuses on essential structures that allow objects that are naively taken for gran ted under a natural attitude to appear in our consciousness (Beyer, 2011). A p henomenological approach focuses on the experience in order to obtain detailed descriptions that reflect the essences of the experience and form the basis for the reflective stru ctural analysis ( Moustakas 1994 ) The argument is that for the sake of understanding objects that appear before individual s they must ret urn to their
63 selves and recognize themselves in the experience. Thus, transcendental phenomenology centers on the per sonal experience as a source of knowledge. The application of transcendental phenomenology in this study followed the guidelines for implementation suggested by Clark Moustakas (1994). Among the main features of transcendental phenomenology are: a focus o n the experience of things (regardless of whether they are tangible or intangible); examination of multiple angles based on both intuition and reflection; commitment to describe experiences rather than to explain; integration between subject and object; an d the ( Moustakas 1994 ) Transcendental phenomenology is integrally bound with the concept of intentionality. Intention refers to the orientation of the mind to the object and, thus, directedness is an inher ent feature of intentionality ( Moustakas 1994 ) Consciousness is intentional as it is directed toward objects and intentionality can be even interpreted as a synonym for the consciousness itself. Based on the concept of intentionality, both the self and t he world are the meaning components. For Husserl, the absolute reality is what appears in consciousness, while what appears in the world is a learning product ( Moustakas 1994 ) Moreover, whether the object actually exists or not makes no difference at al l from the perspective of phenomenology. Perception is the primary source of phenomenological knowledge and is comprised of intentions and sensations. Perceiving is an intentional act as the object of consciousness and the act of consciousness are intentio nally related. For Husserl, the presence of something in consciousness is an act. A n a ct is an intentional experience and the meaning of the phenomenon is embedded in the act of experiencing it (rather that in the object itself). Intentionality incorporat es noema and noesis ( Moustakas 1994 ) the experience. It is not the real object but the experience of the phenomenon (e.g., not the
64 building but the appearance of the building). Noesis is the way in which the noema i s experienced. The noesis brings into being the consciousness of something th r ough thinking, feeling, reflecting, and remembering. Therefore, textures refer to noema, while structures refer to noesis. For every noema there is noesis and for every noesis th ere is noema ( Moustakas 1994 ) For example, perceptions associated with sexual risk taking are noematic meanings (i.e., sexual risk taking that account for the noematic meanings are noetic factors ( i.e., structures). Clarifying the noema noesis relationship and the derivation of meanings is the function of intentionality ( Moustakas 1994 ) The methodology of transcendental phenomenology involves epoche, phenomenological reduction, and imaginative var iation. The researcher adopting the approach of transcendental phenomenology is challenged to describe things as they are, as well as to understand meanings and essences drawing upon intuition and self reflection. transcen dental phenomenology is that the researcher can achieve a state of pure and transcendental ego, a state of complete absence of bias, in order to be undisturbed by habits of the natural world, nave, and receptive to hearing co researchers (Grbich, 2007; Mo ustakas 1994 ) The process of eliminating pre judgments or presupposition is known as the epoche process Coming from Greek, epoche means to refrain from judgment, from taking any position, from perceiving things in an ordinary way ( Moustakas 1994 ) At t he epoche analytical phase every statement dimension quality of the phenomenon has an equal value. Husserl discussed two versions of epoche: the universal epoche and the local epoche (Beyer, 2011). The universal epoche requires the researcher to brac ket i.e., to put aside all her/his existing assumptions regarding the world. The weaker version of local epoche might be more attainable, since it requires bracketing out only particular assumptions relevant to the
65 studied phenomenon. While the requireme nt of the local epoche might be more feasible it is still limited since the choice of what is relevant and what is irrelevant is an assumption in itself. One of the more recent approaches actually suggests substituting bracketing with bridling (Dahlberg, 2006; Vagle et al., 2009). Similarly to bracketing, bridling involves restraining the pre understandings in order to achieve openness and receptiveness as well as helping the researcher to refrain from developing premature understandings and conclusions. T ranscendental phenomenological reduction is called : transcendental because it moves beyond the everyday, natural attitude to pure ego phenomenological attitude where everything is perceived freshly, as if for the first time phenomenological b ecause it translates the world to phenomena reduction because it goes back to the source of the meaning (Beyer, 2011 ; Moustakas 1994). Textural description of the meanings associated with the studied phenomenon is attained via transcendental phen omenological reduction though the transcendental state of freshness free from the habits and knowledge of the un reflected everyday experience. Textural description involves not only what an individual sees in terms of the external object (if such exists) but also an experience itself (i.e., the internal act of consciousness, the relationship between the self and the phenomenon). Conducting textural description requires the researcher to look at the phenomenon again and again to identify the multiplicity and depth of the meanings. This never ending process of discovery is called horizonalization where h orizons are unlimited and do not last indefinitely ( Moustakas 1994). Imaginative variation follows the phenomenological reduction and aim s to identify the structural essences of the studied phenomenon that account for the textural meanings. The product of imaginative variation is structural descriptions of the phenomenon Namely, the ariation is purely
66 imaginative and intuitional (it is not empirical), as it seeks the potential meanings and essences via employing imagination, changing the frames of reference, looking at reversals and polarities, and adopting various angles and perspect ives to look at the phenomenon ( Moustakas 1994) The final step of the phenomenological analysis is the synthesis of the composite textural descriptions and composite structural descriptions. of tr ansformation of empirical individual experiences into essential insights i.e., moving from the ideographic to the nomothetic level (Moustakas, 1994) This synthesis is an intuitive and reflective process resulting in the statement of the essences the c ore meanings of the phenomenon. Data Collection One of the typical ways of obtaining data for a phenome nological study is through open ended questions in an in depth interview or even a series of interviews (Grbich, 2007). This is the strategy adopted fo r this study The research participants are perceived as c o researchers as the aim of the phenomenological analysis is to learn what a phenomenon means for the person who ha s experienced it. For this research, t he individual semi structured, face to face, in depth interview wa s found appropriate because of its flexibility and potential for establishing rapport through personal interaction. This choice allowed the interviewees to address the issues they considered important, while at the same time prompting them to refer to the constant topics establishin g the basis for comparison among the participants This is particularly important for the nomothetic level of the phenomenological analysis Moreover, the potential of personal interaction for establishing r apport is crucial for achieving trustworthiness when dealing with sensitive topics (Gubrium & Holstein, 2002). Particularly, considering the abovementioned double standards of gender appropriate sexual behavior, conducting individual in depth interviews wi th women provide opportunities to probe deeply into meaningful and/or
67 understandings of sexual risk taking in tourism. The participants were recruited via flyers po sted on campus, announcements in classes, and a passive snowball sampling strategy. Snow ball sampling refers to the strategy where a the researcher cannot initiate contact with t he people to whom she was referred; instead, they need to contact her first). The criteria for inviting women to the interviews were that the volunteer had to be 18 or older and had engaged in any kind of sexual risk taking in tourism according to particip subjective perspectives. Fifteen individual interviews were conducted with 13 participants Two of clarifying certain issues. In another case, it was a on the interview and decided to disclose more information as well as to revise some of her initial answers. The phenomenological theoretical perspective of this study implies that each ce is a world of meaning/s. Therefore, the goal wa s to conduct long interviews and the length of the interviews varied between 1.5 hours and 2.5 hours The second interviews conducted with two participants were about one hour each. The interview guide w as initially developed based on the limited literature relevant to the topic of sexual risk taking in tourism and the study goals Expert reviews with two scholars investigating risk taking and/or sexual behavior in tourism were conducted in order to complem ent the ideas, as well as to clarify the wording and the sequence of the questions. In addition, the strength of the in depth interviews stems from the fact that the interviewees bring up unexpected insights. Therefore, the interview guide was constantly d evelop ed throughout the interviews, with question s added, eliminated, and reworded.
68 This strategy allow ed understanding inductively what women perceive to be sexual risk taking, as well as seeing whether it occurs in tourist contexts solely or is part of a behavioral continuum from everyday life. Furthermore, it elucidated the attributes of the tourist experience s that may be conducive to sexual risk taking. Moreover women elaborate d on their motivations for engaging in sexually risky behaviors in tourism self perceptions, and meanings attached to sexual risk taking in tourism. The insights from the answers to these questions serve as a basis e meaningful insights to health promoti on strategies. The IRB app roved informed consent for the first phase and the comprehensive latest version of the interview guide are presented in Appendices A and B, respectively. The questions were complemented with intrinsic probes prompted by the progression of the interview. Th e interviews were audio recorded and transcribed verbatim with the permission of the interviewees whose confidentiality is guaranteed. Considering the sensitive nature of the researched topic, the ethical issues require particular attention. The procedures related to ethical issues in this study followed Binik, Mah, data, and investing maximum effort in making the interviewee feel as comfortable as possible. As for the latter, Catania (1999) suggested a framework of self representation/disclosure to guide a sexological interview in order to maximize an omfort that includes refer ence s demographic characteristics; (2) use of un intimidating wording of the interview questions thereby facilitating trust and disclosure; and (3) the interview context aim ed at maxi mizing the
69 First, the interviewer did her best to achieve a balance between minimizing reactions to a participant would feel accepted, engaging, and actively listened to. As f or the socio demographic characteristics, having a female interviewer to interview female participants minimized the potential risk to trustworthiness bound with tensions and biases linked to gender differences. For instance, Catania, Binson, Canchola, Pol lack, and Hauck (1996) investigated the effects of the research and found that matching interviewers and interviewees on gender elicits better cooperation, particularly on sensitive questions. Moreover, Catania (1999) suggested that greater similarity between the interviewer and interviewee in terms of socio demographic characteristics (such as gender, age, and social class) might elicit higher self disclosure on behalf of the participant. Indeed, the age difference between the i nterviewer and interviewees was not big. Plus, the interviewe r disclosed herself as being a student and based on her subjective perception, the interviews were characterized by quite a favorable degree of disclosure and rapport. Second, the aforementioned expert reviews were helpful in terms of revising the questions and probes in the interview guide to sound as un intimidating as possible. Moreover, there was a learning curve in this respect as questions evolved throughout the interviews. Third, in all the cases, the interviewees were free to choose the time and place of the interview, and all the participants and comfortable place. Participants Prior to the interview, the participants we re asked to read the informed consent, to complete the brief questionnaire about their socio demographic profile, and to pick a pseudonym they would like the researcher to use in any report in order to protect their identities. With a phenomenological appr oach, homogeneity of participants is sought in order to establish common essences of the phenomenon at the nomothetic level. Therefore, the narrow age span of the
70 participants ranging between 19 and 24 (average = 21.5) is compliant with this expectation as well as consistent with the idea of delimiting the study population to late adolescence (as described in the section of literature review). Among the participants, one was married and 12 were single, with three women living with their boyfriends. No one h ad children at the time of the interview. Twelve participants identified themselves as heterosexual and one as lesbian. With respect to the question about highest degree earned, 10 participants indicated having finished high school and three participants Ten participants were enrolled as undergraduates and two interviewees were enrolled in graduate school. As for employment status, seven women indicated being full time students, five women answered being a full time student and having a part time job, and one participant indicated having a full time job, but looking for another one relevant to her career aspirations. Describing their race and ethnicity, three participants self identified as White and/or Cau casian, three as Latin American white (with one women specifying being of Cuban descent and one women being Hispanic Mexican American) three as Asian (with one participant specifying being of Chinese descent and one Khmer and Vietnamese ), three partic ipants self identified as Black (with one participant specifying being Black Haitian and one Black Hispanic ), and one participant as multiracial without further elaboration. Data Analysis The data from the individual interviews were analyzed via a phenomen ological approach provid ing a taking in tourism based on their own subjective perspectives and experiences The implementation of transcendental phenomenology in this study follows the guidelines described by Mous takas (1994). Briefly, the steps of transcendental phenomenology included epoche, phenomenological reduction, imaginative
71 variation and synthesis The data analysis started with ideographic/ individual level and then proceeded to nomothetic/ general / collect ive level. Epoche and p henomenological r eduction Specifically, t he steps of phenomenological reduction were as follows: Epoche / b racketing : The analysis started with the endeavor of attaining epoche i.e., bracketing or at least bridling (Dahlberg, 2006 ; Vagle et al., 2009) placing the focus of the research in brackets an d putting everything else aside This process involved conscious effort of subjective filtering of bias and preconceived notions toward the socially loaded phenomena of sex sexual ris k taking, sexual double standards, STDs and stereotypes about tourist experiences. Horizonalizing : of identifying as many meaningful dimensions (i.e., horizons textural componen ts) as possible, while looking from different angles and dealing with one co researcher at a time. Therefore, this textural stage i.e., the stage of noema was pre reflexive and intuitive. Initially, every statement was treated as having an equal value. Later, repetitive or irrelevant statements were deleted. At this point the researcher focused on the horizons that constitute the invariant qualities of the experience. Starting with horizonalizing, the analysis was supported by the qualitative data analy sis software ATLAS.ti 5. The horizons were identified drawing upon thorough open coding in the software. Then the list of the identified textures was reviewed for duplications and ifferent person in separate list of textures for each interviewee, a separate ATLAS.ti 5 hermeneutic unit was launched for each interviewee.
72 Clustering : At this point, invariant horizons were organized into themes, such as a: sex perceptions cluster, risk perceptions cluster, sexual risk taking dimensions cluster, sexual risk taking cluster, sexual risk taking in tourism cluster, motivations for sexual risk taking in tourism cluster, factors associated with sexual risk taking in tourism cluster ( with sub clusters of facilitators, inhibitors, and context depending factors), etc. In some cases, the same horizon entered more than one cluster, creating some overlap between the clusters. Individual textural descriptions : The purpose of this step was cr eating a coherent, descriptive, textural description of the phenomenon based on the horizons and themes at the ideographic level. Each individual textural description was a couple of pages in length. Each cluster from the previous step had a separate block in the narrative. The approach at this state was purely descriptive. C omposite textural description : Once all of the ideographic textural descriptions were ready, they were integrated into a group collective universal textural description i.e., nomo thetic level composite textural description (presented in Chapter 4). At this point, it was necessary to distinguish between recurring universal and common textures and the less common unique textures. For this purpose, the ideographic documents with clust ered textures (from step 3) were uploaded into the new nomothetic hermeneutic unit and the open coding was conducted on the textures and clusters cutting across the participants and revealing how prevalent the horizons are among the interviewees. Textures discussed by at least nine participants (i.e., 70%) were defined as universal. Textures described by 6 8 participants were determined as common and textures mentioned by five or less participants were defined as unique. The resulting comprehensive list of textures with frequencies and sources participants, in combination with
73 ideographic textural statements, served as the basis for the composite textural description. The focus of this description was on the universal and common textures of the phenomenon. Imaginative v ariation Follow ing the aforementioned steps of transcendental phenomenological reduction, the phenomenological analysis proceeded with imaginative variation and its steps were as follows: Systematic variation : The purpose of this step was syst ematic variation of the possible structural meanings that evoke the textural meanings. Drawing upon extensive textures from the first step, as well as reflection and interpretation, the researcher described the structures that underlie the textures in the the stage of noesis). As opposed to textural description, structural description involved conscious acts of thinking, imagining, recollecting, using fantasy, adopting various perspectives on vie wing the phenomenon, and looking from different vantage points, such as opposite meanings and various roles. The researcher was working from the data, from her personal experience, and from the social narratives. In working with the data, the researcher so rted through the list of horizons identifying textures that could perform as structures. In other words, the researcher was looking for endogenous textures structures. Where such textures str u ctures were identified, the rest of the list was sorted to s ee which textures they underlie. In working from personal experience and social narratives (or in other words, while looking for exogenous structures that might be relevant but are not evolving from the interview per se), the researcher examined a separate cluster of horizons at a time suggesting potential noetic factors for each cluster. For instance, looking at the noematic cluster of motivations for sexual risk taking in tourism of a certain interviewee, the researcher would suggest a number of noetic fa ctors such as experimenting with identity, fulfilling fantasies, self empowerment, reversal of sexual roles, revenge, curiosity,
74 illusion of invincibility, going with the flow, peer pressure, accumulating stories, social status, exercising control, etc. R e cognition of the structures : After the systematic variation that allowed exploration of multiple options, it was necessary to identify the most feasible themes underlying the studied phenomenon. T he most vividly illustrated invariant structures contribut i ng to the development of a structural description of the phenomenon were sought for each interviewee In other words, a list of structural qualities of the experience was compiled for each participant including the items from both working from the data and working from the experience and social narratives Clustering : At this point, the list of the ideographic noetic factors the structural qualities of the experience was organized into major themes clusters, such as the view of sex and sexual risk ta king, different social atmosphere in tourism, sexual risk taking in tourism and perceived lack of social and emotional consequences, sexual risk taking and the characteristics of tourist experiences, etc. Consideration of the universal structures : The ma jor purpose of this step was to consider relevant universal structures i.e., themes that may account for the feelings and thoughts relevant to the phenomenon. In general, s uch universal structures include time, space, bodily concerns, relation to self an d/or others, etc. Among the universal structures complementing this analysis so far were the fun oriented ethos associated with tourist experiences, an instrumental approach to tourism as time out, view of tourism as a rite of passage bound with exploratio n, sexual double standards, gender power differentials and gender passiveness, resistance and sexual role reversals, cultural differences, etc.
75 Individual structural d escriptions : The purpose of this step was creating individual structural description s of the phenomenon based on the structural qualities and themes identified throughout the previous steps Each individual structural description was a couple of pages in l ength. This step elevated the analysis from thick description to interpretative thick description. C omposite structural description : Similarly to the process described with respect to composite textural descriptions, the next step wa s to integrate ideogra phic structural descriptions into the comprehensive group universal structural description i.e., nomothetic level composite structural description (presented in Chapter 4). The same logic of distinguishing between universal (nine or more participants), common (6 8 participants), and unique (five or less participants) noetic factors was applied here as well. For this purpose, the ideographic documents with clusters of structural codes were uploaded to the nomothetic ATLAS.ti 5 hermeneutic unit. These doc uments were open coded cutting across the participants and revealing the common structures among the interviewees The resulting comprehensive list of structures with frequencies alongside with ideographic structural statements, provided the basis for the composite textural description. Finally based on the synthesis of composite textural and structural descriptions, the researcher inferred universal and common meanings and structures reflecting the essence of the studied experience. The process of synth esis wa s intuitive and reflective, where the researcher compile d the essence statement the core meanings of the phenomenon (Moustakas, 1994) The synthesis of the nomothetic textural and structural components reflects the essence of taking in tourism (presented in Chapter 4), pr oviding deep understanding based on their individual voices and subjective perspectives.
76 Trustworthiness The validity and the quality of the analysis were fostered by adhering to the criteria of trustworthin ess in qualitative research i.e., credibility, transferability, dependability, and confirmability (Decrop, 2004; Lincoln & Guba, 1985). Credibility refers to the degree of subjective relevance of the interpretations to the participants as well as to the degree of truthfulness of the data. For the purpose of enhancing the former the participants were invited to review the transcripts of their interviews and ideographic descriptions, to comment on their accuracy, and to suggest changes. Out of the 13 parti cipants, seven responded to the invitation but none suggested changes. The degree of truthfulness of the data was hopefully minimized via efforts to eliminate selective perception and reactive effects during the interviews. Transferability refers to the p ossibility of applying interpretations to similar phenomena. Extensive composite textural and structural descriptions, as well as the essence statement, provided in this dissertation offer transparency and a s such offer a solid ground for evaluating such p interpretations, and what has actually happened in the research setting. This requirement was addressed through preparing, revising, and documenting changes in the interview guide and the research plan. Confirmability is related to the aspiration of providing a variety of relevant explanations to the studied phenomenon without bias. In transcendental phenomenology, the epoche/bracketing/bridling phase (with respect t o bias) and the imaginative variation phase (with respect to variety of explanations) address this aspiration by definition. Furthermore, phenomenology follow
77 Phase 2 Quantitative Method The quantitative phase in this sequential mixed methods design dr e w upon the findings from the qualitative phase and, likewise, was exploratory in nature. One of the advanta ges of mixed methods is implementing triangulation, which allows for looking at the same questions from different v a ntage points and various epistemological approaches (Decrop, 2004). While there wa s an epistemological tension between the qualitative and q uantitative phases, the analytical procedures chosen for this study we re exploratory in nature and in many respects complement ed each other thereby leveraging the potential of triangulation. For instance, the phenomenological approach used in the qualitat ive phase help ed to identify the textures and structures of the phenomenon, while exploratory quantitative analytical techniques in the second phase explore d the dimensions of the phenomenon statistically. Instrument Construction The development of the q uestionnaire proceeded through the following steps. First, the taking in tourism. Second, based on the findings from the interviews and relevant literature, the initial draft of the questionna ire items was developed. Further, the items were arranged in the instrument considering potential order effects. Third, this initial survey draft was pre tested via two cognitive think aloud interviews. Both interviews were audio recorded and extensive not es were taken, generating a brief report of problematic issues with some questions and suggested revisions. The questionnaire items as well as their order were revised based on insights from the interviews. Fourth, a professor specializing in survey design and three other professors having experience with researching sexual behavior in the fields of leisure and tourism provided their feedback on various preliminary drafts of the instrument.
78 In addition to drawing upon empirical findings, the literature was searched for any relevant instruments that could be adopted and/or adapted. While the search did not provide an entire scale that could be adopted for investigating perceptions of sexual risk taking in tourism, two sensation seeking scales and multiple to urism related items were found to be insightful. Guided by the sensation seeking theory in second phase of this study, the sexual sensation seeking (SSS) and nonsexual experience seeking (NES) scales developed and validated by Kalichman et al. (1994) and K alichman and Rompa (1995) were integrated into the instrument. The sexual sensation seeking scale has been found to correlate more highly with sexually risky behaviors than the non sexual sensation seeking scale (Gaither & Sellbom, 2003; Kalichman et al., 1994) and its reliability, validity, and relationship with sexual risk taking has been confirmed in several studies (e.g., Chng & Gliga Vargas, 2000; Kalichman & Rompa, 1995; McCoul & Haslam, 2001; Parsons et al., 2001; Wiederman & Hurd, 1999). Additiona lly, some of the items found in the literature on sexual risk taking were paraphrased and included in this instrument. With respect to ideas about sexual risk taking in and Smith (2007) were found helpful. With respect to sexual risk taking in tourism per se, parts of two instruments were found in the literature. One instrument was partially presented by Eiser and Ford (1995) and Ford and Eiser (1996). Another instrument interpersonal behavior, was developed and implemented/published several times with respect to sexual behavior on spring break by (in chronological order) Maticka Tyndale and Herold (1997), Maticka Tyndale et al. (1998), M attila, Apostolopoulos, Snmez, Yu, and Sasidharan (2001), Apostolopoulos et al. (2002), Maticka Tyndale et al. (2003), and Snmez et al. (2006); and with respect to sexual behavior at Mardi Gras by Milhausen, Reece, and Perera (2006). Finally, a pilot
79 tes t was conducted to refine the instrument (with both empirically based and literature driven scales and items) before the main data collection for the quantitative phase. An email invitation to participate in the study, informed consent for the online surve y, and the final version of the survey instrument are presented in Appendices C, D, and E, respectively. Instrument Validation Validity in the quantitative phase was assessed using c ontent, criterion related, and construct validity (Cronbach & Meehl, 1955 ) Content validity is a non statistical type of validity that involves systematic examination of the instrument content to assess its coverage of the domain under investigation A n inductive qualitative phase before proceeding to cognitive intervie ws, followed by expert content reviews and a pilot study served as the means of establishing content validity for the instrument. Criterion related validity refers to the correlation between the test and a criterion variable/s taken as a valid representati ve of the construct. Since no scale for investigating perceptions of sexual risk taking ( in general or in tourism in particular ) was found in the literature, it precluded establishing concurrent or predictive criterion related validity for this instrument. Construct validity indicates the extent to which the operationalization of the construct of sexual risk taking in tourism actually measures the construct that it is supposed to measure. Statistical analyses for assessing the internal structure of the con struct (i.e., exploratory factor analysis and exploratory structural equation modeling) as well as internal consistency of the (see Chapter 4 and the Tables 4 2, 4 3, 4 5, and 4 6) As part of the construct validation, relationships between the construct of interest and the measures of other relevant constructs should also be assessed i.e., convergent and discriminant validity. To establish convergent validity relationships between sexual risk taking in tourism and the NES and SSS scales (Kalichman et
80 al., 1994; Kalichman & Rompa, 1995) as related measures of non sexual and sexual sensation seeking were analyzed (see Chapter 4 and the Tables 4 3, 4 5, and 4 8) Also, d iscriminant validity was assessed with respect to examining the internal structure of the construct of interest (see Chapter 4 and the Tables 4 2, 4 3, 4 5, and 4 6) Measurement The research instrument for the quantitative phase consisted of five sections: (1) p ersonality traits and sexual behavior (2) vacation as a context for sexual behavior (3) motivations for sexual risk taking in tourism ( 4 ) perceptions and experiences of sexual risk taking in tourism and ( 5 ) s ocio demographic s (Appendix E) Section one was comprised of the nonsexual experience seeking (NES) scale and sexual sensation seeking (SSS) scale, with 11 items each, measured by a 4 ichman & Rompa, 1995). Section two started with two nominal questions investigating the features of tourist experiences perceived as the ultimate touristic contexts facilitating sexual activity, both with a steady and a casual sexual partner (i.e., questio risk taking in the contexts described in questions 3 and 4 using a 5 point Likert scale where attitudes with respect to the incidence and acceptability of sexual risk taking in tourism with 10 items using a 5 f the touristic characteristics that act as facilitators and/or inhibitors of sexual behavior in tourism with 28 items using a 5 point Likert scale where whether and what additional touristic characteristics can affect sexual behavior in tourism.
81 taking in tourism. Question 9 measured motivations for sexual risk taking in tourism with 25 items usi ng a 5 point 10 inquired whether and what additional conditions might motivate participation in sexually risky behaviors in tourism. Section four focused on the with sexual risk perceptions about the vacation activities they considered to be sex (using a 5 point Likert scale perceptions of risk associated with tourist related sexual activities (20 items using a 5 point inquired whether and what additional sexual activities might be considered as risky in tourism. sexual activity in tourism risky (24 items using a 5 point Likert s sexual risk Finally, section five focused on the socio d emographic characteristics of the participants with two numeric open ended questions for age (question 16) and number of sexual partners (question 24), as well as three ordinal questions for education/year at school (question 17), and parental/legal guardi respectively). The rest of the socio demographic characteristics were measured using nominal level questions. Data Collection The data were collected via an online survey using t he Qualtrics software hosted by the University of Florida T his method provides participants with confidentiality and anonymity,
82 which is crucial when investigating such a sensitive topic. In order to encourage participat ion in accordance with Dillman, Sm yth, and Christian 2009 ) advice, the following steps were conducted. First, potential participants received a pre notice email with brief explanations and an invitation to consider participating. After two days, an invitation email with the link to the survey was sent out (Appendix C). After four days, thank you/reminder emails were sent to all the potential participants. The second round of the reminder emails was sent out after a week and a half from the first reminder to all the potential participants The data collection was closed in one month from the second contact i.e., invitation email. Sampling The population for this study consisted of female students enrolled at the University of ffice provided an email list of all 47,647 registered students. The sampling frame consisted of 25,698 female students (53.93%). Excluding the researcher left a target population of 25,697 female students for this study. The sampling process was conducted in Excel. Using a combination of systematic random sampling (with a random entry point and a constant sampling interval) and stratified sampling (with proportional representation according to class standing ) a sample of N = 4,282 was drawn Specifically, the following steps were conducted. After men were eliminated as ineligibles from the sampling frame, it was sorted by the class standing and then by random ID (which was created for each member of the sampling frame) to ensure stratification with proporti onal representation according to class standing. Then, the sampling fraction was dissertation about sexual risk taking and alcohol consumption among the UF studen ts) by the target population (i.e., 25,967), which yielded a result of .15. The inverse of the sampling fraction was 6.49 (i.e., 25,967 divided by 4,000), which meant that either each 6 th or each 7 th
83 woman needed to be invited to participate in the survey. Therefore, each 6 th female student was invited to answer the online questionnaire resulting in the total of 4,282 email invitations having been sent out. Initially, 1,445 women responded to the invitation (preliminary response rate of 33.75%). Yet, elimin ating substantially incomplete responses decreased the effective response rate to 19.92% ( N = 853). Sa mple Table 3 2 presents the socio demographic characteristics of the women who participated in the quantitative phase of this study. Following the hand ling of the missing data (i.e., listwise deletion of cases with substantially incomplete data for section 4 ), the effective sample size used for the data analysis was N = 853. The participants ranged in age from 18 to 62 years, with a mean age of 23.5 year s ( SD = 6.67). The predominant age groups were 18 20 years with 40.4% ( n = 343 ) and 21 23 years with 29.2% ( n = 248 ) Among the participants, 14.0% ( n = 119 ) were freshmen, 15.0% ( n = 128 ) sophomores, 18.3% ( n = 156 ) juniors, 16.5% ( n = 141 ) seniors, and 3 4.2% ( n = 292 ) were enrolled in graduate school. The majority of the participants were born and grew up in the United States (81.5% ( n = 695) and (88.9% ( n = 758), respectively). In terms of the racial or ethnic background, two thirds of the sample 66.6 % ( n = 568 ) identified themselves as white/Caucasian, not of Hispanic origin; 6.4% ( n = 55) as black, not of Hispanic origin; 13.7 % ( n = 117) as Hispanic; 8.3 % ( n = 71) as Asian; 0.4% ( n = 3) as Native Hawaiian/other Pacific Islander; 0.6% ( n = 5) as Ameri can Indian/Native Alaskan, and 3.6 % ( n = 31) as other. With respect to marital status, 84.4% ( n = 720) were never married, 12.8% ( n = 109) married, 2.2% ( n = 19) divorced/separated, and 0.2% ( n = 2) widowed. As for annual income, participants were joint annual household income from all sources before taxes and their personal annual income from all sources before taxes. Regarding 9.1% ( n = 78) reported under $2 5,000;
84 15.9% ( n = 136) between $25,000 $49,999; 17.1% ( n = 146) between $50,000 $74,999; 16.3% ( n = 139) between $75,000 $99,999; 14.4% ( n = 123) between $100,000 $124,999; 8.4% ( n = 72) between $125,000 $149,999; and 16.8% ( n = 143) over $150,00 0. As for personal annual income, 24.0% ( n = 205) reported under $5 ,000; 19.6% ( n = 167) between $5,000 $9,999; 14.1% ( n = 120) between $10,000 $14,999; 11.1% ( n = 95) between S15,000 $19,999; 11.0% ( n = 94) between $20,000 $29,999; 5.4 % ( n = 46) b etween $30,000 $39,999; 3.9% ( n = 33) between $40,000 $49,999; and 9.8% ( n = 84) over $50,000. The women were also asked several questions to establish their sexual characteristics. The majority of the participants 92.8% ( n = 792) identified themse lves as heterosexual, 1.4% ( n = 12) as gay/lesbian, 4.8% ( n = 41) as bisexual, and 0.7% ( n = 6) as other (i.e., pansexual and queer). As for the dating status, 51.8% ( n = 442) reported being in an exclusive relationship, 1.4% ( n = 12) in an open relationsh ip where both partners can see other people, 10.7% ( n = 91) were casually dating, but were not in a relationship, and 35.3% ( n = 301) reported not dating and not being in a relationship. Regarding the number of sexual partners, 20.3% ( n = 173) reported not having had sexual partners, 21.1% ( n = 180) reported having had one sexual partner, 11.4% ( n = 97) reported two sexual partners, 13.6% ( n = 116) had 3 4 partners, 13.6% ( n = 116) had 5 7 partners, 8.3% ( n = 71) had 8 10 partners, and the rest of the sampl e (9.6%, n = 80) had more than 10 sexual partners. As for other characteristics, 8.3% ( n = 71) reported having experienced an unplanned pregnancy (personal or sexual partner's experience). Additionally, 49.2% ( n = 420) reported having been tested for an S TD in the past and 8.2% ( n = 70) reported having been diagnosed with an STD. Furthermore, 18.4% ( n = 157) believed that some of their sexual practices were putting them at risk of contracting HIV/AIDS and 0.8% ( n = 7) admitted having had unsafe sex
85 with so meone they thought might have HIV/AIDS. Finally, 10.7% ( n = 91) reported having had sex without giving their consent and 0.6% ( n = 5) reported having had sex with someone without Data Analysis All the statistical analyses in th is study were conducted using M plus 6 and IBM SPSS Statistics 19. Descriptive statistics were used for research question 1a (w hat sexual activities are perceived as risky in tourism by the female college students? ). Considering the fact that the construct validity of the perceived dimensions of sexual risk in tourism has not been established before exploratory factor analysis (EFA) was conducted EFA is an interdependence technique (i.e., there are no dependent and independent variables) and its purpose is to define the underlying structure among the variables involved in the analysis based on what the data offers rather than using a priori constraints (Hair, Black, Babin, & Anderson, 2010). O blique rotation s (either Direct Oblimin or Geomin) were chosen in order to allow the factors to correlate C onstraining the factors to be orthogonal might not represent the reality and contradict s the exploratory nature of the analysis The number of the retained factors was guided by diagnostics such as Eigenvalues > 1 as a minimum based on the Kaiser criterion, Scree plot, and communalities. The goodness of fit was also assessed based on the goodness of fit test ( 2 ) and goodness of fit indices : (1) comparative fit index ( CFI ) and Tucker Lewis Index ( TLI) with 0.90 reflecting acceptable and 0.95 indicating a good fit for both CFI and TLI; and (2) Standardized Root Mean Square Residual (SRMR) and Root Mean Square Error of Approximation (RMSEA) with values less than 0.05 representing a good fit and values ranging from 0.08 to 0.10 reflecting acceptable fit This approach was used to answer research questions 1b (w hat are the perceived dimensions of
86 sexual ri sk taking in tourism by the female college students? ) and 2 (w hat are the dimensions of motivations for sexual risk taking in tourism among the female college students? ). According to Hair et al. (2010), if the differ ences are expected across the groups of interest in the sample (e.g., sensation seeking propensity levels in this study) separate factor analys e s should be performed for each group Exploratory structural equation modeling ( ESEM ) was utilize d to address this issue. ESEM is a recent approach developed by Asparouhov and Muthn (2009), and Marsh et al. (2009, 2010, 2011 2012 ). Its flexibility stems from its ability to simultaneously incorporate the advantages of exploratory factor analyses, confirmatory factor analyses, and structural equation modeling. An ESEM model can be extended to multiple groups or multiple occasions analysis, where the model is estimated for each group/occasion and some parameters of the model can be constrained to invariance across the groups or the occasions (Marsh et a l., 2009). Marsh et al. (2009, 2010, 2011 2012 ) operationalized and tested a taxonomy of 13 partially nested models of invariance tests that integrated the factor analysis and measurement invariance traditions. The taxonomy allows the investigati on of bot h the invariance of the factor structure (factor loadings and factor variances covariances) and of other measurement parameters (item intercepts and uniquenesses, and latent factor means). The models vary in terms of imposed restrictions from the least res trictive model of configural invariance to the most restrictive model of complete invariance. Tests of c onfigural invariance weak factorial/ measurement invariance, strong factorial/measurement invariance strict measurement invariance latent factor vari ance covariance invariance, and the invariance of the factor means were conducted in this study. The goodness of fit was also assessed based on the goodness of fit test and goodness of fit indices investigating the decrement in the goodness of fit due to
87 imposing various invariance constraints in each model. This approach was used to answer research question 1c (d o the perceived dimensions of sexual risk taking in tourism differ by sexual sensation seeking propensity among the female college students? ). Cl usters of sexual risk takers in tourism were identified though cluster analysis. Cluster analysis is also an exploratory interdependence technique that clusters objects (in this case participants) based on the characteristics they possess (i.e., cluster va riate) into groups with maximum similarity within the groups and maximum heterogeneity between the groups (Hair et al., 2010 ; Mooi & Sarstedt, 2011 ). Cluster analysis is comparable to factor analysis, but while factor analysis groups variables cluster ana lysis groups objects on the basis of distance/ similarity/proximity (Burns & Burns, 2008 ; Hair et al., 2010). Drawing upon Mooi and Means procedure was employed, and, finally, TwoStep clustering was implemented. This sequence allowed an exploratory start and facilitated evaluating the stability of the results (Mooi & Sarstedt, 2011). A hierarchical procedure as a starting point was preferable for this explorator y study since, unlike the nonhierarchical procedure, it does not require setting the number of clusters in advance. Hierarchical cluster analysis was conducted based on the factor scores computed with respect to the research questions 1b and 2. od and squared Euclidean distance were used as the cluster method and a distance measure, respectively. Agglomerative methods were used, where each participant started as its own cluster and at each step the two most similar clusters were united until a s ingle cluster remained. Based on the dendrogram and agglomeration schedule, an elbow/sharp drop in homogeneity was found and used to calculate the number of clusters. Additionally, centroids for the clusters were computed based on the hierarchical
88 solution The centroids and the number of clusters from the hierarchical solution were inserted into the iterative K Means cluster analysis to ameliorate the composition of the clusters. Lastly, a TwoStep cluster procedure was employed based on Bayesian Informatio n Criterion (BIC) and Akaike's Information Criterion (AIC) without setting in advance the number of clusters to evaluate cluster stability and quality, as well as the importance of each classifying variate. The clusters were interpreted and labeled. Th es e procedures were used to answer research question 3a (w hat are the clusters of sexual risk takers in tourism among the female college students based on the perceptions of and motivations for sexual risk taking in tourism? ) Further, the profiles of the c lusters of sexual risk takers in tourism were examined with 2 tests for non metric variables and multivariate analysis of variance (MANOVA) for metric variables. This approach was used to answer research question 3b (w hat are the socio demographic, sexua l, and personality (in terms of sensation seeking) profiles of the clusters of sexual risk takers in tourism among the female college students? ). Descriptive statistics were used to answer research question 4a (w hat are the characteristics of the tourist e xperiences facilitating and/or inhibiting sexual risk taking in tourism among the female college students? ). Finally, MANOVA was used to answer research question 4b (w hat are the perceptions of the risk taking clusters among the female college students reg arding the characteristics facilitating and/or inhibiting sexual risk taking in tourism? ).
