Sexual Permissiveness & Risk Perception Differences between heterosexual and lesbian women in acceptability and risk perceptions associated with sexual permissiveness In many countries, homosexuality is a crime with varying degrees of severity of punishment. In Saudi Arabia and Iran, it is punishable by death. In Jamaica, those convicted are sentenced to 10 years hard labor. In Malaysia, a jail sentence awaits. The United Nations General Assembly, which heard two opposing statements on lesbian, gay, bisexual, transsexual, and intersex (LGBTI) rights on December 18, 2008, shed some light on prevalent global sentiment. It was argued that discrimination, violence, criminal sanctions, and abuse based on sexual or gender identity violate LGBTI human rights. Navanethem Pillay, the United Nations High Commissioner for human rights, commented that laws against homosexuality “are increasingly becoming recognized as anachronistic and as inconsistent both with international law and with traditional values of dignity, inclusion and respect for all” (Macfarquhar, 2008). The response to LGBTI rights among member countries, however, was mixed. Sixty-six nations supported such rights, 57 were against them, and 69 abstained. However, on closer examination, the picture for Asia was more consistent. In Asia, only Armenia, Japan, Nepal, and Timor-Leste supported LGBTI rights. Indonesia, Malaysia, and Brunei opposed. The rest of the South East Asian countries, including Singapore, abstained. Therefore, by their abstinence, the majority of countries in Asia showed a sense of ambivalence towards LGBTI rights. The stance taken by Singapore in the recent UN General Assembly is reflective of its population’s mixed response to the gay and lesbian community. An unsuccessful attempt to 1 Sexual Permissiveness & Risk Perception decriminalize sex acts between two men (section 377A in the penal code) in 2007 brought this issue to the forefront. Under section 377A, it is stated, “penetration is sufficient to constitute the carnal intercourse necessary to the offence described in this section”. Opponents of the repeal of section 377A had spoken against the decriminalization of sex between men. In response to the fierce debate on both sides, the government opted to remain with the status quo (Ramesh, 2007, September 22). Prime Minister Lee Hsien Loong said, “the tone of society, the public, and society as a whole, should be really set by the heterosexuals and that’s the way many Singaporeans feel… what people do in private is their own business; in public, certain norms apply” (Ramesh, 2007, September 22). Therefore, sex acts between males continue to be criminalized, even though there remain strong feelings among both camps. Unlike their gay counterparts, lesbian women were left out of the penal code and newspaper articles related to the acquired immune deficiency syndrome (AIDS). This did not mean, however, that they were spared from being portrayed in a negative light. Some religious groups, for instance, warned that living a lesbian lifestyle would end in misery because of greater promiscuity in the lesbian community (e.g., Harley, 2003). In a brochure entitled “Straight Talk”, given out to secondary school students in Singapore, significant numbers of lesbian women were said to exchange sex for money and drugs (Harley, 2003). It was also written that “lesbian relationships are equally unhealthy and just as life-threatening as gay male relationships”. The sexual health consequences of lesbian activities were emphasized. 2 Sexual Permissiveness & Risk Perception Promiscuity has been defined as irresponsible sexuality, such as having unprotected sex with multiple sexual partners, with adverse health consequences such as sexually transmitted infections (STIs) and unwanted pregnancy (Gershman, 1997; Markey & Markey, 2007; Widom & Kuhms, 1996). Sexual permissiveness appears to cover any sexual behavior that falls outside of a committed marriage. Examples of sexual permissiveness are onenight stands and having threesomes. Certain behaviors are more ambiguous however. For instance, some people perceive premarital sex within committed dating partners as acceptable, but there are others who view the same as sexual permissiveness. Therefore, promiscuity appears to be broadly captured by the concept of sexual permissiveness coupled with irresponsibility and adverse health consequences. The chief emphasis of many current safe sex campaigns is on responsible sex. While safe sex campaigners advocate being faithful to one sexual partner, their work among female sex workers stresses protective measures so that adverse consequences can be avoided. Thus, while promiscuity includes sexual permissiveness, sexual permissiveness can be meaningfully distinguished from promiscuity in that the latter implies both irresponsibility and adverse health consequences, while sexual permissiveness can exist with responsibility (i.e., having multiple partners but always practicing safe sex) that eliminates or at least minimizes adverse health consequences. Are lesbian women sexually permissive? While there is a paucity of research studies on sexual behaviors and attitudes of lesbian women in Asia, research conducted outside of Asian countries suggests that lesbian women can be sexually permissive. In Morrow and Allsworth’s (2000) survey study on sexual risks in 3 Sexual Permissiveness & Risk Perception lesbian women and bisexual women, 13% of their lesbian sample (total lesbian sample size was 436) were then sexually inactive, 70% had a single sexual partner, and 17% had multiple partners. Of those who had been sexually active in the previous year, 30% of their lesbian participants ever had multiple female sexual partners. In Montcalm and Myers’ (2000) study, lesbian participants reported between none to 110 sexual partners in the previous year, with 72% having had only one sexual partner. Therefore, while the majority of lesbian women in these two samples prefer monogamy, approximately a quarter of them do engage in sex with multiple partners. Perceived risk influences the use of safe sex methods. Lesbian women who had a history of STIs, had a friend with human immunodeficiency virus (HIV), or perceived greater risk for the self were more likely to be concerned about STIs and practice safe sex (Diamant, Lever & Schuster, 2000; Montcalm & Myer, 2000). However, it is likely that Singaporean lesbian women think that they are impervious to the effects of STIs. Newspapers frequently allude to HIV and AIDS as a gay man’s illness (e.g., Wong, 2008). Reporting on infected heterosexuals tends to be male-centric, such as a focus on males who participate in the Batam (Indonesia) sex trade. In such instances, married women are portrayed as victims, being infected as a result of and without knowledge of their husbands’ extramarital sexual stints. HIV and AIDS are not typically associated with lesbian women in local news reporting, possibly contributing to the perception that lesbian women are invulnerable. This perception, where many lesbian women see themselves as “safe” from STIs, has been termed “lesbian immunity” (Montcalm & Myer, 2000), which is a perception 4 Sexual Permissiveness & Risk Perception of low or no sexual risk when the sexual partner is a fellow female. Studies conducted in other countries showed that many health practitioners subscribed to this misconception as well (Dworkin, 2005). Local HIV and AIDS infection statistics may also contribute to this perception of lesbian immunity. From Singapore’s Ministry of Health’s statistics on HIV infections or AIDS (Ministry of Health, 2008), the trend, from 1985 to 2007, showed that infected females were increasingly outnumbered by infected males. The number of women who were infected made up approximately 10% of the number of infected men over the last three years. In the year 2007, 31 women were affected as compared to 392 men. In the year 2006, 32 women were affected as compared to 325 men. In 2005, the figures were 30 and 287 for females and males, respectively. Between 2005 and 2007, the odds of married heterosexual women being infected were more than twice that of single women (the statistics for single women do not distinguish between heterosexual and lesbian women). These small numbers of female infections may lead Singaporean lesbian women to perceive themselves as not at risk, since they are female and predominantly unmarried (as same-sex marriage is not allowed in Singapore). Because of this general perception of invulnerability, lesbian women who are aware of safer sexual behaviors may therefore not utilize, even with multiple partners, protective measures (such as dental dams, which are rectangular sheets of latex placed over the genital region during oral stimulation of the genital, or rubber gloves, which are used during finger penetration of the vagina) during risky sexual activities 5 Sexual Permissiveness & Risk Perception (Fishman & Anderson, 2003; Richardson, 2000). Thus, the perception of lesbian immunity may actually be a basis for lesbian women not only being more permissive (having multiple partners) but also being irresponsibly permissive (having unsafe sex with multiple partners). Contrary to this perception of immunity, the wide variety of sexual activities that lesbian women engage in may expose them to STIs. In Montcalm and Myer’s (2000) sample of 248 lesbian participants, the majority practiced receiving digital-vaginal sex (where a partner’s finger is inserted into one’s vagina) (84.3%), giving digital-vaginal sex (where one inserts a finger into a partner’s vagina) (87.1%), receiving oral-vaginal sex (where a partner orally stimulates one’s vagina) (79.0%) and giving oral-vaginal sex (where one orally stimulates a partner’s vagina) (81.5%). Sharing of sex toys such as dildos (33.9%), receiving digital-anal sex (where a partner’s finger is inserted into one’s anus) (23.4%), giving digital-anal sex (where one inserts a finger into a partner’s anus) (24.2%) and fisting (inserting a hand or all fingers into the vagina or anus) (less than 10% each for vaginal and anal fisting) were much less popular. About one-quarter of the women had sex involving genital contact during their partners’ menstruation. A similar preponderance of sexual activities was found in Morrow and Allsworth’s (2000) sample of 504 lesbian women, with approximately 90% of participants practicing oral-vaginal and digital-vaginal sex. Although HIV infections and AIDS are much lower for unmarried females (which would include most lesbian women), exposure to mucous membranes, vaginal secretions, and menstrual blood increases the risk of acquiring STIs such as 6 Sexual Permissiveness & Risk Perception herpes and chlamydia (Fishman & Anderson, 2003; Marrazzo, Coffey, & Bingham, 2005). Some lesbian women acquire female bisexual partners who have had multiple male sexual partners before, thus increasing their risk of STIs (Dworkin, 2005) such as trichomoniasis and anogenital warts which can affect women who never had sex with men (Morrow & Allsworth, 2000). Once infected, complications are greater in women as compared to men, and women face a wider variety of problems, such as infertility and chronic pelvic pain (Diamant, Lever, & Schuster, 2000). Thus, even if the risk for women is indeed lower, it does not mean there is no risk at all. As mentioned earlier, sexual activities that involve infected vaginal fluid and menstrual blood coming into contact with mucosal membranes, open sores, or cuts increase one’s susceptibility to STIs. Therefore, lesbian immunity is really an illusion; lesbian women do not possess immunity to STIs. If lesbian women are indeed more sexually permissive and irresponsibly so because of the perception of immunity even though they really are not immune, this can have important implications for the sexual health of the lesbian community. The main objective of this thesis was therefore to examine perceptions associated with sexual permissiveness, focusing on acceptability (how acceptable) and risk (how risky) ratings associated with such permissiveness among lesbian women, and to contrast these acceptability and risk perceptions against those of heterosexual women. In many studies on lesbian women, lesbian women were recruited on the basis of their self-identification as lesbian women (e.g., Montcalm & Myer, 2000; Morrow & Allsworth, 2000). This self-identification approach can be problematic because 7 Sexual Permissiveness & Risk Perception self-identified lesbian women may not include only lesbian women but also those who are bisexual as well. According to Diamond (2005b), women with attractions to both genders are more likely to change their sexual identity over time because, after acknowledging same-sex attractions, they have the additional step of weighing how strongly they lean towards each gender. Thus, some self-identified lesbian women may actually be bisexuals in the making. Indeed, in Rust’s (1993) retrospective study, 75% of bisexual respondents once identified themselves as lesbian. Alternatively, there may be bisexual respondents who identify themselves as lesbian women because they have a predominant interest in women even though they are attracted (to different extents) to both genders (Diamond, 2005b). Many self-identified lesbian women have also had sex with men (Champion, Wilford, Shain, & Piper, 2005). Thus, self-identification as a lesbian women appears to function more as a tool for making sense of current sexual attractions (Weinberg, Williams, & Proyer, 1994). Additionally, some women may maintain a lesbian identity for reasons irrelevant to their sexuality. For instance, one woman identified as herself as a lesbian simply because of a dislike for the heterosexual culture (Diamond, 2005b). To enhance knowledge and understanding, it is meaningful to make distinctions between bisexual women and lesbian women (Gangestad, Bailey, & Martin, 2000). Diamond’s (2005a) study on subgroups of sexual minority women involving “stable lesbians” (consistently identifying as lesbian), “fluid lesbians” (changing between sexual identities), and “stable nonlesbians” (never adopted lesbian sexual identity, bisexual or unlabeled) is a step forward in this regard. Such a distinction is useful in that it 8 Sexual Permissiveness & Risk Perception was found that “stable lesbians” had the smallest fluctuations in their emotional and physical attraction to the same sex (Diamond, 2005a). In this light, studies that recruit lesbian women may not really have only lesbian women in their samples, which would raise questions regarding the generalizability of their results to lesbian women. One solution to this problem, which would also enhance generalizability to lesbian women, is to flesh out the term “lesbian” in terms of specific criteria, so that potential participants are not left to their own interpretations of who a lesbian is. This therefore was the approach adopted in this thesis. Specifically, lesbian women were defined as women who experienced emotional and physical attractions towards members of the same sex only. Such an approach reduced the lesbian sample to only a subgroup of WSW (women who have sex with women) and helped to sift out bisexual women as well as those who identified themselves as lesbian for reasons irrelevant to their sexuality. In specifying the criteria associated with the term “lesbian” during recruitment, potential participants were therefore provided with objective markers to determine their eligibility for participation. In many studies on lesbian women, heterosexual women were not included as a comparison group. While this may be seen as an acknowledgement of or statement of intent that lesbian women are a group worthy of study in and of themselves, there may be inadvertent knowledge gaps that are created regarding what actually differentiates heterosexual women and lesbian women. From a pragmatic point of view, when comments are made about lesbian women’s sexuality (such as those by religious groups), such comments typically pitch the perceived deviant (lesbian 9 Sexual Permissiveness & Risk Perception women) against the socially acceptable (heterosexual women). Although this may be perceived as biased or prejudiced, heterosexual women do offer a familiar reference point for understanding lesbian women. By asking the same questions of both groups, it is then possible to accurately quantify both differences and similarities between the two groups, and in so doing actually provide for a better understanding of lesbian women. This thesis therefore examined perceptions associated with sexual permissiveness in local lesbian women, with local heterosexual women as a comparison group. Sexual permissiveness is often conceptualized in terms of actual behavior (the number of partners one has). However, such an approach can be severely limiting in recruiting participants, especially in an Asian context where people tend to be more conservative, such that fewer people may actually have multiple partners or are sufficiently bold to be forthcoming about the number of partners they have. To circumvent this, this thesis assessed sexual permissiveness from an attitudinal (how acceptable) perspective and a perceptual (how risky) perspective. Cross-cultural studies have been done examining sexual attitudes of French and American participants, where the French were found to be more permissive (LeGall, Mullet, & Riviere-Shafighi, 2002). There were also gender comparative studies that found women to be less sexually permissive (Hendrick & Hendrick, 1987). Risk perceptions, as examined in this study, pertain to the level of perceived risk associated with sexual permissiveness. It should be noted that perceptions of risk need not correspond with actual risk levels. It is perceived risk, however, that typically influences behavior rather than actual risk levels. As noted 10 Sexual Permissiveness & Risk Perception previously, lesbian women who were conscious of sexual risks were more likely to utilize safe sex methods. Therefore, risk perceptions, not actual levels of risk, were of