《性别、亲密关系与社会》课程教学资源：Romantic Partners, Friends, Friends with Benefits, and Casual Acquaintances as Sexual Partners

JOURNAL OF SEX RESEARCH.48(6).554-564.2011 Copyright The Society for the Scientific Study of Sexuality Routledge ISSN:0022-4499 print/1559-8519 online Taylor &Francis Group D0L:10.1080/00224499.2010.535623 Romantic Partners,Friends,Friends with Benefits, and Casual Acquaintances as Sexual Partners Wyndol Furman Department of Psychology,University of Denver Laura Shaffer Department of Psychology,University of Louisville School of Medicine The purpose of this study was to provide a detailed examination of sexual behavior with different types of partners.A sample of 163 young adults reported on their light nongenital, heavy nongenital,and genital sexual activity with romantic partners,friends,and casual acquaintances.They described their sexual activity with"friends with benefits,"as well as with friends in general.Young adults were most likely to engage in sexual behavior with romantic partners,but sexual behavior also often occurred with some type of nonromantic partner. More young adults engaged in some form of sexual behavior with casual acquaintances than with friends with benefits.The frequencies of sexual behavior,however,were greater with friends with benefits than with friends or casual acquaintances.Interview and questionnaire data revealed that friends with benefits were typically friends,but not necessarily.Nonsexual activities were also less common with friends with benefits than other friends.Taken together, the findings illustrate the value of differentiating among different types of nonromantic partners and different levels of sexual behavior. Most research on sexual behavior has not considered the Furstenberg,1999;Manning,Longmore,Giordano, nature of the relationship in which it occurs.When 2000).Moreover,approximately one half of sexually the context of the relationship has been considered. active adolescents have had intercourse with a non- the research has focused on sexual behavior in romantic romantic partner (Grello,Welsh,Harper,2006; relationships or some subset of romantic relationships, Manning,Giordano,Longmore,2006;Manning, such as marriages or cohabitating couples (e.g.,Kaestle Longmore,Giordano,2005).About one half of these Halpern,2007;O'Sullivan,Mantsun,Harris, incidents with a nonromantic partner occurred only Brooks-Gunn,2007).Yet,the sexual behavior of young once (Manning et al.,2006).Similarly,approximately adults and adolescents often occurs in other contexts. 75%to 80%of college students reported"hooking up" Such sexual activity has been commonly described as or engaging in some form of sexual activity with some- casual sex,nonromantic sexual behavior,or "hookups." one for just one night (England,Shafer,Fogarty, The details of the definitions vary,but they have the 2008;Paul,McManus,Hayes,2000);30%reported common denominator of referring to sexual behavior hooking up with someone for the night and having in uncommitted relationships (Weaver Herold,2000). intercourse (Paul et al.,2000). Sexual intercourse usually occurs first in a romantic Most investigators have not differentiated among or committed relationship,but approximately 25%of different partners within the general category of casual the time,it first occurs with a friend,stranger,or some- or nonromantic sexual partners.Some investigators one the person is occasionally dating (Elo,King, have examined one particular category of nonromantic partners(e.g.,friends [Afifi Faulkner,2000]or friends This research was supported by Grant 50106 from the National with benefits [Bisson Levine,2009;Owen Fincham, Institute of Mental Health(to Wyndol Furman,Primary Investigator) 2010D),but it is not clear if their findings are specific to and Grant HD049080 from the National Institute of Child Health and that category or are applicable to other types of casual Human Development (to Wyndol Furman,Primary Investigator). Appreciation is expressed to the Project STAR staff for their contri- or nonromantic sexual partners. bution to the data collection and to the individuals,families,and In the two studies that did include multiple categories schools who are participating in Project STAR. (Grello et al.,2006;Manning et al.,2005),friends were Correspondence should be addressed to Wyndol Furman,Depart- the most typical type of partner.To date,relatively little ment of Psychology,University of Denver,2155 S.Race St.,Denver, is known about differences in the sexual activity with CO 80208.E-mail:wfurman@nova.psy.du.edu