89 Table 3 1. Socio demographic characteristics of the participants in the qualitative phase Pseudo nym Age Marital status Highest degree earned Current enrollme nt Occupation Racial or ethnic background Cheryl 23 single, cohabitating with boyfriend B.A. graduate student, part time employee Caucasian Carmela 22 single high school undergraduate student, part time employee Latin American white Maya 20 si ngle high school undergraduate student Asian Eileen 19 single high school undergraduate student Latin American white Jade 22 single high school undergraduate student, part time employee Black Faith 24 married B.A. no ne full time e mployee Black Mari a 20 single high school undergraduate student Latin American white Melon 24 single, cohabitating with boyfriend B.A. graduate student Asian Vanessa 21 single, cohabitating with boyfriend high school undergraduate student Black Desy 21 single h igh s chool undergraduate student Multiracial Mariah 20 single high school undergraduate student Caucasian Celine 21 single high school undergraduate student, part time employee Caucasian Selinda 21 single high school undergraduate student, part time employee Asian
90 Table 3 2. Socio demographic characteristics the participants in the quantitative phase Characteristics Frequency (n) Percent (%) Age (n = 849) 18 20 343 40.4 21 23 248 29.2 24 26 93 11.0 27 30 71 8.4 31 40 59 6.9 41 50 2 7 3.2 51 62 8 .9 Year in school (n = 852) Freshman 119 14.0 Sophomore 128 15.0 Junior 156 18.3 Senior 141 16.5 Graduate 292 34.2 Other 16 1.9 Place of birth (n = 849) US 695 81.5 Not US 154 18.1 Place of growing up (n = 851) US 758 88.9 Not US 93 10.9 Racial or ethnic background (n = 850) White/Caucasian, not of Hispanic origin 568 66.6 Black, not of Hispanic origin 55 6.4 Hispanic 117 13.7 Asian 71 8.3 Native Hawaiian/Other Pacific Island er 3 .4 American Indian/Native Alaskan 5 .6 Other 31 3.6 Marital status (n = 850) Never married 720 84.4 Married 109 12.8 Divorced/Separated 19 2.2 Widowed 2 .2 Parents'/legal guardians' joint annual household income from all so urces before taxes (n = 837) Under $25,000 78 9.1 $25,000 $49,999 136 15.9 $50,000 $74,999 146 17.1 $75,000 $99,999 139 16.3 $100,000 $124,999 123 14.4 $125,000 $149,999 72 8.4 Over $150,000 143 16.8 Personal annual i ncome from all sources before taxes (n = 844) Under $5,000 205 24.0 $5,000 $9,999 167 19.6 $10,000 $14,999 120 14.1 S15,000 $19,999 95 11.1 $20,000 $29,999 94 11.0 $30,000 $39,999 46 5.4 $40,000 $49,999 33 3.9 Ove r $50,000 84 9.8
91 Table 3 3. Sexual characteristics of the participants in the quantitative phase Characteristics Frequency (n) Percent (%) Sexual Orientation (n = 851) Heterosexual 792 92.8 Gay/Lesbian 12 1.4 Bisexual 41 4.8 Other 6 .7 Dating status (n = 846) Exclusive relationship 442 51.8 Open relationship where both partners can see other people 12 1.4 Casually dating, but not in a relationship 91 10.7 Not dating and not in a relationship 301 35.3 Number of se xual partners (n = 833) no partners 173 20.3 1 partner 180 21.1 2 partners 97 11.4 3 4 partners 116 13.6 5 7 partners 116 13.6 8 10 partners 71 8.3 11 15 partners 38 4.5 16 20 partners 23 2.7 21 30 partners 15 1.8 31 a nd more partners 4 .5 Unplanned pregnancy (personal or sexual partner's experience) (n = 848) Yes 71 8.3 No 747 87.6 I do not know 30 3.5 Having been tested for STD (n = 846) Yes 420 49.2 No 426 49.9 Having been diagnosed with STD (n = 850) Yes 70 8.2 No 780 91.4 Having had sexual practices putting at risk for HIV/AIDS (n = 849) Yes 157 18.4 No 692 81.1 Having had unsafe sex with someone the participant thought might have HIV/AIDS (n = 850) Yes 7 .8 No 843 9 8.8 Having had sex with someone without giving consent (n = 849) Yes 91 10.7 No 758 88.9 Having had sex with someone without getting their consent (n = 849) Yes 5 .6 No 844 98.9
92 CHAPTER 4 FINDINGS Findings from the Qualitative Pha se Composite Textural Description The women discussed both the variety of tourist experiences a s well as the diversity of sexual experiences in tourism that they perceive as risky. In terms of the palette of tourist experiences, characteristics such as len gth ( n = 9), destination ( n = 9), and trip companions ( n = 12) were universally described as affecting sexual behavior (this issue will be addressed in depth in the section of structures). Among the specific discussed scenarios participants brought up spring breaks, rest and relaxation (R&R) beach vacations, backpacking trips, sightseeing city breaks, study abroad trips, business/academic trips, cruises, etc. As for risky sexual activities in tourism, participants commonly ( n = 8) perceived sexual ri sk The universal scenarios discussed by the women were (1) casual sex (or sex with a stranger, hookup, one night stand, sex with a friend without commitment) either with a fellow tourist and/or with a local sexual par tner ( n = 13) and (2) unprotected penetrative sex (typically referring to not using a condom; n = 11). However, discussions of sexual risks within steady relationships, as well as the risks associated with sex even when proper protection is used, were also common ( n = 8 and n = 6, respectively). The unique scenarios discussed by the women as sexual risk taking in tourism included different types of sexual activities (e.g., vaginal sex, anal sex, oral sex, digital sex, kissing/making out), as well as various actual situations (e.g., being more flirtatious, having sex in semi public spaces, attending a strip club or live sex show, loosing
93 virginity, being raped, feeling pressured/coerced into sex, being harassed/stalked, feeling taken advantage of, being lied to in order to have sex, going with a stranger to his room/car) and potential situations (e.g., attending a swingers club, having sex with another woman, having sex with multiple partners at the same time, being drugged, bring kidnapped, being potentially killed). p erceptions of s exual r isk t aking in t ourism One of the universal textures brought up by the participants was the idea that sexual risk taking is more prevalent, probable, or frequent in certain types of tourist experiences ( n = 10). For in tourist destinations there are more chances that you will find other tourists like you w ho would want to get together in a sexual way [laughs]. ated that on vacation, some people act different from how they normally are, because there's no standard as to how they should behave. [sexual risk taking] definitely more pre valent in tourism, I you have the ability to be s as a tourist going to affect you r life back home Eileen indicated: [At home] I have responsibilities, obligations, people that rely on me, I have to keep my shit together b ut when you are a tourist, let loose, bec ause there is n othing people are prone to taking risks when they are traveling, because they have all this freedom that makes them not really think about any consequences, because they're going to leave all of that behind when the y're done and go back home to live their normal life. When you're traveling you're not living act like you normally would act, because how you normally act is kind of bo ring probably It's like being in some kind of fantasy when you're traveling. You can just do whatever
94 Participants also commonly suggested that sexual risk taking is legitimate, appropriate, or even expected during some tourist experiences, but is not so mething that they would be eager to do at home ( n I'm not really like that, like hook up with anybody, but that's just part of being a tourist and since this was my first night out an d I was okay with this guy and maybe just this whatever attitude. Likewise, Carmela, reminiscing about a vacation, I drank and had sex all the time .. it's not acceptable in everyday life almost taboo [but] it's exciting to do something different, out of the norm and it's just a new experience to life. having risky sex on vacation would if you were to do it on an everyday kind of routine It doesn't mean anything our age, when you go on v acation, it's expected to do crazy stuff if it were to happen like within the normal routine then I'll be like okay, there's a problem. Dimensions of s exual r isk t aking in t ourism In general, participants argued that sexual risk taking is difficult to define ( n = 8) it depends on the person one gets to define on you're a risk adverse perso n [or] a risk taking person taker ll most likely take risks in tourism and go out to meet new people and sleep around. taking in tourism ultimately depends on the person and their personality they are more conservative and mindful t h e n they will probably be more alert, but if they are more carefree, they would probably take advantage of societal leniency and get rowdy oftentimes those are the people who probably are doing the sexual acts and hurting themselves or hurting others Having said that, all the participants discussed sexual risk taking, in general, and in tourism, in particular, as a multidimensional phenomenon and there was a high level of congruence in terms of the perceived dimensions. Among the universal dimension s were
95 physical/sexual health ( n = 13), emotional ( n = 13), social ( n = 13), and mental ( n = 10) dimensions. Another common dimension was cultural ( n = 7) and among the unique dimensions discussed were legal ( n = 4) and medical lack of familiarity with l ocal healthcare ( n = 3). For there is physical, psychological, emotional, cultural lack of awareness, social norms and religious aspects, and then there is this internalization of g u ilt. Like being women we are supposed to be responsible for not attracting unwanted attention. sleeping around with a whole bunch of people mentally you can feel like you have no self kind of uns table because to person P hysically you might catch an STD (sexually transmitted disease) and it might lead to death or something. The classification into dimensions in this study is adopted for purely analytical p urposes, To me, it all ties your mind. I feel like your body is just a reflection of your thoughts and your feelings All the participants were asked to compare sexual risk taking dimensions in tourism vis vis dimensions are basically the same at home and o n vacation, well except for maybe lack of cultural awarene s s because you do have this awareness at home b ut I guess there are different weights and that really depends on the situation As an example, Carmella approached this question from the st andpoint of cost benefit analysis as applied to sexual risk the benefit s are the same of like having an enjoyable experience
96 emotional cost would be higher at home [as in tourism] both parties understand that we are going perceptional cost shrinks i n tourism because sexual risk something normal, everyone does this so I that might make me think, h, what kind of a person am I? physical cost is either equivalent or the only one to become higher in tourism (which was a universally discussed notion; n = 13) because of the lack of familiarity with the area and the culture. Plus, not everybody shared the sentiments about shrinking me ntal and emotional costs or consequences in any time you engage in sexual behavior, I think emotions can come into it, and so I think that emotional consequences aren't minimized just because you are in another city or count ry. dimension. Physical/sexual health dimension of sexual risk taking in tourism The physical dimension of sexual risk taking in tourism was discussed universally ( n = 13) and it was typically emphasized as the most important one. For instance, Eileen argued, I am afraid of being physically hurt more than I am emotionally the re when I think of sexual risk taking. For some women, the physical dimension was identical with the fear of STDs and they called it the dimension of sexual health. In this context, casual sex was the main concern, but some women did not rule out this ris k in steady relationships as well. Selinda used to c heat on her, which diminished her trust in him, and when they celebrated their reunion
97 to have sex with somebody without risk .. even in your relationship a guy could cheat on you any day and just bring anything with him to you In general, in terms of physical endangerment, the universal potential consequences that the women fear ed were contracting STDs or HIV/AIDS in particular ( n = 13), getting pregnant ( n = 12), and facing sexual violence of any sort ( n I have three things in my definition [of the physical dimension]: I have got the pregnancy, the STDs and the potential of rape or brutality t home maybe I would be more worried about pregnancy, and then as a tourist I might be more worried about brutality or STDs. the main points are pregnancy and then ST D s and then anything else that is kind of dangerous, with like sexual escapades, like not knowing the person can be dangerous in terms of like date raping, being drugged, or even killed feeling vulnerable in tou hen we get into the physical aspect of it [casual sex], y ou .. he may force you to have sex without a condom e could be crazy or a potential stalker. if you le ave the US, other countries have really high rates of HIV and stuff like that. So the thing [sex] has a really high risk. Melanie described the consequences of unprotected sex with an unsteady partner on vacation and said that he brought me some kind of a morning after pill a nd said that I should take it to avoid either with a s physical, because sex is
98 e case of rape in tourism not only hurting you physically, but is also hurting you e motionally. Finally, the participants stated that being women (travelling solo or with other female partners) was bound with a sense of vulnerability and being s ubjected to the sexualized male As a female, like rape being overpowered and sexual violence are the top things I need to be the guys there were like super aggressive They were just sexually aggressive. I mean, kissing your and they all just kept trying to lead me out to the b each. take advantage of your vulnerability in a new city. . They know you are not local so like you are a moving target for them a nd that puts you at risk Social dimension of sexual risk taking in tourism The social dimension of sexual risk taking in tourism and everyday life was also universally discussed ( n = 13), but from different angles. Among the social consequences of sexual risk taking ( n = 11), w omen discussed social disapproval, spoiled reputation and tainted professional image, ruined friendships, and breach of privacy. The women emphasized the necessity to maintain their good reputation at home that could be jeopardized if information about the ir sexual risk taking at home or on vacation became public. For example, Jade argued, at home this is where your job is, your reputation is, so you don't want to mess that up because this is the society where you have to live say you go out to a club drink and go crazy, flirting with guys and making out ... someone from your workplace or classroom might see you and that will carry on to how they perceive you and interact with you. social thing is important too
99 ou just gave all these people the reason to talk about your personal life, which it happened to me. judge whether you are a decent person or not especi ally if you are a woman concluded that sexual risk will make me look bad and promiscuous and that's not a really good thing for girls. Yet, when it came to tourism contexts, participants explain ed that this social gaze diminishes and sometimes even disappears, which affects their sexual behavior. For instance, In tourism, I think that people feel more inclined to do things that the loose in another country because nobody would really remember There would be no ramification s or precautions lack of social expectation in tourism: r what about who I social expectations that come from the culture, but then because you are coming from a different culture into that cu lture, you don't necessarily know what their social expectations are a nd anymore anymore I kind of usual ly just throw this out the window. it allows for sexual risk place and make this new persona [ and] it can come back with you [home] if you want it, but it You
100 Furthermore, participants emphasized gender differences ( n = 13) and argued that diminished social con trol in tourism is more important to women compared to men in terms of I think the vacation is actually less of a trigger for the guys, they can easily have all of this [sex] at home, but for the girls, the social and emotional sides become somewhat less of an issue on vacation. W omen have to worry about their reputations and gossips that spread out. So, in this way, tourism offers us some advantages. Like, women might just relax because what someone thinks, or says, or gossips. The fact that they can come home and don't have to worry about having that stigma attached to them because whatever they did happened in that location only and it ca n just be kept there. So I feel like women probably will be more encouraged kind of to have more sexual risk taking behaviors somewhere on vacation. if we are looking at the stereotype that men are more sexually aggressive a nd women are more passive I think that both can become more aggressive when they travel, but I do think that it would be a bigger cha nge for people who are a little more aggressive. taking in tourism vis vis men would either be the same or greater and women would be greater. Like the difference for w more open to sexual experiences while they are traveling because they want it to translat e into a the fantasy of being swept off their feet meeting someone and having Yet, the women explained that the social control over female sexuality does not disappear entirely in tourism contexts and sometimes sexual escapades away from home can find their way to haunt the woman at home. Vanessa discussed vacations with friends and arg become more liberated and trying new things. In a way, it can ma
101 and probably bring it back with them and sometimes p eople [are] just going to exaggerate things Furthermore, for some partici they were all very prudish, to the point of racists .. they were just complete ly talking about is going to get AIDS nd it really hurt, even though I had no respect for these women But still people talking about you, it stinks Celine described a similar women either become openly jealous or just judging .. for expectation that s wrapped up in it L ike, h, now this girl is going to be like this and why not try as well as well as thi nking neg jeopardizes your social status your friends d because they have so much trust in the guy they say, ou are just saying that to get attention the example of women wearing miniskirts that they are opening themselves up to be raped .. society, you can change yourself. ons of the social dimension of sexual risk taking mainly revolved around its importance at home and relative relief from it in tourism
102 contexts, particularly for women. Yet, some element of social judgment was present in tourism, or its potential spillover to everyday life, especially when travelling with friends from home. Emotional dimension of sexual risk taking in tourism Another universal dimension discussed by the women was emotional ( n = 13). Among the emotional consequences described by the women we re emotional distress, trauma, or attachment, as well as the emotional tolls bound with unwanted pregnancy, having to be tested for STDs, and feeling used, humiliated, guilty, unvalued, empty, tainted, mistreated, manipulated, etc. With respect to potentia l emotional attachment in casual sex or other forms of short term relationships, for some women, this aspect of the emotional dimension diminished in those emotional psychological consequences are going to be way high more prolonged relationship, this is r and you sort of set your emotional psychological limits and ended because your life here is open ended. that little thin g that we might get [a] tiny wee bit attached, but it's more of a risk at home. guys just butter you up .. wh en he just stopped talking to me, I was thinking, He just used me I got hu just a bad experience. You kind of say you won't get attached, but as a girl for me it was hard to do that. also emphasized the narrative of romance bound with ome girls are not used to casual sex but when they go over to Europe, they will paint this image and disguise casual sex as some romantic
103 experience. So there is more risk for those girls to get hurt emotionally. of a short term relationship in tourism where she felt that she was pushed into sex in general and t know how to reject .. it when I just got back home, I even cried a lot You start processing what happened and you start crying argued: [On vacation] emotional consequences can happen, definitely. And I think that feeling a new person and pushing the boundaries doesn't save you from emotional consequences I do think that emotional consequences will come back, even with that anonymity, because at the en d of the day you are still They might not affect you the way they would if you were back home, but I think that there is still some kind of emotional consequence definitely that goes in to sexual behavior when you try something new, you really w ant it to want to do it again Another aspect of emotional dimension refers to the emotional toll bound with unwanted pregnancy or contracting an STD. Cheryl, discussing unwan ted pregnancy as a result of sexual risk because you are d you have got to raise a baby o r you have to get an abortion d t hat is horrible emotionally. having an STD can make you feel dirty it's not a fun experience .. I mean, it's physically and emotionally and mentally physical aspect of STDs is obvious and you have to get i a pregna ncy scare takes an emotional toll .. h m an an STD Y ou may have to go get tested and wait days a
104 Mental dimension of sexual risk taking in touris m In addition to the physical, social, and emotional dimensions of sexual risk taking, almost all the participants ( n = 11) discussed various mental/psychological matters bound with sexual risk taking in general and in tourism in particular. Most of the pa rticipants used the label perception, self judgment, self esteem, self respect, clash of identities, feeling uncomfortable, clash with upbringing and morals, internalization of guilt, and feeling you did somethi ng wrong. Among the common mental consequences discussed by the women ( n = 9) were negative impact on self perception/image/esteem/worth, regrets, embarrassment, sense of discomfort, and self questioning as a result of sexual risk taking. The distinction between mental and emotional dimensions is artificial in many respects and participants frequently discussed them in conjunction. Yet, the women explained that there is Emotionally, you know, emotional risk with attachment to the casual sexual partner and mental risk with regrets following the activity: I mean, mentally I was almost like, "Why did I do that?" Do you see what I mean? There's a difference. This hurt is not in the emotional sense like attached to this person." It was more like, "I just shouldn't have done that kind of thing." And since you feel more liberated [in tourism] do sexual which is a mental risk that you Participants offered various examples of the mental dimension bound with sexual risk ng unprotected sex can certainly make me feel unease... like my well
105 may not be judged by other people, but they criticize themselves... I think that's a dimension we da also discussed the mental consequences of rape, where Maya stated, ability thin judgment or image, because you feel like you kind of gave sleeping around with a whole bunc h of people, I guess mentally you can feel like you have no self Interestingly, while for some participants the mental dimension of sexual risk taking became les s relevant in tourism, others described is as a stable feature regardless of the context.
106 Desy stated that having sex with more men than she perceives to be appropriate would make her be there, the fact that I was promiscuous in whichever place. So that would be another thing Yet, talking about mental consequences, participants commonly called to dis tinguish between detrimental and beneficial consequences (n = 7). For instance Maria elaborated: I think that sexual behavior has a lot to do with our mental or emotional health and I think that taking sexual risk can be really positive for your mental emo tional health or it can be very negative I think negative outcomes would definitely be if you feel uncomfortable during or after some kind of sexual activity you feel like I think pos itive would be if you really enjoyed the experience, you found new sexual activity that you really enjoy or you were just really happy that you went outside your comfort zone if there were no regrets. I think that negative risks tend to kind of associate themselves with regret, negative feelings and negative self image, it. Among the positive mental consequences of sexual risk taking in tourism participants discussed posit ive feedback to self image, feeling attractive, enjoying the excitement, gaining means something about me. It tells something about who I am this person finds me attractive, h ow cool is that! It's related to self image I take risks to get that appreciation, acceptance, love, whatever you want to call it. benefit is high er for tourism .. I am having an amazing tri positive consequences, because I found some one that I really liked and I allowed myself to enjoy that, and to try this new thing called casual sex, a nd it was actually really great and it was a lot of fun.
107 Finally, the women explained that even negative experiences can serve as a platform for lea rning a lesson to make their sexual lives more gratifying and/or safe. Melanie concluded her I did learn a lesson and grew up a little bit I learned I should reject people if I and I should be more independent lesson properly. I might go somewhere and have sex or something sexual, but nsensual unprotected casual sex while backpacking in I learned things about myself I had to rethink what I consider acceptable and what I am comfortable with, and find healthy and safe and how easy it was for something like that to happen withou t I was protected so I feel like I kind of learned my lesson. Lastly, Mariah stated that short made me mor e aware of And realizing that led me to understand what I ecause erson, the other person. Cultural dimension of sexual risk taking in tourism The cultural dimension of sexual risk taking was discussed as unique to tourism and irrelevant in everyday life, since it was mainly discussed in terms of the lack of familiarity The cultural dimension was not universal, but common ( n = 7), as not all the experiences of sexual risk taking in tourism occurred in tourist contexts that involved immersion in a cultural environment different from the US. Participants discussed cultura l differences in terms of openness to sex and the status/rights of women, language barrier and fear of offending locals by displays of affection, lack of familiarity with local sexual culture, implications, stereotypes, and mannerisms, as well as potential ly more serious consequences due to such ignorance. The
108 cultural dimension of sexual risk taking in tourism was mostly relevant to casual sexual relationships, but not exclusively. when you're talki ng about tourism and travel, one of the biggest things are the cultural unknowns, those culture barriers like how does this person define sex? How does this person define sexual relationships ? H ow comfortable is this person with it culture there can be much different from the sexual culture you are used to There could be different kinds of mannerisms that may insinuate something that you are not used to just different culture practic es that you may not be aware of and that might potentiall y lead to a that can be totally different somewh ere else to wear a condom could be emasculating to some men could be offensive to them. So like you have to be more conscious, just because ces on an organized I took a huge risk wanderin g off with that guy to his room even speak the language g to call the police the rules and laws are not the same. Rape to them is not rape to us, rape to us may be not rape to them perceive these thi ngs and how they treat women. Participants emphasized that the cultural dimension of sexual risk taking in tourism depends not only on their degree of cultural awareness, but also on the degree of the local it also really depends on the place you are
109 going, because their culture can be more sexually abrasive than you are used to, or more sexually passive, where you might be looking for a sexual endeavor but they are not. cognitive adaptat in Italy, men on the street catcall at women, and our tour guide told us not to be alarmed, like that s how men are if some guy is hollering at me in the US I be like 'Y ou are an asshole b ut in Italy I like, 'A ll right I am sexy today. So it's disrespectable in the US yet it's acceptable in Italy. a respect for the [local] culture, and when you go to a sexually ri sky because you are going to respect that place you are not going to go against what th at culture says. and your husban d or your boyfriend, travel to a v ery conservative country that will put you at if you have like a public display of affection like holding hands or Indeed, some women di scussed legal ( n = 4) and healthcare/medical ( n = 3) dimensions of sexual risk taking in tourism as interwoven with a lack of familiarity with the local culture, as legal and healthcare systems are the reflections of culture in many respects. With respect to healthcare, you don't know the healthcare in that particular country if you do get something [STDs] where do you go? Is there a medical facility near you? Y ou might not get the same kind of insurance in that sense you have to be much more careful in terms of l ike I guess there's the medical kind of risk. taking in tourism mainly revolved around issues related to obtaining just country. Celine discussed ceasing involvement with a local partner in Kenya after hearing from
110 cultural implications and local perception s of women and a woman's place and when it' s okay to beat a woman. countries, wom e n can get raped and that's how the culture is, and there's no social justice for them nd what if you find yourself in a place like that ? T hen you don't have any justice depending on the area you go, it's much more dangerous and you just don't have the kind of rights and freedom that we have here. T but there is no English speaking police there [abroad] Ther Also, you are not even a citizen of that country. So how are you going to punish someone? Are you really going to stay there for a month just to go on trial because you come from this really great country with rules and laws, that other countries are going to be like that, that they are going to treat you the same. In fact, because you are an American. They probably will be like, 'O h, fuck you You raped. Like there is no level of protection. There is nothing to stop anyone from taking advantage of you. Likewise, Selinda suggested that in the case you are probably not even going to bring that to court because he s in another country and bringing him back to the Therefore, Selinda c the probability of the matter being resolved is lower in foreign countries .. b ecause just legal systems, the obstacles of having to get a lawyer and bringing it to court, the cost it takes ourist would not want to go through that beca Composite Structural Description taking in tourism) underlying the textures the aforementione d dimensions) are extremely diverse at the ideographic level, reflecting the variety of circumstances and the plurality of perceptions, motivations, and meanings held by the women At the nomothetic level, the composition of the
111 structures reveals an inter play between the personal characteristics and the attributes of the tourist experiences, as discussed by the participants and supported by the literature and the defini tion of sex ( n = 13), conservative vs. cavalier/open attitude to sex ( n = 10), sexual double standards based on gender ( n = 12), and the effect of age on sexual risk taking ( n = 11). Among the latter, various attributes of the tourist experiences were univ ersally discussed as facilitating and/or inhibiting sexual risk taking in tourism. These included anonymity ( n = 10) and leaving experiences behind ( n = 10), peer support/pressure ( n = 10), touristic atmosphere and fun mentality ( n = 9), alcohol ( n = 9), s ense of vulnerability in tourism ( n = 13), type of the tourist experience ( n = 11), as well as its length ( n = 10), and destination ( n = 9 ). Sex definitions and attitudes underlying perceptions of sexual risk taking Definitions of sex comprise a universa discussions ( n On the one hand, sex can be defined narrowly as vaginal intercourse since it has been historically constructed as a natural expression of sexuality. In this study, some women indeed adopted this perspective. Others treated the concept of penetration somewhat loosely and included anal intercourse as well, but the emphasis was still placed on the idea of penetration. That is not to say that these women did not engage in other sexual activities, but they did not perceive them as sex per se. In terms of the dimensions of sexual risk put an emphasis on the physical dimension of sexu al health, although not exclusively. Th is narrow definition also appeared to be instrumental in terms of circumventing the mental dimension of sexual risk taking when women did not perceive sex activities such as making out, digital and/or oral sex, etc a s sex. It appears that such a strategy allowed the women in this
112 study to alleviate some of the mental weight associated with non penetrative sex, which in turn affected their perceptions of sexual risk taking. On the other hand, some participants prese nted a broader definition of sex by including oral sex as well and analyzing it in terms of risks. Interestingly, some of the participants holding this broader definition were inconsistent throughout the interview, depending on whether they were talking th eoretically or recalling an actual situation. Thus, oral sex travelled in and out of the definition of sex, which might also be related to the aforementioned strategy of lifting the mental dimension of sexual risk taking. Some women distinguished between s ocietal and their personal definitions of sex Typically, personal definitions were more comprehensive and broad than traditional societal definitions, as perceived by the women. Participants also described sometimes the rift between their perceptions and practices. For some women, their theoretical definition s of sex were more inclusive than their actual practice. The rest of the women adopted a broad definition of sex without any caveats interpreting sex as any kind of intimacy Th is broad definition al so appears to provide a more flexible platform for accommodating various dimensions of sexual risk taking. However, the women who adopted narrow or relatively narrow definitions of sex found a strategy to accommodate various dimensions of sexual risk takin g by distinguishing between sex and sexual activities and linking the perceptions of sexual risk taking to the latter rather that to the former. Some women suggested that while the physical risk dimension of sexual health is bound with penetrative sex, the emotional mental, and social dimension s reach out to the entire spectrum of sexual activities Therefore, for these women a multidimensional definition of sexual risk taking was broader than their personal definitions of sex.
113 ve vs. open/cavalier attitudes to sex ( n = 10) appear to underpin their perceptions and behaviors with respect to sexual risk taking. Conservative attitudes toward sex brought the social and mental dimensions of sexual risk taking to the forefront in some of the reported feeling a bad person and/or feeling socially judged as indecent following sexual risk taking in tourism (more of this can be found in the mental and social dimensions of sexual risk taking sections) Conversely, women holding more open attitudes toward sex appeared to feel more comfortable with sex in general and various sorts of sexual risk taking in particular. This seemed to allow for a cognitive accommodation of sexual encounters with various degrees of commitment (or lack of it) without feeling guilty or uncomfortable. The women with such attitudes sometimes discussed the socially constructed nature of the mental and social weight associated with sex Likewise, cavali er attitudes to sex often translated into a knowledgeable relationship with sex in terms of contraception, thereby diminishing sexual health risks. Social control of sexual double standards underlying perceptions of sexual risk taking Sexual double standa rds, both in general, and in tourism in particular, were universally ( n = 12) discussed by the participants as a structure underlying the social dimension of sexual risk taking for these women. The women argued that men are praised for their sexual esc apades or at least are not judged as harshly as women by society for their sexual exploits. The women also discussed the socially constructed origins of sexual double standards where people are acculturated since early childhood into see ing men as more se xually aggressive and women as more sexually passive Some women discussed the divergent semantics underlying the labels applied by society to sexually active men and women outside of committed relationships. The applied to men are not as insulting as when applied to women.