interest in this thesis. Overview of the Current Research This thesis covered 2 studies. The first study examined terms relevant to sexual permissiveness. Examining sexual permissiveness encompassed understanding perceptions of relationships, sex, and virginity. Focus group discussions with heterosexual and lesbian women were employed towards this end. The second study sought to tease out the nuances in acceptability and risk perceptions when different sexual orientations (specifically heterosexual or lesbian), genders and sex types (such as penile-vaginal intercourse or kissing) come into play. This attempt took the form of a survey questionnaire. The terms in the questionnaire were clarified using the results of study 1, so as to maximize similar interpretation of questionnaire items. Study 1 Before examining differences between lesbian and heterosexual women in how they perceived acceptability and risk in sexual permissiveness, it is important first of all to determine whether heterosexual and lesbian women perceive relationships, sex, and virginity similarly. This includes definitions and purposes, such as what it means to be and the purposes of being in a relationship, having sex, as well as the importance of virginity (virginity was included because previous studies had linked it to sexual permissiveness (Pitts & Rahman, 2001). Such a first step is necessary to ensure that when comparisons are subsequently made concerning perceptions, there 11 Sexual Permissiveness & Risk Perception can be confidence that these perceptions apply to concepts and terms that are either interpreted similarly by both groups of women or are known to have different interpretations for the two groups of women. A key question for the two groups of women is the question of what constitutes sex. Answers here have the potential of revealing common sexual activities that both groups of women are familiar with. In doing so, the accompanying perceptions of acceptability and risk of these common sexual activities can then be assessed with confidence. Eliciting responses on the purposes of being in a relationship and having sex might also shed light on differences between lesbian and heterosexual women that could contribute to the negative stereotype of promiscuity. Knowing the purposes can also highlight the boundaries and expectations of being in a relationship for these two groups of women. Available literature suggests that there are possible differences between heterosexual and lesbian women in how they view relationships, sex, and virginity. Heterosexual and lesbian women, for instance, are likely to differ in their characteristics of what a romantic relationship is. There is evidence that heterosexuals equate monogamy with commitment and non-monogamy with infidelity (Smart, 2006). On the other hand, those with same-sex attractions recognize sexual non-exclusivity, which is common among gay-men coupling (Smart, 2006). Therefore, lesbian women are likely to possess a broader definition of “relationship”, recognizing non-monogamous relationships as valid, as compared to their heterosexual counterparts. Given such a discrepancy, lesbian women may consider other forms of relationships when responding to the term “relationship”. This is 12 Sexual Permissiveness & Risk Perception problematic because infidelity is a violation of boundary agreements set between relationship partners. For heterosexuals, these markers are likely to be fixed (non-monogamy equates infidelity). For lesbian women, these markers are likely to be negotiable (non-monogamy may not be a sign of infidelity, depending on prior agreements). Promiscuity, especially that leading to truncated relationships, carries connotation of violation of boundaries. If such a difference in how relationships are viewed surfaced among local lesbian and heterosexual women, there would be a need to take this difference into account before any meaningful comparison could be made in terms of how acceptability of and risks associated with sexual permissiveness are perceived. Secondly, heterosexual and lesbian women may define sex and virginity differently according to different predominant sexual activities, different gender of partner, and presence of orgasm. This has implications for use of the term “sex” for each group of women. Lesbian women have been reported to engage in oral-vaginal sex, oral-anal sex, digital-vaginal sex, digital-anal sex and tribadism (rubbing of one’s genitals against partner’s body part) (Fishman & Anderson, 2003; Morrow & Allsworth, 2000). Penilevaginal intercourse is naturally absent from lesbian women’s repertoire. For heterosexual women, however, penile-vaginal intercourse is possibly the gist of “having sex”. In Pitts and Rahman’s (2001) study, an overwhelming percentage of female participants rated penile-vaginal intercourse as sex, compared to kissing (less than 3%) or oral (less than 35%) or manual (less than 20%) genital stimulation. Trotter and Alderson (2007) found similar results, with their female participants most in agreement 13 Sexual Permissiveness & Risk Perception that penile-vaginal intercourse is sex (more than 87%), compared to deep kissing (less than 4%) or oral (less than 49%) or manual (less than 28%) simulation. Some studies suggest that definitions of sex and virginity loss are defined according to one’s sexual experiences (Bersamin, Fisher, Walker, Hill, & Grube, 2007). For example, female and male heterosexuals were likely to define sex as penile-vaginal intercourse (Trotter & Alderson, 2007). However, other studies found no relationship between sexual experience and definitions of sex. Bersamin et al. (2007), for instance, found among their predominantly heterosexual participants that it was other factors such as gender of partner, length of dating, and presence of orgasm that affected definitions of sex. Specifically, their participants (a) defined a sexual partner as being of an opposite-sex partner, (b) were also more likely to consider a larger number of activities as sex if they were in a dating relationship with the partner for three months, compared to a partner on a single date, (c) were more likely to consider as sex oral sex with orgasm than oral sex without orgasm. Interestingly, however, orgasm was not a criterion for sex when it came to penile-vaginal or penile-anal intercourse. Participants also appeared to have different yardsticks for judgment for oppositesex and same-sex partners. For example, they were more likely to consider oral and manual genital stimulation as sex if the partner was of the same-sex. Thus, although sexual experience did not relate to definitions of sex here, it appeared that gender of partner was nonetheless an important consideration, as is the type of sexual activity being considered. It should be noted that for these two studies, the number of lesbian participants were 14 Sexual Permissiveness & Risk Perception not sufficiently large (less than 3% in Pitts & Rahman’s (2001) sample and less than 2% in Trotter & Alderson’s (2007) sample) to allow for firm conclusions, so that it remains unclear if lesbian and heterosexual women do differ in how they define sex and virginity. The relationship between sex and virginity is also ambiguous (Bersamin et al., 2007). Chambers (2007) showed from her sample of female and male heterosexual college students from the University of Georgia that perceptions towards what constitutes sex and virginity loss differed across the participants. For example, 41.7% of self-identified virgins had engaged oral sex and 39.1% of those who engaged in oral sex but not sexual intercourse called themselves non-virgins. This meant that participants who had oral sex only may or may not classify themselves as virgins, indicating disparate definitions of virginity. Participants often described virginity loss in the context of penile-vaginal intercourse, although some also consider other types of sex as long as genitals were involved (Carpenter, 2001; Trotter & Alderson, 2007). Pitts and Rahman (2001) found that female participants, as compared to male participants, had a narrower definition of sex and made mention of the notion of “technical virginity” (such that one engaging in sexual activities other than penile-vaginal intercourse remains a virgin). There is likely to be more pressure on young unmarried women to retain their virginity. Non-virgin status might be construed as indicating sexual promiscuity. In sum, study 1 was done to examine and compare how lesbian and heterosexual women viewed relationships, sex, and virginity. Method 15 Sexual Permissiveness & Risk Perception Participants The participants consisted of 38 heterosexual and 38 lesbian women. Participants were all English-speaking. There were no prerequisites set for participation regarding experience in relationships and sex because views and opinions were the focus of this study. Heterosexual women were acquired from the Research Participants pool of the Department of Psychology at the National University of Singapore. Their ages ranged from 19 to 31 years (M = 21.40, SD = 2.30). Lesbian women were recruited via posters, from local online portals such as Sayoni.com, and online mailing lists such as Signel yahoogroups. Their ages ranged from 21 to 34 years of age (M = 25.40, SD = 3.70). Lesbian participants were older than heterosexual participants, t(74) = 5.67, p < .01. Participants were predominantly Chinese, in the midst of acquiring or already possessing a degree, and Christian/Catholic, Buddhist, or without religion. Table 1 provides a summary of the demographics of the participants in this study. 16 Sexual Permissiveness & Risk Perception Table 1 Demographics Characteristics of Heterosexual (mean age = 21.40, SD = 2.30) and Lesbian (mean age = 25.40, SD = 3.70) Participants in Focus Group Discussions Heterosexual n Lesbian Percentage n Percentage Race Chinese 27 71.1 35 92.1 Indian 6 15.8 2 5.2 Malay 0 0 1 2.6 Others 5 13.2 0 0 O levels 0 0 1 2.6 Diploma/ 0 0 6 15.8 37 97.4 29 76.3 1 2.6 2 5.2 7 18.4 15 39.5 10 26.3 5 13.2 Hindu 2 5.2 0 0 Islam 3 7.9 1 2.6 Others 1 2.6 2 5.2 Education Level A levels Undergrad Postgraduate Religion Christian/ Catholic Buddhist 17 Sexual Permissiveness & Risk Perception No religion 15 39.5 15 39.5 18 Sexual Permissiveness & Risk Perception Procedure Focus group discussions were utilized to explore participants’ opinions. This method has an added advantage over one-to-one interviews because it allows group interaction and participants to clarify each other’s opinions by asking questions, making comments, and exchanging stories. This, in turn, allows for the gathering of a potentially richer set of data. Permission was first sought from the NUS Institutional Review Board (IRB) to conduct the study. Upon obtaining IRB approval, recruitment for the focus group discussions was done. Heterosexual women and lesbian women were recruited to participate in separate focus group discussions. This was done to take into account the ambivalence towards homosexuality in Singapore and to minimize any sense of discomfort that may arise from the different sexual orientations. Participants signed up for sessions according to their availability. The discussion itself was semi-structured. Specific questions were asked in the same sequence in all sessions (see Appendix A). However, wherever needed, additional questions were asked for clarification or elaboration. The present author facilitated all sessions. To prevent a minority from dominating a group discussion, each participant was given a chance to speak for each of the structured questions drawn up for discussion. These sessions took place in various locations, including tutorial rooms and quiet coffee places, according to the convenience of participants. Each focus group discussion lasted between 1 to 2 hours, with group sizes ranging from 4 to 10. 19 Sexual Permissiveness & Risk Perception Seven sessions were conducted for heterosexual women and 6 were conducted for lesbian women. Before each session, participants were briefed on the purpose of the study, after which they signed the consent form (indicating voluntary participation) and then provided some demographic data. Participants were encouraged to speak freely and permission was obtained from participants for the session to be audio-recorded. The sensitive nature of the study was acknowledged and participants were told that their responses were anonymous and their names would not be used in any report. After each session, participants were thanked. Heterosexual participants were given participation points for partial fulfillment of their course requirements and lesbian participants were given beverages as tokens of appreciation. Results Analytic Approach The present author tape-recorded the focus group discussions to retain exact content. Transcription of tapes to written text was then done. That is, participants’ words were written retaining its original form, including the use of colloquial English, commonly known as “Singlish” in Singapore which is characterized by non-words such as “la” and “lo”). These words did not alter the meanings of the sentences, which were decipherable through the English words they accompanied. For transcription, ellipses (…) were used to indicate pauses. In defining a meaningful unit for analysis, the smallest text component is a full sentence and the biggest text component is a paragraph (Laurenceau, Barrett, & Rovine, 2005; Sanderson, Keiter, Miles, & Yopyk, 2007). A full sentence is defined as an 20 Sexual Permissiveness & Risk Perception expression of one idea, contained within one or a few phrases. A paragraph is defined as a number of sentences that are of the same idea, such as elaboration of the same point in a continuous flow. Questions were divided into distinct segments of the focus group discussions (Schilling, 2006), namely, relationship, sex, and virginity. Responses were thus separated according to segments. Under each segment, responses were further separated according to sub-questions, such as definition and purpose. Categories from responses were then generated for each subquestion. To generate categories, responses were first recoded to reduce material and yet retain the crux of the contents. This included deleting unnecessary words and transforming sentences to short forms (Schilling, 2006). For example, a heterosexual participant said this in response to expectations of a relationship partner: “Someone you can completely be yourself with, like with your bestest bestest best friend, there will be some secrets you will not able to share, but you know you can tell him everything and everything is going to be okay… might fight or whatever… but at the end he will still be there for you.” From this paragraph, the quality “best friend” was placed as one category. Subsequently, responses that included “best friend” defined as “someone you can share everything with” were placed together. Although numerical comparisons are at best tentative in focus group analysis, such comparisons can point to substantive issues. As an extreme example, if all of the lesbian focus groups were to raise a particular point while none of the heterosexual groups were to raise that same point, such a difference in 21 Sexual Permissiveness & Risk Perception numbers may be indicative of important differences between lesbian and heterosexual women. The use of numbers and numerical comparisons was thus adopted, although caution was exercised not to over-interpret the numerical differences. Tapes were reviewed twice to check for consistency and to minimize errors. A second heterosexual male Psychology student (not from the Research Participant pool) also coded one heterosexual and one lesbian group discussion to ensure the reliability of coding. Inter-rater agreement was computed here as total number of agreements minus total number of disagreements, divided by total number of codes. The inter-rater agreement was 91.4%. All names used in the following are pseudonyms to protect participants’ confidentiality. Defining Relationship For all groups, participants were asked: “What is a relationship?” Participants in both lesbian and heterosexual groups defined a relationship as romantic involvement with a partner. There were, however, some differences between groups when discussing relationship characteristics. Monogamy and commitment were generated from all discussion groups, but 3 out of the 6 lesbian groups generated “not necessary monogamous” as a characteristic of a relationship whereas none of the heterosexual groups did. The heterosexual and lesbian women groups also had slightly different ideas about commitment. Heterosexual participants explained commitment as monogamy. For example, Pat (a heterosexual participant) stated, “For me, being in a relationship means… like ultimately commitment… you know… you 22 Sexual Permissiveness & Risk Perception cannot be allowed to go on a date with some other girl… you know… or be too close with some other girl… and for me as well.” Lesbian participants, on the other hand, made a distinction between commitment and monogamy. Commitment was defined as working problems out together and making time for each other. Nic, a lesbian participant, illustrated this as follows, “Commitment is like… like… I know this is our relationship, we are going to try it out… even if we have problems, we are going to work it out between ourselves.” The lesbian women also recognized open relationships (in which partners are allowed to be emotionally and/or sexually involved with more persons outside of the relationship) as valid forms of romantic relationships but did not endorse such relationships. Sharon (a lesbian participant) shared as follows, “Like for me, as long as you and her are on the same page it’s fine… it depends on the kind of partner you are bringing home… how many times you have sex with that person… whether that person is healthy, clean. Basically you have to lay all the cards on the table… not just having random sex, orgies… I am more into… well, I believe in open relationships… but at the same time, I am more into