Romantic Partners, Friends, Friends with Benefits, and Casual Acquaintances as Sexual Partners Wyndol Furman Department of Psychology, University of Denver Laura Shaffer Department of Psychology, University of Louisville School of Medicine The purpose of this study was to provide a detailed examination of sexual behavior with different types of partners. A sample of 163 young adults reported on their light nongenital, heavy nongenital, and genital sexual activity with romantic partners, friends, and casual acquaintances. They described their sexual activity with ‘‘friends with benefits,’’ as well as with friends in general. Young adults were most likely to engage in sexual behavior with romantic partners, but sexual behavior also often occurred with some type of nonromantic partner. More young adults engaged in some form of sexual behavior with casual acquaintances than with friends with benefits. The frequencies of sexual behavior, however, were greater with friends with benefits than with friends or casual acquaintances. Interview and questionnaire data revealed that friends with benefits were typically friends, but not necessarily. Nonsexual activities were also less common with friends with benefits than other friends. Taken together, the findings illustrate the value of differentiating among different types of nonromantic partners and different levels of sexual behavior. Most research on sexual behavior has not considered the nature of the relationship in which it occurs. When the context of the relationship has been considered, the research has focused on sexual behavior in romantic relationships or some subset of romantic relationships, such as marriages or cohabitating couples (e.g., Kaestle & Halpern, 2007; O’Sullivan, Mantsun, Harris, & Brooks-Gunn, 2007). Yet, the sexual behavior of young adults and adolescents often occurs in other contexts. Such sexual activity has been commonly described as casual sex, nonromantic sexual behavior, or ‘‘hookups.’’ The details of the definitions vary, but they have the common denominator of referring to sexual behavior in uncommitted relationships (Weaver & Herold, 2000). Sexual intercourse usually occurs first in a romantic or committed relationship, but approximately 25% of the time, it first occurs with a friend, stranger, or some￾one the person is occasionally dating (Elo, King, & Furstenberg, 1999; Manning, Longmore, & Giordano, 2000). Moreover, approximately one half of sexually active adolescents have had intercourse with a non￾romantic partner (Grello, Welsh, & Harper, 2006; Manning, Giordano, & Longmore, 2006; Manning, Longmore, & Giordano, 2005). About one half of these incidents with a nonromantic partner occurred only once (Manning et al., 2006). Similarly, approximately 75% to 80% of college students reported ‘‘hooking up’’ or engaging in some form of sexual activity with some￾one for just one night (England, Shafer, & Fogarty, 2008; Paul, McManus, & Hayes, 2000); 30% reported hooking up with someone for the night and having intercourse (Paul et al., 2000). Most investigators have not differentiated among different partners within the general category of casual or nonromantic sexual partners. Some investigators have examined one particular category of nonromantic partners (e.g., friends [Afifi & Faulkner, 2000] or friends with benefits [Bisson & Levine, 2009; Owen & Fincham, 2010]), but it is not clear if their findings are specific to that category or are applicable to other types of casual or nonromantic sexual partners. In the two studies that did include multiple categories (Grello et al., 2006; Manning et al., 2005), friends were the most typical type of partner. To date, relatively little is known about differences in the sexual activity with This research was supported by Grant 50106 from the National Institute of Mental Health (to Wyndol Furman, Primary Investigator) and Grant HD049080 from the National Institute of Child Health and Human Development (to Wyndol Furman, Primary Investigator). Appreciation is expressed to the Project STAR staff for their contri￾bution to the data collection and to the individuals, families, and schools who are participating in Project STAR. Correspondence should be addressed to Wyndol Furman, Depart￾ment of Psychology, University of Denver, 2155 S. Race St., Denver, CO 80208. E-mail: wfurman@nova.psy.du.edu JOURNAL OF SEX RESEARCH, 48(6), 554–564, 2011 Copyright # The Society for the Scientific Study of Sexuality ISSN: 0022-4499 print=1559-8519 online DOI: 10.1080/00224499.2010.535623