114 Some women also discussed facing conflicting societal expectations as a result of sexual double standards. Such conflicting ideals bound wi th the Madonna whore dichotomy simultaneously encourag ed sexual restraint and permissiveness, thereby producing an internal conflict for the women. concerns about the social dimension o f sexual risk taking. Participants also discussed power differentials in sex based on gender as underlying sexual double standards. They believed that men have a dominant position in the scenario of penetrative sex. M an as a penetrator was perceived as an active agent/subject, whereas woman as a receptor was perceived as a passive target/object. However, most of the participants saw the genesis of power differential in sex as socially constructed rather than nature based. This asymmetry of power bound with sexual double standards was interpreted by the women as affecting differently, positioning women as more compliant and vulnerable to sexual risk taking, both physically and emotionally. Another example of the socially cons tructed nature of such asymmetry is related to These beliefs often rendered them being unprepared for protected sex or reluctant to insist using condoms. Such findings illustrate that s sex and to be prepared in terms of contraception (i.e., to carry condoms), facing them with ambivalent choice between risks associated with sexual health and social judgment. However, when talking about anonymous tourist contexts, some women felt comfortable carrying their own condoms. In general, women in the current study argued that sexual double standards are subdued in anonymous and/or fun oriented tourist contexts, resulting in a diminished social dimension of sexual risk
115 script. In this study, the findings reveal that in addition to the release from social scrutiny and sexual double standards, some women felt that certain tourist experiences were conducive to some release from internal self scrutiny with respect to sexual behavior, which in turn somewhat attenuated the mental dimension of sexual risk taking in tourism. Although, s elf surveillance was also more pers istent in tourism than social surveillance some women did report losing inhibitions, going against their values, pushing the boundaries, and going easier on themselves for sexual risk taking in tourism. Young age underlying perceptions of sexual risk ta king In addition to gender and sexual double standards, participants discussed their young age ( n = 11) as an important and universal structure underlying sexual risk taking in general and in tourism in particular. Participants stated that they would proba bly not be as prone to sexual experimentation, adventurousness, and risk taking if they were older. Likewise, the combination of age and college culture was discussed as being conducive to creating playgrounds where young adults explore and get to know the mselves, often through taking sexual risks in the context of the hook up sexual script. The women often felt that sexual experimentation is more socially sanctioned and, in some circumstances, even encouraged in young adulthood compared to other life sta ges Talking specifically about tourist experiences during adolescence, the women argued that it is particularly conducive to sexual risk taking. They believed that sexual risk taking in tourism is more exciting and often constitutes the goal of the trip i n young adulthood. The women described adolescence as the time of highest risk taking in general and sexual risk taking in particular due to a sense of invincibility and immortality, which gets amplified in tourism contexts. Participants also argued that v acations are an ultimate setting for sexual risk taking in adolescence because this is the age when the first trip without parental control occur, making this
116 an opportune time to take sexual risks and to venture out. Indeed, the young women in the curren t study explained that sexual risk taking in tourism is underpinned by a sense of adventurousness, experimentation, freedom, and invincibility, contributing to perceptual attenuation of the risks involved. Another crucial factor bound with sexual risk taki ng in tourism is becoming of legal drinking age when the tourists are younger than 21 and this was described as a recipe for trouble Indeed some women explained their past experiences of sexual risk taking in tourism as being associated with young age and a lack of sexual experience and/or agency. B eing young and inexperienced in such novel and anonymous tourist context s they did not feel comfortable reject ing insist ing guid ing negotiat ing and even discuss ing sexual matters. Finally, participants a rgued that mixing going through puberty with attractive ness naivety tourism, and alcohol, creates a Molotov cocktail leading to physical, emotional, social, mental, and cultural sexual risk taking Therefore, any taki ng in tourism should be context sensitive, which brings this discussion to a focus on touristic characteristics associated taking. Facilitating and/or inhibiting factors of sexual risk taking in tourism Various characteristics of tourist experiences were discussed by the women as related to their sexual risk taking. Most of the characteristics could have both facilitating and inhibiting effects depending on the circumstances, individual, and the type of the sexual encounter. Howeve r, participants put substantially more emphasis on the facilitating factors (revealing five universal and seven common facilitating structures as opposed to one universal and two common inhibiting structures), reflecting the conduciveness of many tourist e xperiences to sexual risk taking
117 Destination characteristics as well as type and length of the tourist experience were universally discussed in this study as affecting sexual risk taking patterns. Factors such as anonymity, a fun oriented vacation mentali ty, alcohol, drugs, and nightlife were discussed as facilitators of sexual risk taking in tourism in the context of both casual and steady sexual relationships. Factors such as peer support (or pressure), opportunity to leave experiences behind, limited ti meframe, and the perception of time as being compressed were typically discussed as facilitators of casual sex in tourism. Universal and common inhibiting effects will be discussed later as the flipside of the facilitating effects with respect to relevant touristic characteristics. Anonymity underlying sexual risk taking in tourism Anonymity was a universal structure running across the interviews ( n = 10). It was described as instrumental in diminishing and sometimes even obliterating the social dimension o f sexual risk taking in tourism. Among the common and universal structures discussed in conjunction with anonymity were the absence of social expectations and judgment in tourism ( n = 7), meeting new people whose opinion will not matter after the touri st experience ( n = 7), facing difference and novelty ( n = 11), and having a temporary clean start a fresh playing field and a n = 6). The importance and the liberating effect of a sense of anonymity on sexual behavior translated itself in to fostering sexual exploration diminishing the sense of social surveillance and encouraging sexual experimentation Anonymity was even discussed as a motivation for sexual risk taking in tourism. Moreover, participants highlighted the importance of a nonymity for women as opposed to men, suggest ing that the impact of anonymous tourism environments on sexual behavior is more influential for women compared to men Anonymity allow ed for more sexual freedom and experimentation in the on vacation vis vis everyday life. However,
118 participants also discussed the flipside of anonymity referring to a sense of vulnerability, discomfort, and lack of familiarity in a foreign, new environment ( n = 13). Therefore, anonymity appears to facilitat e consensual sexual risk taking in tourism when it is accompanied by a certain level of comfort and confidence in the new environment. Leaving experiences behind as underlying sexual risk taking in tourism Another universal structure discussed by the wome n in this study was the opportunity to leave behind the experiences of sexual risk taking upon returning home from the tourist experience (n = 10). In a sense, this structure is intricately bound with the aforementioned sense of anonymity. The women discus sed it as diminishing the social, and sometimes even mental and emotional, dimensions of sexual risk taking. With respect to the emotional dimension, participants commonly mentioned that they did not expect to develop relationships on vacation and, thus, n ot growing attached to a sexual partner (n = 6). For example, the women explained that they can run away or distanc e themselves from sexual experiences in tourism that they did not enjoy or did not deem as legitimate. Indeed, the explanations of sexual pe were repeatedly invoked by the participants For some women this arrangement helped to downplay both the social and mental dimensions of sexual ri sk taking. With respect to the social dimension, anonymity was a pivotal matter. With respect to the mental dimension, some women stated that even if self questioning occurred after the sexual experience, the fact that it did not happen at home allowed the m to let go of it relatively easy and to leave it behind. In this sense, tourist experiences appear to have the potential to offer women a safe playground (both socially and mentally) for the exploration of sexual behavior. This arrangement appears to be f urther facilitated by the fact that people tend to leave at home some of the social mental baggage when they travel. This study reveals that some women see the
119 opportunity to leave behind sexual experiences that happened in tourism as an additional benefit or even motivation for sexual risk taking in tourism Atmosphere and fun oriented mentality underlying sexual risk taking in tourism discussions of the tourist atmosp here and vacation mentality (n = 9) as focused on maximizing fun and excitement, seeking a good time, adventure, romance, and euphoria, as well as letting down their guards. This structure was discussed by women as instrumental in terms of downplaying perc eptions associated with all the dimensions of sexual risk taking because the main focus on vacation was on the supremacy of fun and enjoyment and often a rejection of everyday rules and responsibilities This fun oriented mentality translated for many wo men into a mind frame of seeking adventures in general, and sexual adventures in particular. Additionally, participants described taking in tourism. In this respect, taking in tourism were often discussed by the women as comprised of a mix of nightlife, party scene, clubs, dancing alcohol and (less frequently) recreational drugs, euphoria and expectations of having a good ti me on vacation Peer pressure support solidarity vs. judgment underlying sexual risk taking in tourism Participants universally discussed the influence of trip partners on their sexual risk taking patterns. Trip partners were discussed as both a facilitat ing ( n = 10) and inhibiting ( n = 8) factor of sexual risk taking in tourism. Traveling with conservative and judgmental peers, or parents (and other older family members) was described as a situation where the social surveillance was similar to that at hom e and, thus, the social dimension of sexual risk taking was similar to being at home. Conversely, travelling with supportive, like minded, or promiscuous friends was
120 conducive to diminishing social, mental, and even physical dimensions of sexual risk takin g in tourism. However, the flipside of peer support is peer pressure into taking sexual risks. This was discussed by the participants as well. The pressure could be classified into direct and indirect components, where the latter refers to perceiving gen eral promiscuity in tourist environments as a role model. However, the women also discussed direct pressure and the necessity to show off, particularly at a younger age. Yet, of course, peer influence was not always understood as pressure. In many cases, i t was construed as support and a safety network, diminishing not only the social, but also the physical dimension of sexual risk taking ( i.e. a fear of violence). Some women also discussed a sense of solidarity with their female friend / s on vacation and a n understanding that their sexual escapades would be kept discreet upon returning home. Likewise, this sense of solidarity was described as unlikely in a similar scenario at home. For some women, this sense of support and solidarity was also instrumental i n terms of diminishing the mental dimension of sexual risk taking in tourism. They felt that if other women in that context were engaging in sexual risk taking, then they should not feel bad about their own involvement in similar activities. Thus, the char acteristics and behaviors as well as on the ir perceptions of various dimensions of sexual risk taking in tourism. Alcohol and drugs consumption underlying sexual risk taking in tourism Similarly to the fun oriented vacation mentality, alcohol was universally ( n = 9) discussed as a factor facilitating sexual risk taking in tourism and downplaying perceptions associated with all the dimensions of sexual risk taking in tourism. Alcohol was described by the women as helping them to get rid of their inhibitions, impairing their judgment, empowering them with liquid courage, and making them horny. Drugs were also discussed, but as a common rather than a universal structure ( n = 6). The relatio nship between drugs and sexual risk taking
121 was similar to that between alcohol and sexual risk taking, but it was discussed by the women in this study less frequently. When it comes to tourist experiences, the women argued that alcohol consumption typical ly ris es dramatically, often leading to sexual risk taking. Additionally, alcohol in tourism was commonly discussed in conjunction with partying and nightlife ( n = 6). Some tourist experiences were actually described as a long party with a permanent hangov er and drunken (often unprotected) sex. The women explained that partying, drinking, and having sex were perceived as getting the best of what vacation has to offer. Another important consideration with respect to alcohol and sexual risk taking in tourism is the notion of underage drinking. The young women who could legally drink abroad, but not at home, found this experience to be empowering, allowing them to feel independent adults, and alleviating the stigma of underage drinking. W hile such a feeling ma y have a positive capacity, w hat makes the situation of underage drinking abroad particularly dangerous is the lack of experience with consuming alcohol and potential for losing control over the situation in unfamiliar surroundings and, potentially, lack o f a safety network, where someone might take advantage of a young drunk woman. This combination of factors may render these women particularly vulnerable to non consensual sex. Therefore, while the relationship between sex and alcohol in tourism has been described as curvilinear, facilitating sexual behavior up to a certain amount of alcohol consumed, but inhibiting or even preventing it when too much alcohol is consumed the relationship between alcohol and sexual risk taking appears to be linear, as the loss of control due to alcohol consumption increases the exposure for women to all the dimensions of sexual risk taking.
122 Length, destination, and type of the tourist experiences underlying sexual risk taking Participants universally described specific a ttributes of tourist experiences such as length (n = 10), destinations (n = 9), and types of tourist experience (n = 11) as affecting sexual risk taking. The women explained that having a short timeframe is the most conducive to sexual risk taking due to t he aspiration of mak ing the best of the tourist experience and preserv ing the anonymity effect. Yet, there was no agreement as to what is short enough and the estimates ranged from one night to one week, but all were subjectively perceived as short. Additi onally, participants described changes in sexual scripts on vacation as being conducive to the prompt progression of new relationships to sex and perception of time as being compressed into the current moment his arrangement might magnify the physical dimension of sexual risk taking, whereas with emotional and mental dimensions, the effect can be ambivalent. On the one hand, the sense of time compression and the understanding of temporality can facilitate the effect of detach ment from the experience. On the other hand, this arrangement might encourage prompt involvement in the sexual experiences that might eventually cause a sense of discomfort for overstepping personal boundaries. Another structure universally discussed by th e participants was the tourist destination. Among the destination characteristics inhibiting sexual risk taking in tourism, the women brought up the reputation of a place with high STDs and crime rates, as well as conservative sexual norms. The latter were also discussed with respect to the cultural dimension of sexual risk taking in tourism in the composite textural statement. Among the destination characteristics facilitating sexual risk taking in tourism, participants discussed anonymity due to the dista nce from the place of residence, as well as the urban and touristic nature of the destination. Participants also discussed specific tourist destinations as the icons for sexual risk taking, such as
123 Cancun, Amsterdam, and Las Vegas. These destinations were characterized as a playground for adults tailored to taking risks, having fun and no regrets Finally, participants described various types of tourist experiences as conducive (or not so much) to sexual risk taking. The se findings suggest that various typ es of tourist experiences both with steady and casual sexual partner/s Among the discussed scenarios in the current study, were R&R beach vacations, spring breaks, backpacking trips, sightseeing city break s, study abroad trips, cruises, and business/academic trips. The first two scenarios were described as the most conducive to sexual risk taking in tourism. With respect to the R&R beach vacation, participants described beach as a relaxing, romantic, and ar ousing place charged with sexual vibes and fantasies as well as promoting involvement in sexual risks. With respect to spring break, participants argued that sex in general, and casual sex, in particular, are not perceived as taboo in this setting for bot h men and women. Essence Statement taking in tourism involves women engaging in sexual activities in tourism (with a steady and/or casual sexual partner/s) subjectively perceived as putting them at risk, where the ris k is understood very broadly (including physical, psychological, and socio sexual risk taking in tourism reveal that it is a complex and multifaceted phenomenon intricately intertwined wit h the universal structures of sex perceptions, sexual double standards bound with gender and age, as well as the socio cultural fun oriented ethos of the tourist experiences. Among the universal and common dimensions of sexual risk taking in tourism, the i nterviewees described physical/sexual health (e.g., STDs, HIV/AIDS, rape, violence, abduction, and unwanted pregnancy), social (e.g., damaged reputation, being judged, and feeling embarrassed), emotional (e.g., emotional distress, trauma, or attachment), m ental/self perception (e.g., feeling
124 disrespected, uncomfortable, used, rejected, and going against personal values), and cultural (e.g., fear of offending cultural norms, language barrier, and legal issues with obtaining justice if raped) dimensions. Se xual risk taking was perceived by many participants as more prevalent in tourism due to perceived anonymity, detachment from everyday responsibilities, and shrinking social (and in some cases, emotional and/or mental) dimensions of sexual risk taking in to urism. Diminishing importance of the social dimension can be attributed to the subdued sexual double standards in anonymous tourist environments, perceived absence of social judgment, fun oriented vacation mentality, as well as the sense of fleeting tempor ality and perceived legitimacy of sexual experimentation in some tourist environments. However, the social surveillance and sexual double standards frequently assert themselves through internalized self surveillance rendering certain stability to the menta l and emotional dimensions regardless of the sexual risk taking context. Finally, with exception of the cultural dimension related to the lack of familiarity with the local culture as well the legal and medical systems as a tourist, the dimensions of sexua l risk taking in tourism appear to be similar to those in everyday life, but the weights attached to these dimensions seem to be context specific for many participants in this study. Findings from the Quantitative Phase Research Q uestion 1a: W hat S exual A ctivities A re P erceived as R isky in T ourism by the F emale C ollege S tudents? All the risk perceptions were measured on a 5 perce ptions and experiences with sexual risk taking in tourism. Among the sexual activities on vacation perceived as the most risky, the women referred to h aving vaginal or anal sex with an unsteady partner without a condom ( M = 4.79, SD = .64 and M = 4.74, SD = .69, respectively),
125 h aving sex with a sex worker ( M = 4.74, SD = .71 ) h aving sex with multiple partners ( M = 4.63, SD = .71), p erforming or receiving oral sex on /from an unsteady partner without a condom / a latex barrier ( M = 4.52, SD = .82 and M = 4.45, SD = .86, respectively), h aving sex under the influence of drugs ( M = 4.42, SD = .88) and h aving vaginal sex with a steady partner without protection from pregnancy ( M = 4.40, SD = .95). Out of 23 total sex related activities, 16 activities had a mean of 3.50 and higher (Table 4 1). Perceptions of risk were compared across the groups of women who had and had not experienced a particular sexual activity. The actual experiences were obtained through a set of question 15, Appendix E). Independent Samples T Tests were used to test the differences between women with and without experience. With the exception of the 2 activities perceived to be among the riskiest (vaginal sex without a condom and sex with multiple partners), all the rest of the sexual activities (i.e., 16 activities) were perceived as significantly less risky by participants who had actually experienced them (Table 4 1). For 5 sexual activities, the number of women who reported experience with th em was not large enough to conduct meaningful statistical analysis. Thus, it appears that prior experience with sexual activity is related to lower risk perceptions with respect to this activity. Research Q uestion 1b: W hat A re the P erceived D imensions of S exual R isk T aking in T ourism by the F emale C ollege S tudents? The analysis in this section was a multi step process starting with exploratory approach and then demonstrating how confirmatory independent clusters approach would not ameliorate the discriminan t validity, since eliminating cross loadings inflated factor correlations and decreased the goodness of fit of the model. Exploratory factor analysis was used to identify the latent dimensions of sexual risk taking in tourism. Principal components factor a nalysis was conducted with oblique Direct Oblimin rotation. Drawing upon the qualitative findings, it was
126 expected that allowing factors to correlate would be more meaningful than constraining them to be orthogonal. Out of the initial 24 items, 4 items wer e eliminated from the analysis due to low communalities and conflicts with the principles of the simple structure solution. Table 4 2 presents the EFA solution for the sexual risk perceptions model with 3 factors extracted from 20 remaining items and acco unting for 70.29% of the total variance. Based on the Kaiser Meyer p < .001), the relationships in the data lend themselves well to factor analysis. The three factor solution can be interpreted as a combination of the Physical/Sexual Health, Mental/Emotional, and Socio Cultural dimensions of sexual risk taking b, .30 and .32 are regarded as the minimum for acceptable factor loadings (Costello & Osborne, 2005; Tabachnik & Fiddell 2001). Factor one, P hysical/ S exual H ealth ( initial Eigenvalue = 3.7 4; 18.68% of variance ; .92 ) was comprised of 6 pe rceived risk items : the risk of getting sexually M = 4.88, SD = .4 2 M = 4.89 SD = .40 M = 4.83 SD = 0.50 being raped M = 4.88 SD = .42 M = 4.85 SD = 0.49 M = 4.78 SD = .54 ). Factor two, Mental/Emotional ( initial Eigenvalue = 9.25; 46.25% of variance; Cronba the risk of being emotionally hurt ( M = 4.09 SD = .96 ); the risk of feeling uncomfortable after the act ( M = 4.04 SD = .96 ); the risk of being too far out of my comfort zone ( M = 4.06 SD = .95 ); the risk of feeling used ( M = 4.11 SD = .97 ); the risk of having unfulfilled expectations ( M = 3.65
127 SD = 1.10 ); the risk of feeling rejected ( M = 3.76 SD = 1.08 ); the risk of regretting it later ( M = 4.14 SD = .96 ); the risk of doing something against my values ( M = 4.21 SD = .96 ); the risk of embarrassing myself M = 3.96, SD = 1.01); the risk ( M = 3.97 SD = .98 ); and the risk of feeling not respected ( M = 4.10 SD = .97 ) Factor three, Socio Cultural ( initial Eigenvalue = 1.07; 5.36% of variance; the risk of embarrassing myself = .44, M = 3.96, SD = 1.01); ( M = 3.97 SD = .98 ) ; the risk of feeling not respected ( M = 4.10 SD = .97 ) ; the risk of affecting my reputation at home ( M = 4.04 SD = 1.0 3 ) ; the risk of being judged by travelling companions ( M = 3.79 SD = 1.06 ) ; and the risk of offending cultural norms at the destination ( M = 3.85 SD = 1.05 ) Factor one wa s moderately correlated with factor two ( r = .27 ) and factor three ( r = .22). Factor two ha d a relatively high correlation with factor three ( r = .64). The composition of the factors reveals three cross loading items on the Mental/Emotional and Socio Cultural factors (i.e., factors two and three). The content of the the risk of embarrassing myself the the risk of feeling not respected rationale for these items to be related both to social and mental dimensions of sexual risk taking in tourism Moreover, the decision in favor of the three factor model (as opposed to two factor model) was based on the qualitative findings, composition of the factors in terms of unique loadings, relatively simple structure, amount of variance explained, and the p ossibility of retaining the maximum number of items in the model.
128 The goodness of fit indices for the EFA model with three factors show an adequate fit (CFI = .924; TLI = .891; SRMR = .030; RMSEA = .098). The goodness of fit test does not provide evidence in favor of the model ( 2 = 1857.39, df = 151, p = .000), but this test is extremely sample sensitive (it does not perform well with large sample sizes, which is the case here) and it tests whether the model fits the data perfectly (which is never the cas e). Since the aforementioned goodness of fit indices are sample size independent, they are often given a The notion of the cross loadings among the factors in vites comparison between the aforementioned EFA model and the independent clusters model postulated with confirmatory factor analysis (ICM CFA). In ICM CFA, all the cross loading are eliminated and each item is allowed to load on one and only one factor. A is no a priori theory to guide the loadings instruction in ICM CFA, the presented EFA model was used as guidance for the composition of the factors. Among the cross embarrassin on the Mental/Emotional factor only. This comparison is presented in Table 4 3 in the total group models section, where the EFA model is labeled TG ESEM (where all the items are allowed to load on all the factors) and ICM CFA is labeled TG ICM CFA. Marsh et al. (2009, 2010, 2011, 2012) suggested that the ESEM approach has the poten tial to produce a better fit of a model with respect to the multi dimensional constructs (compared to ICM CFA), while resulting in less correlated (i.e., more differentiated) factors, which is important for discriminant validity. In this study, the ICM CFA produced a worse fit (RMSEA = .104; CFI = .892; TLI = .877; SRMR = .066) than the ESEM. This lack of fit would
129 cause a rejection of the model if ICM CFA was adopted, which frequently happens to models of multi dimensional constructs (Marsh et al., 2009, 2010, 2011, 2012). Yet, the corresponding ESEM model allowing for cross loadings produced an adequate fit. Although, the factor loadings in the ICM CFA were somewhat higher on average than in the ESEM model, the differences were typically minor and the fac tor loading patterns were similar. An evaluation of the factor correlations in ESEM and ICM CFA presents a critical advantage of the ESEM approach over the ICM CFA approach. While the factor correlations in the ESEM model range from .22 to .64, in the ICM EFA model the correlations range from .21 to .85. A correlation of .85 between Mental/Emotional and Socio Cultural factors in ICM CFA model is an issue for discriminant validity. In the ICM CFA model, factor correlations are positively biased by not inclu ding the cross loadings, which is not the case in ESEM. In line with similar analyses, in this study the ESEM solution appears to be superior to ICM CFA solution, both in terms of the goodness of fit and the distinctiveness of the factors (Asparouhov & Mut hn, 2009; Marsh et al., 2009, 2010, 2011, 2012). Research Q uestion 1c: D o the P erceived D imensions of S exual R isk T aking in T ourism D iffer by S exual S ensation S eeking P ropensity among the F emale C ollege S tudents? A common approach taken to answer this typ e of question is the use of MANOVA with regressed factor scores or mean scores (based on the EFA solution presented in Table 4 2) as dependent variables and the SSS levels as an independent variable. While this strategy could provide a valuable indication, it is problematic and might be methodologically weak, as drawing interpretations based on manifest mean differences (e.g., factor scores or mean scores), rather than latent variable models, does not allow for the evaluation of measurement invariance and c orrection for potentially complex structures of measurement error (Marsh et al., 2012). Alternatively, the aforementioned ESEM approach extended to multiple groups with the
130 taxonomy of 13 partially nested models ( Marsh et al. 2009, 2010, 2011, 201 2 ) offer s a solution for these issues. Table 4 3 presents the results of the application of this taxonomy in the current study to test the invariance across lower and higher levels of SSS. The SSS scale was used as the grouping variable for ESEM (question 2, Appe ndix E). Descriptive statistics for the SSS items are presented in Table 4 4. The total mean score for the SSS scale was calculated ( M = 2.00, SD ( n = 439, wit h scores below the M n = 414, with scores above the M ). In order to establish the invariance or non invariance of certain parameters across the groups, each time the key relevant models from the taxonomy are being compared based on the goodness of fit test ( 2 ), goodness of indices (CFI, TLI, RMSEA, SRMR), and information criterion indices (AIC, BIC, corBIC). Due to the relatively few applications of ESEM, the relevance of these criteria and the exact cutoff values remain an open researc h area (Marsh et al., 2009, 2011, 2012). Therefore, it is suggested that a comparison of the relative fit of the partially nested models is conducted rather than the absolute level of fit of any one model, as long as the goodness of fit of the best fitting model is acceptable (Marsh et al., 2011). The models are considered to be nested when the set of the estimated parameters in a more restricted model is a subset of the estimated parameters in the model that is less restricted (Marsh et al., 2010). The ad hoc guidelines to evaluate whether the differences in fit are substantial enough to reject a more parsimonious model (i.e., the more constrained one) in favor of a more complex model (i.e., the less constrained one) are as follows: support for the more pa rsimonious model requires a change in CFI (which is monotonic with complexity) of less than .01 or a change in RMSEA of less than .015. However, the indices like TLI and RMSEA penalize for parsimony
131 and, thus, a more conservative guideline is to support a more parsimonious model if TLI or RMSEA is as good or better than the respective index for the more complex model (Marsh et al., 2009, 2010, 2011, 2012). As for the information criterion indices, lower values on these indices reflect a better fit of a mode l to the data compared to the model with higher values. Yet, these indices are sample size dependent by their design and should be interpreted cautiously (Marsh et al., 2009). With respect to the goodness of fit test ( 2 ), being a sample sensitive and a very powerful test, it can detect small and unimportant differences between the groups leading to the rejection of the model. Indeed, in this study 2 test rejected all the models and all the transformations. While the goodness of fit indices are not subject to these issues, they are not flawless and sometimes can miss substantial differences. Therefore, the relevant parameter estimated differences should be evaluated as well in the less constrained model and the invaria nce should be supported if these differences are in some meaningful sense small. While being advised, these rules are only rough guidelines in the novel ESEM context, as the number of the estimated parameters is typically substantially larger in ESEM compa red to confirmatory models (Marsh et al., 2009, 2010, 2011, 2012). The two constraints (MGI1 in Table 4 3) is a model of configural invariance. For the preliminary evaluation of the similarity of parameter estimates over the groups, Marsh et al. (2011) suggested reviewing the pattern and level of factor loadings based on the model MGI1 (Table 4 5). Such a review provides a mixed impression. Typically, the loading has to be at least .3 5 (and preferably, higher than .4) to be considered. For the sake of demonstrating fluctuations in the meaningful loadings across two groups, significant loadings higher than .3 were presented, but the loadings
132 higher than .35 were bolded (reflecting what would actually be considered as a high enough contribution to the factor). With two exceptions, the pattern of the loadings was similar across two groups, rendering certain support for the configural invariance model. However, most of the loadings were sli ghtly higher for the higher SSS group. Indeed, under the requirement of the loading to be at least .35, four items in the Physical/Sexual Health factor did not make the cutoff to be included in the factor. Moreover, on average, the uniquenesses were margin ally higher for the higher SSS group. The factor correlations pattern was also similar across the groups, but these correlations were somewhat higher for the lower SSS group. In addition to this preliminary review, the relative goodness of fit should be e valuated. The MGI1 performed quite well according to CFI = .918 and SRMR = .033. Yet, to establish configural invariance, the MGI1 should be compared to TG ESEM (Table 4 3). The decrement (i.e., the decline) in CFI was less than .01 (.924 vs. .918) and the change in RMSEA was less than .015 (.098 vs. .101). Indeed, the fit statistics in TG ESEM and in MGI1 were approximately the same, but the MGI1 had twice the degrees of freedom (266 vs. 133) and twice the numbers of free parameters (194 vs. 97). Finally, all the information criterion indices were lower for MGI1 vis vis TG ESEM. These results support the configural invariance and indicate that the same ESEM model is able to fit the data for both groups when no additional invariance constraints are postula ted (Marsh et al., 2009). This also suggests that it makes sense to proceed evaluating results from the key models in the taxonomy of the 13 models of invariance (Marsh et al., 2011). Weak factorial/measurement invariance tests the invariance of the facto r loadings across the groups (Marsh et al., 2009, 2010, 2011, 2012). Model MGI2 (Table 4 3) restricted factor loadings to invariance across the lower and higher levels of SSS. The key comparison here was between the MGI2 and MGI1 models. Since the number o f freely estimated factor loadings in
133 ESEM is so high, MGI2 was substantially more parsimonious than MGI1. Indeed, the number of freely estimated parameters decreased from 194 to 143. Yet, MGI2 offered an adequate fit to the data. The drop in CFI was less than .01 (.918 vs. .913), while the indices controlling for parsimony were even better for the more parsimonious MGI2 than for MGI1 (TLI = .895 vs. .883 and RMSEA = .096 vs. .101). Although, the 2 for MGI1 was smaller than that of MGI2, the 2 / df ratio wa s smaller for MGI2 than for MGI1 (4.922 vs. 5.370). The information criterion indices, however, offered a somewhat mixed pattern. While the AIC was slightly higher for MGI2, the BIC and corBIC were lower/better for MGI2. To conclude, the juxtaposition of M GI1 and MGI2 provided support for the invariance of the factor loadings, which is also called weak measurement invariance. Strong factorial/measurement invariance requires the invariance of intercepts in addition to the invariance of the factor loadings ( Marsh et al., 2009, 2010, 2011, 2012), which was the case for MGI5 (Table 4 3). The key comparison for establishing strong factorial/measurement invariance was between MGI2 and MGI5. This strategy tests whether the 20 intercepts can be explained in terms o f 3 latent means. A lack of support for strong factorial/measurement invariance would imply differential item functioning, suggesting that the differences between or means (Marsh et al., 2009). A change of 17 in the df reflects 20 new constraints on the item intercepts minus the three latent factor means that are freely estimated in MGI5. The decline in CFI in MGI5 vs. MGI2 was smaller than .01 (.910 vs. .913). Base d on the fit indices controlling for parsimony, the fit of MGI5 was slightly better than that of MGI2 (TLI = .898 vs. .895, RMSEA = .095 vs. .096). Additionally, while the AIC was slightly higher for the MGI5, the BIC and corBIC were lower/better for MGI5. Therefore, there was a support for strong measurement
134 invariance, indicating that there was no differential item functioning across lower and higher SSS levels, as well as suggesting that the latent means can be compared across the groups. Strict factoria l/measurement invariance requires the invariance of the item uniquenesses in addition to the invariance of the factor loadings and intercepts (Marsh et al., 2009, 2010, 2011, 2012), which was the case in MGI7 (Table 4 3). The key comparison for establishi ng this invariance was between MGI5 and MGI7. A change of 20 in the df reflects 20 new constraints on the item uniquenesses. The decrement in CFI in MGI7 vs. MGI5 was bigger than .01 (.880 vs. .910). Moreover, the TLI and RMSEA were substantially worse in MGI7 vs. MGI5 (TLI = .872 vs. .910 and RMSEA = .106 vs. .095). These results could not support even an adequate fit of MGI7. Moreover, all the information criterion indices were higher/worse for MGI7 vs. MGI5. In summary, there was no support for strict me asurement invariance, meaning that the manifest means should not be compared across the groups in this case. This provides support for the contention that MANOVA with factor scores (as dependent variables) and SSS levels (as independent variable) should no t be used to draw conclusions. Comparisons of all the pairs of models that can be used to test the invariance of uniquenesses (i.e., MGI3 vs. MGI2, MGI6 vs. MGI4, MGI7 vs. MGI5, MGI9 vs. MGI8, MGI11 vs. MGI10, and MGI13 vs. MGI13) led to the same conclusio n. The lack of support for strict invariance suggests that measurement error differs across the groups (Marsh et al., 2009). Indeed, the tendency of the uniquenesses to be slightly but systematically larger for the higher SSS group was identified at the st age where preliminary support for configural invariance was established (Table 4 5). While the latent factor variance covariance invariance is usually not a focus of measurement invariance, it is important for establishing discriminant validity of multidim ensional constructs (Marsh et al., 2009, 2012). For establishing this invariance, the
135 focus of comparison was between the models MGI2 and MGI4, where the former postulated the invariance of factor loadings and the latter required the invariance of factor l oadings as well as of factor variances covariances. A change of 6 in the df reflects 3 factor variances and 3 factor covariances. The decline in fit in MGI4 vs. MGI2 according to CFI was minor and less than .1 (.910 vs. .913). Furthermore, the indices that control for parsimony were similar for MGI4 and MGI2, reflecting a minor decrease in fit (i.e., TLI = .894 vs. .895 and RMSEA = .097 vs. .096). These results provide support for the latent factor variance covariance invariance. However, the higher/worse i nformation criterion indices for MGI4 contradict this support. Therefore further comparisons are warranted. While MGI4 vs. MGI2 is the most basic comparison for testing the invariance of the latent factor variance covariance matrix, it can also be establi shed based on comparisons between MGI3 and MGI6, MGI5 and MGI8, MGI7 and MGI9, as well as MGI10 and MGI9. The first three pairs of comparisons provided support for invariance based on the goodness of fit indices, contradicted by the information criterion i ndices. The last comparison (i.e., between MGI10 and MGI9) provided consistent support for invariance based on all the indices. Considering this pattern and the fact that most of the illustrations of this approach in the literature relied solely on relativ e fit in the goodness of fit indices (not even presenting the information criterion indices as they are considered to be very sample sensitive and performing less well with large samples), it appears that the data offers support for the invariance of the l atent factor variance covariance matrix across the groups. While these findings are encouraging for discriminant validity, conclusions should be drawn cautiously because of the high correlations between the Mental/Emotional and Socio Cultural factors for b oth the low and high SSS groups, thereby pointing to the weak discriminant validity.
136 Lastly, the notion of invariance of the factor means across the groups should be addressed. In the last four models in Table 4 3 (i.e., MGI10 MGI13), the factor means acr oss two groups were constrained to be zero in combination with other invariance restrictions. Several combinations of the models could be used to test factor means invariance across the groups (i.e., MGI5 vs. MGI10, MGI7 vs. MGI11, MGI8 vs. MGI12, and MGI9 vs. MGI13). For example, comparing MGI10 to MGI5 revealed a very minor decrement in the goodness of fit according to CFI of less than .01 (.907 vs. .910) and RMSEA of less than .015 (.096 vs. .095), supporting the invariance of factor means. The TLI also did not change dramatically (.895 vs. .898). Yet, the information criterion indices were higher/worse for MGI10. The comparison results based on all the aforementioned pairs of models revealed the exact same pattern of the goodness of fit indices supportin g the invariance and information criterion indices not supporting it. Furthermore, since (a) the goodness of fit test ( 2 ) indicated a significant change in fit in the comparison between MGI5 and MGI10 ( 2 (3, N = 851) = 48.929, p < .001) and (b) the good ness of fit indices indicated similarity in fit, (a) should be ignored only if the latent factor means are in some meaningful sense small. Indeed, looking at the best fitting model among those used for testing the invariance of the factor means MGI5 (Tab le 4 3), reveals that the latent means for the higher SSS group were systematically smaller than for the lower SSS group on the Physical/Sexual Health ( .54, p = .000) and Mental/Emotional dimensions ( .38, p = .000), but not on Socio Cultural dimension. S ince the SD for the lower SSS group is equal to one and considering the aforementioned support for the invariance of the latent factor variance covariance matrix, these differences can be interpreted as effect sizes reflecting that the mean differences are of substantial size.