monogamy. That’s my personality.” Purpose of Being in a Relationship The question, “What is/are the purpose(s) of being in a relationship?” was asked. Four reasons were generated here: emotional intimacy (“best friend”), physical intimacy (“making love”), marriage, and companionship/support. All groups generated “emotional intimacy”, which they defined as “being able to share everything”, as a reason for being in a relationship and frequently used the term “best friend” to describe this capacity 23 Sexual Permissiveness & Risk Perception to share freely with one’s partner. Nur, a lesbian participant, shared, “Because my soul mate is my very best friend… I am very sure of that. To me it’s like… someone who knows you the most… someone who will be there for your whole life. We know each other very very well.” Indeed, “best friend” was cited as a prerequisite quality for entering a relationship for 5 out of 7 of the heterosexual and 4 out of 6 of the lesbian groups. “Physical intimacy”, which participants defined as sexual activity with emotional attachment involved (“making love”), was commonly raised as a reason for being in a relationship. This purpose of physical intimacy appeared more salient for the lesbian participants. Four of the 6 lesbian groups, compared to 3 out of 7 heterosexual groups, generated physical intimacy as a reason for being in a relationship. Heterosexual participants expressed the notion that women were not as sexual as men. One heterosexual participant, Kathleen, conveyed, “I don’t think that a woman… sexually hungry as a man… I don’t think we have that kind of drive…” On the other hand, the heterosexual groups were more likely to cite marriage as a reason for entering a relationship, with 3 out of 7 groups raising this purpose compared to only 1 lesbian group. Even here, it appears that lesbian women may differ in how they define marriage. Ami, a lesbian participant, expressed, “You don’t have to be married to be married. Canada is probably different but in Singapore you wouldn’t be able to do that right… you can probably have your own ceremony and it will be as meaningful…” Marriage in this context appears defined as going through some kind of traditional or customary marriage ceremony, 24 Sexual Permissiveness & Risk Perception where parents receive cups of tea as respect, rather than obtaining some legal document. There were suggested commonalities in the purpose for being in a relationship. Five of the 7 heterosexual groups and 4 out of the 6 lesbian groups raised “companionship/support” as a reason. A heterosexual participant, Lisa, communicated “I guess being in a relationship means you are sharing your life with somebody… it’s like what’s happening in my life, then I share with you, then he knows what’s happening in my life… be like the supportive role for you.” A lesbian participant also spoke of support as follows, “someone to support me through my bad times and someone to share my happiness la… literally la. I need that person to be involved in my life… and I would like to be involved in that person’s life.” Defining Sex The question “What is sex?” was asked and participants were likely to cite examples that were relevant to them. For example, all 7 heterosexual groups brought up penile-vaginal intercourse as sex compared to just 4 out of the 6 lesbian groups. On the other hand, more lesbian groups than heterosexual groups brought up oral sex (all of the lesbian groups vs. none of the heterosexual groups), digital sex (all lesbian groups vs. 4 out of 7 heterosexual groups) and external genital stimulation (4 out of the 6 lesbian groups and none of the heterosexual groups). Only one heterosexual group and one lesbian group cited anal sex. All heterosexual groups and 5 out of the 6 lesbian groups mentioned that orgasm was not a necessary criterion for sex. However, 3 out of the 6 lesbian groups generated “yes, orgasm is a criteria for sex”, compared to none of the heterosexual groups. 25 Sexual Permissiveness & Risk Perception These seemingly inconsistent numbers for the lesbian groups could reflect differing opinions among lesbian women. Purpose of Having Sex The question “What is/are the purpose(s) of having sex?” and participants differentiated between having sex and making love when giving their reasons for having sex or making love. Having sex was described as sexual activity without emotional involvement. Making love was described as sexual activity with emotional involvement (love). For having sex, all participants cited physical gratification as the main purpose and physical attraction as the pre-requisite quality. Five out of the 7 heterosexual groups generated “thrill” or “fun” for purposes of having sex, while 5 out of the 6 lesbian groups did so. For making love, participants indicated that they would only sleep with their relationship partners, indicating that the qualities for a partner with whom they make love are the same as the qualities they seek in a relationship partner. All participants groups mentioned expression of love as a purpose here. Lesbian groups were more likely to cite “satisfy partner” as a purpose of lovemaking as compared to heterosexual groups (5 out of 6 lesbian groups and 3 out of 7 heterosexual groups). In their elaboration, they brought up the “prevalence of one-way sex” among lesbians, where one party receives sexual pleasure and the other party gives sexual pleasure without reciprocation. Jocie (a lesbian participant) spoke as such, “Especially when your partner wants it and you are having your menstruation, then you give your partner la and vice verse. Or when I am very tired from work... but she wants it… then do lo. After that sleep.” Ying, another lesbian participant, said, “Low mood for the person who’s 26 Sexual Permissiveness & Risk Perception performing the one-way. Low mood means you don’t feel like having sex but you want to satisfy your partner.” There appeared to be greater differences in purpose for having sex for the reasons of “bring relationship to next level” and “to procreate”. Three of the 7 heterosexual participant groups brought up “bring relationship to next level” but none of lesbian groups did so. This may be attributed to more prolonged dating periods and later marriages nowadays as compared to older times, such that heterosexual couples use sex (traditionally reserved for marriage) as a marker of greater intimacy (Carpenter, 1998). For lesbian couples, this boundary between the dating period and a more committed period such as marriage is likely to be less clear, as marriage was not an option until recent days. As for “to procreate”, 3 out of the 7 heterosexual groups brought this up but none of the lesbian groups did so. This is most likely because lesbian women cannot procreate with their partners. Three out of the 6 lesbian groups indicated that they would need to know if their sex partner was “clean” (i.e., free from STIs) when participating in sex without emotional involvement. In contrast, only 2 out of the 7 heterosexual groups brought this up. This could be a result of the recent promotion of AIDS awareness program by AFA’s WSW branch (Action for Aids, Women who have sex with women). Defining Virginity Participants were asked: “What is virginity?” and “Is virginity important?” Participants defined virginity loss in a few ways: presence of a partner as pre-requisite, tearing of the hymen, and sexual experience. All groups generated these 27 Sexual Permissiveness & Risk Perception definitions. Definitions that were very stringent included sexual experience of any nature, such as masturbation and sexual fantasies. It was unanimously agreed by all participants that penilevaginal intercourse would render someone a non-virgin. Some lesbian participants claimed that they did not endorse the medical definition for virginity (which is having one’s hymen intact). Heterosexual participants mentioned the importance of mutual genital involvement, including penile-vaginal intercourse, for transition to non-virgin status. Participants perceived virginity as “important”, “unimportant”, or “not relevant at all”. Some heterosexual participants (found in 4 out of the 7 heterosexual groups) found virginity (on the part of both parties) to be important and sacred, and expressed that sex should be reserved for marriage. One heterosexual participant, Feng, shared, “If your boyfriend is a virgin, I think it’s respect… This is the person who you want to spend the rest of your life with… Giving your virginity is a good gift to your partner.” Other heterosexual participants (found in 5 out of the 7 heterosexual groups) reported virginity to be unimportant, indicating differing views among the heterosexual participants. On the other hand, only 2 out of the 6 lesbian groups found virginity important and none of the lesbian participants expressed that sex should be reserved for marriage. There appeared to be a differentiation between virginity loss and sexual permissiveness. Sex within the context of a relationship was considered acceptable among participants. Aisha (a heterosexual participant) explained: “There is a difference between being a virgin and being promiscuous… I… I think… 28 Sexual Permissiveness & Risk Perception basically, people tend to think that if you lost your virginity, you are promiscuous… you had a lot of sex with a lot of people. I think it’s important (to differentiate) between the two… you probably don’t want a very loose partner… but I don’t see why you should have a problem with having a non-virgin… probably she really did love him.” For lesbian women who suggested that virginity is important, their argument was linked to sexual permissiveness as well, not virginity per se. This is similar to that made by heterosexual women who perceived virginity to be unimportant. Irene, a lesbian participant, stated, “I think it’s (virginity) important… in the sense as… how was their attitude towards it… when they you know… if it’s like anything anything one… then I will be wa lao (a colloquial term to indicate disapproval or displeasure). It’s important how they treat it at first. Not like die die must be virgin, married then you can. Must still treat it as something sacred… for the right person.” All lesbian groups suggested that the concept of virginity is borne out of a patriarchal society. A lesbian participant, Jer, said, “I think for me… like ya… Over-rated? Because I think virginity is a social construct. It’s created by a patriarchal system to perpetrate certain norms. I mean there is no definition… to me… I don’t really have a definition.” Discussion On the whole, heterosexual and lesbian participants had similar as well as dissimilar views on relationships, sex, and virginity. Even within similar views, however, there were nuances within the groups suggesting that terminologies may need to be better specified. For instance, while both groups desired commitment and monogamy in their relationships, they explained 29 Sexual Permissiveness & Risk Perception these terms differently. Heterosexual participants saw commitment as monogamy, but lesbian participants saw commitment and monogamy as distinct. Lesbian participants were cognizant of nonmonogamous relationships as valid, presumably as long as the two parties agreed on it. This, in turn, has implications for the term “relationship”. Specifically, the term that provides a common platform for both groups is “monogamous relationship”. Similarities between heterosexual and lesbian participants can be seen in their reasons for entering into a romantic relationship. Both groups cited emotional and physical intimacy as reasons. Intimacy has been found to be a goal of dating and relationships (Gable, Reis, Impett, & Asher, 2004; McNulty & Karney, 2004) and is often an experienced result of an interpersonal process, which includes self-disclosure and partner-responsiveness. Other researchers have found communication and partner’s responsiveness when sharing both good and bad news to be indicative of relationship longevity (2000). Similarly, in Mackey, Diemer, and O’Brien’s study (Dolan, 2005), lesbian women described their relationships as “psychologically intimate”, which they explained as having “a best friend” (Mackey, Diemer & O’Brien, 2000, p. 215). This suggests that qualities that facilitate self-disclosure and minimize the negative effects of conflicts are desired in relationship partners, whether such partners are of the same or different sex. Interestingly, more lesbian groups mentioned physical intimacy as a reason for entering a romantic relationship. Here, however, heterosexual participants could have been more restrained at expressing sexual desires, as participants stated that women are less sexual than men. 30 Sexual Permissiveness & Risk Perception In terms of how relationships are viewed, the most salient group difference was marriage. The small number of lesbian groups that brought up marriage could be due to factors such as parental disapproval, current lack of legal recognition for same-sex pairings, and the use of different ways to indicate a deeper level of commitment (e.g., purchase of matching “couple rings”). The political and social climate of Singapore may also have an impact on the lesbian women’s views of marriage. Absent from the penal code of Singapore, lesbian women do not face legal persecution for same-sex sexual activities. However, unions (civil unions or marriages) between same-sex partners are not recognized and child adoption is not extended to same-sex partners. Therefore, while lesbian women in Singapore have the liberty to date and form meaningful relationships without legal persecution, marriage and the prospects of having children would less likely be part of their considerations of relationships. For definitions of sex, participants were likely to cite examples that were relevant to them. This is consistent with past findings where heterosexuals defined sex as penile-vaginal intercourse (Berasmin & Fisher, 2007; Trotter & Alderson, 2007). Lesbian women were likely to engage in activities such as cunninlingus (oral sex), digital sex, and tribadism (stimulation of genitals via rubbing against body parts) (Edward S. Herold & Mewhinney, 1993; Wallerstein, 1996). It should be noted although lesbian participants did not participate in penile-vaginal intercourse, they did cite such intercourse as sex. This points to lesbian women knowing societal norms and definitions of sex. This is not surprisingly so, as such norms and definitions are transmitted through peer gossip, confidences between friends, and 31 Sexual Permissiveness & Risk Perception formal sex education that takes place in the schools. A similar rationale can be applied to understand lesbian participants’ definitions of virginity. Although lesbian participants did not find virginity useful as a concept, they were nevertheless able to generate similar criteria for what constitutes virginity. Although there is some ambiguity here, it is interesting to note that lesbian participants from 3 of the 6 groups indicated that orgasm is important as a criterion for sex, whereas none of the heterosexual participant groups did so. For lesbian women who sometimes participate mainly in external stimulation of the genitals (such as oral or manual stimulation of genitals) during sexual activity, orgasm may be an important marker of when foreplay stops and when the actual sex act (perhaps an equivalent to penetration) begins. Heterosexual and lesbian participants had similar purposes for sex, such as physical gratification and making love. Women have been found to engage in sexual activity for pleasure (Metz & McCarthy, 2007; Patrick, Sells, Giordano, & Tollerud, 2007) and for a physical expression of intimacy (Carpenter, 1998). Lesbian participants, however, differed from heterosexual participants on the purpose of partner satisfaction. This is seen in the prevalence of one-way sex being cited among lesbian women, which added to their purposes of sex. The primacy of penetration in virginity loss was found. This is similar to previous findings (Carpenter, 2001; Trotter & Alderson, 2007). As with Chambers’ (2007) heterosexual participants, in which those who had oral sex only may or may not classify themselves as virgins, some heterosexual participants in this study brought up the need for mutual genital involvement for virginity loss. Therefore, some participants perceived that some 32 Sexual Permissiveness & Risk Perception forms of sex would not take away one’s virginity. Lesbian participants, on the other hand, do not require involvement of both genitals to lose one’s virginity. This is likely because lesbian women’s sex experiences do not usually contain mutual genital involvement. Instead, they take turns to give and receive pleasure, or either receive or give pleasure (one-way sex). None of the lesbian groups expressed that sex should be retained for marriage, probably because there is currently no such thing as same-sex marriages in the local context. For heterosexual groups, there were mixed opinions on virginity, depending on whether it reflects sexual permissiveness. This is consistent with the observed shift, which took place gradually over the last few decades, from a traditional to a relational orientation towards sexuality (Carpenter, 1998). People who hold a traditional orientation to sexuality believe that sex should only take place within marriage and discourage all forms of sexual activity that does not lead to procreation (Laumann, Gagnon, Michael, & Michaels, 1994). People who hold a relational orientation to sexuality are in favor of sexual activity as long as it takes place within monogamous, committed, and loving relationships (Edward S. Herold & Mewhinney, 1993). As a result, people are increasingly more accepting of pre-martial sex in the context of a loving relationship (Fishman & Anderson, 2003; Montcalm & Myer, 2000). Regardless of whether sex should be reserved for marriage, both heterosexual and lesbian participants were concerned about sexual permissiveness. Virginity was deemed important if it served as an indicator of sexual permissiveness and perceived as unimportant if it was not. This indicates that sexual permissiveness, not virginity, is of interest. 