137 Another way to look at it is to estimate the probability that a randomly chosen woman from the lower SSS group has a score on a given factor that exceeds the same factor score for a randomly chosen woman from the higher SSS group. Th is can be estimated based on dividing the difference in the factor means across the groups by the square root of the sum of variances ( M for lower SSS M for higher SSS Var for lower SSS + Var for higher SSS ) where the result is a z score. For exam ple, the calculation with respect to the Physical/Sexual Health factor is as follows: (0.00 ( z score is .64, meaning that the probability that a randomly chosen person from the lower SSS group has a higher Physical/Sexual Health factor score than a randomly chosen person from the higher SSS group is 64%. Likewise, with respect to the Mental/Emotional factor: (0.00 ( The area below this z score is .60, meaning that the probabili ty that a randomly chosen person from the lower SSS group has a higher Mental/Emotional factor score than a randomly chosen person from the higher SSS group is 60%. Finally, when the latent means are equal the probability should be 50%, which is almost the case with respect to the Socio Cultural factor: (0.00 ( z score. As for the substantive meaning of this investigation, the findings provide some evidence sexual risk taking perceptions in tourism is relatively invariant across the lower and higher levels of SSS. In other words, there was no evidence that the nature of the construct varies across the groups. This interpretation draws primarily on the establ ished weak factorial/measurement invariance. However, support for the invariance of the latent factor variance covariance matrix across the groups also contributes to this conclusion. Additionally, support for the invariance across the groups in terms of f actor loadings and item intercepts (i.e., strong factorial/measurement invariance) offered justification for the
138 interpretation of the differences across the lower and higher levels of SSS based on the latent factor means. The lack of support for the stric t factorial/measurement invariance in this investigation suggests that the measurement error is different across the groups, which precludes the comparison of the manifest means (e.g., factor scores). Finally, the results suggest that there are significant differences across the groups in terms of the Physical/Sexual Health and Mental/Emotional factor means. Yet, the latent means on the Socio Cultural factor did not differ by SSS propensity among the women. Furthermore, conclusions regarding the differences in risk perceptions across the SSS propensity groups should be drawn somewhat cautiously since the differences in factor means were not big enough to decrease the goodness of fit indices substantially when constrained to invariance across the groups. Rese arch Q uestion 2: W hat A re the D imensions of M otivations for S exual R isk T aking in T ourism A mong the F emale C ollege S tudents? To identify the latent dimensions of the motivations for sexual risk taking in tourism, exploratory principal components factor ana lysis was conducted with oblique Direct Oblimin rotation. Of the initial 25 items, six items were eliminated from the analysis due to low communalities and conflicts with the principles of the simple structure solution. Table 4 6 presents the EFA solution for the motivations model with three factors extracted from 19 remaining items and accounting for 61.89% of the total variance. Based on the Kaiser Meyer 2 = 8, 786.82, p < .001), the relationships in the data provide a good basis for factor analysis. In the three factor solution, the motivational factors can be interpreted as follows: factor one Anonymity/Detachment/Experimentation, factor two Safe Exploratio n/ Thrill/Empowerment, and factor three Fun/Less Inhibitions/Opportunity. Factor loadings ( presented in Table 4 6.
139 Factor one, Anonymity/Detachment/Experimentation ( initial Eigenvalue = 8.83 ; 46.48% of variance ; .91 ) was comprised of 10 items referring to the motivations for T = .81, M = 3.46, SD = 1.0 8 ); = .80, M = 3.28, SD = 1.15); I can enact sexual fantasies not = .79, M = 3.15, SD = 1.14); = .7 5 M = 3.8 7, SD = 1.1 7 ); = .74, M = 3.44, SD = 1. 10 ); = .57, M = 2.68, SD = 1.07); Casual sex = .4 4 M = 2.10, SD = 1.08); T aking sexual risks on = .39, M = 2.68, SD = 1.11); = .57, M = 3.84, SD = .99); and = .40, M = 3.33, SD = 1.11). Factor two, Safe Explo ration/Thrill/Empowerment ( initial Eigenvalue = 1.9 2; 10.09% of variance ; 90) included 10 = .4 9 M = 2.10, SD = 1.08); = 50 M = 2.68, SD = 1.11); = .71, M = 2.27, SD = 1.0 2 ); = 50 M = 2.83, SD = 1.18); = .6 6, M = 2.28, SD = 1.16); Sex is a way to get to know the locals = .8 4 M = 1.76, SD = 0.89); = 80 M = 1.76, SD = .95); What happens on vacation stays on vacation = .45, M = 2.79, SD = 1.27); Vacatio n is the time for sexual exploration = .6 2 M = 2.62, SD = 1.0 2 ); and Availability of my type of sexual partner on vacation = .41, M = 3.07, SD = 1.15). Factor three, Fun/Less Inhibitions/Opportunity ( initial Eigenvalue = 1.0 1; 5.31% of variance ; .82 ) Vacation is the time for sexual exploration
140 = 41 M = 2.62 SD = 1.02 ); Availability of my type of sexual partner on vacation = 53 M = 3.07 SD = 1.15 ); Sex feels good = 75 M = 3.98 SD = 1.00 ); Vacation is the time to have fun = 48 M = 3.84 SD = .99 ); and I have less inhibitions on vacation = 42 M = 3.33 SD = 1.11 ). Factor one wa s moderately correlated with factor two ( r = .49 ) and factor three ( r = 40 ). Factor two ha d a low correlation with factor three ( r = .1 9 ). The composition of the factors reveal ed several cross loading items, which is an issue for the simple structure criteria Yet, an examination of the content of the cross loading items in conjunction with the i nterpretations of the factors suggests that conceptually these items could be related to both factors. Clearly, this is a challenge for discriminant validity. However, a decision in favor of the three factor model (as opposed to two factor model) was guide d by the qualitative findings and the amount of explained variance (which was dropping dramatically with only two factors and/or further amputation of the items). Moreover, the goodness of fit indices for the EFA model with three factors reflected a good f it ( CFI = .949 ; TLI = .926 ; SRMR = .028 ; RMSEA = .068 ) Research Q uestion 3a: W hat A re the C lusters of S exual R isk T akers in T ourism A mong the F emale C ollege S tudents B ased on the P erceptions of and M otivations for S exual R isk T aking in T ourism? Regressed factor scores were calculated when running the EFA for the perceived dimensions of and motivations for sexual risk taking in tourism (presented in the previous sections). These factor scores served as the cluster variate for the cluster analysis in this s tudy. Factor cluster segmentation based on the factor scores is recommended when exploring the data structure (Mooi & Sarstedt, 2011). Dolnicar and Grn (2008), in an examination of factor cluster segmentation (i.e., conducting factor analysis and plugging in resulting factor scores into cluster has a history as long as the history of segmenting tourism markets itself (p. 64) dominates data driven segmentation in tourism
141 research (p. 65) Indeed, numerous empirical examples and conceptual methodological discussions of factor cluster segmentation (based on the factor scores) can be found throughout the leisure and tourism literature (e.g., Frochot & Morrison, 2001; Havitz, Dimanche, & Bogle, 1994; Kyle, Kerstetter, & Guadagnolo, 2002; Park, Yang, Lee, Jang, & Stokowski, 2002; Shoemaker, 1994). Furthermore, using regressed factor scores was a superior option for this study, since being one of the refined methods for calculating factor scores i t considers shared variance between the factor and the item and maximizes validity by producing highly correlated factor scores with a given factor ( DiStefa ). Conversely, n on refined methods for computing scores to reflect the factors (e.g., sum scores, mean scores) ignore factor loadings disregarding the strength of the item, typically giving equal weight to all the items on a factor and raise additional issues when accommodating cross loading items (DiStefano et al., 2009). Regression factor scores predict the location of the participant on the latent factor. R egressed factor scores are standardized (i.e., having M = 0.00 and SD = 1 .00). A score of zero on a factor implies that this participant s rating of the items included in the factor is close to the means of those items in the sample For factors with positive item loadings, a positive factor score indicates responses higher tha n the mean s for th e item s included in the factor For negative loadings the relationship is inverted and positive factor scores correspond with responses lower than the mean s for the items included in the factor Following the procedures described in the methods section (i.e., starting with Hierarchical cluster analysis, calculating centroids and the number of clusters, and then proceeding to K Means and TwoStep cluster analyses), five clusters of sexual risk takers in tourism among the women were identif ied. Table 4 7 presents the mean values/final cluster
142 centers based on the K Means cluster analysis These final cluster centers reflect the characteristics of the typical case for each cluster. The higher is the number in the table, the higher is the aver age score of this cluster on this factor relative to the mean in the sample (i.e., zero) In other words, s coring zero indicates scoring like everybody else on average. Conversely, s cor es substantially higher or lower than zero reflect the unique features of the cluster. Additionally, w hat i s considered to be substantial could be suggested based on the range Table 4 7 also presents the results from the analysis of variance performed on each dimension as part of internal validation reflecting that each cl uster variate included in this analysis distinguished well between the clusters (i.e., each F value wa s high and significant at p < .001). motivated by anonymity and exp dimensions of sexual risk taking as relatively important (mental/emotional : M = .55 SD = .61; physical/sexual health: M = .30, SD = .18; and socio cultural: M = .59 SD = .59 ) and they were motivate d to get involved in sexual risk taking in tourism by anonymity/detachment/ experimentation ( M = .66 SD = .60 ) as well as by exploration/thrill/empowerment ( M = .77 SD = .77 ) Cluster two ( n = oderate and dive rse risk In this cluster, the women perceived physical/sexual health dimension as important ( M = .29 SD = .24 ) as well as the rest of the dimensions as moderately important (mental/emotional: M = .29 SD = .64; and socio cult ural: M = .30 SD = .68 ) In turn, they were relatively motivated to engage in sexual risk taking in tourism by fun/less inhibitions/opportunity ( M = .32, SD = .74). Cluster three ( n = 180, 21.1%) mo
143 sexual risk taking dimensions as important (except the physical dimension) and they were moderately motivated to engage in sexual risk taking by all that the touri st experiences have to offer (anonymity/ detachment/experimentation: M = .39 SD = .71 ; safe exploration/thrill/ empowerment: M = .29 SD = .93 ; and fun/less inhibitions/opportunity: M = .37 SD = .74 ) Cluster four ( n = In this group, the women did not perceive any of the sexual risk taking dimensions as important and they were motivated to get involved in it by the safe exploration/thrill/empowerment ( M = .65 SD = .86 ) Finally cluster five ( n = unmotivated high and diverse risk perceivers Relative to other clusters, the women in this group were not motivated by any of the motivational dimensions and they perceived all of the sexual risk t aking dimensions as important (mental/emotional: M = .51 SD = .79; physical/sexual health: M = .21, SD = .49; and socio cultural: M = .40 SD = .83 ) Research Q uestion 3b: W hat A re the P ersonality S exual and S ocio D emographic P rofiles of the C lusters o f S exual R isk T akers in T ourism A mong the F emale C ollege S tudents? Profiling clusters with respect to nonsexual experience seeking (NES) and sexual sensation seeking (SSS) was conducted based on the MANOVA presented in Table 4 8. The NES trait (question 1, Appendix E) and SSS trait (question 2, Appendix E ), were measured on a 4 point Likert scale, where 1 = Not at all like me to 4 = Very much like me The total mean scores computed for each trait were moderately correlated ( r = .30, p = .000). In general, t he levels of sensation seeking traits in this sample were moderate (NES: M = 2.39, SD = .53; SSS: M = 2.00, SD = .60). Table 4 8 demonstrates significant differences among the clusters in terms of sensation seeking ( = .813, F (8, 1,594) = 21.77, p < .001), with significant differences on NES trait ( F (4, 798) = 14.34, p < .001) and SSS trait ( F (4, 798) = 39.24, p < .001).
144 Post Hoc tests revealed with respect to NES ( p < .05) that: (1) cluster five (i.e ., u nmotivated high and diverse risk perceivers ; M = 2.06, SD = .46) was significantly lower on NES than clusters one ( M = 2.50, SD = .56), two ( M = 2.36, SD = .50), three ( M = 2.50, SD = .52), and four ( M = 2.32, SD = .44); and (2) cluster one (i.e., h igh and diverse risk perceivers motivated by anonymity and exploration ; M = 2.50, SD = .56) and cluster three (i.e., l ow risk perceivers with diverse motivations ; M = 2.50, SD = .52) were significantly higher on NES than cluster two (i.e., m oderat e and diverse risk perceivers motivated by fun ; M = 2.36, SD = .50). As for the SSS, the Post Hoc tests indicated ( p < .05) that: (1) cluster five (i.e., u nmotivated high and diverse risk perceivers ; M = 1.49, SD = .48) was significantly lower on SS S than clusters one ( M = 2.09, SD = .60), two ( M = 1.90, SD = .48), three ( M = 2.33, SD = .53), and four ( M = 1.98, SD = .69); (2) cluster three (i.e., l ow risk perceivers with diverse motivations ; M = 2.33, SD = .53) was significantly higher on SSS than the rest of the clusters; and (3) cluster one ( M = 2.09, SD = .60) was significantly higher on SSS than cluster two ( M = 1.90, SD = .48). Cross tabulations were used to develop sexual profiles of the clusters (Table 4 9). In terms of se xual characteristics, the clusters were significantly different based on the number of sexual partners ( 2 = 87.35, p = .000), having been tested for STDs ( 2 = 9.76, p = .045), and having had sexual practices putting the participant at risk for HIV/AIDS ( 2 = 18.39, p = .001). Heterosexual women were the predominant group in this study ( n = 7,892, 92.8%) an d, as such, they comprised the major portion in each cluster. However, relative to the other clusters, cluster five (i.e., u nmotivated high and diverse risk perceivers ) had the highest proportion of
145 gay/lesbian women (2.2%) and cluster three (i.e., l ow ri sk perceivers with diverse motivations ) had the highest proportion of bisexual women (7.8%). Regarding dating status, women being in exclusive relationships were the biggest groups in all the clusters. Yet, relative to the other clusters, cluster one (i.e ., h igh and diverse risk perceivers motivated by anonymity and exploration ) had the lowest proportion of women involved in an exclusive relationship (49.5%) and the highest percentage of casually dating women, but not being in a relationship (15.3%). Also, cluster five was characterized by the relative extremes with the biggest proportion of women in exclusive relationship (56.2%) and one of the biggest proportions of the women who are not dating and are not relationship (37.1%). With respect to the number of sexual partners, clusters five (i.e., u nmotivated high and diverse risk perceivers ) and four (i.e., l ow risk perceivers motivated by exploration ) had the highest proportions of virgins (37.1% and 28.8%, respectively). Cluster two (i.e., m oderate and d iverse risk perceivers motivated by fun ) had the highest proportion of women who had one sexual partner (25.4%). Additionally, relative to the other clusters, cluster four had the highest proportion of women who had 3 4 partners (20.3%), cluster two had th e highest proportion of women who had 5 7 partners (18.3%), cluster one had the highest proportion of women who had 8 10 partners (13.8%), and cluster three had the highest proportion of women who had 11 15 partners (10.3%). In general, cluster three had r elatively more women with higher numbers of the sexual partners compared to other clusters. Most of the participants did not have experience with unplanned pregnancy (n = 747, 87.6%) and, as such, comprised the predominant proportions in all the clusters. Yet, clusters one, three, and five had more women with such experience (10.7%, 10.1%, and 10.2%, respectively) than clusters two and four (6.3% and 6.6%, respectively). As for having been tested for STDs,
146 cluster three had the highest proportion of women w ho had this experience (57.5%) and cluster five had the highest proportion of women who did not (61.8%). Likewise, most of the participants in this study had never been diagnosed with an STD ( n = 780, 91.4%), comprising the majority in all the clusters. However, cluster three (i.e., l ow risk perceivers with diverse motivations ) had a substantially higher proportion of women who had been diagnosed with an STD (14.0%) compared to other clusters. Cluster three also had the highest proportion of women who had engaged in sexual practices putting them at risk for HIV/AIDS (26.8%). Conversely, cluster five (i.e., u nmotivated high and diverse risk perceivers ) had the lowest proportion of women with experience of such practices (6.7%). Finally, while most of the w omen in this study did not report having had sex without giving their consent ( n = 758, 88.9%), the proportions of women who reported having had such an experience were slightly higher in clusters one, three, and four (11.1%, 13.4%, and 11.5%, respe ctively) compared to clusters two and five (9.4% and 9.0%, respectively). The following section of the analysis was conducted on an exploratory basis as little support exists in the literature to substantiate such relationships. Although, Hair et al. (2010 ) suggest that in socio demographic profiling following cluster analysis, no theoretical rationale is required for expecting the differences across the clusters as long as there is some practical importance associated with these variables. In terms of the socio demographic characteristics (Table 4 10), the clusters were significantly different based on place of birth ( 2 = 21.55, p = .000), place of growing up ( 2 = 14.95, p = .005), racial or ethnic background ( 2 = 55.44, p = .000), and personal annual income ( 2 = 42.27, p = .041). The majority of the sample ranged in age from 18 to 23 (n = 591, 69.6% ). Therefore, this age group comprised the predominant proportions in all the clusters. However, cluster one (i.e., h igh and diverse risk perceivers
147 motivated by anonymity and exploration ) and cluster three (i.e., l ow risk perceivers with diverse motivatio ns ) had the highest proportions of women aged between 18 20 (47.2% and 45.9%, respectively). In the other three clusters, the distributions between the age groups of 18 20 and 21 23 were more even. Relative to the other clusters, the proportion of the age group 24 26 was the highest (16.4%) in cluster four (i.e., l ow risk perceivers motivated by exploration ). Cluster one had the highest proportions of freshmen and sophomores relative to the other clusters. The proportions of freshmen were the lowest for th e clusters of the low risk perceivers clusters three and four (11.1% and 11.5%, respectively). Likewise, the proportion of juniors was the highest for these clusters (21.7% and 23.0%, respectively), relative to the other clusters. The proportion of senio rs was the highest for cluster two (18.4%) and the proportion of the graduate students was the highest for cluster five (40.4%). In terms of the place of birth, the proportions between the US born and non US born participants were similar across the clust ers (ranging for the US born participants from 82.9% to 85.2%), with exception of cluster four (i.e., l ow risk perceivers motivated by exploration ) where the proportion of the non US born participants was 40%. A very similar tendency could be noted with re spect to the place of growing up, where cluster four had substantially more members who did not grow in the US (24.6%) compared to other clusters. As for the racial or ethnic background, the highest proportion of white women (73.8%) was in cluster two (i. e., m oderate and diverse risk perceivers motivated by fun ), followed by cluster five of the u nmotivated high and diverse risk perceivers (67.0%). However, the white women were the biggest groups in all the clusters. Relative to the other clusters, the hig hest proportion of black women (9.8%) was in cluster four (i.e., l ow risk perceivers motivated by exploration ). Likewise, the highest proportion of Hispanic women was in cluster two (14.4%)
148 and the highest proportion of Asian women was in cluster four (24. 6%). Regarding the marital status, clusters three and four had the highest proportions of the never married women (88.2% and 88.5%, respectively). Yet, the never married women were the predominant group in all the clusters. Relative to other clusters, the highest proportion of the married women was in cluster one of h igh and diverse risk perceivers motivated by anonymity and exploration (15.7%). distribution appears to b e rather similar across clusters one, two, three, and five, with substantial numbers of participants dispersed throughout the middle income categories (as presented Table 4 10). However, relative to the other clusters, cluster four (i.e., l ow risk perceive rs motivated by exploration ) had the highest proportions in lower income categories (15.0% in under $25,000 and 26.7% in $25,000 $49,999) and the lowest proportions in the higher income categories (1.7% in $125,000 $149,999 and 11.7% in over $150,000). As for the personal annual income before taxes, cluster four had the highest proportion of the income category of under $5,000 (40.0%). Yet, this income category was dominant in all the clusters. Clusters five and one had the biggest proportions in the income category of $5,000 9,999 (25.8% and 24.4%, respectively), relative to the other clusters. Likewise, cluster three had the biggest proportion in the income category of $10,000 $14,999 (20.8%). Research Q uestion 4a: W hat A re the C haracteristics of the T ouri st E xperiences F acilitating and/or I nhibiting S exual R isk T aking in T ourism A mong the F emale C ollege S tudents? Participants were presented with a list of the touristic characteristics potentially facilitating or inhibiting sexual risk taking in tourism (5 M = 4.31, SD = M = 4.15, SD M = 4.12, SD loose
149 M = 4.10, SD M = 4.09, SD M = 3.99, SD ( M = 3.95, SD = .73). The factors reported as the most important inhibitors of sexual risk taking in tourism M = 1.61, SD M = 1.83, SD M = 1.94, SD = .83). The descriptive statistics for the complete list of the facilitating inhibiting touristic characteristics of sexual risk taking in tourism is presented in the Table 4 11. All the facilitators positively and substantially correlated with each other (.43 < r < .66, p < .001). Likewise, all the inhibitors were positively related to each other (.21 < r < .42, p < .001). With one exception, all the facilitators negatively correlated with all the inh ibitors ( .42 < r < .28, p < .01). Research Q uestion 4b: W hat A re the P erceptions of the R isk T aking C lusters A mong the F emale C ollege S tudents R egarding the C haracteristics F acilitating and/or I nhibiting S exual R isk T aking in T ourism? Differences in the perceptions of the five clusters with respect to the most influential facilitating and inhibiting characteristics were tested by MANOVA, where clusters were used as independent variables and the chosen touristic characteristics served as dependent variabl es. Due to the listwise deletion of some missing data in the dependent variables, the sizes of the clusters changed slightly, as reflected in Table 4 12. This analysis revealed significant differences among the clusters in terms of their perceptions of the touristic characteristics ( = .846, F (40, 2,940) = 3.31, p < .001), with significant F tests on each included facilitating and inhibiting item: consuming alcohol ( F (4, 784) = 6.52, p < .001), party scene at the destination ( F (4, 7 84) = 7.28, p < .001), being detached from everyday norms ( F (4, 784) = 10.93, p < .001), being in a break loose mood ( F (4, 784) = 8.69, p < .001), feeling more sexually confident ( F (4, 784) = 7.98, p < .001), having lots of free leisure ti me ( F (4, 784) = 6.91,
150 p < .001), scene where revealing clothing is appropriate ( F (4, 784) = 4.60, p < .005), being in a country with strict religious beliefs ( F (4, 784) = 4.05, p < .005), having judgmental trip partners ( F (4, 784) = 8.26, p < .001), and having a steady relationship at home ( F (4, 784) = 10.89, p < .001). Post Hoc tests revealed the following differences (focusing on the results significant at the .05 level). C luster one (i.e., h igh and diverse risk perceivers motivat ed by anonymity and exploration ) rated party scene at the destination ( M = 4.31, SD = .65), detachment from everyday norms ( M = 4.26, SD = .67), break loose mood ( M = 4.28, SD = .68), increased sexual confidence ( M = 4.25, SD = .67), lots of free leisure t ime ( M = 4.09, SD = .67), and the scene where revealing clothing is appropriate ( M = 4.09, SD = .62) as significantly more facilitating sexual risk taking in tourism than cluster four (i.e., l ow risk perceivers motivated by exploration ; M = 3.84, SD = .78; M = 3.74, SD = .72; M = 3.70, SD = .80; M = 3.80, SD = .73; M = 3.74, SD = .66; M = 3.80, SD = .70; respectively) and cluster five (i.e., u nmotivated high and diverse risk perceivers ; M = 3.93, SD = 1.00; M = 3.84, SD = 1.00; M = 3.97, SD = 1.02; M = 3.85, SD = 1.05; M = 3.74, SD = 1.02; M = 3.80, SD = 1.03; respectively). Cluster one also perceived having a steady relationship at home as significantly less inhibiting ( M = 1.70, SD = .94) than cluster two (i.e., m oderate and diverse risk perceiv ers motivated by fun ; M = 1.44, SD = .81). Cluster two reported alcohol ( M = 4.33, SD = .82), party scene ( M = 4.15, SD = .79), and break loose mood ( M = 4.09, SD = .75) to be significantly more facilitating than cluster four ( M = 3.92, SD = .82; M = 3 .84, SD = .78; M = 3.70, SD = .80; respectively). Cluster two reported detachment from everyday norms ( M = 4.14, SD = .71) and increased sexual confidence ( M = 4.12, SD = .65) to be significantly more facilitating than clusters four ( M = 3.74, S D = .73; M = 3.80, SD = .73; respectively) and five ( M = 3.84, SD = 1.00; M = 3.85, SD = 1.05;
151 respectively). The women in cluster two perceived free leisure time as significantly more facilitating ( M = 4.01, SD = .72) than those in cluster five ( M = 3.74, SD = 1.02), but they perceived the appropriateness of revealing clothing as significantly less facilitating ( M = 3.90, SD = .68) than the women in cluster one ( M = 4.09, SD = .62). Cluster three (i.e., l ow risk perceivers with diverse motivations ) report ed the party scene ( M = 4.25, SD = .73), detachment from everyday norms ( M = 4.26, SD = .69), increased sexual confidence ( M = 4.15, SD = .65), and lots of free leisure time ( M = 4.12, SD = .65) to be significantly more facilitating than clusters four (i.e ., l ow risk perceivers motivated by exploration ; M = 3.84, SD = .78; M = 3.74, SD = .73; M = 3.80, SD = .73; M = 3.74, SD = .66; respectively) and five (i.e., u nmotivated high and diverse risk perceivers ; M = 3.93, SD = 1.00; M = 3.84, SD = 1.00; M = 3.8 5, SD = 1.05; M = 3.74, SD = 1.02; respectively). Additionally, the women in cluster three reported alcohol consumption ( M = 4.41, SD = .76) and break loose mood ( M = 4.22, SD = .69) to be significantly more facilitating than those in cluster four ( M = 3.92, SD = .82; M = 3.70, SD = .80; respectively). Cluster four (i.e., l ow risk perceivers motivated by exploration ) reported consuming alcohol ( M = 3.92, SD = .82), party scene ( M = 3.84, SD = .78), detachment from everyday norms ( M = 3.74, SD = 73), break loose mood ( M = 3.70, SD = .80), and sexual confidence ( M = 3.80, SD = .73) to be significantly less facilitating than clusters one ( M = 4.46, SD = .65; M = 4.31, SD = .65; M = 4.26, SD = .67; M = 4.28, SD = .68; M = 4.25, SD = .67; resp ectively), two ( M = 4.33, SD = .82; M = 4.15, SD = .79; M = 4.14, SD = .71; M = 4.09, SD = .75; M = 4.12, SD = .65; respectively), and three ( M = 4.41, SD = .76; M = 4.25, SD = .73; M = 4.26, SD = .69; M = 4.22, SD = .69; M = 4.15, SD = .65; respectivel y). The women in cluster four also perceived
152 having lots of free leisure time on vacation as significantly less facilitating sexual risk taking ( M = 3.74, SD = .66) than clusters one ( M = 4.09, SD = .67) and three ( M = 4.12, SD = .65). However, cluste r four perceived strict religious beliefs at the destination country ( M = 2.30, SD = .67), judgmental trip partners ( M = 2.31, SD = .77), and having a steady relationship at home ( M = 2.18, SD = .90) as significantly less inhibiting sexual risk taking than clusters one ( M = 1.93, SD = .87; M = 1.83, SD = .84; M = 1.70, SD = .94; respectively), two ( M = 1.93, SD = .85; M = 1.71, SD = .65; M = 1.44, SD = .81; respectively), three ( M = 1.83, SD = .80; M = 1.81, SD = .80; M = 1.52, SD = .84; re spectively), and five ( M = 1.82, SD = .75; M = 1.72, SD = .62; M = 1.49, SD = .70; respectively) Thus, relative to other clusters, cluster four was significantly less susceptible to both facilitating and inhibiting touristic characteristics. Finally, clus ter five ( i.e., unmotivated high and diverse risk perceivers) perceived d etachment from everyday norms ( M = 3.84, SD = 1.00) increased sexual confidence ( M = 3.85, SD = 1.05) and lots of free leisure time ( M = 3.74, SD = 1.02) as significantly less faci litating than clusters one ( M = 4.26, SD = .67; M = 4.25, SD = .67; M = 4.09, SD = .67; respectively), two ( M = 4.14, SD = .71; M = 4.12, SD = .65; M = 4.01, SD = .72; respectively), and three ( M = 4.26, SD = .69; M = 4.15, SD = .65; M = 4.19, SD = .65; respectively). The women in cluster five also reported the party scene ( M = 3.93, SD = 1.00), b reak loose mood ( M = 3.97, SD = 1.02) and appropriateness of revealing clothing ( M = 3.80, SD = 1.03) to be significantly less facilitating than the wom en in cluster one ( M = 4.31, SD = .65; M = 4.28, SD = .68; M = 4.09, SD = .62; respectively)
153 Table 4 1. Perceptions and experiences of sexual risk taking on vacation Perceptions as sexual risk taking Experience with sexual activity on vaca tion Independent samples T Test Sexual Activities on Vacation N M SD N Yes (1) No (2) M t n % n % (1) (2) Having vaginal sex with an unsteady partner without a condom 850 4.79 .640 847 87 10.2 760 89.1 .049 .681 ^^^^ Having anal sex wi th an unsteady partner without a condom 848 4.74 .692 847 17 2.0 830 97.3 .402 1.562 ^^^^ Having sex with a sex worker (e.g., prostitute) 847 4.74 .710 849 1 0.1 848 99.5 Not possible Having sex with multiple partners 849 4.63 .713 847 44 5 .2 803 94.1 .272 1.947 ^^^^ Performing oral sex on an unsteady partner without a condom 848 4.52 .817 844 145 17.0 699 81.9 .331 3.979 *** Receiving oral sex from an unsteady partner without a latex barrier 851 4.45 .863 847 127 14.9 720 84.4 .377 4.599 *** Having sex under the influence of drugs 848 4.42 .886 847 98 11.5 749 87.8 .548 2.879 *** Having vaginal sex with a steady partner without protection from pregnancy 849 4.40 .947 848 198 23.3 650 76.2 .354 4.152 *** Att ending a swinger club 849 4.22 .939 847 12 1.4 835 97.9 .050 .183 ^^^^ Having sex in semi public spaces (like a restroom or elevator) 851 4.05 .912 848 226 26.5 622 72.9 .366 4.979 *** Having vaginal sex with an unsteady partner with a condom 85 0 3.90 1.029 849 146 17.1 703 82.4 .251 2.545* ^^^ Having anal sex with an unsteady partner with a condom 850 3.87 1.055 849 15 1.8 864 97.8 .412 1.170 ^^^^ Having sex in unfamiliar surroundings 848 3.84 1.094 849 325 38.1 524 61.4 .513 6.6 47 *** Having sex under the influence of alcohol 851 3.79 1.037 848 396 46.4 452 53.0 .544 7.778 *** Performing oral sex on an unsteady partner with a condom 849 3.51 1.124 849 46 5.4 803 94.1 .394 2.292* ^^^ Receiving oral sex from an unstead y partner with a latex barrier 851 3.48 1.145 849 25 2.9 824 96.6 .078 .335 ^^^^ Going to live sex shows 849 3.38 1.246 845 31 3.6 814 95.4 .64 0 2.726 * ^^ Going to a strip club 851 3.01 1.182 848 133 15.6 715 83.8 .374 3.370 ** ^ Using sexu al enhancers with your steady partner 848 2.91 1.275 848 122 14.3 726 85.1 .698 5.729 *** Fondling with an unsteady partner 849 2.73 1.196 849 298 34.9 551 64.6 .563 6.690 *** Trying sexual experimentation with your steady partner 850 2.72 1.25 7 848 355 41.6 493 57.8 .340 3.916 *** Making out with/kissing an unsteady partner 848 2.54 1.168 849 395 46.3 454 53.2 .502 6.369 *** Being more flirtatious 850 2.46 1.108 847 582 68.2 265 31.1 .469 5.821 *** Note: **** p < .001, *** p < .005, ** p < .01, p < .05 P erceptions were measured on a 5 point Li kert scale, where 1 = trongly disagree and 5 = trongly agree
154 Table 4 2. EFA solution for the sexua l risk perceptions model M SD Factor l oadings Items F1 F2 F3 S exual activity on vacation is risky if it entails: the risk of getting sexually transmitted diseases 4.88 .419 .935 the risk of contracting HIV/AIDS 4.89 .401 .961 the risk of getting pregnant 4.83 .503 .775 the risk of being raped 4.88 .420 .920 the risk of being kidnapped 4.85 .487 .828 the risk of getting physically hurt 4.78 .540 .709 the risk of being emotionally hurt 4.09 .961 .723 the risk of feeling uncom fortable after the act 4.04 .962 .934 the risk of being too far out of my comfort zone 4.06 .948 .871 the risk of feeling used 4.11 .975 .930 the risk of having unfulfilled expectations 3.65 1.102 .794 the risk of feeling rejected 3.76 1.078 789 the risk of regretting it later 4.14 .963 .775 the risk of doing something against my values 4.21 .965 .703 the risk of embarrassing myself 3.96 1.010 .420 .438 3.97 .984 .497 .352 the risk of feeling not respected 4.10 .968 .423 .478 the risk of affecting my reputation at home 4.04 1.026 .864 the risk of being judged by travelling companions 3.79 1.056 .844 the risk of offending cultural norms at the destination 3.85 1.052 .706 Initial E igenvalues 3.736 9.250 1.072 Cronbach's alpha .921 .946 .837 Factor correlations F1 F2 F3 F1 1.000 F2 .275 1.000 F3 .221 .644 1.000 Note : F1 = Factor 1 Physical/Sexual Health F2 = Factor 2 Mental/Emotional F3 = Factor 3 Socio Cultural All the items were measured on a 5 point Likert scale, where 1 = trongly disagree 5 = trongly agree
155 Table 4 3 Summary of goodness of fit statistics for all models with s exual s ensation s eeking (SSS) index as the grou ping variable Model 2 /df NF Parm CFI TLI RMSEA SRMR AIC BIC corBIC Description Total group (TG) model s TG ESEM 1217.701/133 97 .924 .891 .098 .030 27242.195 27702.597 27394.554 T G ESEM with 3 factors TG ICM CFA 1704.236/167 63 .892 .877 .104 .066 2766 0.730 27959.754 27759.684 T G ICM CFA with 3 factors Multiple (two) group invariance (MGI) across low and high levels of sexual sensation seeking (SSS) MGI1 1428.404/266 194 .918 .883 .101 .033 26689.238 27610.042 26993.955 IN = none MGI2 1560.563/317 1 43 .913 .895 .096 .056 26719.396 27398.133 26944.007 IN = FL MGI3 2012.447/337 123 .882 .867 .108 .129 27131.280 27715.089 27324.477 IN = FL, Uniq MGI4 1608.499/323 137 .910 .894 .097 .132 26755.333 27405.591 26970.519 IN = FL, FVCV MGI5 1608.298/334 1 26 .910 .898 .095 .060 26733.131 27331.179 26931.040 IN = FL, INT MGI6 2080.492/343 117 .878 .865 .109 .210 27187.326 27742.656 27371.098 IN = FL, Uniq, FVCV MGI7 2058.100/354 106 .880 .872 .106 .135 27142.934 27346.054 27309.429 IN = FL, Uniq, INT MG I8 1656.779/340 120 .908 .897 .095 .135 26769.613 27339.182 26958.098 IN = FL, FVCV, INT MGI9 2126.230/360 100 .876 .869 .107 .214 27199.064 27673.705 27356.134 IN = FL, FVCV, INT, Uniq MGI10 1657.227/337 123 .907 .895 .096 .078 26776.061 27359.870 26969 .258 IN = FL, INT, FMn MGI11 2106.855/357 103 .877 .869 .107 .160 27185.688 27674.569 27347.471 IN = FL, Uniq, INT, FMn MGI12 1706.223/343 117 .904 .894 .097 .164 26813.057 27368.387 26996.830 IN = FL, FVCV, INT, FMn MGI13 2175.361/363 97 .873 .867 .1 08 .244 27242.195 27702.597 27394.554 IN = FL, FVCV, INT, Uniq, FMn Note: CFI = Comparative Fit Index; TLI = Tucker Lewis Index; NFParm = number of free parameters; AIC = Akaike's Information Criterion; BIC = Bayesian Information Criterion; corBic = sample size adjusted BIC; RMSEA = Root Mean Squared Error of Approximation; SRMR = Standar dized Root Mean Square Residual; ESEM = exploratory structural equation modeling; ICM CFA = independent clusters model confirmatory factor analysis. For multiple group invariance models, IN = means the number of parameters constrained to be invariant across multiple groups: FL = factor loadings; FVCV = factor variance covariances; INT = item intercepts; Uniq = item uniquenesses; FMn = factor means. All the items were measured on a 5 point Likert scale, where 1 = trongly disagree 5 = trongly agree. The source of the taxonomy and the description colu mn is Marsh et al. (2009, 2010, 2011, 201 2 ).