33 Sexual Permissiveness & Risk Perception The findings of the present study show that even though there are differences, the views and opinions put forth on relationships, sex, and virginity by the lesbian and heterosexual participants are in many ways similar. Indeed, there appears to be a prevailing thread of sexual permissiveness among the responses of both groups of participants, taking on the form of a distinction between monogamy versus commitment in how relationships are viewed as well as the importance of virginity and how it is defined in the context of sex. This primacy of sexual permissiveness, as well as the sexual activities cited by (and therefore familiar to) both heterosexual and lesbian groups in the present study, indicate the possibility of a common framework for comparing lesbian and heterosexual women with respect to different aspects of sexual permissiveness (such as how acceptable and how risky such permissiveness is). Study 2 Study 2 was conducted to examine and compare sexual permissiveness, in terms of acceptability and perceived risk, of heterosexual and lesbian women. Drawing on the prevailing environment (e.g., religious groups warning that lesbian women are sexually permissive and that significant numbers of lesbian women exchange sex for money or drugs), the documentation that lesbian women do participate in sexually permissive activities (in having multiple partners) (e.g., Morrow & Allsworth, 2000), as well as the perception of lesbian immunity (Montcalm & Myer, 2000), it was thus expected that lesbian women are likely to be more accepting of sexual permissiveness (hypothesis 1a) and to perceive less risk in being sexually permissive (hypothesis 1b) than heterosexual women. 34 Sexual Permissiveness & Risk Perception Although researchers had examined self-perceived risk for HIV in lesbian women and the types of sexual activities that lesbian women engage in, these studies did not identify the specific perceived risk associated with each sexual activity (Fishman & Anderson, 2003; Montcalm & Myer, 2000). Neither was the acceptability of these various sexual activities explored. On the other hand, participants are likely to distinguish between different types of sexual activities and which activities would be regarded as “sex”. In Pitts and Rahman’s (2001) study on what constitutes sex with 190 UK female and 353 US female undergraduate students (predominantly heterosexual), students differentiated between deep kissing (involving the tongue), receiving and giving manual touch to genitals, receiving and giving oral-genital contact, and penile-vaginal intercourse. Less than 3% deemed kissing as sex. For receiving manual touch to own genitals, 15.8% of UK participants and 12.2% of US women considered it sex. For giving manual touch to partner’s genitals, 15.3% of UK and 11.6% of US participants considered it sex. For receiving and giving oral sex, 32.1% to 37.7% of the participants considered it sex. The majority of participants (more than 98%) deemed penile-vaginal intercourse as sex. The same hierarchy appeared in Randell and Byers’ (2003) sample of 164 heterosexual university participants. 1.9% of the female participants considered tongue kissing as sex. 11% considered receiving, and 15% considered giving, manual touch to the point of orgasm as sex. Slightly less than 25% of the female participants regarded receiving and giving oral contact with genitals as sex. More than 95% of participants regarded penile-vaginal intercourse as sex. Hence, the hierarchy, from the least considered as sex to the 35 Sexual Permissiveness & Risk Perception most, appeared to be as such: kissing, digital sex, oral sex, and finally penile-vaginal intercourse. The mode (receiving and giving) of sexual activities did not appear to make a major difference. Participants from Trotter and Alderson’s (2007) study also appeared to have different yardsticks for same-sex activities, being more likely to consider oral and manual stimulation as sex for same-sex partners. Thus, any examination of acceptability and risk perceptions would need to take into account the type of sexual activity as well as the gender of the partner involved. It should be noted that studies on sex-related issues, and not just those focusing on lesbian women, rarely specify what the relationship context surrounding the sex-related issues was, such that it is not clear whether the results apply to, for instance, sex in the context of a monogamous relationship. Results from Study 1 indicate that relationship context is important and needs to be taken into account when examining sex-related differences between lesbian and heterosexual women (e.g., while both lesbian and heterosexual women in Study 1 desired commitment and monogamy in their relationships, lesbian women did not equate monogamy with commitment the way heterosexual women did). To accurately examine sexual permissiveness as applied to both groups of women, it is thus imperative to specify the relationship context. Since the focus of this thesis was on sexual permissiveness, the relationship context was therefore restricted to only those settings that are arguably considered as sexually permissive by both lesbian and heterosexual women. Such settings of permissive sexual activities include engaging in various sexual activities 36 Sexual Permissiveness & Risk Perception outside of a love relationship and/or with multiple partners and having one-night stands. It can be expected that lesbian women differ from heterosexual women in how acceptable they see sex with same-sex partners in sexual situations that can be regarded as permissive (such as one-night stands). Heterosexual participants are less exposed to same-sex sexuality and possibly less accepting therefore of such sexual expressions. On the other hand, lesbian women who step forward to participate in research (i.e., make it known that they are lesbian) are typically those who have “come out” and are more accepting of same-sex sexuality than closeted lesbian and heterosexual women. They are thus more likely to see sexual activities with female partners—even those that might be regarded as permissive—as a natural expression of sexuality. Thus, it was expected that lesbian participants are likely to be more accepting of permissive sexual activities involving a samesex (female) partner than heterosexual women (hypothesis 2a). Trotter and Alderson (2007) also found that their heterosexual participants were more likely to judge an activity as sex when the sexual partner is of the opposite sex. Since sexual permissiveness is generally not sanctioned, permissiveness of activities that are classified as “sex” should be less accepted than what is classified as “non-sex”. This also implies that permissiveness with partners of the opposite sex is likely to be something more greatly frowned upon among heterosexual women, as seen in typical labels such as being “cheap” or being a “slut”. On the other hand, the current trend of “pseudo lesbian acts” among heterosexual women in mainstream media where nonlesbian women behave intimately with other women (such as kissing 37 Sexual Permissiveness & Risk Perception in music videos) may also increase the acceptability of same-sex activities among heterosexual women. Thus, it was expected that both lesbian and heterosexual women see permissive sexual activities involving a same-sex (female) partner as more acceptable than those involving an opposite-sex (male) partner (hypothesis 2b). In comparing lesbian and heterosexual women, there are potential control variables to be considered. Negative attitudes towards lesbians, for instance, may bias participants towards lower ratings of acceptability and higher ratings of risk for permissive sexual activities involving partners of the same sex. A similar, though less likely, argument could be made regarding negative attitudes towards heterosexual women. Such negative attitudes may therefore need to be controlled for if the purpose of the study is to attribute differences in acceptability and risk perceptions to differences in sexual orientation. In the same manner, knowledge of STI (e.g., STI rates among men and women) may need to be controlled for because it may affect acceptability and risk perceptions of permissive sexual activities. Those with greater STI knowledge may be less accepting of and may perceive greater risk then those with lesser STI knowledge, especially in the present context of permissive sexual activities. In terms of risk perceptions, the concept of lesbian immunity (where lesbian women see themselves as safe from STIs) and popular notions of men being more sexually permissive than women are likely to result in lesbian women perceiving lower risk with same-sex partners, as compared to heterosexual women. Thus, it was expected that lesbian women, as compared to heterosexual 38 Sexual Permissiveness & Risk Perception women, would perceive permissive sexual activities with same-sex (female) partners as less risky as compared to engaging in the same sexual activities with opposite-sex (male) partners (hypothesis 3a). On the other hand, given the various sex education campaigns (such as safe sex), it is likely that both lesbian and heterosexual women are somewhat familiar with how STIs are passed on to another. Thus, both lesbian and heterosexual women were expected to perceive risk according to the type of permissive sexual activity. Specifically, it was expected that both lesbian and heterosexual women would rate those sexual activities with higher likelihood of transmission of bodily fluids (namely, oral sex and penile-vaginal intercourse) as riskier than those activities with lower likelihood of transmission of bodily fluids (namely, kissing and digital sex), with penile-vaginal intercourse rated as the riskiest and kissing as the least risky. (hypothesis 3b). In the present study, the link between acceptability of and risk associated with the various permissive sexual activities was also explored, in general and separately by each group. As mentioned previously, perceived risk can influence the use of safe sex methods. This, in turn, can mean that the activity with greater risk (one that carries a higher likelihood of transmission of bodily fluids) is also seen as less acceptable. Thus, it was expected that, in general, there would be a negative correlation between perceived risk of permissive sexual activities and the acceptability of those permissive activities (hypothesis 4a). However, given that lesbian participants may think that they are invulnerable to STIs and therefore tend towards disregarding risk, the negative correlation between risk 39 Sexual Permissiveness & Risk Perception and acceptability for lesbian women was expected to be lower than that for heterosexual participants (hypothesis 4b). This lower correlation between risk and acceptability for lesbian women compared to heterosexual women was expected to apply across different sexual activities (oral sex, digital sex, etc.) (hypothesis 4c). In summary, the hypotheses for this study were as follows: 1a. Lesbian participants are more likely than heterosexual participants to be accepting of sexual permissiveness. 1b. Lesbian participants are more likely than heterosexual participants to perceive less risk associated with being sexually permissive. 2a. Lesbian participants are more likely than heterosexual participants to be accepting of permissive sexual activities involving a same-sex (female) partner. 2b. Both heterosexual and lesbian participants are likely to see permissive sexual activities involving a same-sex (female) partner as more acceptable than those involving an oppositesex (male) partner. 3a. Lesbian participants are more likely than heterosexual participants to perceive permissive sexual activities with same-sex (female) partners as less risky compared to engaging in the same sexual activities with opposite-sex (male) partners. 3b. Both heterosexual and lesbian participants are likely to perceive oral sex and penile-vaginal intercourse as riskier than kissing and digital sex with penile-vaginal intercourse as the riskiest and kissing as the least risky. 40 Sexual Permissiveness & Risk Perception 4a. The correlation between perceived risk and acceptability of permissive sexual activities is negative, such that as level of risk perceived increases, acceptability decreases. 4b. The negative correlation between risk and acceptability for lesbian women was expected to be lower than that for heterosexual participants. 4c. The negative correlation between perceived risk and acceptability for lesbian women of each sexual activity is expected to be lower than that for heterosexual participants. Method Participants One hundred and one heterosexual women (mean age = 20.02, SD = 1.33) and 100 lesbian women (mean age = 20.64, SD = 1.81), between ages 19 to 26, were recruited for this study. All heterosexual women and 5 lesbian women were recruited from the Research Participants pool of the Department of Psychology at the National University of Singapore. These participants completed the study for research participation credit. The remaining 95 lesbian women were recruited via posters and local online portals such as Sayoni, “an online blog and forum for queer women”. Participants recruited outside of the Research Participants pool were given $5 as compensation for their time. To better distinguish lesbian women as well as heterosexual women and especially to minimize the likelihood of bisexual women being recruited, recruitment advertisements specified that potential participants had to be romantically attracted to only members of the same sex to be considered lesbian, and 41 Sexual Permissiveness & Risk Perception romantically attracted to only members of the opposite sex to be considered heterosexual. As a checking procedure for sexual orientation, checking items were included in the survey, asking participants who identified themselves as lesbians if they had experienced physical attraction to men in the last two years or had sexual relationships with men in the past two years. Similar checking items were included for those identifying themselves as heterosexual. These checking items confirmed that there were no bisexual women in the study. Table 2 provides a summary of the demographics of participants from this sample. 42 Sexual Permissiveness & Risk Perception Table 2 Demographics Characteristics of Heterosexual (mean age = 20.02, SD = 1.33) and Lesbian Participants (mean age = 20.64, SD = 1.81) in Survey Study Lesbian Heterosexual n Percentage n Percentage Race Chinese 94 93.1 94 94.0 Indian 2 2.0 4 4.0 Malay 4 4.9 0 0.0 Others 1 1.0 2 2.0 O levels 0 0.0 6 6.0 Diploma/ 99 98.0 77 77.0 Undergrad 2 2.0 15 15.0 Others 0 0.0 2 2.0 Catholic 6 5.9 7 7.0 Christian 30 29.7 32 32.0 Buddhist 20 19.8 14 13.2 Muslim 5 5.0 4 4.0 Others 6 5.9 3 3.0 34 33.7 40 40.0 95 94.1 77 77.0 Education Level A levels Religion No religion Employment Status Studying 43 Sexual Permissiveness & Risk Perception Working 6 5.9 23 23.0 A set of preliminary analyses were conducted to see if heterosexual and lesbian participants differed on religion, ethnicity, educational level, employment status, and age. The categories of some of these demographics variables had to be reduced in view of low frequencies in certain categories. Thus, for religion, categories comprised only Buddhists, Catholics, Christians, freethinkers, and Hindus. Likewise, race was reduced to “Chinese” and “non-Chinese”, employment status reduced to “Working” and “Not working”, and educational level reduced to “Acquired tertiary education” and “Did not acquire tertiary education”. Lesbian participants were more likely to be employed, X2(1) = 12.06, p < .01; to have acquired tertiary education, X2(1) = 11.31, p < .01; and to be older, t(199) = -2.77, p < .001, as compared to heterosexual participants. Participants did not differ on religion, X2(5) = 2.79, p = .73, and ethnicity, X2(1) = .07, p = .79. Employment status and age were related, r(201) = .34, p < .001, such that those who were working were older than those who were not. Educational status and age were also related, r(201) = .44, p < .001, such that participants who had acquired tertiary qualifications were older. Lesbian women (M = 20.64, SD = 1.81) were older than heterosexual women (M = 20.02, SD = 1.33). Therefore, because they were older, lesbian women were more likely to have graduated with a degree, whereas heterosexual participants were mostly undergraduates with the highest qualification of ‘A’ levels (general pre-requisite for university entry). 