156 Table 4 4 Sexual sen s ation seeking (SSS) level s Items N M SD 852 1.67 .865 The physical sensations are the most important thing about having sex 849 1.97 .815 I enjoy the sensation of intercourse without a condom 849 2.44 1.104 My sexual 844 1.67 .785 When it comes to sex, physical attraction is more important to me than how well I know the person 851 1.56 .721 849 2.18 .879 rated 850 1.90 .936 I have said things that were not exactly true to get a person to have sex with me 850 1.25 .545 I am interested in trying out new sexual experiences 852 2.51 .964 I feel like exploring my sexuality 852 2.36 .973 I like to ha ve new and exciting sexual experiences and sensations 852 2.53 .951 Note: All the SSS items were measured on a 4 and
157 Table 4 5 Perceived sexual risk dimensions f or lower and higher levels of Sexual Sensation Seeking (SSS) based on MGI1 Factor loadings Items Lower levels of SSS Higher l evels of SSS Intercepts Uniquenesses Sexual activity on vacation is risky if it entails: F1 F2 F3 F1 F2 F3 Lower Hig her Lower Higher the risk of getting sexually transmitted diseases .374* .436* 4.895* 4.860* .015* .018* the risk of contracting HIV/AIDS .363* .440* 4.911* 4.877* .006* .005* the risk of getting pregnant .327* .370* 4.872* 4.790* .073* .177* the risk of being raped .337* .396* 4.897* 4.857* .032* .048* the risk of being kidnapped .346* .366* 4.874* 4.833* .079* .124* the risk of getting physically hurt .306* .334* 4.828* 4.724* .106* .209* the risk of bei ng emotionally hurt .545* .663* 4.270* 3.903* .233* .368* the risk of feeling uncomfortable after the act .768* .871* 4.249* 3.809* .181* .303* the risk of being too far out of my comfort zone .706* .762* 4.249* 3.868* .218* .370* t he risk of feeling used .742* .930* 4.307* 3.892* .158* .231* the risk of having unfulfilled expectations .742* .763* 3.770* 3.519* .496* .484* the risk of feeling rejected .671* .807* 3.903* 3.601* .452* .404* the risk of regrettin g it later .603* .614* 4.348* 3.921* .247* .488* the risk of doing something against my values .585* .530* .401* 4.459* 3.950* .225* .513* the risk of embarrassing myself .427* .336* .400* .439* 4.135* 3.773* .397* .534* the risk of h .377* .398* .460* .361* 4.092* 3.845* .340* .512* the risk of feeling not respected .448* .340* .550* 4.265* 3.925* .241* .368* the risk of affecting my reputation at home .799* .828* 4.185* 3.876* .216* .488 the risk of being judged by travelling companions .883* .861* 3.962* 3.599* .176* .417* the risk of offending cultural norms at the destination .363* .554* 4.017* 3.665* .675* .752* Factor correlations F1 F2 F3 F1 F2 F3 F1 1.000 1.000 F2 .342* 1.000 .276* 1.000 F3 .285* .720* 1.000 .213* .578* 1.000 Note: at least p < .05; F1 = Factor 1 Physical/Sexual Health; F2 = Factor 2 Mental/Emotional; F3 = Factor 3 Socio Cultural A ll the items were measured on a 5 point Likert scale, where 1 = trongly disagree 5 = trongly agree
158 Table 4 6. EFA solution for the motivations model M SD Factor loadings Items F1 F2 F3 Tourism offers anonymity 3.46 1.079 .810 Nobody will judge me 3.28 1.150 .804 I can enact sexual fantasies not available at home 3.15 1.141 .795 Risk is a motivation in itself 3.87 1.168 .749 I feel detached from everyday social norms 3.44 1.097 .745 Tourism is a scene for sexual conquest 2 .68 1.072 .572 Casual sex on vacation has no consequences 2.10 1.083 .438 .487 Taking sexual risks on vacation is empowering 2.68 1.114 .392 .499 Vacation mentality is about taking sexual risks 2.27 1.017 .713 Taking sexual risks on vacation is t hrilling 2.83 1.184 .499 It is cool to brag about sexual risks taken on vacation 2.28 1.156 .664 Sex is a way to get to know the locals 1.76 .890 .836 1.76 .950 .799 What happens on vacation stays on vacation 2.79 1. 270 .450 Vacation is the time for sexual exploration 2.62 1.017 .615 .409 Availability of my type of sexual partner on vacation 3.07 1.154 .412 .533 Sex feels good 3.98 1.002 .749 Vacation is the time to have fun 3.84 .991 .572 .476 I have less inhibitions on vacation 3.33 1.113 .405 .420 Initial Eigenvalues 8.832 1.917 1.009 Cronbach's alpha .910 .895 .820 Factor correlations F1 F2 F3 F1 1.000 F2 .489 1.000 F3 .395 .186 1.000 Note: F1 = Factor 1 Ano nymity/Detachment/Experimentation F2 = Factor 2 Safe Exploration/Thrill/Empowerment F3 = Factor 3 Fun/Less Inhibitions/Opportunity All the items were measured on a 5 point Likert scale, where 1 = trongly disagree and 5 = trongly agree
159 Tabl e 4 7. Classification of sexual risk takers in tourism Range Sexual risk taking clusters ( N = 803) F Cluster Variate Min Max C1 (n=217) C2 (n=256) C3 (n=180) C4 (n=61) C5 (n=89) Perceived risk dimensions Mental/Emotional 3.678 1 .642 .552 .291 1.155 .529 .514 167.532 *** Physical/Sexual Health 9.831 .445 .297 .291 .079 2.773 .207 386.388 *** Socio Cultural 3.792 2.122 .586 .299 1.191 .426 .398 176.498 *** Perceived motivations Anonymity/Detachment/Experim entation 2.723 2.397 .664 .208 .388 .043 1.801 222.652 *** Safe Exploration/Thrill/Empowerment 1.927 3.664 .765 .641 .294 .652 1.055 176.501 *** Fun/Less Inhibitions/Opportunity 2.978 2.706 .132 .323 .369 .344 1.753 144.187 *** Note: * ** p < .001 C1 = Cluster 1 High and diverse risk perceivers motivated by anonymity and exploration C2 = Cluster 2 Moderate and diverse risk perceivers motivated by fun C3 = Cluster 3 Low risk perceivers with diverse motivations C4 = Cluster 4 Low risk perceivers motivated by exploration C5 = Cluster 5 Unmotivated high and diverse risk perceivers
160 Table 4 8. Sensation seeking profile of five clusters Sexual risk taking clusters Wilks' F C1 (n=217) C2 (n=256) C3 (n=180) C4 (n=61) C5 (n=89) Sensation seeking indices .813 21.771**** Nonsexual Experience Seeking (NES) 2.50 2.36 2.50 2.32 2.06 14.339**** Sexual Sensation Seeking (SSS) 2.09 1.90 2.33 1.98 1.49 39.237* *** Note : **** p < .001 C1 = Cluster 1; C2 = Cluster 2; C3 = Cluster 3; C4 = Cluster 4; C5 = Cluster 5 All the items were measured on a 4 mean endorsements fo r NES and SSS were computed.
161 Table 4 9. Sexual profile of five clusters Sexual risk taking clusters (%) C1 (n=217) C2 (n=256) C3 (n=180) C4 (n=61) C5 (n=89) Sexual Orientation 8.753 ^^^^ Heterosexual 93.5 94.1 90.6 93.4 94.4 Gay/Lesbian .9 1.6 .6 1.6 2.2 Bisexual 5.1 3.5 7.8 3.3 3.4 Other .5 .8 1.1 1.6 .0 Dating s tatus 11.349 ^^^^ Exclusive relationship 49.5 52.8 53.1 52.5 56.2 Open relationship .9 1.2 2.2 3.3 1.1 Casually dating, but not in a relationship 15.3 9.9 9.5 6.6 5.6 Not dating and not in a relationship 34.3 36.1 35.2 37.7 37 .1 Number of sexual partners 87.352 *** no partners 18.6 19.4 14.4 28.8 37.1 1 partner 23.8 25.4 14.4 20.3 23.6 2 partners 13.8 10.3 9.2 10.2 10.1 3 4 partners 12.4 13.9 16.1 20.3 11.2 5 7 partners 11.0 18.3 15.5 11. 9 7.9 8 10 partners 13.8 4.8 10.3 1.7 3.4 11 15 partners 2.9 4.4 10.3 .0 2.2 16 20 partners 2.4 2.8 4.0 3.4 2.2 21 30 partners 1.0 .8 4.6 3.4 1.1 31 and more partners .5 .0 1.1 .0 1.1 Unplanned pregnancy (persona l or sexual partner's experience) 7.056 ^^^^ Yes 10.7 6.3 10.1 6.6 10.2 No 84.7 91.0 85.5 91.8 87.5 I do not know 4.7 2.7 4.5 1.6 2.3 Having been tested for STD 9.757* ^^^ Yes 47.4 50.0 57.5 46.7 38.2 No 52.6 50.0 42.5 5 3.3 61.8 Having been diagnosed with STD 9.189 ^^^^ Yes 6.9 7.4 14.0 6.6 5.6 No 93.1 92.6 86.0 93.4 94.4 Having had sexual practices putting at risk for HIV/AIDS 18.394* ** ^ Yes 20.4 16.9 26.8 13.1 6.7 No 79.6 83.1 73.2 86.9 93.3 Having had unsafe sex with someone the participant thought might have HIV/AIDS 2.936 ^^^^ Yes .9 1.6 .6 .0 .0 No 99.1 98.4 99.4 100.0 100.0 Having had sex with someone without giving consent 2.110 ^^^^ Yes 11.1 9.4 13.4 11.5 9 .0 No 88.9 90.6 86.6 88.5 91.0 Having had sex with someone without getting their consent 3.377 ^^^^ Yes .9 .0 1.1 .0 1.1 No 99.1 100.0 98.9 100.0 98.9 Note: **** p < .001, *** p < .005, p < .05 C1 = Cluster 1; C2 = Clus ter 2; C3 = Cluster 3; C4 = Cluster 4; C5 = Cluster 5
162 Table 4 10 Socio demographic profile of five clusters Sexual risk taking clusters (%) C1 (n=217) C2 (n=256) C3 (n=180) C4 (n=61) C5 (n=89) Age 24.539 ^^^^ 18 20 47.2 37.5 35.6 45.9 38.2 21 23 26.6 31.6 31.1 24.6 33.7 24 26 8.4 11.7 10.6 16.4 9.0 27 30 7.5 8.6 11.7 4.9 6.7 31 40 7.0 5.9 8.3 6.6 4.5 41 50 2.3 3.5 2.8 1.6 4.5 51 62 .9 1.2 .0 .0 3 .4 Year in school 24.897 ^^^^ Freshman 19.0 12.5 11.1 11.5 14.6 Sophomore 19.0 12.5 12.2 16.4 15.7 Junior 18.5 18.4 21.7 23.0 12.4 Senior 16.2 18.4 15.6 14.8 16.9 Graduate 26.4 35.9 36.7 34.4 40.4 Other .9 2.3 2.8 .0 .0 Place of birth 21.550 *** US 82.9 84.4 83.2 60.0 85.2 Not US 17.1 15.6 16.8 40.0 14.8 Place of growing up 14.951 * ^^ US 88.0 91.0 92.2 75.4 89.9 Not US 12.0 9.0 7.8 24.6 10.1 Racial or ethnic background 55.442 *** W hite/Caucasian, not of Hispanic origin 64.2 73.8 66.1 52.5 67.0 Black, not of Hispanic origin 7.0 2.7 7.8 9.8 6.8 Hispanic 14.4 16.0 13.3 11.5 10.2 Asian 10.2 4.3 6.7 24.6 6.8 Native Hawaiian/Other Pacific Islander .9 .0 .0 0 1.1 American Indian/Native Alaskan .9 .8 .0 .0 1.1 Other 2.3 2.3 6.1 1.6 6.8 Marital status 9.546 ^^^^ Never married 80.6 85.5 88.2 88.5 86.5 Married 15.7 12.9 9.0 9.8 10.1 Divorced/Separated 3.2 1.2 2.8 1.6 3.4 Wid owed .5 .4 .0 .0 .0 Parents'/legal guardians' joint annual household income from all sources before taxes 21.966 ^^^^ Under $25,000 7.1 8.6 11.3 15.0 6.9 $25,000 $49,999 16.1 15.7 14.1 26.7 17.2 $50,000 $74,999 16.6 20.4 15.3 16 .7 18.4 $75,000 $99,999 16.6 17.6 14.7 18.3 17.2 $100,000 $124,999 17.5 12.2 16.9 10.0 12.6 $125,000 $149,999 8.5 7.5 8.5 1.7 11.5 Over $150,000 17.5 18.0 19.2 11.7 16.1 Personal annual income from all sources before t axes 42.269* ^^^ Under $5,000 27.7 24.7 17.4 40.0 24.7 $5,000 $9,999 24.4 17.6 18.0 13.3 25.8 $10,000 $14,999 11.7 14.5 20.8 8.3 6.7 S15,000 $19,999 10.3 10.2 10.7 15.0 10.1 $20,000 $29,999 8.9 12.9 10.1 13.3 11.2 $30,000 $39,999 6.6 5.9 5.1 5.0 3.4 $40,000 $49,999 3.8 3.5 6.2 1.7 4.5 Over $50,000 6.6 10.6 11.8 3.3 13.5 Note: **** p < .001, ** p < .01, p < .05 C1 = Cluster 1; C2 = Cluster 2; C3 = Cluster 3; C4 = Cluster 4; C5 = Cluste r 5
163 Table 4 11. Facilitating and inhibiting touristic characteristics of sexual risk taking in tourism Items N M SD Consuming alcohol 851 4.31 .823 Party scene at the destination 846 4.15 .788 Being detached from everyday norms 851 4.12 .762 Being in a break loose mood 850 4.10 .797 Feeling more sexually confident 846 4.09 .748 Having lots of free leisure time 850 3.99 .745 Scene where revealing clothing is appropriate 851 3.95 .730 Consuming recreational drugs 852 3.92 1.097 Feelin g invincible in terms of risk consequences 846 3.90 .882 Permissive sexual atmosphere 847 3.86 .846 Touristic fun oriented mentality 847 3.83 .749 Being anonymous 852 3.83 .991 Lack of social expectations 846 3.80 .828 Leaving bad experiences beh ind 847 3.79 .763 Beach scene at the destination 851 3.75 .763 Feeling a different person 846 3.71 .827 The length of the vacation 851 3.68 .778 Not expecting a lasting relationship 848 3.67 1.029 Meeting new people 846 3.65 .802 Sharing a room 848 3.64 1.097 Being in a country with more tolerant religious beliefs 848 3.44 .820 Experiencing peer pressure 850 3.39 1.034 The size of your travel party 850 3.19 .902 Being in a country that speaks a different language 850 2.86 .884 Having l imited or no safety network 847 2.55 1.140 Being in a country with strict religious beliefs 850 1.94 .832 Having judgmental trip partners 849 1.83 .775 Having a steady relationship at home 847 1.61 .887 Note: All the items were measured on a 5 poi
164 Table 4 1 2 Clusters' perceptions of facilitating and inhibiting touristic characteristics Sexual risk taking clusters Wilks' F C1 (n=212) C2 (n =252) C3 (n=176) C4 (n=61) C5 (n=88) Facilitating and inhibiting characteristics .846 3.310 *** Consuming alcohol 4.46 4.33 4.41 3.92 4.20 6.516 *** Party scene at the destination 4.31 4.15 4.25 3.84 3.93 7.280 *** Being detached from ev eryday norms 4.26 4.14 4.26 3.74 3.84 10.933 *** Being in a break loose mood 4.28 4.09 4.22 3.70 3.97 8.688 *** Feeling more sexually confident 4.25 4.12 4.15 3.80 3.85 7.975 *** Having lots of free leisure time 4.09 4.01 4.12 3.74 3.74 6.908 *** Scene where revealing clothing is appropriate 4.09 3.90 4.03 3.80 3.80 4.599 ** ^ Being in a country with strict religious beliefs 1.93 1.93 1.83 2.30 1.82 4.053 ** ^ Having judgmental trip partners 1.83 1.71 1.81 2.31 1.72 8.262 *** Having a st eady relationship at home 1.70 1.44 1.52 2.18 1.49 10.887 *** Note: *** p < .0 0 1 *** p < .005 C1 = Cluster 1; C2 = Cluster 2; C3 = Cluster 3; C4 = Cluster 4; C5 = Cluster 5 All the items were measured on a 5 p oint Likert scale, where 1 = trongly inhibits 5 = trongly facilitates
165 CHAPTER 5 DISCUSSION Taking in Tourism The findings from both the qualitative and quantitative phases point to a lack of consistency in te rms of the specific sexual behaviors perceived as risky by the women in tourism. In the qualitative phase, the women referred to various sexual activities where the degree and the nature of the risks involved were dependent on the context and the type of t he sexual relationships (i.e., casual vs. steady sexual partner). Furthermore, even when presented with a list of potentially risky sexual activities in tourism in the quantitative phase, the women rated them quite differently, which is reflected in the su bstantial amount of variance around the means. However, at least two nuances require attention in this respect. First, the quantitative results provide some evidence for suggesting that having had an actual experience with a certain sexual activity in tou perceptions of the degree of risk involved in it. Second, penetrative sex (especially unprotected) was portrayed in the qualitative phase and rated in the quantitative phase as involving the highest degree of risk, with a p articular concern for sexual health consequences. On the one hand, this focus on penetrative sexual activities resonates with the narrow definition of sex for understanding sexual risk taking in tourism among women (Thomas, 2000). On the other hand, the di versity of sexual activities in tourism perceived as risky suggests that a broader definition of sex is needed that allows for a variety of meanings and understandings of sex and sex related risk (Eiser & Ford, 1995; Maticka Tyndale & Herold, 1997; Opperma nn, 1998, 1999; Ryan & Martin, 2001). Additionally, for some women, personal definitions of sex were more comprehensive and broad than traditional societal definitions and for some women, their theoretical definitions of
166 sex were more inclusive than act ual practice. These distinctions can be explained based on the interpersonal, and intrapsychic levels) and the empirical studies that found discrepancies between perceptio ns of sexual matters at the cultural level and personal experiences and interpretations at the interpersonal and intrapsychic levels (McCabe et al., 2010; Milhausen & Herold, 1999). Despite the aforementioned lack of consistency in terms of the specific ri sky sexual activities in tourism, there was consensus among the women that the risk involved is multifaceted. Both the qualitative and quantitative findings reveal that sexual risk taking in tourism is a multidimensional and complex phenomenon/construct, w hich is not acknowledged sufficiently in the current tourism literature. The juxtaposition of the qualitative and quantitative findings with respect to the perceived dimensions of sexual risk taking in tourism reflects both commonalities and differences. T he qualitative findings suggest physical/sexual health, social, emotional, mental, and cultural dimensions as the universal (or at least common) textures of sexual risk taking in tourism. The quantitative results, in turn, provide a somewhat different comp osition of the factors. The p hysical/sexual health factor remained consistent across the phases, which is in line with the current state of literature in the field mainly revolving around STDs as a consequence of sexual risk taking in tourism (Bloor et al ., 2000; Clift & Forrest, 2000; Hart & Hawkes, 2000). In this study, however, the physical component was constructed by the women as broader than STDs, including any sort of violence or unwanted physical outcome (e.g., rape, abduction, or pregnancy). Women consciousness in light of the sexualized gaze are frequently constructed as gendered constraints and expressions of patriarchy 08;
167 risk with respect to sexual behavior in leisure travel contexts. While the importance of the physical dimension was generally rated as the highest in both research phases, the rest of the dimensions were not neglected by the women. Moreover, while the physical dimension was a major concern with casual sexual partners, other dimensions were more relevant with respect to sexual risk taking in tourism in long term relationships. In other words, women were typically not afraid of contracting STDs and/or of the violent behavior on behalf of a steady partner, but did bring up socio cultural risks associated with violating the destination specific norms regarding d isplays of affection as well as the mental emotional risks related to overstepping personal boundaries in sexual exploration while in a liminoid tourist state of mind (Lett, 1983; Selnniemi, 2003; Turner, 1974). Likewise, casual sexual encounters in touri sm might have important emotional implications for women (Thomas, 2000). Thus, the importance of this study stems from going beyond the physical dimension and considering the nuances bound with the characteristics of tourist environments and types of relat ionships. In the qualitative phase, the women discussed social and cultural aspects of sexual risk taking in tourism as distinct, yet, overlapping dimensions. The social dimension referred to avelling companions. In this respect, women frequently alluded to sexual double standards as a social norm restricting their sexual behavior at home (Jonason & Fisher, 2009; McCabe et al., 2010) and often to a considerably lesser degree in tourism (Maticka Tyndale & Herold, 1997; Mewhinney et al., 1995). Indeed, sex has been historically wrapped up with silence, taboos, and double standards (Foucault, 1976), which are more restrictive for women (Eaton & Rose, 2011; McCabe et al., 2010; Muehlenhard et al., 2 003).
168 In this respect, masculinity. Moreover, conflicting ideals bound with the Madonna whor e dichotomy encourag e both sexual restraint and permissiveness, leaving women with a n internal conflict (Crawford & Popp, 2003). Contradicting social expectations for women to be sexually available, but not active, limit their opportunity for sexual pleasu re, autonomy, and self determination (Blanc, 2001). Furthermore, the asymmetry of power bound with compliant and vulnerable to sexual risk taking, both physically and emotionally (Impett & Peplau, 2003). The qualitative findings in this study illustrate that sexual double standards ( i.e., to carry condoms), facing them with an ambivalent choice between risks associated with sexual health and social judgment (Crawford & Popp, 2003; Higgins & Browne, 2008; Tolman, Striepe, & Harmon, 2003). Yet, it is important to mention that with two e xceptions, women in this study referred to transitional double standards rather than orthodox double standards (Crawford & Popp, 2003; Reiss, 1964) whereby premarital sex is deemed acceptable in committed, but not casual, relationships Additionally, s exua l double standards as a form social surveillance/control over vis everyday life, frequently loosening their grip in the liminoid tourist environment ( Berdychevsky, Gibson, & Poria, 2013b ). In deed, the tourism literature provides examples illustrating a decline in the relevance of sexual double standards on vacation, particularly with respect to casual sex (Eiser & Ford, 1995; Maticka Tyndale et al., 2003). Moreover, some women in the qualitati ve phase reported feeling
169 comfortable having their own condoms on vacation, but not at home. A similar tendency regarding the willingness to carry condoms during tourist experiences was found in another vsky et al., 2013b). Furthermore, in general, different types of tourist experiences are constructed in Western culture as socially tolerated periods for sexual wish fulfillment (Ryan & Kinder, 1996; McKercher & Bauer, 2003) and contra normative settings ( Apostolopoulos et al., 2002). Yet, of course, this is not always the case, and judgmental attitudes of the travelling party can preserve the influence of social control and double standards even in tourist contexts (Ragsdale et al., 2006; Thomas, 2000). In turn, the cultural dimension was related to the lack of familiarity with the local culture, the degree of its openness to various issues related to sexual behavior, the status of women in the local culture, a sense of visibility, lack of awareness to the mannerisms insinuating sexual connotations, and lack of a social, legal, and medical safety network as a foreigner. The combination of this sense of vulnerability and fear of violence (from the physical dimension) can be understood based on the concept of vulnerability s erves as a spatial expression of patriarchy and constrains their leisure experiences (Valentine, 1989). This sense of fear and vulnerability is enhanced by a lack of familiarity and comfort in shoulders for being in the wrong place at the wrong time and, maybe, in the wrong outfit (Bialeschki, 2005; Valentine, 1992). While both the qualitative and quantitative findings reflect such concerns with respect to sexual risk taking in tourism, the qual itative findings also reveal that women discipline themselves
170 accordingly, particularly in foreign spaces, due to the sense that society cannot be changed and, thus, they should adapt themselves and their sexual comportment for the sake of their safety. In deed, the fear of sexual assault and concerns for safety are so ubiquitous and omnipresent in experiences (Henderson, 1991). This could be explained using Sheffiel the actions. Sheffield defined sexual terrorism as a system where fear of violence is used to control and to dominate women, 110). Two of the major characteristics of sexual terrorism are its pervasiveness cutting across socio economic lines and the tendency to blame the victim in the crimes of violence aga inst women. The findings in this study illustrated both the pervasiveness of sexual terrorism (i.e., responsibility for their safety as well as disciplining their behavior in general, and sexual behavior in particular. While in the qualitative phase the social and cultural dimensions of sexual risk taking in tourism were discussed as relatively distinct, they emerged as one factor in the quantitative phase. At least two pote ntial reasons could be suggested for this union. First, conceptually, social norms as the customary rules governing behavior in groups and societies are inextricably linked with cultural conventions in general, and with respect to sexual behavior in partic ular (Bicchieri & Muldoon, 2011; Crawford & Popp, 2003; Lear, 1997). Mutually constructing and feeding on each other, they are like two mirrors placed in front of each other, reflecting one another. Second, statistically, after deleting items for the sake of the goodness of fit for the optimum three
171 factor solution, perhaps, there was not enough variance left in the retained items to extract a separate cultural factor. While the combination of these reasons offers a plausible explanation, further research i n this direction is desirable to bring about more clarity. Emotional and mental dimensions of sexual risk taking in tourism were frequently discussed by the women together or even interchangeably in the qualitative phase. As such, it is not surprising tha t these dimensions were united into one factor in the quantitative phase. Another potential reason for expecting these dimensions to unite stems from the qualitative findings suggesting that both emotional and mental aspects are intricately bound with, and revolve around a sense of guilt, regrets and self questioning, self worth/value/esteem, clash of identities, autonomy, attachment, rejection, etc. Most if not all of these sentiments lend themselves well to the explanation suggested by the concept of self surveillance as an internalized projection of social surveillance (Foucault, 1977 1980 ). In general, the distinction between social and self scrutiny with respect to sexual behavior can be understood using a post structuralist lens of power relations ba sed on the concepts of social and self surveillance (Foucault, 1977, 1980). According to Foucault, power is exercised through social surveillance for the purpose of maintaining discipline in the established status quo. Social surveillance is internalized b y the individuals and transformed into self surveillance permeating and governing all t he aspects of everyday life. Berychevsky et al. (2013b), in their study of behavior into social and self surveillance. Similar to the findings in the current st udy, w hile
172 social surveillance was perceived as diminishing or even vanishing during some tourist experiences, self surveillance was more stable. worth to their patterns of sexual behavior is exactly the purpose of sexual d ouble standards as a form of social surveillance aimed at disciplining nferior social status (Grimshaw, 1993). While the internalized self surveillance somewhat loosens its grip for some women with respect to sexual behavior in tourism (Berdychevsky et al., 2013b), it still asserts itself as a more stable vehicle of control c ompared to social surveillance. Its efficiency stems from the underlying emotional and psychological risks associated with going against social norms as perceived by women. Perceived Dimensions of Sexual Risk Taking in Tourism across the SSS Levels Follow ing the rationale suggested by the sensation seeking theory (Zuckerman, 1994, 2007; Zuckerman et al., 1964; Zuckerman & Kuhlman, 2000), adopted as a theoretical framework for the quantitative phase of this dissertation, differences could be expected betwee n various groups of sensation seekers with respect to their perceptions of sexual risk taking in tourism. In the qualitative findings of this study, however, sensation seeking propensity (or, more generally, the influence of personality on sexual risk taki ng on tourism) was seldomly discussed by the participants, thereby suggesting that among the interviewed women a thrill seeking trait was not a major influence underpinning their sexual risk taking in tourism. This finding somewhat contradicts the body of knowledge on the sensation seeking trait in general (Hoyle et al., 2000; Parsons et al., 2001; Zuckerman & Kuhlman, 2000), and SSS in particular (Kalichman et al., 1994; Kalichman & Rompa, 1995; McCoul & Haslam, 2001), as the correlates or even predictors of sexual risk taking in various contexts.
173 However, at least three caveats should be mentioned in this respect. First, most of the aforementioned literature did not focus on women per se. Perhaps, the lack of attention to the sensation seeking trait in the qualitative phase could be attributed to the influence of gender since women on average typically have a lower sensation seeking propensity compared to men (Zuckerman, 1994, 2007). Second, being a latent personality trait, sensation seeking propensity mig ht be not the most intuitive explanation for the interviewee to identify and talk about in an unstructured, open references to a sense of excitement, thrill, exploration, adrenaline, experimen tation, conquest, and empowerment associated with sexual risk taking in tourism could be explained to a certain degree by their sensation seeking propensity. Indeed, sensation seeking in general and sexual sensation seeking in particular have been found to correlate with sexual risk taking behaviors in various populations and contexts (Gaither & Sellbom, 2003; Greene et al., 2000; Kalichman et al., 1994; Wagner, 2001; Zuckerman et al., 1993). Third, a self selection/volunteer bias in sex research (Dunne et al., 1997 ; Fenton et al., 2001; Wiederman, 1999 ) might play a role in this respect. While sensation seeking propensity of the interviewees is unknown, the literature suggests that volunteers in sex research typically have higher levels of sensation seekin g than non volunteers (Dunne et al., 1997 ; Wiederman, 1999 ). Yet, such a tendency appears to be counterintuitive to the modest amount of attention given by the interviewees to sensation seeking in the qualitative phase. In this study, the results from the quantitative phase reflect moderate levels of NES and SSS traits (Kalichman et al., 1994; Kalichman & Rompa, 1995) with a relatively modest amount of variability in the overall mean scores for each of the scales. As such, the combination of the gender effe ct and these quantitative
174 estimates may serve as an indication for explaining the limited amount of attention the interviewees gave to sensation seeking propensity during the qualitative phase. The role of sensation seeking propensity was more substantial in the quantitative phase investigated with the ESEM approach. Although, t he interpretations of the results based on the ESEM, as a relatively new and exploratory approach, should be cautious (Marsh et al., 2009, 2010, 2011, 2012). The results revealed a r elatively stable factor structure across the lower and higher SSS groups (i.e., supported weak/factorial measurement invariance), suggesting that the composition of the perceived risk dimensions in terms of included items is relatively similar in these gro ups of women. Likewise, the ESEM results pointed to an absence of the differential item functioning across the SSS groups and supported the rationale for comparison of the latent means across the groups (i.e., supported strong factorial/measurement invaria nce). Since there was no support for the strict factorial/measurement invariance (i.e., the item uniquenesses differed across the lower and higher SSS groups), the differences in manifested means should not be interpreted as such a comparison may be flawed due to a disregard of the complex structure of measurement error (Marsh et al., 2009, 2010, 2011, 2012). Therefore, it is the latent means across the groups that should be compared in this study. The latent means were significantly higher for the lower S SS group on the Physical/Sexual Health and Mental/Emotional dimensions compared to the higher SSS group, suggesting that the perceptions of these dimensions differ by SSS propensity. Substantial effect sizes for the differences on these dimensions also sup port this conclusion. Indeed, high sensation seekers tend to downplay the risks involved in order to achieve the levels of desired stimulation (Zuckerman et al., 1964). As such, the sensation seeking trait has been described as a correlate of risk taking i n many contexts in general (Zuckerman, 1994, 2007) and in the context of sexual
175 risk taking in particular (Bancroft et al., 2003, 2004; Cohen & Fromme, 2002; Donohew et al., 2000; Kalichman & Rompa, 1995; Katz et al., 2000). Yet, such a tendency of downpl aying risks among the women in the higher SSS group vs. lower SSS group was not found with respect to the Socio Cultural dimension, suggesting that it is more stable across the groups of women with various SSS levels. Perhaps, the social stigma associated with certain patterns of sexual behavior for women (Eaton & Rose, 2011; Jonason & dimension of sexual risk taking in tourism regardless of their SSS levels. Ther efore, while sexual double standards become somewhat subdued in some tourist contexts vis vis everyday life (Eiser & Ford, 1995; Maticka Tyndale & Herold, 1997; Mewhinney et al., 1995), in a cross sectional comparison between the groups of women with low er vs. higher levels of SSS, sexual double standards assert themselves as a relatively stable form of social surveillance/control over In general, however, the conc lusions about the differences across sensation seeking groups in terms of the perceived dimensions of sexual risk taking in tourism should be drawn cautiously. At least two nuances could provide support for this cautiousness. First, restricting latent fact or means to invariance across the SSS groups did not decrease substantially the goodness of fit indices for the model with three factors. Second, the composition of the sample in the quantitative phase of this study might play a role. The sample characteri stics revealed moderate levels of SSS and NES traits across the participants with modest standard deviations. Had the sample been more heterogeneous with respect to SSS and NES, the results of the ESEM may have been different.
176 al Risk Taking in Tourism It has been proposed that the social atmosphere in many types of tourist experiences is conducive to sexual exploration, serving as a liminoid sanctuary and cathartic break (Lett, 1983; Ryan & Kinder, 1996; Shields, 1990), a heter otopia and contra normative setting (Andriotis, comportment in tourism in various ways, often encouraging experimentation, risk taking, and resistance to social stereot ypes of appropriate sexual behavior for women (Berdychevsky et al., for sexual risk taking in tourism require clarity. Hence, while the focus in the previous sec tions of this discussion was on the perceived risk dimensions, this section switches to another aspect of taking in tourism despite of (and, occasionally, due to) the ris ks involved. Both the qualitative and quantitative findings in this study reveal a variety of motivations taking in tourism environments. The themes of anonymity, detachment from everyday rules and social gaze, exploration and opportunity to enact sexual fantasies, invincibility and loss of inhibitions, vacation mentality and fun were portrayed by the women as motivations for sexual risk taking, both in the qualitative and quantitative phases. Interestingly, sometimes, risk itself was described as a motivation, which is surprising since Zuckerman (1994, 2007) argued that risk is not a motivation in itself, but is a price that people are willing to pay (to different degrees) to achieve desired level of arousal. Another motivat ion for women in the qualitative phase was constructing sexual risk taking in tourism as adventure. Indeed, s ex in tourism has been associated with adventure in the tourism literature because tourism has been characterized by the primacy of the senses and supremacy of enjoyment, as well as the promise of sexual adventure (Littlewood, 2002; Selnniemi, 2003). Sexual escapades
177 have been found to be part of the adventurous ethos in some tourist experiences where sex is perceived as making the vacation and cele brating the moment (Berdychevsky et al., 2010; Wickens & Snmez, 2007). for sexual risk taking in tourism could be grouped under three dimensions: (1) anonymity/ detach ment/experimentation, (2) safe exploration/thrill/empowerment, and (3) fun/less inhibitions/opportunity. Interestingly, some of these motivations are reflected in the elements of the four sensation seeking dimensions developed by Zuckerman et al. (1978). S pecifically, motivations such as exploration, experimentation, and thrill are relevant to the dimensions of thrill and adventure seeking, as well as experience seeking. Additionally, losing inhibitions in tourism as a motivation for sexual risk taking in t ourism is relevant to the disinhibition dimension of sensation seeking. Finally, the women in the qualitative phase frequently sexual risk taking in tourism, whic h could be conceptualized as associated with impulsive sensation seeking (Zuckerman et al., 1993). This could be a reasonable explanation since impulsive sensation seeking was found to be related to risk taking in general and sexual risk taking in particul ar (Smith 2007; Turchik, 2007; Zuckerman & Kuhlman, 2000). In turn, the framework for understanding risk in tourism based on the components of context, likelihood, and consequence (Elms, 1998; Ryan, 2003) is instrumental for grasping the interaction betwee sexual risk taking in tourism. First, the uniqueness of the touristic context as a motivation for taking is reflected in the components of anonymity, displacement fr om everyday life and detachment from everyday rules, fun oriented vacation mentality, and
178 loosening inhibitions. Second, likelihood refers to: (1) opportunity for sexual risk taking in tourism, which was perceived by the participants as increased compared to everyday life; and (2) probability of negative consequences that was perceptually diminished by the sense of invincibility in the ephemeral touristic context. Third, the lack of perceived consequences in the motivations for sexual risk taking in tourism refers to the sense of safe exploration, thrill, empowerment, sense of conquest, and fun. Below is a more extensive explanation for each of the components. taking in tourism when they can offer anonymity (Berdychevsky et al., 2013a, Ragsdale et al., 2006; Thomas, 2005; Wickens, 1997). Indeed, the qualitative findings in this study portray perceived anonymity as one of the crucial structures helping to diminish the importance of the social d imension of sexual risk taking in tourism. Likewise, anonymity as a motive was rated relatively high by the participants in the quantitative stage. This certainly concurs with existing studies in that the importance and liberating effect of a sense of anon ymity on sexual behavior has been discussed in the tourism literature (Pritchard & Morgan, 2006; Shields, 1990; Wickens, 1997). Anonymity in tourism has been conceptualized as fostering sexual exploration (Bauer & McKercher, 2003), diminishing the sense o f social surveillance and contributing to sexual experimentation (Berdychevsly et al., 2013b), underlying the sense of situational disinhibition (Ford & Eiser, 1995), contributing to the temporary approval of casual sex (Mewhinney et al., 1995), and facili tating sexual risk taking, promiscuity, and increased rates of casual sex (Apostolopoulos et al., 2002; Black, 1997; Maticka Tyndale et al., 2003; Thomas, 2005). Moreover, the literature suggests that the impact of anonymous tourism environments on sexual behavior is more influential for women compared to men (Eiser & Ford, 1995; Mewhinney et al.,
179 vacation vis vis everyday life (Berdychevsky et al., 2013a; Maticka Tynd ale & Herold, 1997; Poria, 2006; Thomas, 2000, 2005). However, the qualitative findings also revealed that in the contemporary world with the pervasiveness of technology and saturation of social media, anonymity is often imperfect. Thus, sometimes, the ill usion of anonymity encouraged the women to become involved in sexual behaviors that eventually haunted them upon returning home, or even before returning home, through technologically mediated channels, such as the internet. These findings illustrate the i ncreasing spillover between tourism and everyday life contexts, contributing to the argument positioning tourist experiences as a part of a rather than a liminoid cathartic break (Lett, 1983; Turner, 1974). However, conclusions should be drawn cautiously since the liberating effect of anonymity is conducive to the detachment from everyday rules, routine responsibilities, and rd & Morgan, 2006; Ryan, 2003). Many women in both phases of this study found the sense of being betwixt and between the social worlds of home and the tourist destination as motivating and instrumental for sexual exploration and risk taking. This feeling o f displacement and detachment lends itself well to the explanation offered by the concept of liminoid (Turner, 1974) as a state of transition where an individual is temporarily detached from everyday social rules, and yet not integrated into another social order. Various tourist experiences have been found to offer liminoid contexts where people can fulfill sexual fantasies and express repressed feelings and desires (Lett, 1983; Ryan & Kinder, 1996). In general, the anonymous social atmosphere of tourism h as been discussed as liminal and
180 Liminoid tourist experience s can be understood as a fourfold tra nsition conducive to detachment in terms of space and time as well as mental and sensual states (Selnniemi, 2003). Drawing upon the qualitative findings, the perception of space shifted from the arena of social surveillance in everyday life to the more an onymous playground in tourism (Lett, 1983) Additionally, the perception of time was distorted by the previously referenced time compression effect conducive to sexual risk taking (Thomas, 2005). In turn, these perceptual changes with respect to space and time Both the qualitative and quantitative findings provide evidence that a combination of anonymity and detachmen t was also associated with loosening inhibitions with respect to sexual behavior on vacation among the women. Disinhibition as one of the four sensation seeking dimensions was conceptualized by Zuckerman et al. (1978) as reflecting seeking sensations throu gh indulgence in a hedonistic lifestyle involving wild parties, drinking, sexual variety, etc. Likewise, the situational disinhibition effect is intricately related to sexual experimentation and risk taking in tourism (Apostolopoulos et al., 2002; Berdych evsky et al., 2013b; Ford & Eiser, 1995). Under conditions of situational disinhibition, an individual might feel a different person, often, a more liberated one, as well as less responsible for any transgressive actions. This sentiment was frequently desc ribed in the qualitative findings and loosening inhibitions as well detachment from everyday social norms were relatively highly rated as motives for sexual risk taking in the quantitative phase.
181 In turn, these aforementioned contextual characteristics no t only increased the perceived opportunity for sexual risk taking in tourism, but they also decreased the perceived probability of negative consequences. Particularly, the sense of situational disinhibition (Apostolopoulos et al., 2002; Ford & Eiser, 1995) was associated by the women in both phases in this study with a certain sense of invincibility and a sentiment that unsuccessful sexual experiences in tourism can be left behind taking on vacation does stay on vacation and has no consequences beyond it. Indeed, the explanations of literature (Maticka Tyndale et al., 1998; Thomas, 2005; Yeoman, 2008). This arrangement appears to be facilitated by the fact that people tend to leave at home some of the ir social mental baggage when they travel (McKercher & Bauer, 2003). Specifically fo r women, tourism can offer a self exploration unavailable at home (Berdychevsky et al., 2013a; Ragsdale et al., 2006; Thomas, 2005). Moreover, some women navigate around the mental weight t hat they would typically assign to sexual escapades using a strategy of mentally leaving it behind and/or not taking it personally (Berdychevsky et al., 2013b). However, the quantitative results suggest that this illusion of invincibility acts as a motivat ion for sexual risk taking in tourism for some women, but not for others. Indeed, a sense of invincibility and immortality is also a characteristic inherent in young adulthood and it is associated with risk taking, thrill seeking, and the perceptual downpl aying of risk in general (Levinson, 1996) and in tourism in particular (Gibson, 1996; Gibson & Yiannakis, 2002).
182 Some women in both phases in this study were motivated by the expected positive outcomes of sexual risk taking in tourism, reflecting the compo nent of consequences in the adopted framework to understanding risk in tourism comprised of the context, likelihood, and consequences (Elms, 1998; Ryan, 2003). Such outcomes from both research phases included excitement/thrill, sense of exploration, empowe rment, curiosity, self challenge, conquest, fun, and collecting memories/stories. Tourism has indeed been explored as a context for sexual wish fulfillment (Ryan & Kinder, 1996). Furthermore, empowerment has been found to be an outcome of reversing sexual roles in tourism as well as resisting social stereotypes of appropriate sexual behavior for women (Berdychevsky et al., 2013b). Finally, many of the aforementioned benefits have also been delineated in the literature about female solo travelling as an eman cipatory practice (Jordan & Gibson, 2005; Obenour, 2005). Linking these outcomes and the emancipatory taking in tourism poses a challenge to the construction of female sexuality as passi ve, subordinate, or inferior ( Grimshaw, 1993 ; & Holland 1 993 surveillance (Foucault, 1977) with respect to their sexual behavior in tourism (discussed earlier as underpinning the mental/emotional dimensi on of sexual risk taking) and the fact that this challenge occurs in a liminoid setting undermine its potential for social transformation. Yet, the sites of resistance are often spontaneous, sporadic, and local, but the accumulating effect of the localized resistances may eventually accrue into a challenge to the existing status quo (Jordan & Gibson, 2005; McLaren, 2002). Additionally, some women in the qualitative phase emphasized the search for romance as a motivation for sexual risk taking in tourism, w hich resonates with the motivations described in the literature with respect to romance tourism (Dahles & Bras, 1999; Jefreys, 2003; Pruitt &
183 Lafont, 1995). However, the difference is that in this study the women reported sexual encounters that did not inv olve financial transaction and role reversals in terms of power differentials based on economic privilege or racial background. Finally, the qualitative findings in this study suggest that sex and sexual risk taking in themselves can serve as a motivation to partake in some tourist experiences. The role of sex as a motive for tourist experiences has been reiterated in the context of commercial sex tourism and romance tourism (Herold et al., 2001; McKercher & Bauer, 2003; Oppermann, 1998, 1999; Pruitt & LaF ont, 1995). This study shed additional light on sex as a motive in the context of non commercial sex in tourism (Carter & Clift, 2000). However, the importance of non commercial sex as a motive for going on vacation depends on the type of the tourist exper ience and the type of the sexual relationship involved (Berdychevsky et al., 2013a). Clusters and Profiles of Sexual Risk Takers in Tourism among Women The idea that various types of sexual risk takers in tourism exist was evident in the ideographic stage of the qualitative analysis as taking patterns, perceptions, and motivations varied Yet, no generalizable conclusions could be made based on the number of the participants in the qualitative phase. Conversely, the quantitative phase allowed this issue to be addressed by clustering the women based on their perceived dimensions of sexual risk taking and motivations for it. This procedure identif ied five clusters: (1) h igh and diverse risk perceivers motivated by anonymity and exploration; (2) m oderate and diverse risk perceivers motivated by fun; (3) low risk perceivers with diverse motivations; (4) low risk perceivers motivated by exploration; and (5) unmotivated high and diverse risk perceivers. In terms of profiling the clusters, both indice s of sensation seeking trait i.e., NES and SSS (Kalichman et a., 1994; Kalichman & Rompa, 1995), reflected significant differences among the clusters. As for sexual characteristics, the number of sexual partners as well as
184 previously being tested for STD s and having had sexual practices putting the participant at risk for HIV/AIDS distinguished significantly among the clusters. These indicators were frequently used in the literature on sensation seeking as the proxies for sexual experience/activeness and sexual risk taking (Bancroft et al., 2003, 2004; Donohew et al., 2000; Greene et al., 2000; Hoyle et al., 2000; McCoul & Haslam, 2001; Zuckerman & Kuhlman, 2000). In this study, these sexual everyday sexual practices contextualizing their sexual risk perceptions and motivations in tourism. Additionally, it is important to mention that age did not play a significant role since most of the sample, as expected, was in the young adulthood stage ( n = 814, 95.9%), with the vast majority in the novice young adulthood phase ( n = 755, 88.9%), reflecting the aspiration to focus on the life stage of peak thrill seeking and risk taking in general and in tourism in particular (Gibson, 1996; Gibson & Yianna kis, 2002; Levinson, 1996). Therefore, the brief profiles of the clusters were constructed based on sensation seeking propensity and sexual characteristics. C luster one, the high and diverse risk perceivers motivated by anonymity and exploration while hav ing substantial proportions of the participants with no or small numbers of sexual partners, relative to other clusters, had the highest proportion of women having had eight to ten sexual partners, with almost half of the members having been tested for STD s and the second highest proportion of the participants with sexual practices involving risk for contracting HIV/AIDS. The high and diverse risk perceivers motivated by anonymity and exploration also had the highest NES and the second highest SSS average l evels. Their relatively high scores on the perceived risk dimensions as well as NES and SSS challenge the idea that high sensation seekers tend to underestimate the risks (Zuckerman 2007; Zuckerman et al., 1964). The propensity of the women in this cluster to be both high risk perceivers and high
185 sensation seekers suggests that perceived risks, even if not downplayed, do not necessarily serve to deter women from sexual risk taking. Additionally, their highest susceptibility (relative to the other clusters ) to anonymity, detachment, exploration, thrill, and empowerment (Apostolopoulos et al., 2002; Berdychevsky et al., 2013b; Eiser & Ford, 1995) as the motivating forces for sexual taking in tourism may put the high and diverse risk perceivers motivated by a nonymity and exploration at increased risk for negative consequences that they are so vulnerable to (based on their high perceptions of the dimensions of sexual risk taking in tourism). In other words, if anything goes wrong during a sexual risky activity in tourism, the high and diverse risk perceivers motivated by anonymity and exploration believed that they are the most likely to get hurt physically, mentally, and socially. Cluster two, the m oderate and diverse risk perceivers motivated by fun had the highest proportions of women who had one sexual partner and five to seven partners. Half of the members in this cluster had been tested for STDs. Relative to the other clusters, it had the third highest proportion of women with sexual practices potentially risky for HIV/AIDS. Women in this cluster had a moderate NES, but among the lowest SSS levels compared to other clusters. The m oderate and diverse risk perceivers motivated by fun had relatively high perceptions of the physical dimension of sexual risk ta king in tourism. Yet, their perceptions with respect to the mental/emotional and socio cultural dimensions were only moderate. Plus, with the exception of the fun/less inhibitions/ opportunity as a motivation for sexual risk taking in tourism, the rest of the touristic characteristics had a relatively low motivational impact on the women in this cluster. Thus, the m oderate and diverse risk perceivers motivated by fun generally took more of an opportunistic approach to situational disinhibition in tourism f or the sake of fun (Apostolopoulos et al., 2002; Eiser & Ford, 1995). On the one hand, the m oderate and diverse
186 risk perceivers motivated by fun (cluster two) might be less at risk than the high and diverse risk perceivers motivated by anonymity and explor ation (cluster one) since they reported lower motivations for sexual risk taking in tourism and they were less concerned about the mental/ emotional and socio cultural dimensions of risk. On the other hand, Ragsdale et al. (2006) argued that women with an opportunistic approach to sexual risk taking in tourism are actually at a higher risk (particularly, in terms of sexual health) as they are not prepared for sexual encounters. Cluster three, the l ow risk perceivers with diverse motivations while having s ubstantial proportions of the women with no or a small number of sexual partners, relative to the other clusters, had by far the highest proportion of women having had ten to fifteen sexual partners. The l ow risk perceivers with diverse motivations had the highest proportion of women having been tested for STDs and having engaged in sexual practices potentially putting them at risk for contracting HIV/AIDS. The l ow risk perceivers with diverse motivations also had the highest levels of NES and SSS. These re latively sexually active women with a history of sexual risk taking and the highest sensation seeking propensity tended to disregard the dimensions of sexual risk taking in tourism (with a small caveat for the physical dimension) and were moderately motiva ted by all that tourist experiences have to offer. For the l ow risk perceivers with diverse motivations high scores on sensation seeking indices were significantly associated with downplaying the mental/emotional and socio cultural dimensions of sexual r isk taking in tourism, which is consistent with the literature on sensation seeking (Zuckerman, 1994, 2007; Zuckerman et al., 1964). The l ow risk perceivers with diverse motivations might also be at risk, but for different reasons than the previous two clu sters. They appear to be motivated to various degrees by all three motivational factors and, as such, the sense
187 of anonymity, detachment, situational disinhibition, and excitement (Apostolopoulos et al., 2002; Bauer & McKercher, 2003; Ford & Eiser, 1995; T homas, 2005) may magnify their tendency to downplay the risks. This, in turn, may augment the illusion of invincibility which is already in its zenith at this young age based on a life span perspective (Gibson, 1996; Gibson & Yiannakis, 2002; Levinson, 199 6). These potential effects in tourism may encourage women in this cluster to get involved in sexual behaviors that may eventually turn out to be beyond their comfort level. Yet, their lack of susceptibility to perceived risks (particularly, mental/emotion al and socio cultural) might lend them certain psychological armor. In other words, not believing in mental/ emotional and socio cultural dimensions of sexual risk taking in tourism might make the l ow risk perceivers with diverse motivations somewhat imper vious to the influence of these risks. Cluster four, the l ow risk perceivers motivated by exploration had the second highest proportion of virgin women, relative to the other clusters, and had substantial proportions of women who had one partner or three to four partners (the highest proportion among the clusters for the latter). Additionally, this cluster had the second lowest proportions of women who had been tested for STDs and had experienced sexual practices putting them at risk for HIV/AIDS. Likewise the l ow risk perceivers motivated by exploration had a moderate NES and relatively low SSS levels compared to other clusters. This relatively less active group of women in terms of sex and sexual risk had low perceptions with respect to all the dimensio ns of sexual risk taking in tourism and they were only motivated by one factor, that of safe exploration/thrill/empowerment. Perhaps, the argument about the sense of invincibility in tourism could be applied to this cluster to some extent as well. Yet, the l ow risk perceivers motivated by exploration did not score high on sensation seeking and were susceptible only to one motivational factor in tourism. For this cluster, the sense of
188 exploration and empowerment related to sexual risk taking in tourism (Berd ychevsky et al., 2013b) motivated them to take part in these experiences. Furthermore, similar to the l ow risk perceivers with diverse motivations (cluster three), the lack of sensitivity of the l ow risk perceivers motivated by exploration (cluster four) t o various risks involved might render them impervious to mental/emotional and socio cultural dimensions. However, such a defense mechanism might be dubious with respect to the physical dimension of sexual risk taking in tourism since ignoring it is not lik ely to make it disappear. Cluster five, the u nmotivated high and diverse risk perceivers had by far the highest proportion of virgin women. Likewise, this cluster had by far the lowest proportion of members who had been tested for STDs and engaged in sex ual practices putting them at risk for HIV/AIDS. This cluster also had the lowest average NES and SSS scores compared to the other clusters. Apparently, the u nmotivated high and diverse risk perceivers seem to be the least likely to engage in sexual risk taking in tourism since they appear to be neither susceptible to the motivating potential of touristic characteristics nor do they tend to downplay the various risks involved. The lowest (relative to other clusters) sensation seeking propensity of the u nmo tivated high and diverse risk perceivers may explain their risk aversion and lack of motivation since sensation seeking tendency has been found to positively correlate with, and sometimes even to predict, sexual risk taking (Bancroft et al., 2003, 2004; D onohew et al., 2000; Greene et al., 2000; Hoyle et al., 2000; McCoul & Haslam, 2001; Zuckerman & Kuhlman, 2000). Finally, since all the cluster variates in this study were latent constructs (i.e., perceived dimensions of and motivations for sexual risk ta king in tourism), profiling the clusters in terms of the observable and/or easily obtainable socio demographic characteristics is essential for the sake of the future segmentation of the participants whose perceptions and motivations might be
189 still unknown (Hair et al., 2010; Mooi & Sarstedt, 2011). This would also be instrumental for offering practical implications for information campaigns and health programs aimed at taking in tourism. In this study, these profiles were developed from a purely exploratory standpoint to identify the characteristics distinguish ing among the clusters of sexual risk takers in tourism to provide some directions for future research. In terms of the socio demographic characteristic s, racial or ethnic background, place of birth and growing up, and personal income differed significantly among the clusters of sexual risk takers. Race has received some research attention in the context of sexual risk taking and sexual health, typically portraying Black people and especially Black adolescents as being at particularly high risk (Broman, 2007; Chng & Gliga Vargas, 2000; Fasula, Miller, & Wiener, 2010; Miller, Farrell, Barnes, Melnick, & Sabo, 2005; Varga, 1997). In this study, the proport ions of Black women were the highest in the clusters of low risk perceivers (i.e., clusters three and four), but the numeric differences were not dramatic for this group. Moreover, the literature on sensation seeking suggests that, typically, no race diffe rences were found on the disinhibition dimension of sensation seeking, which is the most relevant dimension for sexual behavior, drinking, partying, and hedonistic lifestyles (Zuckerman, 1994, 2007). Furthermore, no literature was found on the role of race with respect to sexual risk taking for other racial minorities. Place of birth and growing up is a proxy for nationality, which has been found to play a significant role in the relationship between risk taking, sensation seeking, and tourist behavior amo ng young adults (Pizam et al., 2004). In this study, relative to other clusters, the low risk perceivers motivated by exploration (cluster four) had a substantially higher proportion of non
190 roportion of American respondents scoring high on risk taking and sensation seeking was among the highest compared based on a mixed gender sample, while this s tudy has focused on women only, and women typically score lower on sensation seeking than men (Roberti, 2004; Zuckerman, 1994, 2007; Zuckerman et al., 1978). Finally, personal income is one of the proxies for social class, which has received limited atten tion in the context of sexual risk and is problematic since sexual risk taking is intricately Social class is terms of sexual appetites as well as their biological vs. socially constructed origins, rendering various degrees of legitimation and justification for sexual risk taking depending on these views ( Higgins & Browne 2008) Yet, socioeconomic factors typical ly have a relatively weak effect on sensation seeking propensity (Zuckerman, 2007). In this research, the women in the low risk perceivers with diverse motivations cluster were on average the most economically affluent, while the women in the low risk perc eivers motivated by exploration cluster were the least economically affluent The current state of the literature does not appear to offer evident explanations to the differences between the clusters in terms of the aforementioned socio demographic charact eristics. Hence, future research is needed to shed substantially more light on these socio demographic characteristics in the context of sexual risk taking in tourism. Facilitating and Inhibiting Characteristics of Sexual Risk Taking in Tourism There is a lot of congruence between the qualitative and quantitative findings in this respect, but the quantitative results offer additional insights, particularly, with respect to the cluster specific perceptions of the characteristics that facilitate or inhibit s exual risk taking in
191 tourism. Most of the characteristics could have both facilitating and inhibiting effects depending on the circumstances, individual, and the type of the sexual encounter, which was also the case in Alcohol consumption in general, and underage drinking in particular, was a universal taking in tourism in the qualitative phase of this study. I t is important to mention that sexual risk taking in adolescence frequently occurs in conjunction with alcohol consumption ( Brookmeyer & Henrich, 2009; Cho & Span, 2010; Wade & Heldman, 2012). The findings also emphasized the expectancy effect, as opposed to the pharmacological effect (Cho & Span, 2010), providing a psychological excuse and disinhibiting effect. This expectancy effect is partially alleviating responsibility for sexual ouble standards (Eaton & Rose, 2011; Jonason & Fisher, 2009; McCabe et al., 2010). risk taking in the quantitative phase. This tends to confirm the triplex of alcohol, s ex, and tourism that has been widely highlighted in the literature (Apostolopoulos et al., 2002; Berdychevsky et al., 2010; Diken & Laustsen, 2004; Josiam et al., 1998; Maticka Tyndale & Herold, 1997; Maticka Tyndale et al., 2003; Mewhinney et al., 1995; T homas, 2000). T he relationship between sex and alcohol in tourism has been described as curvilinear, in a sense that it facilitat es sexual behavior up to a certain amount of consumed alcohol, yet, inhibit s or even pre cludes it if the amount of consumed alc ohol is too high ( Berdychevsky et al., 2010; Thomas, 2000) Conversely the qualitative findings in this research suggest a linear relationship between alcohol and sexual risk taking since the loss of control associated with intensive alcohol consumption m ay expos e women to all the sexual risk taking dimensions.
192 Alcohol was also closely related to other touristic characteristics that the participants in the quantitative phase highly rated as facilitating their sexual risk taking in tourism : the party scene, detachment from everyday norms, and being in a break loose mood. Likewise, the women in the qualitative phase described a combination of alcohol, nightlife, and letting their hair down as contributing to sexual risk taking in tourism. The detaching and di sinhibiting effects of the transient, liminoid, touristic environments (Eiser & Ford, 1995; Lett, 1983; Ryan & Kinder, 1996; Turner, 1974) have been discussed in this dissertation with respect to the motivations fo r sexual risk taking. The p arty scene and break loose mood are integral features of the social mytho logy about spring break facilitating sexual risk taking (Josiam et al., 1998; Maticka Tyndale & Herold, 1997), which has also been frequently raised in this study. Yet, these characteristics do not appear to be totally unique to spring break as tourism in general is tourism hedonism [is] enjoyed on a massive scale well as the pursuit of transgression and unlimited enjoyment ( Dicken & Laust s en, 2004 p. 99). Likewise, in a study offering a typology of leisure based tourist roles, the action seeker emerged as a distinct role associated with partying, night clubs, and uncomplicated sexual experiences (Yiannakis & Gibson, 1992). Another facilit ating characteristic discussed by women in the qualitative phase and highly rated as such in the quantitative phase was having lots of free leisure time. Therefore, tourism as an ultimate form of leisure is conducive to sexual risk taking due to one of its fundamental
193 and leisure are one of the axes for defining tourist experience based on the theoretical framework chosen for this study (Poria et al., 2003). Mor tourist experiences as leisure vs. non leisure were found to affect their sexual behavior, with leisure based experiences being more conducive to various changes in sexual behavior in tourism vis vis every day life ( Berdychevsky et al., 2013a ). Despite the multiplicity of the definitions of leisure, the concepts of freedom, choice, intrinsic value, and playfulness are the cornerstones of leisure meaning (Kelly, 1987). Such leisure qualities explain the poten tial of leisure experiences behavior. However, the concept of freedom in leisure is relative, especially for women (Wimbush & Talbot, 1988), which is reflected self surveillance exercised with respect to their sexual behavior as a relatively stable mechanism of social control (Foucault, 1977). where revealing clothing is appropriate were also rated highly as facilitators of sexual risk taking in tourism. Increased sexual confidence and self awareness in terms of sensuality and sexuality are related to sexual experimentation and the courage to go beyond the perceived boundaries in tourism for the sake of excitement and self empowerment among women (Berdychevsky et al., 2013a, 2013b). Yet, different transgressions might also involve various risks. As for the appropriateness of revealing clothing, i t brings the ideas of sexual body and the social gaze to the forefront (Jordan, 2007) The concept of the body can be approached broadly involving aspects of its physicality, sensuality, and appearance (Small, 2007; Veijola & Jokinen, 1994). The role of th e body in tourist experiences until recently was relatively ignored in the tourism literature (Pritchard, Morgan, Ateljevic, & Harris, 2007; Small, 2007). Yet, embodiment plays an essential
194 role in tourism reflecting emotional and sensual aspects of touris t experiences (Ateljevic & Hall, 2007; St John, 2001), particularly, when it comes to sexual behavior as indicated in this study. All these facilitating characteristics were perceived differently by the various clusters of sexual risk takers found in this study. For instance, the high and diverse risk perceivers motivated by anonymity and exploration (cluster one ) and the low risk perceivers with diverse motivations (cluster three) systematically rated all of the facilitating characteristics the highest. Th taking were consistent with their high perceptions of the motivating potential of anonymous, liminoid, and disinhibiting touristic environments (Dicken & Laustsen, 200 4; Eiser & Ford, 1995; Lett, 1983). Likewise, the women in these clusters also ranked the highest on sensation seeking T his may help to explain their susceptibility to the facilitating characteristics of sexual risk taking in tourism as sensation seeking has been frequently confirmed to be related to sexual risk taking (Chng & Gliga Vargas, 2000; Gaither & Sellbom, 2003; Hoyle et al., 2000; Zuckerman & Kuhlman, 2000). The rest of the clusters had lower sensation seeking propensity, particularly in te rms of their average SSS scores (Kalichman et al., 1994; Kalichman & Rompa, 1995), which might explain their lower susceptibility to the facilitating potential of touristic environments. The m oderate and diverse risk perceivers motivated by fun (cluster tw o) were not too far behind the previous two clusters in terms their susceptibility to the facilitating characteristics. However, the l ow risk perceivers motivated by exploration (cluster four) systematically scored the lowest on all the potentially facilit ating characteristics. Likewise, the u nmotivated high and diverse risk perceivers (cluster five) scored as low as the l ow risk perceivers motivated by exploration on the
195 effect of having leisure time and appropriateness of revealing clothing, but they sco red higher on everything else. Indeed, the u nmotivated high and diverse risk perceivers had relatively high scores on alcohol (this one particularly stands out), party scene, and break loose mood as facilitators (Apostolopoulos et al., 2002; Josiam et al ., 1998; Maticka Tyndale et al., 2003; Thomas, 2000) which is inconsistent with their reported unmotivated state and might suggest that still waters run deep. Perhaps, admitting the facilitating capacity of the touristic contexts is easier than recogniz ing their motivational potential for sexual risk taking in tourism. This discrepancy and, particularly, the highly rated facilitating role of alcohol can be explained by the expectancy effect (Cho & Span, 2010), suggesting that alcohol consumption, party s cene, and break loose mood may offer a disinhibiting psychological excuse for sexual risk taking in tourism. (as discussed earlier) and their vulnerability to all three risk dimensi ons may put them at increased risk. The inhibiting factors were considerably less prominent than the facilitators, both in the qualitative and quantitative phases, which may reflect the conduciveness of tourist environments taking i n tourism. Among the reported inhibiting factors in the quantitative phase, the women rated having a steady relationship at home, having judgmental trip partners, and being in a country with strict religious beliefs as the most influential. The inhibiting effect of commitments associated with steady relationships at home was raised in the literature perceived by women as something beyond the boundaries that they would be willing to transgress.
196 Additionally, the results of this study support the idea that the characteristics of the 2010; Ragsdale et al., 2006), as well as on the p erceptions of various dimensions of sexual risk taking in tourism. Having judgmental trip companions brings back the element of social control into tourism environments as well as diminishes a sense of anonymity and its disinhibiting effect. Conversely, ha taking in tourism. Indeed, direct and indirect peer support and/or pressure to participate in a sexual environment was found to be an integral part of the sexual script on sprin g break ( Maticka Tyndale & Herold, 1997; Maticka Tyndale et al., 2003; Mewhinney et al., 1995) The findings in this study show that this tendency might be more universal and apply to a variety of tourist experiences. These influences have been found to af as well as the planning of the tourist experience Indeed, women anticipating casual sexual adventure/s on vacation tend to either choose travelling alone, or with a very few supportive female friends (Berdychevsky e t al., 2010; Ragsdale et al., 2006). Interestingly, having judgmental trip companions was rated as more inhibiting than being in a country with strict religious beliefs, suggesting that the opinion of an immediate social circle is more important to women. Perhaps, Tyndale et al., 1998; Thomas, 2005; Yeoman, 2008), while the judgment of their immediate social circle might haunt women at home through traveling companions or social media. This disciplining effect of the immediate social circle illustrates the dynamic interaction and sometimes even a blurring of the boundaries between tourism contexts and everyday life (Larsen, 2008; Uriely, 2010).