44 Sexual Permissiveness & Risk Perception Measures A survey was constructed to assess acceptability of sexual permissiveness, perceived risks associated with sexual permissiveness, acceptability of and risks perceived for various permissive sexual activities involving partners of both genders, as well as negative attitudes towards lesbian and heterosexual women and knowledge of STIs. (See Appendix E for all survey questions and instructions.) Acceptability of sexual permissiveness. To assess how acceptable being sexually permissive was, the Permissiveness subscale of the Brief Sexual Attitudes Scale (1983) was used (see Appendix B for items). In this study, items from this subscale were clarified in parenthesis to maximize similar interpretations between the two groups of women, as informed by the results of Study 1 (see Appendix C for these clarified items under ‘General Sexual Permissiveness’). Thus, for example, an item that originally read “I do not need to be in a monogamous relationship with a person to have sex with that person” was edited to “I do not need to be in a monogamous relationship with a person (where there is an agreement between the 2 partners to be sexually and/or romantically involved with only each other) to have sex with that person”. There were 10 items in the scale and both heterosexual and lesbian women rated the items on an 8-point Likert-type scale (1 = absolutely disagree and 8 = absolutely agree). A mean score was derived for each participant, such that the highest attainable score for each variable was 8 and the lowest attainable score was 1. Higher scores indicated greater acceptability of sexual permissiveness. The Cronbach’s alphas 45 Sexual Permissiveness & Risk Perception were .85 for heterosexual participants and .90 for lesbian participants. Perceived risk associated with sexual permissiveness. To assess perceived risk, items in the previously described acceptability of sexual permissiveness scale were reworded to capture the perceived risk of being sexually permissive. For example, an item that originally read “I would like to have sex with many partners” was edited to “having sex with many partners is risky”. Risk here was explicitly defined as sexual risks, which are sex-related consequences (e.g., STIs) that could arise from sexual activity. This risk perception was measured on an 8point Likert-type scale (1 = absolutely disagree and 8 = absolutely agree). There were again 10 items in the scale and a mean score was derived for each participant, with higher scores indicating greater perceived risk associated with sexual permissiveness. The Cronbach’s alphas were .88 for heterosexual participants and .92 for lesbian participants. See Appendix C for these items under ‘General Risk Perception’. Acceptability and risk perception of different permissive sexual activities. As mentioned, all sexual activities used for this study denoted sexually permissive settings (multiple partners, one-night stands, etc.). Although most sexual activities apply regardless of gender of partner (e.g., oral sex can be performed with a same-sex or an opposite-sex partner), there are instances where a sexual activity is restricted to a particular gender of partner. In particular, penile-vaginal intercourse requires an opposite-sex partner while the giving of digital sex can only be done for a same-sex (female) partner. In all, there were 5 sexual activities in which a same-sex (female) 46 Sexual Permissiveness & Risk Perception partner was involved (French kissing, receiving oral sex, giving oral sex, receiving digital sex, and giving digital sex) and 5 sexual activities in which an opposite-sex (male) partner was involved (French kissing, receiving oral sex, giving oral sex, receiving digital sex, and penile-vaginal intercourse). The choice of these sexual activities was informed by the findings of Study 1. Thus, for instance, anal sex was not included in this study because of the low frequency in which it was generated among both the heterosexual and lesbian focus groups. To ensure familiarity with the terms as well as similar interpretations of each sexual activity, definitions of the sexual activities were provided. Thus, French kissing was specified as kissing that involved the tongue, oral sex was defined as oral stimulation of genitals, digital sex was defined as penetration of a vagina with finger(s), and penile-vaginal intercourse was defined as penetration of a vagina with a penis. To create the scale for acceptability of the various permissive sexual activities, each of the 10 items from the acceptability of sexual permissiveness scale was placed on separate page with all 5 same-sex and 5 opposite-sex sexual activities (see Appendix D). As before, clarifications were provided (from “I do not need to be in a monogamous relationship with a person to have sex with that person” to “I do not need to be in a monogamous relationship with a person [where there is an agreement between the 2 partners to be sexually and/or romantically involved with only each other] to have sex with that person”) to maximize similar interpretations between the two groups of women. To allow for acceptability and risk perception ratings to be provided by both groups of women, some of whom may 47 Sexual Permissiveness & Risk Perception not find a particular sexual activity relevant, the term “target woman” was used instead of a personal “I” pronoun. This allowed heterosexual participants to also rate how they perceived the acceptability and risk associated with permissive same-sex sexual activities and lesbian women to also provide their perceptions with respect to permissive opposite-sex activities. The use of “target woman” also helped reduce social desirability as participants were likely to feel less threatened by having to rate someone else instead of themselves. Participants were then asked to rate how okay (acceptability) and how risky (risk perception) each of the 10 sexual activities was for the target woman. Thus, every permissive sexual activity was rated for acceptability and risk perception. To minimize order effects and to encourage participants to rate carefully, four versions of this measure (randomly ordering the list of permissive sexual activities) were created and used. Acceptability was measured on an 8-point Likert-type scale (1 = absolutely not okay and 8 = absolutely okay). Likewise, risk perception was measured on an 8-point Likert-type scale (1 = absolutely safe and 8 = absolutely risky). A mean acceptability score and a mean risk perception score were then derived for each participant, with the highest attainable score being 8 and the lowest attainable score being 1. Higher scores for acceptability meant greater acceptability of that particular permissive sexual activity, while higher scores for risk perception meant a greater risk perceived for that particular permissive sexual activity. For heterosexual participants, the Cronbach’s alphas for acceptability ranged from .90 to .95 across the different 48 Sexual Permissiveness & Risk Perception permissive sexual activities, while the alphas for risk perception ranged from .89 to .98. For lesbian participants, the Cronbach’s alpha for acceptability ranged from .94 to .96, while the range for risk perception was from .93 to .97. Negative attitudes towards lesbian and heterosexual women. Five items were used to measure negative attitudes towards lesbian women and a corresponding 5 items were used to measure negative attitudes towards heterosexual women. All participants completed these 10 items. These items were developed for this study and derived from negative stereotypes of lesbian women. Thus, an example item here was “homosexual women are mentally ill”. Items measuring negative attitudes towards lesbian women were mirrored in items measuring negative attitudes towards heterosexual women, such that there was a corresponding “heterosexual women are mentally ill” item. Ratings were made on an 8-point Likert-type scale (1 = absolutely disagree and 8 = absolutely agree). Two mean scores were obtained for each participant, one reflecting her attitude towards lesbian women and the other reflecting her attitude towards heterosexual women. Higher mean scores indicated more negative attitudes. The Cronbach’s alphas for negative attitudes towards lesbian women for heterosexual and lesbian participants were .69 and .44 respectively. The Cronbach’s alpha for lesbian participants was rather low, possibly reflecting greater diversity of views on what constitutes negative attitudes towards lesbian women. Results emerging from the use of this scale for lesbian participants should therefore be interpreted with caution. The Cronbach’s alphas for negative attitudes towards heterosexual women for heterosexual and lesbian participants were both .71. 49 Sexual Permissiveness & Risk Perception See Appendix C for these 10 items under ‘Negative Attitudes towards Lesbian and Heterosexual Women’. STI knowledge. There were 10 items in this self-developed scale, which were derived from official HIV statistics in Singapore. Example items here were “more men than women get STI” and “women who are affected by STI are more likely to have had female sexual partners than male sexual partners”. Ratings were made on 8-point Likert-type scales (1 = absolutely disagree and 8 = absolutely agree). Because items were based on official statistics, each participant’s responses were dichotomized into right and wrong. Thus, for an item which was correct, participants who rated from 5 to 8 (from slightly agree to absolutely agree) were given a score of 1, while those who rated from 1 to 4 (from absolutely disagree to slightly disagree) were given a score of 0. Thus, the highest attainable score was 10 and the lowest attainable score was zero. Higher scores indicated greater STI knowledge. See Appendix C for all items. Demographics. Participants also provided information about their age, ethnicity, religion, employment status, and highest educational qualification attained. Procedures Participants were first briefed on the study and then asked to sign their consent if they were willing to participate. They were then given an instruction sheet on how to complete the survey and asked to complete the survey independently. Upon completion, participants were debriefed and participants from the Research Participant pool were given their 2 credit points, while other participants were given $5 as compensation for their time. 50 Sexual Permissiveness & Risk Perception All materials and procedures used in this study were approved by the NUS IRB. Results Table 3 shows the means and standard deviations of all of the study variables. 51 Sexual Permissiveness & Risk Perception Table 3 Means, Adjusted Means (in parenthesis and where applicable), and Standard Deviations for All Variables in Survey Study Heterosexual Lesbian M SD M SD NALes 4.07 0.96 2.88 0.64 NAStra 2.23 0.85 2.40 0.88 STI knowledge 7.15 1.59 8.27 1.46 2.62 1.07 3.49 1.33 6.79 .85 6.22 1.17 General sexual permissiveness General risk perception Acceptability Female Kiss 3.54 (3.80) 1.64 5.22 (4.96) 1.79 rO 2.37 (2.59) 1.21 4.05 (3.84) 1.78 gO 2.35 (2.57) 1.20 3.95 (3.73) 1.79 rD 2.44 (2.64) 1.23 4.11 (3.91) 1.80 gD 2.46 (2.65) 1.22 4.19 (4.01) 1.83 Kiss 3.98 (4.11) 1.65 4.80 (4.67) 1.91 rO 2.53 (2.68) 1.30 3.52 (3.37) 1.86 gO 2.50 (2.64) 1.28 3.28 (3.13) 1.89 rD 2.58 (2.72) 1.31 3.59 (3.44) 1.92 PVI 2.18 (2.32) 1.19 2.96 (2.82) 1.89 Male Risk perceived 52 Sexual Permissiveness & Risk Perception Female Kiss 3.96 1.67 3.48 1.73 rO 5.80 1.40 5.11 1.69 gO 6.00 1.30 5.34 1.61 rD 5.51 1.40 4.75 1.61 gD 5.37 1.49 4.53 1.70 Kiss 4.03 1.65 3.86 1.88 rO 6.01 1.32 5.75 1.65 gO 6.24 1.18 6.16 1.47 rD 5.63 1.36 5.32 1.74 PVI 7.33 .65 7.14 .88 Male Note. Negative attitudes towards lesbian women and negative attitudes towards heterosexual women were abbreviated to “NALes” and “NAStra” respectively. Sex types were abbreviated to fit the table. Gender of partner was indicated as “F” or “M”, mode was represented with “r” (receiving) or “g” (giving), type of sex was abbreviated to “kiss” (kissing), “O” (oral sex), “D” (digital sex) and “PVI” (penile-vaginal intercourse). Potential Control Variables Negative attitudes towards lesbian women, negative attitudes towards heterosexual women, and STI were assessed first to see if they were correlated with acceptability and risk perceptions, which would then mean that they should be controlled for in the main analyses. As a rule of thumb, only correlations .40 and above were regarded as substantially meaningful and therefore required control in the main analyses (Durrheim & Tredoux, 2004). 53 Sexual Permissiveness & Risk Perception For the various acceptability ratings, Pearson correlations involving negative attitudes towards lesbian women ranged from .21 to -.42, those involving negative attitudes towards heterosexual women ranged from .01 to .07, and those correlations involving STI knowledge ranged from .02 to .12. Table 4 shows all correlations for acceptability ratings for heterosexual and lesbian participants. 54 Sexual Permissiveness & Risk Perception Table 4 Correlations for Acceptability Ratings NALes NALes NAStra STIkno Fkiss FrO FgO FrD FgD Mkiss MrO MgO MrD PVI .26** .07 -.34** -.37** -.38** -.37** -.36** -.15 -.25* -.24* -.25* -.19 .06 -.09 .00 .00 -.02 -.02 -.03 .00 .02 -.01 .07 -.08 -.12 -.12 -.10 -.09 -.08 -.16 -.15 -.14 -.06 .75** .75** .77** .76** .82** .58** .58** .61** .46** .99** .98** .97** .63** .85** .84** .83** .68** .97** .96** .64** .85** .85** .83** .69** 1.00** .64** .84** .82** .85** .65** .63** .83** .81** .85** .65** .71** .71** .73** .64** 1.00** .98** .88** .97** .89** NAStra .17 STIkno .15 -.03 Fkiss -.05 -.07 -.00 FrO -.04 -.03 -.05 .75** FgO -.04 -.04 -.05 .72** .96** FrD -.01 -.05 -.05 .74** .97** .93** FgD .00 -.05 -.01 .78** .98** .95** .97** Mkiss -.04 -.15 .01 .90** .71** .70** .71** .75** MrO -.02 -.12 -.03 .62** .84** .85** .81** .84** .75** MgO -.02 -.11 -.08 .51** .73** .79** .74** .73** .67** .89** MrD .00 -.13 -.06 .60** .79** .78** .82** .81** .75** .95** .88** .86** 55 Sexual Permissiveness & Risk Perception PVI -.09 -.13 -.13 .43** .66** .72** .67** .64** .59** .85** .88** .86** Note. Negative attitudes towards lesbian women and negative attitudes towards straight women were abbreviated to “NALes” and “NAStra” respectively. Knowledge about sexually transmitted infections was abbreviated to “STIkno”. Sex types were abbreviated to fit the table. Gender of partner was indicated as “F” or “M”, mode was represented with “r” (receiving) or “g” (giving), type of sex was abbreviated to “kiss” (kissing), “O” (oral sex), “D” (digital sex) and “PVI” (penile-vaginal intercourse). The upper diagonal portion represents correlation values of heterosexual participants. The lower diagonal portion represents correlation values of lesbian participants. Heterosexual and lesbian participants numbered 101 and 100 respectively. *p < .05. **p < .01. 56 Sexual Permissiveness & Risk Perception For the various risk perception ratings, Pearson correlations involving negative attitudes towards lesbian women ranged from .07 to .20, those involving negative attitudes towards heterosexual women ranged from .02 to .19, and those involving STI knowledge ranged from .03 to .22. Table 5 shows all correlations for risk perception ratings for heterosexual and lesbian participants. 57 Sexual Permissiveness & Risk Perception Table 5 Correlations for Risk Perception Ratings NALes NALes NAStra STIkno Fkiss FrO FgO FrD FgD Mkiss MrO MgO MrD PVI .26** .07 .17 .22* .24* .19 .19 .13 .22* .18 .19 .15 .06 .25* .14 .08 .18 .18 .23* .13 .06 .17 -.09 .03 -.06 -.02 -.05 -.06 -.05 -.02 .01 -.04 .13 .62** .56** .67** .69** .97** .60** .54** .66** .18 .91** .84** .82** .56** .95** .84** .77** .35** .78** .77** .50** .82** .90** .67** .40** .96** .62** .78** .70** .94** .32** .64** .77** .72** .94** .30** .58** .52** .64** .16 .85** .80** .39** .68** .47** NAStra .17 STIkno .15 -.03 Fkiss -.03 .09 -.14 FrO -.09 .01 -.23* .56** FgO -.08 .01 -.17 .51** .86** FrD -.17 .03 -.19 .59** .83** .76** FgD -.19 .05 -.23* .62** .82** .73** .93** Mkiss -.02 .17 -.11 .93** .47** .41** .53** .55** MrO -.08 .07 -.15 .48** .83** .65** .70** .69** .56** MgO .05 .09 -.05 .37** .58** .69** .52** .52** .47** .75** MrD -.18 .10 -.15 .49** .65** .51** .81** .77** .60** .84** .61** .32** 58 Sexual Permissiveness & Risk Perception PVI .08 .00 .30** .04 .26** .34** .25** .16 .17 .44** .57** .34** Note. Negative attitudes towards lesbian women and negative attitudes towards straight women were abbreviated to “NALes” and “NAStra” respectively. Knowledge about sexually transmitted infections was abbreviated to “STIkno”. Sex types were abbreviated to fit the table. Gender of partner was indicated as “F” or “M”, mode was represented with “r” (receiving) or “g” (giving), type of sex was abbreviated to “kiss” (kissing), “O” (oral sex), “D” (digital sex) and “PVI” (penilevaginal intercourse). The upper diagonal portion represents correlation values of heterosexual participants. The lower diagonal portion represents correlation values of lesbian participants. Heterosexual and lesbian participants numbered 101 and 100 respectively. *p < .05. **p < .01. 59 Sexual Permissiveness & Risk Perception Thus, only negative attitudes towards lesbian women (with a correlation above .40) was included as a covariate in the main analyses for acceptability. As a point of supplementary interest, the two groups of women were also compared on these negative attitudes and STI knowledge variables. Here, heterosexual participants (M = 2.23) and lesbian participants (M = 2.40) did not differ in their negative attitudes towards heterosexual women, t(199) = -1.39, p = .166. However, heterosexual participants (M = 4.07) possessed more negative attitudes towards lesbian women as compared to lesbian participants (M = 2.88), t(199) = 10.28, p < .001. On the other hand, lesbian participants (M = 8.27) received higher scores for STI knowledge as compared to heterosexual participants (M = 7.15), t(199) = -5.22, p = .001. Acceptability of Sexual Permissiveness and Perceived Risk Associated with Sexual Permissiveness An alpha level of .05 was used for this and all subsequent analyses. Comparison of the two groups revealed that heterosexual participants (M = 2.62) were less sexually permissive than lesbian participants (M = 3.49), t(199) = -5.05, p < .001, thus supporting hypothesis 1a. Heterosexual participants (M = 6.79) also perceived more risk associated with sexual permissiveness as compared to lesbian participants (M = 6.22), t(199) = 3.95, p < .001, thus supporting hypothesis 1b. Acceptability of Permissive Sexual Activities To circumvent the problem of missing cells in a completelycrossed design involving sexual activity and gender of partner as two separate factors (since women cannot engage in penile-vaginal intercourse and a woman cannot give a male partner digital sex), 60 Sexual Permissiveness & Risk Perception the 10 sexual activities were collapsed into a single factor. This factor was therefore made up of French kissing a female partner, French kissing a male partner, receiving oral sex from a female partner, receiving oral sex from a male partner, giving oral sex to a female partner, giving oral sex to a male partner, receiving digital sex from a female partner, receiving digital sex from a male partner, giving digital sex to a female partner, and penile-vaginal intercourse. A 2 (sexual orientation) X 10 (sexual activity) repeatedmeasures ANCOVA, with sexual activity as the within-subjects factor and sexual orientation as the between-subjects factor, was conducted to assess whether there were differences in acceptability ratings between heterosexual and lesbian participants, controlling for negative attitudes towards lesbian women. The Greenhouse-Geisser adjustment was used in instances where the sphericity assumption was not met. Table 3 presents the adjusted and unadjusted group means for acceptability ratings. Results indicated a main effect of sexual activity, F(9, 1782) = 9.31, p < .001, ŋ 2 = .05, with Greenhouse- Geisser adjusted dfs = 3.00, 593.64, p < .001. There was also a main effect of sexual orientation, F(1, 198) = 13.69, p < .001, ŋ 2 = .07. However, these main effects were qualified by an interaction between sexual orientation and sexual activity, F(3.00, 593.64) = 7.00, p < .001, ŋ 2 = .03. Simple effect tests were conducted to follow up on this interaction. Differences between sexual orientation groups for each permissive sexual activity. One set of simple effects testing was conducted to explore whether heterosexual and lesbian participants differed in how acceptable they saw each of the 10 61 Sexual Permissiveness & Risk Perception permissive sexual activities. To control for inflated Type I error, Bonferroni corrections were used to adjust the level of significance used. Because there were 10 comparisons to be made here, simple effects testing thus proceeded with an alpha of .005. Lesbian participants rated French kissing a female partner (F(1, 198) = 15.55, p < .001, ŋ 2 = .07), receiving oral sex from a female partner (F(1, 198) = 22.88, p < .001, ŋ 2 = .10), giving oral sex to a female partner (F(1, 198) = 19.90, p < .001, ŋ 2 = .09), receiving digital sex from a female partner (F(1, 198) = 23.19, p < .001, ŋ 2 = .11) and giving digital sex to a female partner (F(1, 198) = 25.55, p < .001, ŋ 2 = .11) as more acceptable compared to heterosexual participants. The two groups did not differ for French kissing a male partner (F(1, 198) = 3.21, p = .08, ŋ 2 = .02), receiving oral sex from a male partner (F(1, 198) = 6.03, p = .02, ŋ 2 = .03), giving oral sex to a male partner (F(1, 198) = 3.10, p = .08, ŋ 2 = .02), receiving digital sex from a male partner (F(1, 198) = 6.40, p = .01, ŋ 2 = .03) and penile-vaginal intercourse (F(1, 198) = 3.24, p = .07, ŋ 2 = .02). Therefore, lesbian participants rated all same-sex (female) sexual activities in a permissive setting as more acceptable as compared to heterosexual women, but the two groups did not differ on all opposite-sex (male) sexual activities in a permissive setting. This supports hypothesis 2a. Differences among permissive sexual activities within each sexual orientation group. A second set of simple effects tests was conducted to explore how acceptability ratings differed among the 10 permissive sexual activities within each sexual orientation group. 62 Sexual Permissiveness & Risk Perception Two one-way repeated-measures ANCOVAs (again controlling for negative attitudes towards lesbian women) were conducted, one for each sexual orientation group. For the heterosexual group, results indicated a main effect of sexual activity (F(9, 891) = 6.93, p < .001, ŋ 2 = .07, with Greenhouse-Geisser adjusted dfs = 2.62, 259.19, p < .001. Each sexual activity was then tested against every other sexual activity, resulting in 45 pairwise comparisons to be performed. Bonferroni corrections were again used to adjust for Type 1 error, such that each pairwise comparison was tested with an alpha of .0011. For heterosexual participants, French kissing a female partner was rated as more acceptable than giving oral sex to a female partner (t(1, 99) = 11.81, p < .001) and giving oral sex to a male partner (t(1, 99) = 12.28, p < .001), while penile-vaginal intercourse was rated as less acceptable than kissing a female partner (t(1, 99) = 17.40, p < .001) and kissing a male partner (t(1, 99) = 11.69, p < .001). All other pairwise comparisons did not yield significance. For lesbian participants, the one-way ANCOVA indicated a main effect of sexual activity (F(9, 882) = 2.96, p < .05, ŋ 2 = .03, with Greenhouse-Geisser adjusted dfs = 3.06, 300.30, p = .03. The same 45 pairwise comparisons were then done, at an alpha of .0011 to control for inflated Type I error. Using this significance level, however, none of the pairwise comparisons emerged as significant. Thus, across both heterosexual and lesbian participants, it was not the case that permissive sexual activities involving a same-sex (female) partner were seen as more acceptable than those involving an opposite-sex (male) partner. Hypothesis 2b was not supported. Risk Perceptions Associated with Permissive Sexual Activities 63 Sexual Permissiveness & Risk Perception A 2 (sexual orientation) X 10 (sexual activity) repeatedmeasures ANOVA, with sexual activity as the within-subjects factor and sexual orientation as the between-subjects factor, was conducted to assess whether there were differences between heterosexual and lesbian participants in risk perceptions associated with permissive sexual activities. There was a main effect of sexual activity, (F(9, 1791) = 246.76, p < .001, ŋ 2 = .55, with Greenhouse-Geisser adjusted dfs = 4.00, 796.32, p < .001. There was also a main effect of sexual orientation, F(1, 199) = 6.60, p < .05, ŋ 2 = .03. These main effects, however, were qualified by an interaction between sexual orientation and sexual activity, F(4.00, 796.32) = 4.461, p < .001, ŋ 2 = .022. Simple effect tests were thus conducted to follow up on the interaction. Differences between sexual orientation groups for each permissive sexual activity. One set of simple effects testing was conducted to explore whether heterosexual and lesbian participants differed in how risky they saw each of the 10 permissive sexual activities. To control for inflated Type I error, Bonferroni corrections were used to adjust the level of significance used. Because there were 10 comparisons to be made here, simple effects testing thus proceeded with an alpha of .005. Heterosexual participants rated receiving oral sex from a female partner (t(199) = 3.14, p > .01), giving oral sex to a female partner (t(199) = 3.04, p < .01), receiving digital sex from a female partner (t(199) = 3.57, p > .001), and giving digital sex to a female partner (t(199) = 3.75, p < .001) as riskier compared to lesbian women. The two groups, however, did not differ in how risky they perceived French kissing a female 64 Sexual Permissiveness & Risk Perception partner (t(199) = 2.00, p = .05), French kissing a male partner (t(199) = .65, p = .52), receiving oral sex from a male partner (t(199) = 1.24, p = .22), giving oral sex to a male partner (t(199) = .39, p = .70), receiving digital sex from a male partner (t(199) = 1.42, p = .16), and penile-vaginal intercourse (t(199) = 1.67, p = .10). Therefore, with the exception of French kissing a female partner, lesbian participants rated permissive sexual activities with a same-sex (female) partner as less risky when compared to heterosexual participants. Thus, hypothesis 3a was mostly supported, except for French kissing a female partner. Differences among permissive sexual activities within each sexual orientation group. A second set of simple effects tests was conducted to explore how risk perceptions differed among the 10 permissive sexual activities within each sexual orientation group. A one-way repeated-measures ANOVA with sexual activity as the factor was first conducted for heterosexual participants only. Results indicated a main effect of sexual activity, F(9, 900) = 157.03, p < .001, ŋ 2 = .61, with Greenhouse-Geisser adjusted dfs = 3.36, 335.95, p < .001. Each sexual activity was then tested against every other sexual activity, resulting in 45 pairwise comparisons to be performed. Bonferroni corrections were again used to adjust for Type 1 error, such that each pairwise comparison was tested with an alpha of .0011. For heterosexual participants, perceived risk for French kissing a female partner and French kissing a male partner did not differ from each other, but they were perceived as less risky than all the other sexual activities, with French kissing a female partner being perceived as less risky than giving digital sex to a female partner (t(100) 65 Sexual Permissiveness & Risk Perception = -11.24, p = .001), receiving digital sex from a female partner (t(100) = -12.24, p = .001), receiving digital sex from a male partner (t(100) = -13.08, p = .001), receiving oral sex from a female partner (t(100) = -13.62, p = .001), giving oral sex to a female partner (t(100) = -14.10, p = .001), receiving oral sex from a male partner (t(100) = -14.92, p = .001), giving oral sex to a male partner (t(100) = -15.97, p = .001), and penile-vaginal intercourse (t(100) = -20.14, p = .001). Similarly, French kissing a male partner was perceived as less risky than giving digital sex to a female partner (t(100) = 10.00, p = .001), receiving digital sex from a female partner (t(100) = -11.00, p = .001), receiving digital sex from a male partner (t(100) = 12.40, p = .001), receiving oral sex from a female partner (t(100) = 12.34, p = .001), giving oral sex to a female partner (t(100) = 12.92, p = .001), receiving oral sex from a male partner (t(100) = -14.30, p = .001), giving oral sex to a male partner (t(100) = -15.34, p = .001), and penile-vaginal intercourse (t(100) = -19.81, p = .001). With regards to digital sex, giving digital sex to a female partner differed from the other sexual activities. Giving digital sex to a female partner was rated as less risky than receiving digital sex from a male partner (t(100) = -5.01, p = .001), receiving oral sex from a female partner (t(100) = 4.97, p = .001), giving oral sex to a female partner (t(100) = 6.25, p = .001), receiving oral sex from a male partner (t(100) = -6.61, p = .001), giving oral sex to a male partner (t(100) = -8.38, p = .001), and penile-vaginal intercourse (t(100) = -13.66, p = .001). On the other hand, receiving digital sex from a female partner was perceived as less risky than receiving oral sex from 66 Sexual Permissiveness & Risk Perception a female partner (t(100) = 3.73, p = .001), giving oral sex to a female partner (t(100) = 5.15, p = .001), receiving oral sex from a male partner (t(100) = -5.60, p = .001), giving oral sex to a male partner (t(100) = -7.21, p = .001), and penile-vaginal intercourse. Receiving digital sex from a male partner was perceived as less risky than receiving oral sex from a male partner (t(100) = 4.42, p = .001), giving oral sex to a male partner (t(100) = 5.93, p = .001), and penile-vaginal intercourse (t(100) = -13.05, p = .001). With regards to oral sex, receiving oral sex from a female partner was perceived as less risky than receiving oral sex from a male partner (t(100) = -4.53, p = .001), giving oral sex from a male partner (t(100) = -5.76, p = .001), and penile-vaginal intercourse (t(100) = -11.59, p = .001). On the other hand, giving oral sex to a female partner was perceived as less risky than giving oral sex to a male partner (t(100) = -4.62, p = .001) and penile-vaginal intercourse (t(100) = -11.35, p = .001). Receiving oral sex from a male partner was perceived as less risky than penile-vaginal intercourse (t(100) = -10.87, p = .001), while giving oral sex to a male partner was perceived as less risky than penile-vaginal intercourse (t(100) = -10.48, p = .001). For heterosexual participants, therefore, French kissing and digital sex were rated as less risky than oral sex (within each gender) and penile-vaginal intercourse, with kissing perceived as the least risky and penile-vaginal intercourse as the riskiest. A hierarchy of perceived risk associated with the different permissive sexual activities, from the least risky to the 67 Sexual Permissiveness & Risk Perception riskiest, can be deduced from these results and is depicted in Figure 1. 68 Sexual Permissiveness & Risk Perception Figure 1. Hierarchy of Sex Type Risk Ratings. Sex types within a single box did not differ from each other. Sex types were abbreviated. Gender of partner was indicated as “F” or “M”, mode was represented with “r” (receiving) or “g” (giving), type of sex was abbreviated to “kiss” (kissing), “O” (oral sex), “D” (digital sex) and “PVI” (penile-vaginal intercourse). FrD MKiss MrD FgD FrO FKiss FgO Least Risky MrO MgO PVI Most Risky Heterosexual Participants MrD FrO FKiss MKiss Least Risky FgD FrD FgO Lesbian Participants MrO MgO PVI Most Risky 69 Sexual Permissiveness & Risk Perception A similar one-way repeated-measures ANOVA was conducted for lesbian participants only and results indicated a main effect of sexual activity, F(9, 891) = 106.75, p < .001, ŋ2 = .52, with Greenhouse-Geisser adjusted dfs = 4.28, 423.53, p < .001. The same 45 pairwise comparisons were then done, at an alpha of .0011 to control for inflated Type I error. Results here indicated that for lesbian participants, French kissing a female partner was perceived as least risky, less so than French kissing a male partner (t(99) = -5.56, p = .001), giving digital sex to a female partner (t(99) = 7.03, p = .001), receiving digital sex from a female partner (t(99) = -8.34, p = .001), receiving oral sex from a female partner (t(99) = -10.20, p = .001), giving oral sex to a female partner (t(99) = 11.24, p = .001), receiving digital sex from a male partner (t(99) = -10.55, p = .001), receiving oral sex from a male partner (t(99) = 13.19, p = .001), giving oral sex to a male partner (t(99) = -14.83, p = .001), and penile-vaginal intercourse (t(99) = -19.27, p = .001). On the other hand, French kissing a male partner was perceived as less risky than giving digital sex to a female partner (t(99) = 3.90, p = .001), receiving digital sex from a female partner (t(99) = 5.16, p = .001), receiving oral sex from a female partner (t(99) = 6.79, p = .001), giving oral sex to a female partner (t(99) = 7.77, p = .001), receiving digital sex from a male partner (t(99) = -8.94, p = .001), receiving oral sex from a male partner (t(99) = -11.32, p = .001), giving oral sex to a male partner (t(99) = -13.06, p = .001), and penile-vaginal intercourse (t(99) = -16.97, p = .001). 70 Sexual Permissiveness & Risk Perception With regards to digital sex, giving digital sex to a female partner was perceived as less risky than receiving digital sex from a female partner (t(99) = 3.51, p = .001), receiving oral sex from a female partner (t(99) = 5.81, p = .001), giving oral sex to a female partner (t(99) = 6.74, p = .001), receiving digital sex from a male partner (t(99) = -6.85, p = .001), receiving oral sex from a male partner (t(99) = -9.25, p = .001), giving oral sex to a male partner (t(99) = -10.47, p = .001), and penile-vaginal intercourse (t(99) = -14.73, p = .001). Receiving digital sex from a female partner was perceived as less risky than receiving oral sex from a female partner (t(99) = 3.80, p = .001), giving oral sex to a female partner (t(99) = 5.35, p = .001), receiving digital sex from a male partner (t(99) = -5.54, p = .001), receiving oral sex from a male partner (t(99) = -7.89, p = .001), giving oral sex to a male partner (t(99) = -9.33, p = .001), and penile-vaginal intercourse (t(99) = 14.65, p = .001). Receiving digital sex from a male partner was rated as less risky than receiving oral sex from a male partner (t(99) = 4.47, p = .001), giving oral sex to a male partner (t(99) = 5.86, p = .001), and penile-vaginal intercourse (t(99) = -11.00, p = .001). With regards to oral sex, receiving oral sex from a female partner was perceived as less risky than receiving oral sex from a male partner (t(99) = -6.42, p = .001), giving oral sex to a male partner (t(99) = -7.21, p = .001), and penile-vaginal intercourse (t(99) = -12.07, p = .001). Giving oral sex to a female partner was rated as less risky than giving oral sex to a male partner (t(99) = -6.74, p = .001) and penile-vaginal intercourse (t(99) = -11.60, p = 71 Sexual Permissiveness & Risk Perception .001). Receiving oral sex from a male partner was rated as less risky than giving oral sex to a male partner (t(99) = -3.70, p = .001) and penile-vaginal intercourse (t(99) = -9.42, p = .001). Giving oral sex to a male partner was rated as less risky than penile-vaginal intercourse (t(99) = -8.11, p = .001). Thus, similar to heterosexual participants, lesbian participants perceived kissing and digital sex as less risky than oral sex (within each gender) and penile-vaginal intercourse, with kissing perceived as the least risky and penile-vaginal intercourse as the riskiest. A similar hierarchy of perceived risks associated with the different permissive sexual activities can be derived for lesbian participants; this hierarchy is depicted in Figure 1. For both heterosexual and lesbian participants, hypothesis 3b was therefore supported. Relationship between perceived risk and acceptability of permissive sexual activities Pearson correlations were computed to examine this relationship at a general level and with respect to each permissive sexual activity. Perceived risk associated with and acceptability of sexual permissiveness. The Pearson correlation between risk and acceptability here was negative as expected, r(201) = -.42, p < .01. Thus, as risk perceived increased, acceptability of sexual permissiveness decreased, supporting hypothesis 4a. About 17.39% of variance overlapped between risk perceived and acceptability of sexual permissiveness. 