197 ow risk perceivers motivated by exploration (cluster four) seemed to be by far the least inhibited by all three factors. ions with respect to sexual risk taking in tourism were consistent with their relative disregard of the inhibiting factors of sexual risk taking in tourism. Conversely, the u nmotivated high and diverse risk perceivers (cluster five) appeared to be among t he most inhibited by all three factors. Additionally, the l ow risk perceivers with diverse motivations (cluster three) seemed to be as inhibited as cluster five by the strict religious beliefs at the destination and a steady relationship at home. Likewise, the m oderate and diverse risk perceivers motivated by fun (cluster two) appeared to be about as inhibited as cluster five by the judgmental trip partners and a steady relationship at home. In contrast to the facilitating factors, the average levels of sex ual and non sexual sensation seeking propensity (Kalichman et al., 1994; Kalichman to the influence of the inhibiting factors. Perhaps, further research is needed to shed additional light on this issue. Limitation s and Delimitation s Among the potential limitations in the qualitative phase it is important to mention the possibility of social desirabil ity and memory decay affecting the data Social desirability was h opefully minimized by establishing rapport and guaranteeing confidentiality to the research participants. Memory decay was hopefully minimized by probing and subsequent member check where participants were invited to eliminate or to add any data they found appropriate Additionally, the qualitative sample was homogenous in terms of gender, age, and university phenomenological approach searching for commonalities be at the nomothetic level. Additionally, in order to minimize the limitations associated with the
198 quality of the data analysis and interpretations, the rigorousness of the analysis was enhanced through adhering to the cann ons of trustworthiness in qualitative research i.e., credibility, transferability, dependability, and confirmability (Decrop, 2004; Lincoln & Guba, 1985), as discussed in Chapter 3. As for the quantitative phase the anonymity and confidentiality associ ated with online surveys hopefully minimize d th e s ocial desirability effect. Additionally, use of unintimidating wording of the questions in the study of such a sensitive topic is particularly important (Fenton, Johnson, McManus, & Erens, 2001). This pot ential limitation was addressed via developing the items based on the qualitative findings and revising several versions of the instrument following Alth ough, working with a previously untested scale also gives rise to another potential limitation in that the scale was not underpinned by a recognized theory, having been derived based on inductive findings this possible threat is somewhat diminished Thi nking further about the limitations and delimitations, it is important to consider all four types of error: measurement error coverage error, sampling error, and non response error (Dillman et al. 2009). Dillman et al. argue d that conducting quality self administered surveys implies holding error levels of all four types low simultaneously. M easurement error was hopefully, minimized via adherence to sound analytical procedures described earlier. Coverage error occurs if any unit in the survey population does not have a known, nonzero chance of being included in the sample. This issue was addressed by implementing probability sampling. Sampling error occurs due to collecting the data from only a subset of the members in the sampling frame. This issue was a ddressed by obtaining an up to date sampling frame of the
199 students registered in the UF in spring 2013 and adhering to sound sampling procedures that were described earlier. Additionally, h igh non response rate increases the sampling error. Thus, the pote ntial participants receive d two rounds of follow up email s to encourage their participation and they were guaranteed confidentiality. Yet, the response rate was still quite modest. Another type of error is non response error, which is not a direct function of low response rates. Non response error results from the differences between those who do not return the survey and those who do. This type of error is directly related to delimitations related to generalizability. To verify this potential source of err or, the demographic characteristics of the sample were compared to the UF official data on the demographics of the UF student body (UF Annual Accountability Report, 2010 2011 ; UF Admissions, 2013 ). The comparison revealed that the sample reflects well the UF female student body in terms of the class standing (undergraduate vs. graduate) and the proportion of international students. The sample also adequately reflects the population in terms of the racial or ethnic background (with rather similar proportions of Black and Hispanic women, but a slight over representation of Caucasian women and substantial under representation for Finally, it should be mentioned that the results are delimited to young, college educated, In this respect, investigations of the differences between participants and non participants in existing studies of sexual behavior indicate that the former might not represent the later, since people who consent to participate in sex related research are typically less religious ; have less conservative sexual attitudes ; are more sexually experienced and sexually confident; more likely to smoke cigarettes and to con sume alcohol ; are more novelty seeking sensation seeking, and reward
200 dependent as well as less harm avoidant (Dunne et al., 1997 ; Fenton et al., 2001; Wiederman, 1999 ). While no such estimates are available for the target population to conduct a comparis on with a sample, it is important to mention that the women in this study were rather diverse in terms of the levels of sexual experience and had moderate levels of sexual and non sexual sensation seeking propensities. To conclude, based on this informatio n and the comparison of the available socio demographic characteristics between the target population and the sample, it appears that the results of the quantitative phase in this study could be generalized at least to other samples drawn from the UF femal e student body and, perhaps, to other female populations in similar Southeastern US Universities as well. Recommendations for Future Research and Practical Implications There are several directions for future studies suggested by this study. For instance, a risk taking in tourism. Considering the effects of sexual double standards, the results for men might be different in terms of the perceived risks and/or the role of tourist experiences as a motivational trigger for sexual experimentation. Additionally, comparative studies based on gender can be a source of meaningful findings, revealing nuanced complexities of the cross gender differences and similarities. Su ch a comparison could be done based on the quantitative cross sectional studies and/or via conducting mixed gender focus groups where men and women can dialogue with each other, reflecting the socially constructed nature of sexual matters in general, and s exual double standards in particular. taking in tourism was essential for the purposes of this study, investigations of sexual risk taking in leisure and tourism contexts across the lifespan shoul d be conducted. Sex research has gained some attention with respect to adolescence and young adulthood, but the life stages of middle and late adulthood have been
201 virtually ignored in respect, which is problematic since sex, expression of intimacy, and sex ual risk taking do not necessarily disappear in these life stages (Carpenter & DeLamater, 2012). Ideally, longitudinal studies could shed the most light on the role of the life stage as they would allow the intrapersonal comparison across time. Yet, cross sectional studies with various cohorts would be a source of valuable findings as well. This would allow for further investigation of the sense of immortality, invincibility, and disinhibition as related to sexual behavior in tourism inherent in the finding s in this study across the lifespan. Likewise, the focus on sexuality is important in the investigations of sexual behavior and risk taking in tourism since risk perceptions, motivations, and different social, psychological, and health implications might v ary according to the sexual orientation of men and women (Poria, 2006). T his study revealed a diversity of sexual risk takers in tourism among women with different risk taking and sensation seeking propensities as well as various levels of susceptibility to touristic characteristics as motivational triggers for sexual risk taking. For example, drawing upon the quantitative results, the h igh and diverse risk perceivers motivated by anonymity and exploration ( cluster one ) and the low risk perceivers with div erse motivations (cluster three ) based on their risk perceptions, motivations, and susceptibility to facilitating factors, might be the highest priority for health programs and/or information campaigns, particularly when budgets are limited. Further resea rch of the socio demographic profiles of the identified clusters could facilitate targeting these women effectively. Likewise, the cluster of the unmotivated high and diverse risk perceivers (cluster five) appears to be controversial. On the one hand, th ese women seem to be the least likely to engage in sexual risk taking in tourism due to the relatively low susceptibility to the motivating potential of touristic characteristics and an acute awareness of the risks involved. On the other hand, the
202 women in this cluster rated relatively high the touristic characteristics as facilitators of sexual risk taking, particularly increased alcohol consumption. The combination of this tendency with a relative lack of sexual experience reported by the women in this cl uster and their vulnerability to all three risk dimensions raises potential concerns about the susceptibility of these women to risk and, therefore, renders this cluster relevant to the information campaigns and/or health programs addressing sexual risk ta king in tourism. Furthermore, future research is needed to shed more light on the identified clusters as well as their characteristics and risk taking propensities Lastly, a more d irect approach to investigating preferences regarding specific str ategies and messages for designing effective information campaigns and/or health programs addressing sexual risk taking in leisure and tourism contexts could be a source of invaluable practical health recommendations. Since various clusters have different risk perceptions and motivations, they may respond differently to the messages highlighted by the health program s and/or information campaigns ifferent levels of susceptibility to various inhibiting as well as facilitating factors might off er some clues as to what could be effectively emphasized in the informational efforts aimed at addressing sexual risk taking in tourism. This diversity may decrease the efficacy of the health program that approaches various clusters of sexual risk takers w ith a homogeneous message. Conversely, leveraging this diversity through a polymorphic approach to health programs that accounts for the differences among the clusters of sexual risk takers and tailors various information messages accordingly might be a mo re effective option. In other words, matching health information messages to specific profiles of sexual risk takers can potentially boost the effectiveness and efficiency of the health program aim ed at addressing sexual risk taking among female travelers.
203 Conclusions The contribution of this study stems from addressing sexual risk taking in tourism, which is believed to be more prevalent in tourism vis vis everyday life (Black, 1997 ; Clift & Forrest, 2000). Furthermore, this study drew attention to the less researched topic of sex in tourism as opposed to the frequently discussed commercial sex tourism (Carter & Clift, 2000). Moreover, this study shed additional light on the topic of sexual risk taking in adolescence/young adulthood, which constitutes a priority for public health in the United States (CDC, 2008), where women have been found to be at a particular risk for sexual health concerns ( Broaddus et al., 2010; Padian et al., 1997). However, while the existing literature on sexual risk taking in to urism mainly focuses on STDs (Bloor et al., 2000; Hart & Hawkes, 2000), the value of this study is also embedded in its taking in tourism, reflecting the complex and multidimensional nature of this phenom enon and considering a variety of potential consequences. The qualitative phase enabled an identification of the following universal and common dimensions of sexual risk taking in tourism : physical/sexual health (e.g., STDs, rape, violence, and unwanted pr egnancy), social (e.g., damaged reputation, peer judg ment and embarrassm ent ), emotional (e.g., emotional distress, trauma, or attachment), mental/self perception (e.g., feeling disrespected, used, rejected, and going against personal values), and cultura l (e.g., fear of offending cultural norms and legal issues with obtaining justice). The quantitative phase enabled statistical exploration of this matter identifying three dimensions of sexual risk taking in tourism as perceived by women: physical/sexual health, mental/emotional, and socio cultural. Additionally, the findings suggest that sexual sensation seeking propensity (Kalichman et al., 1994; Kalichman & Rompa, 1995) appears to be related to taking in tourism. Speci fically, the women with lower sexual
204 sensation seeking propensity rated physical/sexual health and mental/emotional dimensions significantly higher than women with higher sexual sensation seeking propensity. This could be explained by the tendency of the h igh sensation seekers to downplay the risks involved to achieve the levels of desired stimulation (Zuckerman, 2007; Zuckerman et al., 1964). Additionally, this exploration provided meaningful motivations for becoming involved in sexual risk behaviors in tourism as well as the facilitating and inhibiting potential of various touristic characteristics. The themes of anonymity, detachment from everyday rules and social gaze, exploration and opportunity to enact sexual fantasies, inv incibility and loss of inhibitions, vacation mentality and fun were portrayed by the women as motivations for sexual risk taking both in the qualitative and quantitative phases in this study. The quantitative approach enabled capturing the aforementioned v for sexual risk taking in tourism in three dimensions: anonymity/detachment/ experimentation ; safe exploration/thrill/empowerment ; and fun/less inhibitions/opportunity. The findings suggest that anonymity and the liberating li minoid potential of tourist experiences (Lett, 1983; Pritchard & Morgan, 2006; Selnniemi, 2003) subdue the influence of sexual double standards as a form of social control allowing for more experimentation in s which in turn was asso ciated with sexual risk taking (Maticka Tyndale & Herold, 1997; Ragsdale et al., 2006). Furthermore, diminished social surveillance under conditions of touristic situational disinhibition (Apostolopoulos et al., 2002; Eiser & Ford, 1995) appeared to be som flexible and forgiving self surveillance with respect to sexual risk taking in tourism. Although self surveillance was found to be a more as it
205 remained an influential force on dimension of sexual risk taking. T his study identified clusters of sexual risk takers in tourism among the female college students based on the perceived dime nsions of and motivations for sexual risk taking in tourism as well as their personality, sexual, and socio demographic profiles. The five clusters identified are as follows : (1) h igh and diverse risk perceivers motivated by anonymity and exploration; ( 2) moderate and diverse risk perceivers motivated by fun; (3) low risk perceivers with diverse motivations; (4) low risk perceivers motivated by exploration; and (5) unmotivated high and diverse risk perceivers. These insights not only contribute to the l iterature by addressing an under researched aspect of human health and behavior in tourist contexts, but also provide an initial platform for a line of research formulating practical implications for the health programs and/or information campaigns aimed a t taking in tourism. To conclude, while methodological difficulties associated with researching sex as well as perceptions of this topic as inappropriate and not deserving academic attention seem to have kept it away from th e spotlight in the fields of leisure and tourism (Poria & Carr, 2010), this study joined the voices calling to shed light on the sexual matters in leisure and tourist experiences. Indeed, this study revealed that the topic taking in tourism should not be mental health, as well as social reputation, indicating that what happens on vacation may spillover to everyday life. T he synergy of the qu alitative and quantitative methods in a sequential mixed methods design (Cameron, 2009; Leech & Onwuegbuzie, 2009) used in this study enabled a triangulation (Decrop, 2004), juxtaposing the findings from different phases, and brought to light the studied
206 phenomenon in its fuller complexity. In this dissertation, both methodological phases were integrated across research purposes, core foci in the data, exploratory analytical techniques, and interpretations. This combination enabled the constructi on of mean ingful theoretical and taking in tourism that would not have been possible without the implemented methodological synergy. Finally, in looking forward towards the eventual development of a sexual risk taking in tourism theory that could be implemented in various tourism and leisure contexts, leaning back on phenomenology and the quantitative exploration conducted in this study, a strong direction seems to be evident in the interaction among the physical/sexual h ealth, mental, emotional, and socio cultural dimensions, as well as the context specific importance assigned by women to each dimension Inherent in this direction appears to be the potential for explaining and/or predicting isk taking in various leisure and tourism contexts. Moreover, the clusters of sexual risk takers introduced in this study, based on perceived risks and motivations, ensure that the eventual sexual risk taking theory will be flexible in terms of accommodati ng various types of sexual risk takers as well as their sensation seeking propensities.
207 APPENDIX A INFORMED CONSENT FOR THE QUALITATIVE PHASE Informed Consent for the Interview Protocol Title: The study is being c onducted by Yelyzaveta Berdychevsky a Ph.D. student at the Department of Tourism, Recreation and Sport Management, at the University of Florida under the supervision of Dr. Heather J. Gibson. Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to : (1) explore the characteristics of the tourist experiences as a context for sk taking in tourism and meanings attached to it; and to (4) identify potentially effective strategies for mediating sexual risk taking in tourism wit hout eliminating the element of fun from the tourist experience What you will be asked to do in the study: You will be asked to participate in the individual face to face interview You will be requested to talk about your experiences of sexual risk tak ing in tourism including the topics like the reasons why you chose to engage in those experiences, your perceptions of those acts and meanings attached to them. The conversation will develop in a free manner. Please note, there is no right answer and your opinion is of utmost importance. You are free not to answer any question. You will be asked to give permission to audio record the interview but you may request that the voice recorder is stopped at any time. The interview will be conducted at the time an d place according to your maximum convenience. Time requ ested : 2 2.5 hours Risks and Benefits: No risks are anticipated. I do not anticipate that you will benefit directly by participating in this research although your participation might help to she were ignored until now. In addition, you might contribute valuable insights to the management of tourist experience. Compensation: Unfortunately, there is no compensation for being involved in this research However if you wish I will provide you with a summary of the findings
208 Confidentiality: Your identity will be kept confidential to the extent provided by law. Your name will not be used in any report. Your information will be assigned a p seudonym (that you will choose prior the interview) and a code number. There will be no list connecting your name to the pseudonym and/or to the code number. Nobody, except for the investigators, will have an access to audio recordings of the interviews. T he audio files will be kept in a secured area at the home of the principal investigator until the transcription is conducted The transcription of the audio files will be done as soon as practicable, and the files will be erased at that point. Voluntary pa rticipation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at anytime without consequence. You have right not to answe r any question/s. You have right to ask not use the information revealed in the course of the interview. Whom to contact if you have questions about the study: Yelyzaveta Berdychevsky, Graduate Student, Department of Tourism, Recreation and Sport Manageme nt, Room 300, Florida Gym PO Box 118208 Gainesville, FL 32611 8208, phone: (352) 392 4042 Ext. 1311, email: firstname.lastname@example.org Supervisor: Dr. Heather J Gibson, Ph.D. / Associate Professor, Department of Tourism, Recreation and Sport Management, Room 304, Florida Gym PO Box 118208 Gainesville, FL 32611 8208 phone: ( 352) 392 4042 Ext. 1249, email: email@example.com Whom to contact about your rights as a researc h participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; phone 392 0433. Agreement: I have read the procedure described above. I voluntarily agree to participate in the procedure and I have received a cop y of this description. Principal Investigator: ___________________________________ Date: _________________ Thank you very much for your cooperation!
209 APPENDIX B INTERVIEW GUIDE FOR THE QUALITATIVE PHASE What do you think is meant by sexual risk takin g? What does it mean to you? Probes: What kind of activities could be considered as sexual risk taking in tourism? ; How do you understand the risk? ; Is it related to emotional involvement / sexually transmitted diseases/ reputation/safety issues? ; Is it rela ted to potential consequences, if any? What do you think about sexual risk taking in tourism? Probes: Is sexual risk taking more or less prevalent in tourism compared to everyday life? Why? ; Is it related to the type of the tourist experience? ; Is it re lated to travelling companions/duration of the trip/tourist destination? ; Are there any gender differences in terms of sexual risk taking in tourism? What are the specific characteristics of the tourist environment as a context for sexual risk taking by wo men? ; What Could you share with me any experiences you have had of taking such risks? Why did you decide to get invo lved in those experiences? Probes: How did they make you feel? ; Is there a particular meaning that you assign to those experiences? ; What were your expectations from those experiences? ; Can you think of something that you had expected to derive from those experiences but have not derived? Why? If you were a marketing consultant hired to design an information campaign addressing sexual risk taking in tourism, what would be your main message? If you were conducting this interview instead of me, what else wou ld you ask? What
210 APPENDIX C EMAIL INVITATION TO PARTICIPATE IN THE QUANTITATIVE PHASE Dear Fellow Student, My name is Liza Berdychevsky. I am a Ph.D. candidate in the College of Health a nd Human Performance at the University of Florida. Your email address was obtained from a random sample provided by the University of Florida Registrar. I would like to invite you to participate in a study I am conducting for my doctoral dissertation inves tigating the perceptions of sexual risk taking on vacation. Participation in my study will take about 15 minutes and will involve answering an online questionnaire. The study is anonymous and your opinion is very importan t Following is the link to the inf ormed consent for this study: http://goo.gl/QgRAI If a link above does not work, please copy and paste the link into your browser's address bar (i.e., the URL address line). Please read the informed consent and keep a co py of the contact information. The access to the online survey is provided at the end of the informed consent form. Please click on the button at the bottom of the informed consent to access the online survey. Your help is much appreciated and extremely im portant. Your responses will be used to provide for my dissertation research which has the goal of gaining insights on an under researched topic that has implications for the health programs. If you would like to know more about this study or if you would like to see a summary of the results please contact me at firstname.lastname@example.org or by phone at 352 870 1287. Thank you very much. Liza Berdychevsky, Ph.D. Candidate Department of Tourism, Recreation and Sport Managemen t 300 Florida Gym, P.O. Box 118208 University of Florida Gainesville, FL 32611 8208 ; Email: email@example.com ; Phone: 352 870 1287
211 Your responses are ANONYMOUS APPENDIX D INFORMED CONSENT FOR THE QUANTITATIVE PHASE Welcome to the 2012 Sexual Behavior in Tourism Survey. Your email address is part of a this study. The purpose of the survey is to examine the link between tourism, risk taking, and se xual behavior according to your current perceptions The study is a dissertation conducted in the College of Health and Human Performance at the University of Florida. Please take about 15 minutes to complete this survey. Your opinion is valuable to us. T here are no anticipated risks for participating in this study. Your identity will be unknown to us and your responses will be anonymous. There is no way to connect your identity to your responses and your IP address or email will not be collected. There is a minimal risk that security of any online data may be breached, since the online host Qualtrics.com uses password protection, secure connections, several layers of encryption and firewalls in compliance with the U.S. and E.U. Safe Harbor Framework ag reement, and is subject to U.S. privacy laws Your data will be removed from the server soon after you complete the survey. It is unlikely that a security breach of the online data will result in any adverse consequence for you. For the Qualtrics Security Statement, please see http://www.qualtrics.com/security statement There is no compensation for completing the questionnaire, but your input is very important. Furthermore, you may benefit from le arning about sexual risk taking and/or from reminiscing about your tourist experiences. Your participation in this study is completely voluntary and you have right to withdraw at any time and/or not to answer any question without consequences. There are no
212 learn about your current perceptions. This study has been reviewed and approved by the University of Florida Institutional Review Board for human subject participation. If you have ques tions about the study, please contact: Liza Berdychevsky, Ph.D. Candidate Department of Tourism, Recreation and Sport Management 300 Florida Gym, P.O. Box 118208, University of Florida Gainesville, FL 32611 8208; Email: firstname.lastname@example.org Phone: 352 870 1287 Any further questions can be directed to my academic supervisor: Dr. Heather Gibson, Department of Tourism, Recreation and Sport Management, 304 Florida Gym, P.O. Box 118208, Gainesville, FL 32611 8208, Email: email@example.com Phone: 352 294 1649. If you have questions about your rights as a participant in this study, please contact the University of Florida IRB office, P.O. Box 112250, University of Florida, Gain esville, FL32611 2250, Email: firstname.lastname@example.org Phone: 352 392 0433. Thank you very much for your time and consideration! PLEASE CLICK ON THE BUTTON BELOW TO ACCESS THE SURVEY
213 APPENDIX E INSTRUMENT FOR THE QUANTITATIVE PHASE Section 1: Personality traits and sexual behavior 1. In this question you will find a series of statements that persons might use to describe themselves. Please, read each statement and indicate to what extent the statement is or is not like you. Not at all like me Not like me Like me Very much like me a I get bored and restless if I have nothing to do b people c I would like parachute jumping d I sometimes like to do things that are a little frightening e I enjoy the feeling of fast driving or riding in a speeding car f I get bored seeing the same old faces g I in advance h i I would enjoy the sensations of skiing very fast down a high mountain slope j While driving, I will sometimes try to run yellow lights for the thrill of it k I would like 2. In this question you will find a series of statements that persons might use to describe themselves. Please, read each statement and indicate to what extent the statement is or is not like you. If you feel that the sta Not at all like me Not like me Like me Very much like me a b The physical sensations are the most important thing about having sex c I enjoy the sensation of intercourse without a condom d e When it comes to sex, physical attraction is more important to me than how well I know the person f I enjoy the comp g h I have said things that were not exactly true to get a person to have sex with me i I am interested in trying out new sexual experiences j I feel like exploring my sexuality k I like to have new and exciting sexual experiences and sensations
214 Section 2 : Vacation as a context for sexual behavior 3 If you were to describe a vacation that provided maximum opportunit y for sexual activity with a steady p artner, what characteristics would it have in terms of the: a. type of vacation (you may check more than one option) r est and r elaxation s ightseeing b ackpacking group tour o ther_________________________________(please, specify) b length of vacation ___________days c. destination (i.e., where) ________ _____________________________ (please, specify) d. trip partners (you may check more than one option) no other trip partners except for a steady partner _____ single fema le trip partners (please indicate the number) _____ single male trip partners (please indicate the number) _____ couples (please indicate the number) other ___ ____________________________ (please, specify) 4 If you were to describe a vacation t hat provided maximum opportunit y for casual sex, what characteristics would it have in terms of the: a. type of vacation (you may check more than one option) r est and r elaxation s ightseeing b ackpacking group tour o ther____________ _____________________(please, specify) b. length of vacation ___________days c. destination (i.e., where) ________ _____________________________ (please, specify) d. trip partners ( you may check more than one option) no trip partners travelling alon e steady partner _____ single female trip partners (please indicate the number) _____ single male trip partners (please indicate the number) _____ couples (please indicate the number) other ____ ___________________________ (please, spec ify) 5 How likely would you be to get involved in behaviors you perceive as sexually risky if you were going on a vacation that provided maximum opportunit y for: Very unlikely Unlikely Neither likely nor unlikely Likely Very likely a sexual ac tivity with a steady partner (the one you described in question 3 )? b casual sex (the one you described in question 4 )?
215 6 To what extent do you agree or disagree with the following statements when you think of a vacation that provid ed maximum opportunit y for behaviors you perceive as sexually risky : Strongly disagree Dis agree Neither agree nor disagree Agree Strongly Agree a Casual sex is more frequent on vacation b For a person of my gender, it is more acceptable to take sexual risks on vacation than at home c Alcohol is liquid courage for taking sexual risks on vacation d Most unattached males are looking for casual sex on vacation e e For a person of my age, it is more acceptable to take sexual risks on vacation than at home f Alcohol is a psychological excuse for taking sexual risks on vacation g Most unattached females are looking for casual sex on vacation h h For a person of my relationship status, it is more acceptable to take sexual risks on vacation than at home i People engage in more sexual risk taking on vacation than in everyday life j j I am more prone to sexual risk taking on vacation than in everyday life k 7 To what extent does each of the following vacation related characteristics facilitate or inhibit behaviors you perceive as sexually risky on vacation? Strongly inhibits Inhibits Neither inhibits nor facilitates Facilitates Strongly facilitates a Being anonymous b Having lots of free leisure time c Being detached from everyday norms d Consuming alcohol e Consuming recreational drugs f Experiencing peer pressure g Being in a break loose mood h Lack of social expectations i Having a steady relationship at home j Having limited or no safety network k Permissive sexual atmosphere l Feeling a different person m Feeling more sexually confident n Meeting new people o Leaving bad experiences behind p Not expecting a lasting relationship q Feeling invincible in terms of risk consequences r Touristic fun oriented mentality
216 s Party scene at the dest ination t Sharing a room u Being in a country with more tolerant religious beliefs v Having judgmental trip partners w Beach scene at the destination x Scene where revealing clothing is appropriate y T he length of the vacation z The size of your travel party aa Being in a country that speaks a different language bb Being in a country with strict religious beliefs 8 Aside from those mentioned in question 7 what ot her vacation characteristics could be related to behaviors you perceive as sexually risky on vacation ? Section 3 : Motivations for sexual risk taking on vacation 9 To what extent do you agree or disagree that the following factors motivate your involvement in behaviors you consider to be sexually risky as a tourist? Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree a Tourism offers anonymity b Nobody will judge me c I can enact sexual fantasie s not available at home d Risk is a motivation in itself e I feel detached from everyday social norms f Vacation is the time to have fun g Tourism is a scene for sexual conquest h Friends encourage ment to have sex on vacation i Casual sex on vacation has no consequences j I have more opportunities to be sexually active k It is easier to meet people on vacation l Taking sexual risks on vacation is empowering m Sex on v acation is more romantic than in everyday life n A foreign sexual partner is exotic o I have less inhibitions on vacation p Vacation mentality is about taking sexual risks q Taking sexual risks on vacation is thrilling r It is cool to brag about sexual risks taken on vacation
217 s I am not so worried about what other people think of me when I am on vacation t Sex is a way to get to know the locals u v Sex feels good w Vacation is the time for sexual exploration x Availability of my type of sexual partner on vacation y What happens on vacation stays on vacation 10 Aside from those mentioned in question 9 what ot her factors could motivate your involvement in behaviors you consider to be sexually risky as a tourist? Section 4 : Perceptions and experiences of sexual risk taking on vacation 11 To what extent do you agree or disagree that the following acti vities on vacation could be considered as sex? Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree a Vaginal sex b Receiving oral sex c Performing oral sex d Anal sex e Fondling f Making out /kissing 12. To what extent do you agree or disagree that the following activities on vacation could be considered as sexual risk taking? Strongly disagree Disagree Neither agree nor disagree Agree Strongly agree a Having v aginal sex with an unsteady partner with a condom b Receiving oral sex from an unsteady partner with a latex barrier (such as a dental dam or plastic wrap) c Performing oral sex on an unsteady partner with a condom d Ha ving a nal sex with an unsteady partner with a condom e Fondling with an unsteady partner f Making out with/kissing an unsteady partner g Having sex with multiple partners h Using sexual enhancers with your steady partner i Attending a swinger club j Trying sexual experimentation with your steady partner
218 k Being more flirtatious l Going to live sex shows m Going to a strip club n Having sex in semi public spaces (like a restroom or elevator) o Having sex under the influence of alcohol p Having vaginal sex with an unsteady partner without a condom q Receiving oral sex from an unsteady partner without a latex barrier (such as a dental dam or p lastic wrap) r Performing oral sex on an unsteady partner with out a condom s Having anal sex with an unsteady partner without a condom t Having vaginal sex with a steady partner without protection from pregnancy u Havin g sex under the influence of drugs v Having sex in unfamiliar surroundings w Having sex with a sex worker (e.g., prostitute) 1 3 Aside from those mentioned in question 11 what other activities on vacation could be conside red as sexual risk taking? 1 4 Strongly disagree Disagree Neither agree nor disagree Agree Strong ly agree a the risk of getting sexually transmitted diseases b the risk of contracting HIV/AIDS c the risk losing control over the situation d the risk of getting pregnant e the risk of getting physically hurt f the risk of being raped g the risk of being kidnapped h the risk of affecting my reputation at home i the risk of being judged by travelling companions j the risk of feeling not respected k the risk of hurting l the risk of being emotionally hurt m the risk of feeling uncomfortable after the act n the risk of being too far out of my comfort zone o the risk of feeling used p the risk of having unfu lfilled expectations
219 q the risk of feeling rejected r the risk of not being able to procure a condom s the risk of facing legal consequences t the risk of offending cultural norms at the destination u the risk o f embarrassing myself v the risk of not being able to perform sexually up to my standards w the risk of regretting it later x the risk of doing something against my values 1 5 Have you ever experienced the following sexual activities on vacation? Please keep in mind that your responses are anonymous and your honesty is greatly appreciated. Yes No a Having v aginal sex with an unsteady partner with a condom b Receiving oral sex from an unsteady partner with a latex barrier (such as a dental dam or plastic wrap) c Performing oral sex on an unsteady partner with a condom d Having a nal sex with an unsteady partner with a condom e Fondling with an unsteady partner f Making out with/kissing an unsteady partn er g Having sex with multiple partners h Using sexual enhancers with your steady partner i Attending a swinger club j Trying sexual experimentation with your steady partner k Being more flirtatious l Going to live sex shows m Going to a strip club n Having sex in semi public spaces (like a restroom or elevator) o Having sex under the influence of alcohol p Having vaginal sex with an unsteady partner without a condom q Receiving oral sex from an unsteady partner without a la tex barrier (such as a dental dam or plastic wrap) r Performing oral sex on an unsteady partner without a condom s Having anal sex with an unsteady partner without a condom t Having vaginal sex with a steady partner without protection from pregna ncy u Having sex under the influence of drugs v Having sex in unfamiliar surroundings w Having sex with a sex worker (e.g., prostitute)
220 Section 5 : Demographics Now a few questions about you to help interpret your answers. Please keep in mind that you responses are anonymous and complete the following: 1 6 What is your age? __________ 1 7 What is your current year in school? Freshman Sophomore Junior Senior Graduate Other ________________________________ _______________(please, specify) 1 8 Were you born in the United States? Yes No 19. Did you grow up in the United States? Yes No 20 What is your rac ial or ethnic background ? White/Caucasian not of Hispanic origin Blac k not of Hispanic origin Hispanic Asian Native Hawaiian/Other Pacific Islander American Indian/Native Alaskan Other _______________________________________________(please, specify) 21 What is your sexual orientation? Hetero sexual Gay/Lesbian Bisexual Other _______________________________________________(please, specify) 2 2 What is your current marital status? Never married Married Divorced /Separated Widowed 2 3 What is your current datin g status? Exclusive relationship Open relationship where both partners can see other people Casually dating, but not in a relationship Not dating and not in a relationship
221 2 4 How many sexual partners have you had? __________ 2 5 Have you or any of your sexual partners ever experienced an unplanned pregnancy? Yes No I do not know 2 6 Please refer to the following: Yes No a Have you ever been tested for a sexually transmitted disease? b b Have you ever been diagnosed wi th a sexually transmitted disease? c Do you think any sexual practices in your past have put you at risk for HIV/AIDS? d Have you ever had unsafe sex with someone you thought might have HIV/AIDS? e Have you ever had sex with someone without gi ving your consent? f Have you ever had sex with someone without getting their consent? 2 7 from all sources before taxes? Under $25,000 $25,000 $49,999 $50,000 $74,999 $75,000 $99,999 $100,000 $124,999 $125,000 $149,999 Over $150,000 2 8 Approximately what is your personal annual income from all sources before taxes (including scholarship s, assistantships financial aid, support)? Under $ 5,000 $ 5,000 $9,999 $10,000 $14,999 $15,000 $19,999 $20,000 $29,999 $30,000 $39,999 $40,000 $49,999 Over $50,000 Thank you very much for your time and help Your participation in this study is greatly appreciated.
222 LIST OF REFERENCES Andriotis, K. (2010). Heterotopic erotic oases: The public nude beach experience. Annals of Tourism Research, 37 (4), 1076 1096. Apostolopoulos, Y., Snmez, S., & Yu, C. H. (2 002). HIV risk behaviours of American spring break vacationers: A case of situational disinhibition? International Journal of STD & AIDS, 13 (11), 733 743. Arnett, J. J. (2010). Adolescence and emerging adulthood (4th ed.). Upper Saddle River, NJ: Prentice Hall. Asparouhov, T., & Muthn, B. (2009). Exploratory structural equation modeling. Structural Equation Modeling, 16 (3), 397 438. Aston, E. (2008). A fair trade?: Staging female sex tourism in "Sugar mummies" and "Trade". Contemporary Theatre Review, 18 (2 ), 180 192. Ateljevic, I., & Hall, D. (2007). The embodiment of the macho gaze in South eastern Europe: Performing femininity and masculinity in Albania and Croatia. In A. Pritchard, N. Morgan, I. Ateljevic & C. Harris (Eds.), Tourism and gender: Embodimen t, sensuality and experience (pp. 138 157). Wallingford, UK: CABI. Attwood, F. (2007). Sluts and Riot Grrrls: Female identity and sexual agency. Journal of Gender Studies, 16 (3), 233 247. AVERT ( 2009 ) Worldwide AIDS & HIV Statistics. Retrieved from the A VERT website on March 29 th 2011. http://www.avert.org/worldstats.htm Baldwin, J. D., & Baldwin, J. I. (1988). Factors affecting AIDS related sexual risk taking behavior among college students. The Journal of Sex Research, 25 (2), 181 196. Bancroft, J., Janssen, E., Carnes, L., Goodrich, D., & Strong, D. (200 4 ). Sexual activity and risk taking in young heterosexual men: The relevance of sexual arousability, mood, and sensation seeking. Journal of Sex Research 41 (2), 181 192. Bancroft, J., Janssen, E., Strong, D., Carnes, L., Vukadinovic, Z., & Long, S. (2003). Sexual risk taking in gay men: The relevance of sexual arousability, mood, and sensation seeking. Archives of Sexual Behavior, 32 (6), 555 572. Bauer, T G., & McKercher, B. (Eds.). (2003). Sex and tourism: Journeys of romance, love, and lust New York: The Haworth Hospitality Press. Bazeley, P. (2009). Integrating data analyses in mixed methods research. Journal of Mixed Methods Research, 3 (3), 203 207. Bell, N. J., O'Neal, K. K., Feng, D., & Schoenrock, C. J. (1999). Gender and sexual risk. Sex Roles, 41 (5/6), 313 332.
223 Belliveau, J. (2006). Romance on the road: Traveling women who love foreign men Baltimore: Beau Monde Press. Bender, K., & Furman, R. (2 004). The implications of sex tourism on men's social, psychological, and physical health. The Qualitative Report, 9 (2), 176 191. Berdychevsky, L., Poria, Y., & Uriely, N. (2010). Casual sex and the backpacking experience: The case of Israeli women. In N. Carr & Y. Poria (Eds.), Sex and the sexual during people's leisure and tourism experiences (pp. 105 118). Newcastle upon Tyne: Cambridge Scholars Publishing. Berdychevsky, L., Gibson, H., & Poria, Y. (2013b). Women's sexual behavior in tourism: Loosening t he bridle. Annals of Tourism Research (in press) 65 86. doi: http://dx.doi.org/10.1016/j.annals.2013.01.006 st experiences: Motivations, behaviors, and meanings. Tourism Management, 35 144 155. Beyer, C (2011). Edmund Husserl The Stanford Encyclopedia of Philosophy (Winter 2011 Ed ition ) Edward N. Zalta (ed.), url: http://plato.stanford.edu/archives/win2011/e ntries/husserl Bialeschki, D. M. (2005). Fear of violence: Contested constraints by women in outdoor recreation activities. In E. L. Jackson (Ed.), Constraints to leisure (pp. 103 114). State College, Pennsylvania: Venture Publishing, Inc. Bicchieri, C ., & Muldoon, R (2011). Social Norms The Stanford Encyclopedia of Philosophy (Spring 2011 Ed ition ), Edward N. Zalta (ed.), url: http://plato.stanford.edu/archives/spr2011/entries/ social norms Binik, Y. M., Mah, K., & Kiesler, S. (1999). Ethical issues in conducting sex research on the internet. The Journal of Sex Research, 36 (1), 82 90. Black, P. (1997). Sexual behaviour and travel: Quantitative and qualitative perspectives on the behaviour of genito urinary medicine clinic attenders. In S. Clift & P. Grabowski (Eds.), Tourism and health: Risks, research and responses (pp. 165 180). London: Pinter. Black, P. (2000). Sex and travel: Making the links. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 250 264). London: Pinter. Blanc, A. (2001). The effect of power in sexual relationships on sexual and reproductive health: An examination of the evidence. Studies in Family Planning, 32 (3), 189 213. Bloo r, M. (1995). HIV related risk behavior among international travellers. In D. Fitzimons, V. Hardy & K. Tolley (Eds.), Economic and social impact of AIDS in Europe (pp. 189 198). London: Cassell and National AIDS Trust.