72 Sexual Permissiveness & Risk Perception Correlations were also computed for each sexual orientation group. For heterosexual participants, the correlation between risk perceived and acceptability of sexual permissiveness was -.40 (p < .01), while that for lesbian participants was -.34 (p < .01). Using Fisher’s r-Z transformation, these two correlations were found not to differ, Ζ = 0.34, ns. Therefore hypothesis 4b was not supported. Perceived risk associated with and acceptability of the different permissive sexual activities. The correlations across the different permissive sexual activities were all negative (from -.32 to -.53), indicating that as risk perceived increased, acceptability decreased. For heterosexual participants, correlations ranged from .32 to -.53, while for lesbian participants, correlations ranged from -.34 to -.48. Using Fisher’s r-Z transformation, the correlations of heterosexual and lesbian participants for each permissive sexual activity were found not to differ. Therefore hypothesis 4c was not supported. Discussion The present study was conducted to examine acceptability of sexual permissiveness as well as perceived risks associated with sexual permissiveness for lesbian and heterosexual women. Generally, both heterosexual and lesbian participants were not accepting of sexual permissiveness. However, lesbian participants were comparatively more accepting of sexually permissiveness than heterosexual participants. They also perceived less risk associated with sexual permissiveness. For both groups of women, as perceived risk increases, acceptability of sexual permissiveness decreases, 73 Sexual Permissiveness & Risk Perception whether in general or with respect to specific permissive sexual activities. There was no difference between the two groups in the association between acceptability of sexual permissiveness and risk perception, whether in general or with respect to specific permissive sexual activities. This suggests that both heterosexual and lesbian participants take risk into account when forming an attitude towards sexual permissiveness in general or towards a particular permissive sexual activity. Thus, contrary to what might be expected from the concept of lesbian immunity, lesbian participants in this sample seemed to be in tune with the risks involved in sexual permissiveness and their acceptability of such permissiveness varied accordingly. There were no differences between heterosexual and lesbian participants in their ratings of the acceptability and risk perception of permissive sexual activities with male partners. However, heterosexual participants rated all female same-sex sexual activities as less acceptable. They also rated most female same-sex sexual activities as riskier (except French-kissing). Therefore, differences between the two sexual orientation groups arose for female same-sex sexual activities only. One possible explanation for this result is that heterosexual participants might be less familiar with female same-sex sexual activities and therefore rated these as less acceptable and riskier. Differences in the acceptability of the different permissive sexual activities within the heterosexual group of participants possibly reflect varying levels of relationship commitment. Heterosexual participants differentiated between kissing (partners 74 Sexual Permissiveness & Risk Perception of both genders) and penile-vaginal intercourse. They also differentiated between kissing a female partner and giving oral sex (to partners of both genders). The relationship commitment required for oral sex is likely greater than that for kissing. In Herold and Way’s (Billy & Tanfer, 1993) sample of tertiary female students, oral sex (with men) correlated with level of dating commitment. The greater the commitment perceived, the more likely they were to engage in oral sex with their partners. A study on men revealed that oral sex was also more likely to occur in serious relationships as well (Halpern-Felsher, Cornell, Kropp, & Tschann, 2005). Oral sex was seen as something one can engage in before intercourse (Chambers, 2007). Some studies showed that it was more acceptable compared to intercourse among teenagers (C. Hendrick et al., 2006). Thus, oral sex signifies greater commitment than kissing, but less than penile-vaginal intercourse. In Trotter and Alderson’s (2007) heterosexual sample, the percentage of female participants who included each sexual activity in their definition of “sexual partner” increased from the least intimate to the most (4% for deep kissing, 40 to 50% for manual stimulation, 60% for oral stimulation and 80% for penile-vaginal intercourse). That means that 80% of the participants would consider someone with whom they engage in penilevaginal intercourse with as a sexual partner, but only 4% of the participants would consider someone with whom they engage in deep kissing (French kissing) with as a sexual partner. In the current study, French kissing was similarly rated as most acceptable and penile-vaginal intercourse as least acceptable. There was also some distinction between French kissing (female partners only) and giving 75 Sexual Permissiveness & Risk Perception oral sex. However, the rest of the sexual activities were not rated differently. This implies that the acceptability of other permissive sexual activities were similar to that of penile-vaginal intercourse. Lesbian participants did not make any distinctions for acceptability of sex types at all, suggesting that lesbian participants did not judge based on types of sexual activities. This is probably because those with same-sex attractions recognize sexual non-exclusivity (Smart, 2006), and other considerations, such as prior agreements, that can come into play. The differences in risk perception ratings within each sexual orientation group suggest that gender of sexual partner, mode of sexual activities and type of sexual activities make a difference in risk perceived for sexual permissiveness. However, there were similarities and differences in the way both groups rated. There were two similarities in the ratings of both groups of women. The first similarity was in the general hierarchy of sexual activities. Kissing was rated as least risky, followed by digital sex, oral sex and penile-vaginal intercourse. The second similarity was gender difference. Sexual activities with female partners were generally perceived as less risky than with male partners. However, lesbian women had more nuanced distinctions with gender, differentiating between modes (receiving or giving) as well. Similar distinction between genders occurred for receiving and giving oral sex. Here, both groups of women rated giving and receiving oral sex from either gender differently (with female partners deemed as less risky). However, lesbian participants further rated kissing and receiving digital sex from either gender 76 Sexual Permissiveness & Risk Perception differently (with female partners deemed as less risky), but heterosexual participants did not. Heterosexual participants made no distinction between modes (receiving and giving) at all. They rated three pairs of sex types equally: receiving and giving digital sex to a female partner, receiving and giving oral sex to a female partner, receiving and giving oral sex to a male partner. Lesbian participants, on the other hand, distinguished between receiving and giving digital sex to a female partner (with receiving digital as riskier), and receiving and giving oral sex to a male partner (with giving oral sex as riskier). Therefore, heterosexual and lesbian women might have slightly different criteria for risk. It appears that heterosexual participants decided on risk level predominantly by sexual activity and then by gender (for oral sex only). For example, kissing and digital sex were rated similarly regardless of gender and mode. Oral sex, although differentiated by gender, were not differentiated by mode. Lesbian participants appear to decide risk level by all three variables equally (sexual activity, gender and mode). Sexual activity was differentiated for different pairs of mode and gender. For example, receiving from a male partner digital sex was rated differently from receiving from a male partner oral sex. There were no overlaps. Gender was differentiated for different modes of every sexual activity. For example, receiving oral sex from a female partner was rated differently from receiving oral sex from a male partner. There were no overlaps. Finally, with the exception of female oral sex, mode was differentiated for various pairs of sexual activity and gender. For example, receiving oral sex from a male partner was rated 77 Sexual Permissiveness & Risk Perception differently from giving oral sex to a male partner. There were no other overlaps. Both heterosexual and lesbian participants might have judged level of risk of type of sexual activities based on exposure to bodily fluids (such as saliva, semen and vaginal fluids). French kissing involves exchange of saliva and is least associated with STI. This is because saliva does not contain large amounts of HIV (Minnesota AIDS Project, 2008), unless cuts, sores or bleedings gums are present in the mouth (Action for AIDS Singapore). Therefore, chances of infection through French kissing are low. Digital sex was rated as less risky than oral sex likely because it involves bodily fluid from one party only. Infection through digital sex is possible if there are cuts on the finger(s) used. Both saliva and semen or vaginal fluid are exchanged in oral sex, therefore it is rated as riskier than kissing (only saliva) and digital sex (contact with vaginal fluid only). Penile-vaginal intercourse involves exchange of vaginal fluid and semen. Local safe-sex campaigns’ focus on penilevaginal intercourse and promotion on condom use could have contributed to the perception that one is most susceptible to STI during penile-vaginal intercourse. Therefore it is rated as the riskiest. For lesbian participants, the level of control could have been important in risk perception as well. For digital sex, for example, it is probably easier to check for cuts on one’s finger/hand than to ensure that the finger/hand of one’s sexual partner has no open cuts. Therefore, it is riskier to receive digital sex than to give digital sex. Giving oral sex to a male partner might expose one to 78 Sexual Permissiveness & Risk Perception semen, whereas receiving oral sex from a male partner does not. Since saliva is likely perceived as less dangerous (like in French kissing), receiving oral sex is rated as less risky than giving oral sex. General Discussion Heterosexual and lesbian participants had many similarities and differences. From study 1, lesbian participants held similar views of relationships, such as the desire for commitment and monogamy, emotional and physical intimacy. The main difference between heterosexual and lesbian participants centered on marriage, which is currently not permitted by law for same-sex couples. With regard to sex, the two groups of participants were more dissimilar. Like previous studies (Berasmin & Fisher, 2007; Trotter & Alderson, 2007), heterosexual groups were more likely to generate penilevaginal intercourse. Lesbian groups were more likely to generate some forms of sex (such as digital sex) other than penile-vaginal intercourse. Thus, the propensity of participants to cite what is salient or relevant to them applies to lesbian women as well. Orgasms and partner satisfaction were featured more prominently among lesbian groups, indicating that lesbian women had different yardsticks for what constitutes sex. In fact, none of the heterosexual groups indicated orgasm as important for an activity to be described as “sex”. However, Randall and Byers (2003) found that their heterosexual participants were more likely to include a behavior in their definition of sex if orgasm occurred than if it did not. Trotter and Alderson (2007) had similar results. Perhaps Asian heterosexual women have a narrower definition of sex. Study 2 79 Sexual Permissiveness & Risk Perception shows that lesbian women also have different yardsticks for other aspects of sex (acceptability and risk perception) according to gender of partner. From study 1, both heterosexual and lesbian participants were concerned about sexual permissiveness. However, lesbian participants scored higher on acceptability of sexual permissiveness in study 2. On closer examination, the difference between groups in acceptability appears to be the result of sexual partner’s gender. Both heterosexual and lesbian participants found sexual permissiveness of sexual activities with male partners equally acceptable. However, lesbian participants were a lot more accepting of sexual permissiveness with female partners than heterosexual participants. The higher acceptability of sexual permissiveness coupled with a preference for monogamy in relationships probably mean that lesbian women are likely to be more sexually permissive outside of a relationship as compared to heterosexual women. Out of a relationship, lesbian women are no longer bound by pre-discussed relationship boundaries. Therefore, they are at liberty to have sexual activities with any woman. The lack of distinction for specific permissive sexual activities suggests that prior agreement on relationship boundaries is of paramount importance to lesbian women. Without additional information provided, lesbian participants were unable to judge acceptability of specific permissive sexual activities. Results from focus group discussions in study 1 support this argument. For instance, although lesbian participants preferred monogamy for themselves, they did not invalidate other forms of 80 Sexual Permissiveness & Risk Perception relationships (such as open relationships) as long as partners do not violate prior agreements of relationship boundaries. As a lesbian participant (Sharon) said, it is a matter of laying the cards on the table. Like lesbian participants, heterosexual participants also preferred monogamy for themselves. However, heterosexual participants defined relationships as monogamous. So, heterosexual participants would not recognize a non-monogamous arrangement as a relationship. These suggest that heterosexual and lesbian women have different markers of morality. Lesbian participants decide morality based on explicit agreements made between two parties, but heterosexual women do so based on whether monogamy is violated. This is congruent with Smart’s (2006) findings where heterosexuals equate non-monogamy with infidelity. During singlehood, heterosexual women are likely to be less sexually permissive than lesbian women. This is not surprising given the negative connotations associated with female sexual permissiveness in the heterosexual world. Words like “slut” are gendered, carrying meaning for sexually permissive heterosexual women, but not for their lesbian counterparts or men. Therefore, lesbian participants were likely more accepting of sexual permissiveness than heterosexual participants as judged by differing markers of morality and a lack of social judgment for lesbian permissiveness. Lesbian women also scored lower on perceived risk associated with sexual permissiveness. Upon examining group differences on different sex types, this appears to be the result of gender of sexual partner and type of sexual activity. There were no differences between heterosexual and lesbian participants on the 81 Sexual Permissiveness & Risk Perception risk perceived for permissive sexual activities with male partners. Lesbian participants, however, found permissive sexual activities with female partners less risky than heterosexual participants. Lesbian participants also perceived specific permissive sexual activities with a woman as less risky than with a man. Lesbian immunity, which is the perception of lesbian women being safe from STI, might be at work here (Montcalm & Myer, 2000). The way heterosexual participants differentiated perceived risk of specific permissive sex activities suggest that this viewpoint of women being safe from STI is not restricted to lesbian women only. Except for French kissing and receiving digital sex, heterosexual participants rated receiving and giving oral sex to a female partner as less risky than receiving and giving oral sex to a male partner. Lesbian participants appeared to be more entrenched in lesbian immunity. This sense of security is not due to disregard for risk, but arises out of low levels of perceived risk. The low perceived risk might mean that lesbian women are unlikely to use safe sex methods. Although lesbian women are likely to be more sexually permissive than their heterosexual counterparts, there is no evidence that they are likely to be unfaithful (a consequence of sexual permissiveness that can result in relationship misery for a monogamous arrangement). Their preference for monogamy suggests that lesbian women are likely to form monogamous relationships. As there were no differences between sexual orientation groups for all correlations between acceptability of and risk perceived of specific permissive sexual activities, and no difference between the two groups for the correlation between acceptability of sexual 82 Sexual Permissiveness & Risk Perception permissiveness and its perceived associated risk, lesbian women are likely to assign similar weight to risk as heterosexual women. These correlations were negative for both heterosexual and lesbian participants. Therefore, it is unlikely that most local lesbian women exchange sex for money or drugs, as these activities are highly risky and require high levels of sexual permissiveness. There is no evidence that lesbian women would perceive such activities to be low in risk. Heterosexual participants distinguished between French kissing and penile-vaginal intercourse, as well as French kissing (a female partner) and the giving of oral sex. The level of commitment might affect how heterosexual participants rate specific permissive sexual activities. From the focus group discussions in study 1, heterosexual participants shared that engaging in penile-vaginal intercourse is bringing the relationship to a higher level. This suggests that penile-vaginal intercourse requires the most relationship commitment of all sexual activities, followed by oral sex and digital sex, with kissing as requiring the least. As commitment increases, women are likely to move from self-protection to maintaining the relationship, and are less likely to engage in safe sex because they perceive their partners as monogamous (Tucker, Elliott, Wenzel, & Hambarsoomian, 2007). Therefore, with the move towards greater intimacy, sexual permissiveness (which heterosexual women perceive as a lack of commitment) might be less acceptable because it violates their expectations of monogamy and its perceived accompanying low sexual risk. Thus the hierarchy could be perceived as one of progressive intimacy and commitment. 83 Sexual Permissiveness & Risk Perception The current research extended previous studies by pointing to similar trends between acceptability of sexual permissiveness (for heterosexual women only), perceived risk of sexual permissiveness and definitions of sex for sexual activities. Specifically, what was least acceptable to be sexually permissive about (penile-vaginal intercourse), considered riskiest to be sexually permissive about (penile-vaginal intercourse), was also what had the highest agreement among participants that it is sex. The hierarchy (for sexual permissiveness, risk perception and sex definitions) for sexual activities is as follows: kissing, digital sex (or manual stimulation), oral sex and penile-vaginal intercourse. In a committed and monogamous relationship, these sexual activities would have been acceptable, seen as non-risky and would have been defined as part of the sexual repertoire. This is supported by Trotter and Alderson’s (2007) study, in which partners who had dated three months or more were considered by more participants as sexual partners, those who were not dating or had only one date. The trend was similar across all sexual activities. For example, more participants considered three-month dating partners as sexual partners when they deep-kissed (French kissed) than non-dating or one-date partners who deep-kissed. Therefore, the underlying mechanism for differences between sexual activities for definitions of sex appears to be the level of perceived required commitment that accompanies each sexual activity. This piece of research is the first comparative research on acceptability of sexual permissiveness and perceived risk associated with sexual permissiveness in heterosexual and lesbian women. It 84 Sexual Permissiveness & Risk Perception highlighted factors that are linked to sexual permissiveness, such as gender of partner and type of sexual activity. It also shows that lesbian women assign similar weight to risk as heterosexual women, which is different from saying that lesbian women are sexually permissive because of inherent sexual permissiveness. Conclusion and Future Research The results of these studies have implications for the local lesbian community. Lesbian women are more sexually permissive than heterosexual women and at the same time, they perceive less risk. Such perceptions might contribute towards lesbian women not taking protective measures during sex. If lesbian women are indeed more likely to be sexually permissive outside of a relationship, they will benefit from taking safety measures during sex with their sexual partners. Since the association between general sexual permissiveness and general risk perception did not differ between sexual orientation groups, the campaign for safe sex may be effective. Although female same-sex sexual activities are likely to hold less risk than heterosexual sexual activities, there is still an element of risk. If nothing is done, there may be a higher infection rate among lesbian women in future. Therefore, the results point to education regarding safe sex among local lesbian women. Future research should examine actual instances of sexual permissiveness and local lesbian HIV rates. It is also worthwhile to explore the link between sexual permissiveness and relationship longevity in the lesbian community, as well as what lesbian women consider as sexually permissive. For instance, experiencing a few 85 Sexual Permissiveness & Risk Perception one-night stands might not be considered sexually permissive in the lesbian community. Limitations This study did not cover the whole spectrum of women who are sexually attracted to women. The findings of this study are restricted to those who identify as lesbian and who are physically and romantically attracted to members of the same sex only. Respondents in this study were generally well educated and between ages 19 to 26. Heterosexual participants were acquired from the research pool of the department of Psychology. Majority of lesbian participants were in the midst of acquiring or had acquired tertiary education. Their sexual experience might differ from those who had not acquired such qualifications. Hence, different findings might occur with different samples of students or other age groups. 86 Sexual Permissiveness & Risk Perception References Action for AIDS Singapore. (n.d.). Frequently asked questions Retrieved December 30, 2008, from http://www.afa.org.sg/faq.asp Bersamin, M. M., Fisher, D. A., Walker, S., Hill, D. L., & Grube, J. W. (2007). Defining virginity and abstinence: Adolescents' interpretations of sexual behaviors. Journal of Adolescent Health, 41(2), 182-188. Billy, J. O. G., & Tanfer, K. (1993). The sexual behavior of men in the United States. Family Planning Perspectives, 25, 52-60. Carpenter, L. M. (1998). From girls into women: Scripts for sexuality and romance in Seventeen magazine, 1974-1994. Journal of Sex Research, 35(2), 158-168. Carpenter, L. M. (2001). The ambiguity of "having sex": The subjective experience of virginity loss in the United States. Journal of Sex Research, 38(2), 127-139. Chambers, W. C. (2007). Oral sex: Varied behaviors and perceptions in a college population. Journal of Sex Research, 44(1), 2842. Champion, J. D., Wilford, K., Shain, R. N., & Piper, J. M. (2005). Risk and protective behaviours of bisexual minority women: a qualitative analysis. International nursing review, 52, 115122. Diamant, A. L., Lever, J., & Schuster, M. A. (2000). Lesbians' sexual activities and efforts to reduce risks for sexually transmitted diseases. Journal of the Gay & Lesbian Medical Assn, 4(2), 41-48. 87 Sexual Permissiveness & Risk Perception Diamond, L. M. (2005a). A new view of lesbian subtypes: stable versus fluid identity trajectories over an 8-year period. Psychology of Women Quarterly, 29, 119-128. Diamond, L. M. (2005b). What we got wrong about sexual identity development: unexpected findings from a longitudinal study of young women. In A. M. Omoto & H. S. Kurtzman (Eds.), Sexual orientation and mental health (pp. 73- 94). Washington, DC: American Psychological Association. Dolan, K. A. (2005). Lesbian women and sexual health: the social construction of risk and susceptibility. New York: Haworth Press. Durrheim, K., & Tredoux, C. (2004). Numbers, hypotheses and conclusions: a course in statistics for the Social Sciences. Juta & Company Limited. Dworkin, S. L. (2005). Who is epidemiologically fathomable in the HIV?AIDS epidemic? Gender, sexuality, and intersectionality in public health. Culture, Health & Sexuality, 7(6), 615-623. Fishman, S. J., & Anderson, E. H. (2003). Perception of HIV and safer sexual behaviors among lesbians. JANAC: Journal of the Association of Nurses in AIDS Care, 14(6), 48-55. Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when things go right? The intrapersonal and interpersonal benefits of sharing positive events. Journal of Personality and Social psychology, 87(2), 228-245. Gangestad, S. W., Bailey, J. M., & Martin, N. G. (2000). Taxometric analyses of sexual orientation and gender identity. Journal of Personality and Social Psychology, 78, 1109-1121. 88 Sexual Permissiveness & Risk Perception Gershman, H. (1997). Sexual permissiveness and its consequences. Springfield, IL: Charles C Thomas Publisher. Halpern-Felsher, B. L., Cornell, J. L., Kropp, R. Y., & Tschann, J. M. (2005). Oral versus vaginal sex among adolescents: perceptions, attitudes, and behavior. Pediatrics, 115, 845851. Harley, M. (2003). Straight talk: the myths and facts about homosexuality: Focus on the Family. Hendrick, C., Hendrick, S. S., & Reich, D. A. (2006). The Brief Sexual Attitudes Scale. Journal of Sex Research, 43(1), 76-86. Hendrick, S. S., & Hendrick, C. (1987). Love and sex attitudes and religious beliefs. Journal of Social and Clinical Psychology, 5, 391-398. Herold, E. S., & Mewhinney, D.-M. K. (1993). Gender differences in casual sex and AIDS prevention: A survey of dating bars. Journal of Sex Research, 30(1), 36-42. Herold, E. S., & Way, L. (1983). Oral-genital sexual behavior in a sample of university females. Journal of Sex Research, 19, 327-338. Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organization of sexuality: sexual practices in the United States. Chicago: University of Chicago Press. Laurenceau, J. P., Barrett, L. F., & Rovine, M. J. (2005). The interpersonal process model of intimacy in marriage. Journal of family Psychology, 19(2), 314-323. 89 Sexual Permissiveness & Risk Perception LeGall, A., Mullet, E., & Riviere-Shafighi, S. (2002). Age, religious beliefs, and sexual attitudes. The Journal of Sex Research, 39, 207-216. Macfarquhar, N. (2008, December 18). In a first, gay rights are pressed at the U.N. The New York Times. Retrieved January 15, 2009, from http://www.nytimes.com/2008/12/19/world/19nations.html?ref=wor ld Mackey, R. A., Diemer, M. A., & O'Brien, B. A. (2000). Psychological intimacy in the lasting relationships of heterosexual and same-gender couples- statistical data included. Sex roles, 43(3-4), 201-227. Markey, P. M., & Markey, C. N. (2007). The interpersonal meaning of sexual promiscuity. Journal of Research in Personality, 41, 1199-1212. Marrazzo, J. M., Coffey, P., & Bingham, A. (2005). Sexual practices, risk perception and knowledge of sexually transmitted disease risk among lesbian and bisexual women. Perspectives on Sexual and Reproductive Health, 37(1), 6-12. McNulty, J. K., & Karney, B. R. (2004). Positive expectations in the early years of marriage: should couples expect the best or brace for the worst? Journal of Personality and Social psychology, 86(5), 729-743. Metz, M. E., & McCarthy, B. W. (2007). The "good-enough sex" model for couple sexual satisfaction. Sexual and relationship therapy, 22(3), 351-362. 90 Sexual Permissiveness & Risk Perception Ministry of Health, S. (2008). Ministry of Health, Singapore: HIV stats. Retrieved. from http://www.moh.gov.sg/mohcorp/statistics.aspx?id=246#table1 Minnesota AIDS Project. HIV Transmission Retrieved December 30, 2008, from http://www.mnaidsproject.org/learn/transmission.htm Montcalm, D. M., & Myer, L. L. (2000). Lesbian immunity from HIV/AIDS: Fact or fiction? Journal of Lesbian Studies, 4(2), 131-147. Morrow, K., & Allsworth, J. (2000). Sexual risk in lesbians and bisexual women. Journal of the Gay and Lesbian Medical Association, 4(4), 159-165. Patrick, S., Sells, J. N., Giordano, F. G., & Tollerud, T. R. (2007). Intimacy, differentiation, and personality variables as predictors of marital satisfaction. The family journal, 15(4), 359-367. Pitts, M., & Rahman, Q. (2001). Which behaviors constitute "having sex" among university students in the UK? Archives of Sexual Behavior, 30(2), 169-176. Ramesh, S. (2007, September 22). PM Lee fielded questions on gays, foreign talent at NUS forum. Retrieved December 24, 2008, from http://www.channelnewsasia.com/stories/singaporelocalnews/view /301420/1/.html Randall, H. E., & Byers, E. S. (2003). What is sex? Students' definitions of having sex, sexual partner, and unfaithful sexual behaviour. The Canadian journal of human sexuality, 12(2), 87-96. 91 Sexual Permissiveness & Risk Perception Richardson, D. (2000). The social construction of immunity: HIV risk perception and prevention among lesbians and bisexual women. Culture, Health & Sexuality, 2(1), 33-49. Rust, P. (1993). Coming out in the age of social constructionism: sexual identity formation among lesbians and bisexual women. Gender and Society, 7, 50-77. Sanderson, C. A., Keiter, E. J., Miles, M. G., & Yopyk, D. J. (2007). The association between intimacy goals and plans for initiating dating relationships. Personal Relationships, 14(2), 225-243. Schilling, J. (2006). On the pragmatics of qualitative assessment, designing the process for content analysis. European journal of psychological assessment, 22(1), 28-37. Smart, C. (2006). Review of the state of affairs: explorations in infidelity and commitment. Sexualities, 9(2), 259-262. Trotter, E. C., & Alderson, K. G. (2007). University students' definitions of having sex, sexual partner, and virginity loss: the influence of participant gender, sexual experience, and contextual factors. The Canadian journal of human sexuality, 16(1-2), 11-29. Tucker, J. S., Elliott, M. N., Wenzel, S. L., & Hambarsoomian, K. (2007). Relationship commitment and its implications for unprotected sex among impoverished women living in shelters and low-income housing in Los Angeles County. Health Psychology, 26(5), 644-649. Wallerstein, J. S. (1996). The psychological tasks of marriage: part 2. American journal of orthopsychiatry, 66(2), 217-227. 92 Sexual Permissiveness & Risk Perception Weinberg, M. S., Williams, C. J., & Proyer, D. W. (1994). Dual attraction: understanding bisexuality. New York: Oxford University Press. Widom, C. S., & Kuhms, J. B. (1996). Childhood victimization and subsequent risk for promiscuity, prostitution, teenage pregnancy: a prospective study. American Journal of Public Health, 86(11), 1607-1612. Wong, K. H. (2008, December 7). Surge in HIV cases among gay men The Straits Times, p 16. 93 [...]... comparing lesbian and heterosexual women with respect to different aspects of sexual permissiveness (such as how acceptable and how risky such permissiveness is) Study 2 Study 2 was conducted to examine and compare sexual permissiveness, in terms of acceptability and perceived risk, of heterosexual and lesbian women Drawing on the prevailing environment (e.g., religious groups warning that lesbian women. .. Study 1 indicate that relationship context is important and needs to be taken into account when examining sex-related differences between lesbian and heterosexual women (e.g., while both lesbian and heterosexual women in Study 1 desired commitment and monogamy in their relationships, lesbian women did not equate monogamy with commitment the way heterosexual women did) To accurately examine sexual permissiveness. .. terms in the questionnaire were clarified using the results of study 1, so as to maximize similar interpretation of questionnaire items Study 1 Before examining differences between lesbian and heterosexual women in how they perceived acceptability and risk in sexual permissiveness, it is important first of all to determine whether heterosexual and lesbian women perceive relationships, sex, and virginity... sexual permissiveness (hypothesis 1a) and to perceive less risk in being sexually permissive (hypothesis 1b) than heterosexual women 34 Sexual Permissiveness & Risk Perception Although researchers had examined self-perceived risk for HIV in lesbian women and the types of sexual activities that lesbian women engage in, these studies did not identify the specific perceived risk associated with each sexual. .. encompassed understanding perceptions of relationships, sex, and virginity Focus group discussions with heterosexual and lesbian women were employed towards this end The second study sought to tease out the nuances in acceptability and risk perceptions when different sexual orientations (specifically heterosexual or lesbian) , genders and sex types (such as penile-vaginal intercourse or kissing) come into play... activities other than penile-vaginal intercourse remains a virgin) There is likely to be more pressure on young unmarried women to retain their virginity Non-virgin status might be construed as indicating sexual promiscuity In sum, study 1 was done to examine and compare how lesbian and heterosexual women viewed relationships, sex, and virginity Method 15 Sexual Permissiveness & Risk Perception Participants... women are familiar with In doing so, the accompanying perceptions of acceptability and risk of these common sexual activities can then be assessed with confidence Eliciting responses on the purposes of being in a relationship and having sex might also shed light on differences between lesbian and heterosexual women that could contribute to the negative stereotype of promiscuity Knowing the purposes can... boundaries and expectations of being in a relationship for these two groups of women Available literature suggests that there are possible differences between heterosexual and lesbian women in how they view relationships, sex, and virginity Heterosexual and lesbian women, for instance, are likely to differ in their characteristics of what a romantic relationship is There is evidence that heterosexuals... other hand, more lesbian groups than heterosexual groups brought up oral sex (all of the lesbian groups vs none of the heterosexual groups), digital sex (all lesbian groups vs 4 out of 7 heterosexual groups) and external genital stimulation (4 out of the 6 lesbian groups and none of the heterosexual groups) Only one heterosexual group and one lesbian group cited anal sex All heterosexual groups and 5... acceptability of and risks associated with sexual permissiveness are perceived Secondly, heterosexual and lesbian women may define sex and virginity differently according to different predominant sexual activities, different gender of partner, and presence of orgasm This has implications for use of the term “sex” for each group of women Lesbian women have been reported to engage in oral-vaginal sex, oral-anal ... acceptable) and risk (how risky) ratings associated with such permissiveness among lesbian women, and to contrast these acceptability and risk perceptions against those of heterosexual women In many... 2005 and 2007, the odds of married heterosexual women being infected were more than twice that of single women (the statistics for single women not distinguish between heterosexual and lesbian women) ... possible differences between heterosexual and lesbian women in how they view relationships, sex, and virginity Heterosexual and lesbian women, for instance, are likely to differ in their characteristics