224 Bloor, M., Thomas, M., Abeni, D., Gou jon, C., Hausser, D., Hubert, M., et al. (2000). Sexual risk behaviour in a sample of 5676 young, anaccompanied travellers. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 157 167). London: Pinter. Bradshaw, C., Kahn, A S., & Saville, B. K. (2010). To hook up or date: Which gender benefits? Sex Roles, 62 661 669. Brennan, D. (2004). Dominican Republic Durham: Duke University Press. Broaddus, M. R., Morris, H., & Bryan, A. D. (2010). 'It's not what you said, it's how you said it': Perceptions of Condom proposers by gender and strategy. Sex Roles, 62 603 614. Broman, C. L. (2007). Sexual risk behavior among black adolescents. Journal of African American Studies, 11 180 188. Brookmeyer, K. A., & Henrich, C. C. (2009). Disentangling adolescent pathways of sexual risk taking. The Journal of Primary Prevention, 30 677 696. Browning, J. R., Kessler, D., Hatfield, E., & Choo, P. (1999). Power, gende r, and sexual behavior. The Journal of Primary Prevention, 36 (4), 342 347. Burns, R. B., & Burns, R. A. (2008). Business research methods and statistics using SPSS London: Sage Publi cations. Butler, J. (1990). Gender trouble: Feminism and the subversion o f identity New York: Routledge. Butler, J. (1996). Sexual inversions. In S. J. Hekman (Ed.), Feminist interpretations of Michel Foucault (pp. 59 75). University Park, Pennsylvania: The Pennsylvania State University Press. Cabezas, A. C. (2004). Between lo ve and money: Sex, tourism, and citizenship in Cuba and the Dominican Republic. Journal of Women in Culture and Society, 29 (4), 987 1015. Cameron, R. (2009). A sequential mixed model research design: Design, analytical and display issues. International Jou rnal of Multiple Research Approaches, 3 (2), 140 152. Carpenter, L. M. (2010). Gendered sexuality over the life Course: A conceptual framework. Sociological Perspectives, 53 (2), 155 178. Carpenter, L. M., & DeLamater, J. (2012). Sex for life: From virginity to Viagra, how sexuality changes throughout our lives New York/London: New York University Press. Carr, N., & Poria, Y. (Eds.). (2010). Sex and the sexual during people's leisure and tourism experiences Newcastle upon Tyne: Cambridge Scholars Publishing
225 Carter, S., & Clift, S. (2000). Tourism, international travel and sex: Themes and research. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 1 19). London: Pinter. Catania, J. A. (1999). A framework for conceptualizing reporting bias and its antecedents in interviews assessing human sexuality. The Journal of Sex Research, 36 (1), 25 38. Catania, J. A., Binson, D., Canchola, J., Pollack, L. M., & Hauck, W. (1996). Effects of interviewer gender, interviewer choice, and ite m wording on responses to questions concerning sexual behavior. The Public Opinion Quarterly, 60 (3), 345 375. Centers for Disease Control. (2007). Trends in reportable sexually transmitted diseases in the United States, 2007: National surveillance data for chlamydia, gonorrhea, and syphilis. Retrieved from the Centers for Disease Control and Prevention website on July 7 th 2012. http://www.cdc.gov/std/stats07/trends.pdf Centers for Disease Control. (2008). Nationally representative CDC study finds 1 in 4 teenage girls has a sexually transmitted disease. Retrieved from the Centers for Disease Control and Prevention website on July 7 th 2012. http://www.cdc.gov/stdconference/2008/press/release 11march2008.pdf Cho, Y., & Span, S. A. (2010). The effect of alcohol on sexual risk taking among young men and women. Addictive Behaviors, 35 779 785. Chng, C. L., & Gliga Vargas J. (2000). Ethnic identity, gay identity, sexual sensation seekingand HIV risk taking among multiethnic men who have sex with men. AIDS Education and Prevention, 12 (4), 326 339. Clift, S., & Carter, S. (Eds.). (2000). Tourism and sex: Culture, commerce a nd coercion London: Pinter. Clift, S., & Forrest, S. (2000). Tourism and the sexual ecology of gay men. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 179 199). London: Pinter. Cohen, E. (1982). Thai girls and farang men: The edge of ambiguity. Annals of Tourism Research, 9 (3), 403 428. Cohen, S. E., & Fromme, K. (2002). Differential determinants of young adult substance use and high risk sexual behavior. Journal of Applied Social Psychology, 32 (6), 1124 1150. Costello A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10 (7), 1 9. Crawford, M., & Popp, D. (2003). Sexual double standards : A review and methodological critique of two decades of research. The Journal of Sex Research, 40 (1), 13 26.
226 Crick, M. (1989). Representations of international tourism in the social sciences: Sun, sex, sights, savings, and servility. Annual Review of Anth ropology, 18 307 344. Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52 (4), 281 302 Dahlberg, K. (2006). The essence of essences the search for meaning structures in phenomenological analysis of lifeworld phenomena. International Journal of Qualitative Studies on Health and Well being, 1 11 19. Dahles, H., & Bras, K. (1999). Entrepreneurs in romance: Tourism in Indonesia. Annals of Tourism Research, 26 (2), 267 293. De Albuquerque, K. (1998). S ex, beach boys and female tourists in the Caribbean. Sexuality and Culture, 2 87 111. Decrop, A. (2004). Trustworthiness in qualitative tourism research. In J. Phillimore & L. Goodson (Eds.), Qualitative research in tourism: Ontologies, epistemologies and methodologies (pp. 156 169). London: Routledge. Dellinger, A. B., & Leech, N. L. (2007). Toward a unified validation framework in mixed methods research. Journal of Mixed Methods Research, 1 (4), 309 332 Diken, B., & Laustsen, C. B. (2004). Sea, sun, sex and the discontents of pleasure. Tourist Studies, 4 (2), 99 114. Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). Internet, mail, and mixed mode surveys: The tailored design method (3rd ed.). Hoboken, NJ: John Wiley & Sons, Inc. DiStefano, C., Zhu, Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14 (20), 1 11. Dolnicar, S., & Grn, B. (2008). Challenging ''factor cluster segmentation''. Journal of Travel Research, 47 (1), 63 71. Donohew, L., Zimmerman, R., Cupp, P. S., Novak, S., Colon, S., & Abell, R. (2000). Sensation seeking, impulsive decision making, and risky sex: Implications for risk taking and design of interventions. Personality and Indivi dual Differences, 28 1079 1091. Dunne, M. P., Martin, N. G., Bailey, J. M., Heath, A. C., Bucholz, K. K., Madden, P. A. F., et al. (1997). Participation bias in a sexuality survey: Psychological and behavioural charcteristics of responders and non respond ers. International Journal of Epidemiology, 26 (4), 844 854. Eaton, A. A., & Rose, S. (2011). Has dating become more egalitarian? A 35 year review using Sex Roles Sex Roles, 64 (11 12), 843 862.
227 Eiser, J. R., & Ford, N. (1995). Sexual relationships on holid ay: A case of situational disinhibition? Journal of Social and Personal Relationships, 12 (3), 323 339. Elms, D. (1998). Risk management: General issues. In D. Elms (Ed.), Owning the future: Integrated risk management in practice (pp. 43 56). Cristchurch, N ew Zealand: Centre for Advanced Engineering, University of Canterbury. Elsrud, T. (2001). Risk creation in traveling: Backpacker adventure narration. Annals of Tourism Research, 28 (3), 597 617. Emmers Sommer, T. M., & Allen, M. (2005). Safer sex in persona l relationships: The role of sexual scripts in HIV infection and prevention New Jersey: Lawrence Erlbaum Associates, Inc., Publishers. Falconer, E. (2011). Risk, excitement and emotional conflict in women's travel narratives. Recreation and Society in Afr ica, Asia & Latin America, 1 (2), 65 89. Fasula, A. M., Miller, K. S., & Wiener, J. (2010). Sexual risk and the double standard for African American adolescent women. In M. Stombler, D. W. Baunach, C. Burgess, D. Donelly, W. Simonds & E. J. Windsor (Eds.), Sex matters: The sexuality and society reader (3rd ed., pp. 194 201). Boston: Allyn & Bacon. Feilzer, M. Y. (2010). Doing mixed methods research pragmatically: Implications for the rediscovery of pragmatism as a research paradigm. Journal of Mixed Methods Research, 4 (1), 6 16. Fenton, K. A., Johnson, A. M., McManus, S., & Erens, B. (2001). Measuring sexual behavior: Methodological challenges in survey research. Sexually Transmitted Infections, 77 (2), 84 92. Ford, N., & Eiser, J. R. (1996). Risk and liminali ty: The HIV related socio sexual interaction of young tourists. In S. Clift & S. Page (Eds.), Health and the international tourist (pp. 152 179). London: Routledge. Foucault, M. (1976). The history of sexuality: An introduction (Vol. 1). New York: Pantheon Foucault, M. (1986). Of other spaces. Diacritics, 16 (1), 22 27. Frochot, I., & Morrison, A. M. (2001). Benefit segmentation: A review of its applications to travel and tourism research. Journal of Travel & Tourism Marketing, 9 (4), 21 45. Gagnon, J. H., & Simon, W. (1973). Sexual Conduct Chicago: Aldine Pub. Co. Gaither, G. A., & Sellbom, M. (2003). The sexual sensation seeking scale: Reliab i lity and validity within a heterosexual college student sample. Journal of Personality Assessment, 81 (2), 157 167.
228 Gibson, H. J. (1996). Thrill seeking vacations: A life span perspective. Loisir et Socit / Society and Leisure, 19 (2), 439 458. Gibson, H. J. (2001). Gender in tourism: Theoretical perspectives. In Y. Apostolopoulos, S. Snmez & D. J. Timothy (Eds.), Wom en as producers and consumers of tourism in developing regions (pp. 19 43). Westport, CT: Praeger. Gibson, H. J., & Yiannakis, A. (2002). Tourist roles: Needs and the Lifecourse. Annals of Tourism Research, 29 (2), 358 383. Grbich, C. (2007). Qualitative da ta analysis: An introduction London: S age Publications. Greene, J. C. (2007). Mixed methods in social inquiry San Francisco, CA: Jossey Bass. Greene, J. C. (2008). Is mixed methods social inquiry a distinctive methodology? Journal of Mixed Methods Resear ch, 2 (1), 7 22. Greene, K., Krcmar, M., Walters, L. H., Rubin, D. L., & Hake, J. L. (2000). Targeting adolescent risk taking behaviors: The contributions of egocentrism and sensation seeking. Journal of Adolescence, 23 439 461. Grello, C. M., Welsh, D. P. & Harper, M. S. (2006). No strings attached: The nature of casual sex in college students. The Journal of Sex Research, 43 (3), 255 267. Up against Foucault: Explorations of some tensions between Foucault and feminism (pp. 51 72). London: Routledge. Gubrium, N. J., & Holstein, J. A. (2002). Handbook of interview research: Context and method Thousand Oaks CA : Sage Publications. Hair, J. F., Black, W. C., Babin, B. J., & Ande rson, R. E. (2010). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall. Hall, C. M. (1996). Gender and economic interests in tourism prostitution: The nature, development and implications of sex tourism in South east Asia. In Y. Ap ostolopoulos, S. Leivadi & A. Yiannakis (Eds.), The sociology of tourism: Theoretical and empirical investigations (pp. 265 280). London: Routledge. Harrits, G. S. (2011). More than method? A discussion of paradigm differences within mixed methods research Journal of Mixed Methods Research, 5 (2), 150 166. Hart, G., & Hawkes, S. (2000). International travel and the social context of sexual risk. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 168 178). London: Pinter. Ha vitz, M. E., Dimanche, F., & Bogle, T. (1994). Segmenting the adult fitness market using involvement profiles. Journal of Park and Recreation Administration, 12 (3), 35 56.
229 Henderson, K. A. (1991). The contribution of feminism to an understanding of leisure constraints. Journal of Leisure Research, 23 (4), 363 377. Herold, E., Garcia, R., & DeMoya, T. (2001). Female tourists and beach boys: Romance or sex tourism? Annals of Tourism Research, 28 (4), 978 997. Higgins, J. A., & Browne, I. (2008). Sexual needs, c ontrol, and refusal: how "doing" class and gender influences sexual risk taking. Journal of Sex Research, 45 (3), 233 245. Hobson, J. S. P., & Dietrich, U. C. (1994). Tourism, health and quality of life: Challenging the responsibility of using the tradition al tenets of sun, sea, sand, and sex in tourism marketing. Journal of Travel & Tourism Marketing, 3 (4), 21 38. Hovell, M. F., Hillman, E. R., Blumberg, E., Sipan, C., Atkins, C., Hofstetter, C. R., et al. (1994). A behavioral ecological model of adolescent sexual development: A template for AIDS prevention. The Journal of Sex Research, 31 (4), 267 281. Hoyle, R. H., Fejfar, M. C., & Miller, J. D. (2000). Personality and sexual risk taking: A quantitative review. Journal of Personality, 68 (6), 1203 1231. Huss erl, E. (1913). Ideas pertaining to a pure phenomenology and to a phenomenological phylisophy (F. Kersten, Trans.). The Hague: 1983 Martinus Nijhoff Publishers. Impett, E. A., & Peplau, L. A. (2003). Sexual compliance: Gender, motivational, and relationshi p perspectives. The Journal of Sex Research, 40 (1), 87 100. Jeffreys, S. (1999). Globalizing sexual exploitation: Sex tourism and the traffic in women. Leisure Studies, 18 (3), 179 196. Jeffreys, S. (2003). Sex tourism: Do women do it too? Leisure Studies, 22 (3), 223 238. Johnson, B. (2008). Living with tensions: The dialectic approach. Journal of Mixed Methods Research, 2 (3), 203 207. Johnson, B., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research paradigm whose time has come. Educational Resea rcher, 33 (7), 14 26. Johnson, B., Onwuegbuzie, A. J., & Turner, L. A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 1 (2), 112 133. Jonason, P. K., & Fisher, T. D. (2009). The power of prestige: Why young men repo rt having more sex partners than young women. Sex Roles, 60 (3), 151 159. ( Eds.), Tourism and gender: Embodiment, sensuality and experience (pp. 92 106). Wallingford, UK: CABI.
230 Jordan, F., & Aitchison, C. C. (2008). Tourism and the sexualisation of the gaze: Solo female bodiment. Leisure Studies, 27 (3), 329 349. Experiences of solo women travelers. Tourism Review International, 9 (2), 195 211. Josiam, B. M., Hobson, J. S. P., Dietrich, U. C., & Smeaton, G. (1998). An analysis of the sexual, alcohol and drug related behavioural patterns of students on spring break. Tourism Management, 19 (6), 501 513. Kalichman, S. C., Johnson, J. R., Adair, V., Rompa, D., Multhauf, K., & Kelly, J. A. (1994). Sexual sensation seeking: Scale development and predicting AIDS risk behavior among homosexually active men. Journal of Personality Assessment, 62 (3), 385 397. Kalichman, S. C., & Rompa, D. (1995). Sexual sensation seeking and sexual compulsivity scales: Reliability, validity, and predicting HIV risk behavior. Journal of Personality Assessment, 65 (3), 586 601. and risky sexual behavior. Cognitive Therapy and Research, 24 (1), 1 22. Kelly, J. R. (1987). Freedom to be: A new sociology of leisure New York: Macmillan. Kempadoo, K. (Ed.). (1999). Sun, sex, and gold: Tourism and sex work in the Caribbean Lanham: Row man & Littlefield Publishers, Inc. Kim, H., & Jamal, T. (2007). Touristic quest for existential authenticity. Annals of Tourism Research, 34 (1), 181 201. Kyle, G. T., Kerstetter, D. L., & Guadagnolo, F. B. (2002). Market segmentation using participant invo lvement profiles. Journal of Park and Recreation Administration, 20 (1), 1 21. Larsen, J. (2008). De exoticizing tourist travel: everyday life and sociality on the move. Leisure Studies, 27(1), 21 34. Lear, D. (1997). Sex and sexuality: Risk and relationshi ps in the age of AIDS Thousand Oaks, CA: Sage Publications. Leech, N. L., & Onwuegbuzie, A. J. (2009). A typology of mixed methods research designs Quality and Quantity, 43 265 275. Lepp, A., & Gibson, H. J. (2003). Tourist roles, perceived risk and inte rnational tourism. Annals of Tourism Research, 30 (3), 606 624.
231 Lepp, A., & Gibson, H. (2008). Sensation seeking and tourism: Tourist role, perception of risk and destination choice. Tourism Management, 29 (4), 740 750. Lett, J. W. (1983). Ludic and liminoid aspects of charter yacht tourism in the Caribbean. Annals of Tourism Research, 10 (1), 35 56. Levinson, D. J. (1996). New York: Knopf. Levinson, D. J., Darrow, C., Klein, E., Levinson, N., & McKee, B. (1978). The seasons of a man's life New York: Knopf. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry Beverly Hills, CA: Sage Publications. Littlewood, I. (2002). Sultry climates: Travel and sex Cambridge: Da Capo Press. Litvin, S. W. (2010). A comparison of student spring break and their "normal" behaviors: Is the hype justified? Tourism Review International, 13 (3), 173 181. Marsh, H. W., Liem, G. A. D., Martin, A. J., Morin, A. J. S., & Nagengast, B. (2011). Methodological measurement fruitfulness of exploratory str uctural equation modeling (ESEM): New approaches to key substantive issues in motivation and engagement. Journal of Psychoeducational Assessment, 29 (4), 322 346. Marsh, H. W., Ldtke, O., Muthn, B., Asparouhov, T., Morin, A. J. S., Trautwein, U., et al. ( 2010). A new look at the big five factor structure through exploratory structural equation modeling. Psychological Assessment, 22 (3), 471 491. Marsh, H. W., Muthn, B., Asparouhov, T., Ldtke, O., Robitzsch, A., Morin, A. J. S., et al. (2009). Exploratory structural equation modeling, integrating CFA and EFA; Application to students' evaluations of university teaching. Structural Equation Modeling, 16 (3), 439 476. Marsh, H. W., Nagengast, B., & Morin, A. J. S. (2012). Measurement invariance of Big Five fact ors over the life span: ESEM tests of gender, age, plasticity, maturity, and la dolce vita effects. Developmental Psychology (in press), doi: 10.1037/a0026913 Mason, P., Augustyn, M., & Seakhoa King, A. (2010). Exploratory study in tourism: Designing an initial, qualitative phase of sequenced, mixed methods research. International Journal of Tourism Research, 1 2 432 448. Maticka Tyndale, E., & Herold, E. S. (1997). The scripting of sexual behaviour: Canadian university students on spring break in Florida. The Canadian Journal of Human Sexuality, 6 (4), 317 328. Maticka Tyndale, E., & Herold, E. S. (1999). Condom use on spring break vacation: The influence of intent i ons, prior use, and context. Journal of Applied Social Psychology, 29 (5), 1010 1027.
232 Maticka Tyndale, E., Herold, E. S., & Mewhinney, D. M. (1998). Casual sex on spring break: Intentions and behaviors of Canadian students. The Journal of Sex Research, 35 (3), 254 264. Maticka Tyndale, E., Herold, E. S., & Oppermann, M. (2003). Casual sex among Australian schoolies. The Journal of Sex Research, 40 (2), 158 169. Mattila, A. S., Apostolopoulos, Y., Snmez, S ., Yu, L., & Sasidharan, V. (2001). The impact of Journal of Travel Research, 40 (2), 193 200. McCabe, J., Tanner, A. E., & Heiman, J. R. (2010). The impact of gender expectations on meanings o f sex and sexuality: Results from a cognitive interview study. Sex Roles, 62 (3 4), 252 263. McCoul, M. D., & Haslam, N. (2001). Predicting high risk sexual behaviour in heterosexual and homosexual men: The roles of impulsivity and sensations seeking. Perso nality and Individual Differences, 31 1303 1310. McKercher, B., & Bauer, T. G. (2003). Conceptual framework of the nexus between tourism, romance, and sex. In T. G. Bauer & B. McKercher (Eds.), Sex and tourism: Journeys of romance, love, and lust (pp. 3 1 7). New York: Haworth Hospitality Press. Mewhinney, D. M., Herold, E. S., & Maticka Tyndale, E. (1995). Sexual scripts and risk taking of Canadian university students on spring break in Daytona Beach, Florida. The Canadian Journal of Human Sexuality, 4 (4), 273 288. 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. Milhausen, R. R., Reece, M., & Perera, B. (2006). A theory based approach to unde rstanding sexual behavior at Mardi Gras. The Journal of Sex Research, 43 (2), 97 106. Miller, K. E., Farrell, M. P., Barnes, G. M., Melnick, M. J., & Sabo, D. (2005). Gender/racial differences in jock identity, dating, and ad o lescent sexual risk. Journal of Youth and Adolescence, 34 (2), 123 136. Mooi, E., & Sarstedt, M. (2011). A concise guide to market research: The process, data, and methods using IBM SPSS Statistics Berlin Heidelberg: Springer. Morgan, D. L. (2007). Paradigms lost and pragmatism regained : Methodological implications of combining qualitative and quantitative methods. Journal of Mixed Methods Research, 1 (1), 48 76. Moustakas, C. (1994). Phenomenological research methods Thousand Oaks CA : Sage Publications.
233 Muehlenhard, C. L., Peterson, Z. D., Karwoski, L., Bryan, T. S., & Lee, R. S. (2003). Gender and sexuality: An introduction to the special issue. The Journal of Sex Research, 40 (1), 1 3. Tourism Review Internat ional, 9 (2), 213 227. Onwuegbuzie, A. J., & Johnson, R. B. (2006). The validity issue in mixed research. Research in the Schools, 13 (1), 48 63 Oppermann, M. (1998). Sex tourism and prostitution: Aspects of leisure, recreation, and work Elmsford, NY: Cogn izant Communication Corporation. Oppermann, M. (1999). Sex tourism. Annals of Tourism Research, 26 (2), 251 266. Padian, N. S., Shiboski, S. C., Glass, S. O., & Vittinghoff, E. (1997). Heterosexual transmission of human immunodeficiency virus (HIV) in North ern California: Results from a ten year study. American Journal of Epidemiology, 146 (4), 350 357. Padian, N. S., Shibowski, S. C., & Jewell, N. P. (1991). Female to male transmission of human immunodeficiency virus. Journal of the American Medical Associat ion, 265 (12), 1664 1667. Park, M., Yang, X., Lee, B., Jang, H., & Stokowski, P. A. (2002). Segmenting casino gamblers by involvement profiles: A Colorado example. Tourism Management, 23 55 65. Parsons, J. T., Bimbi, D., & Halkitis, P. N. (2001). Sexual co mpulsivity among gay/bisexual male escorts who advertise on the Internet. Sexual Addiction & Compulsivity, 8 101 112. Paul, E. L., McManus, B., & Hayes, A. (2000). "Hookups": Characteristics and correlates of college students' spontaneous and anonymous se xual experiences. The Journal of Sex Research, 37 (1), 76 88. Peterson, Z. D. (2010). What is sexual empowerment? A multidimensional and process oriented approach to adolescent gir l s' sexual empowerment. Sex Roles, 62 307 313. Piscitelli, A. (2007). Shifting boundaries: Sex and money in the North East of Brazil. Sexualities, 10 (4), 489 500. Pizam, A., Jeong, G., Reichel, A., van Boemmel, H., Lusson, J. M., Steynberg, L., et al. (2004). The relationship between risk taking, sensa tion seeking, and the tourist behavior of young adults: A cross cultural study. Journal of Travel Research, 42 (3), 251 260. Pizam, A., Reichel, A., & Uriely, N. (2002). Sensation seeking and tourist behavior. Journal of Hospitality and Leisure Marketing, 9 (3/4), 17 33. Poria, Y. (2006). Tourism and spaces of anonymity: An Israeli lesbian woman's travel experience. Tourism, 54 (1), 33 42.
234 Tourism Geographies, 5 (1 ), 26 38. Poria, Y., & Carr, N. (2010). What have we done and where should we go? In N. Carr & Y. Poria (Eds.), Sex and the sexual during people's leisure and tourism experiences (pp. 181 196). Newcastle upon Tyne: Cambridge Scholars Publishing. Pritchard, A., & Morgan, N. J. (2006). Hotel Babylon? Exploring hotels as liminal sites of transition and transgression. Tourism Management, 27 (5), 762 772. gender, embodi ment and experience. In A. Pritchard, N. Morgan, I. Ateljevic & C. Harris (Eds.), Tourism and gender: Embodiment, sensuality and experience (pp. 1 12). Wallingford, UK: CABI. Pruitt, D., & LaFont, S. (1995). For love and money: Romance tourism in Jamaica Annals of Tourism Research, 22 (2), 422 440. Ragsdale, K., Difranceisco, W., & Pinkerton, S. D. (2006). Where the boys are: Sexual expectations and behaviour among young women on holiday. Culture, Health & Sexuality, 8 (2), 85 98. J. (1993). Women's sexuality and men's appropriation of desire. In Up against Foucault: Explorations of some tensions between Foucault and feminism (pp. 239 264). London: Routledge. Rathus, S. A., Nevid, J. S., & Fichne r Rathus, L. F. (1993). Human sexuality in a world of diversity Boston: Allyn and Bacon. Reiss, I. L. (1964). The scaling of premarital sexual permissivness. Journal of Marriage and Family, 26 (2), 188 198. Roberti, J. W. (2004). A review of behavioral and biological correlates of sensation seeking. Journal of Research in Personality, 38 256 279. Russell, S. T., Van Campen, K. S., & Muraco, J. A. (2012). Sexuality development in adolescence. In L. M. Carpenter & J. DeLamater (Eds.), Sex for life: From virg inity to viagra, how sexuality changes throughout our lives (pp. 70 87). New York/London: New York University Press. Ryan, C. (2000). Sex tourism: Paradigms of confusion? In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 23 40). London: Pinter. Ryan, C. (2003). Risk acceptance in adventure tourism paradox and context. In J. Wilks & S. J. Page (Eds.), Managing tourist health and safety in the new millennium (pp. 55 65). Amsterdam: Pergamon.
235 Ryan, C., & Hall, C. M. (2001). Sex tourism: Marginal people and liminalities London: Routledge. Ryan, C., & Kinder, R. (1996). Sex, tourism and sex tourism: Fulfilling similar needs? Tourism Management, 17 (7), 507 518. Ryan, C., & Martin, A. (2001). Tourists and strippers: Liminal the ater Annals of Tourism Research, 28 (1), 140 163. Sale, J. E. M., Lohfeld, L. H., & Brazil, K. (2002). Revisiting the quantitative qualitative debate: Implications for mixed methods research. Quality and Quantity, 36 43 53. Snchez Taylor, J. (2001). Doll ars are a girl's best friend? Female tourists' sexual behaviour in the Caribbean. Sociology, 35 (3), 749 764. Snchez Taylor, J. (2006). Female sex tourism: A contradiction in terms? Feminist Review, 83 (1), 42 59. Sawicki, J. (1996). Feminism, Foucault, and "subjects" of power and freedom. In S. J. Hekman (Ed.), Feminist interpetations of Michel Foucault (pp. 159 178). University Park, P A : The Pensylvania State University Press. Selnniemi, T. (2002). Couples on holiday: (En)gendered or endangered experience s. In M. B. Swain & J. H. Momsen (Eds.), Gender/tourism/fun (?) (pp. 15 23). Elmsford, NY: Cognizant Communication Corporation. Selnniemi, T. (2003). On holiday in the liminoid playground: Place, time, and self in tourism. In T. G. Bauer & B. McKercher (E ds.), Sex and tourism: Journeys of romance, love, and lust (pp. 19 31). New York: The Haworth Hospitality Press. Sheffield, C. J. (1997). Sexual terrorism. In L. L. O'Toole, J. R. Schiffman & M. L. Kiter Edwards (Eds.), Gender violence: Interdisciplinary p erspectives (2nd ed., pp. 110 127). New York: New York University Press. Theory, Culture and Society, 7 (1), 39 72. Shoemaker, S. (1994). Segmenting the U.S. travel market according to benefits realized. Journal of Travel Research, 32 (3), 8 21. Small, J. (2007). The emergence of the body in the holiday accounts of women and girls. In A. Pritchard, N. Morgan, I. Ateljevic & C. Harris (Eds.), Tourism and gender: Embodim ent, sensuality and experience (pp. 73 91). Wallingford, UK: CABI. Smith, M. (2007). Personality, affect and risky sexual behaviour: An examination of individual differences and their relationship to sexual risk taking. Unpublished master thesis, Acadia Un iversity, Wolfville, Nova Scotia.
236 Snmez, S., Apostolopoulos, Y., Yu, C. H., Yang, S. Y., Mattila, A., & Yu, L. C. (2006). Binge drinking and casual sex on spring break. Annals of Tourism Research, 33 (4), 895 917. St John, G. (2001). Aternative cultural he terotopia and the liminoid body: Beyond Turner at ConFest. The Australian Journal of Anthropology, 12 (1), 47 66. Steiner, C. J., & Reisinger, Y. (2006). Understanding existential authenticity. Annals of Tourism Research, 33 (2), 299 318. Struckman Johnson, C., Struckman Johnson, D., & Anderson, P. B. (2003). Tactics of sexual coercion: When men and women won't take no for an answer. The Journal of Sex Research, 40 (1), 76 86. Swain, M. B. (1995). Gender in tourism. Annals of Tourism Research, 22 (2), 247 266. Sylwester, K., & Pawlowski, B. (2011). Daring to be darling: Attractiveness of risk takers as partners in long and short term sexual relationships. Sex Roles, 64 (9 10), 695 706. Tabachnick, B. G., & Fidell, L. S. (2001). Using Multivariate Statistics Bos ton: Allyn and Bacon. Tashakkori, A., & Teddlie, C. (2003). The past and future of mixed methods research: From data triangulation to mixed model designs. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp 671 701). Thousand Oaks, CA: Sage Publications Thomas, M. (2000). Exploring the contexts and meanings of women's experiences of sexual intercourse on holiday. In S. Clift & S. Carter (Eds.), Tourism and sex: Culture, commerce and coercion (pp. 200 220). London: Pinter. sexual behaviour on holiday. Culture, Health & Sexuality, 7 (6), 571 584. Tolman, D. L. (2003). Gender matters: Constructing a model of adolescent sexual health. The Journal of Sex Research, 40 (1), 4 12. Turchik, J. A. (2007). Identification of sexual risk behaviors among college students: A new measure of sexual risk. Unpublished master thesis, Ohio University, Athens. Turchik, J. A., & Garske, J. P. (200 9). Measurement of sexual risk taking among college students. Archives of Sexual Behavior, 38 (6), 936 948. Turner, J. S., & Rubinson, L. (1993). Contemporary Human Sexuality Englewood Cliffs, NJ: Prentice Hall Inc. Turner, V. (1974). Liminal to liminoid, in play, flow, and ritual: An essay in comparative symbology. Rice University Studies, 60 (3), 53 92.
2 37 University of Florida (2011). University of Florida 2010 2011 Annual Accountability Report. Retrieved from the University of Florida website on May 7 th 2 013. http://www.ir.ufl.edu/oirapps/factbooktest/bog/2010 11/UF_2010 11_Annual_Report_FINAL_REVISED.pdf University of Florida Admissions (201 3). Stats and f acts for p rospective s tudents. Retrieved from the University of Florida website on April 26 th 2013. http://www.admissions.ufl.edu/ufprofile.html Annals of Tourism Research, 37(3), 854 857. Vagle, M. D., Hughes, H. E., & Durbin, D. J. (2009). Remaining skeptical: Bridling for and with one another. Field Methods, 21 (4), 347 367. Valentine, G. (1989). The geography of women's fear. Area, 21 (4), 385 390. Valentine, G. (1992). Images of danger: Women's sources of information about the spatial distribution of male violence. Area, 24 (1), 22 29. Varga, C. A. (1997). Sexual decision making and negotiation in the midst of AIDS: Youth in Kw aZulu Natal, South Africa. Health Transition Review, 7 (3), 45 67. Veijola, S., & Jokinen, E. (1994). The body in tourism. Theory, Culture & Society, 11 (1), 125 151. Wade, L., & Heldman, C. (2012). Hooking up and opting out: Negotiating sex in the first yea r of college. In L. M. Carpenter & J. DeLamater (Eds.), Sex for life: From virginity to viagra, how sexuality changes throughout our lives (pp. 128 145). New York/London: New York University Press. Wang, N. (1999). Rethinking authenticity in tourism experi ence. Annals of Tourism Research, 26 (2), 349 370. Wang, N. (2000). Tourism and modernity: A sociological analysis Oxford: Pergamon Press. Webb, M. C. (2011). Specific sexual risk behaviors of college students and the role of alcohol intoxication in the in tention to participate. Unpublished doctoral dissertation, University of Florida, Gainesville. Wickens, E. (1997). Licensed for thrills: Risk taking and tourism. In S. Clift & P. Grabowski (Eds.), Tourism and health: Risks, research and responses (pp. 151 164). London: Pinter. Wickens, E., & Snmez, S. (2007). Casual sex in the sun makes the holiday: Young tourists' perspectives. In Y. Apostolopoulos & S. Snmez (Eds.), Population mobility and infectious disease (pp. 199 214). US: Springer.
238 Wiederman, M. W. (1999). Volunteer bias in sexuality research using college student participants. Journal of Sex Research, 36 (1), 59 66. Wiederman, M. W., & Hurd, C. (1999). Extradyadic involvement during dating. Journal of Social and Personal Relationships, 16 (2), 265 27 4. Current Issues in Tourism, 11 (2), 167 186. Wimbush, E., & Talbot, M. (Eds.). (1988). Relative freedoms Milton Keynes, England: Open Univer sity Press. Yeoman, I. (2008). Tomorrow's tourist: Scenarios and trends Amsterdam: Elsevier. Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking Cambridge: Cambridge University Press. Zuckerman, M. (2007). Sensation seek ing and risky behavior Washington, DC: American Psychological Association. Zuckerman, M., Eysenck, S., & Eysenck, H. J. (1978). Sensation seeking in England and America: Cross cultural, age, and sex Comparisons. Journal of Consulting and Clinical Psycholo gy, 46 (1), 139 149. Zuckerman, M., Kolin, E. A., Price, L., & Zoob, I. (1964). Development of a sensation seeking scale. Journal of Consulting Psychology, 28 (6), 477 482. Zuckerman, M., & Kuhlman, D. M. (2000). Personality and risk taking: Common biosocial factors. Journal of Personality, 68 (6), 999 1029. Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, M. (1993). A comparison of three structural models for personality: The Big Three, the Big Five, and the Alternative Five. Journal of Persona lity and Social Psychology, 65 (4), 757 768.
239 BIOGRAPHICAL SKETCH Liza Berdychevsky began her Ph.D. program working with Prof. Heather Gibson in the Department of Tourism, Recreation and Sport Management at the University of Florida in 2009. Prior to t his, s he graduated from Ben Gurion University of the Negev in Israel with a Bachelor of Arts in h otel and t ourism m anagement and a Master of Business Administration. Her research interests include: gender issues and identity in leisure and tourism; sexual behavior and risk taking in tourism; leisure, well being, and mental health; and research methods quantitative, qualitative, and mixed. In her dissertation, Liza investigated taking in tourism implementing a sequential mixed methods d esign, starting with a qualitative phase, drawing upon phenomenology, and proceeding to a quantitative phase. In the course of her graduate studies, Liza has gained teaching experience (as a lecturer or teaching assistant) in the following courses : r esear ch m ethods and m arketing r esearch, h otel f inance and h otel a ccounting, t ourism s ociology, i ntroduction to l eisure s ervices m anagement, as well as p rinciples of t ravel and t ourism, and e valuation in l eisure s ervices. Working with Prof. Gibson, Liza has been involved in various research projects, working on different aspects of the research process. While at UF, Liza has worked on three research teams involving scholars from the US, Israel, and UK. These collaborations have resulted in several conference pres entations and publications in leading journals such as Tourism Management Annals of Tourism Research and Journal of Leisure Research