SEXUAL PARTNERS different partners.Grello et al.(2006),however,found then asked about sexual behavior with friends with that more affectionate sexual behavior (e.g.,handhold- benefits (see rationale in the Methods section).We ing,hugging,kissing,and massaging)occurred when distinguished among types of sexual behavior: they were friends than when they were acquaintances or strangers.Thus,the limited research suggests that sexual activity may vary across different kinds of non- 1."Light"nongenital acts (kissing on the lips, romantic partners. cuddling,and "making out"). Not only have most investigators failed to differen- 2. "Heavy"nongenital acts (light petting,heavy tiate among categories of nonromantic partners,but petting,and dry sex). they also have not typically distinguished among differ- 3. Genital acts (oral sex,vaginal intercourse,and ent types of sexual behaviors.Intercourse does not occur anal intercourse). in approximately 60%of hookups (Paul et al.,2000). Different sexual behaviors involve different levels of risk Based on the existing literature (e.g.,Grello et al, of sexually transmitted diseases.The type of sexual 2006;Manning et al.,2006),we predicted that young behavior that commonly occurs also varies as a function adults would be more likely to engage in light nongeni- of the type of sexual partner (Grello et al.,2006). tal,heavy nongenital,and genital sexual behaviors with Finally,genital,heavy nongenital,and light nongenital romantic partners than with nonromantic partners of sexual behaviors are differentially related to representa- any type (Hla).Moreover,we expected that the fre- tions of romantic relationships (Jones Furman,2010). quencies of all types of sexual behavior would be greater These findings suggest that it is important to distinguish with romantic partners than with any type of nonro- among different types of sexual behaviors. mantic partners because romantic relationships in early adulthood are more intimate in nature (Furman Buhrmester,1992;HIb).Based on prior research Friends with Benefits (Grello et al.,2006;Manning et al.,2006),we also pre- dicted that a greater proportion of young adults would Recently,the idea of "friends with benefits"has engage in sexual behaviors with friends than with casual received considerable attention in the mass media (e.g., acquaintances (H2a).The frequencies of sexual beha- Denizet-Lewis,2004).This relationship is commonly viors,especially light sexual behaviors such as kissing, described by laypersons as friends engaging in sexual cuddling,and making out,were also expected to be behavior without a monogamous relationship or any greater in friendships because of the affectionate nature kind of commitment (see http://www.urbandictionary. of the relationships (H2b).The limited literature on com/define.php?term=friends+with+benefits).Social friends with benefits provided little basis for predictions, scientists have similarly described them as friends engag- but we expected fewer participants would report engag- ing in sex or sexual activity (e.g.,Bisson Levine,2009). ing in sexual behavior with friends with benefits than What is less clear.however,is whether friends with with friends or casual acquaintances because a signifi- benefits are typically seen as a distinct category of sexual cant proportion of sexual activity with a nonromantic partners-that is,it is not apparent if all friends one has partner only occurs on one occasion,whereas being engaged in sexual activity with are considered friends friends with benefits may require establishing a relation- with benefits;for example,being a friend with benefits ship that involves some ongoing opportunities for sexual may imply some ongoing opportunities for sexual beha- behavior (H3a).When young adults have friends with vior,rather than a single episode.Some types of sexual benefits,however,we expected the frequency of sexual activity behavior may also be necessary to be considered behavior with friends with benefits to be higher than a friend with benefits.In addition,it is unclear if it is even the frequencies with friends or casual acquaintances necessary to first be a friend in the traditional sense of a because of the ongoing opportunities with friends with friend to be considered a friend with benefits.For benefits (H3b). example,it is not apparent if a casual acquaintance could Past work has consistently found that males have be considered a friend with benefits or not.A clearer greater interest in sexual behavior with nonromantic understanding of the nature of friends with benefits is partners(see Okami Shackelford,2001).To date,how- needed. ever,distinctions among different types of nonromantic partners have not been made.Gender differences may be less pronounced in friendships than in casual acquaint- This Study anceships,as friendships entail some level of intimacy that encounters with casual acquaintances may not. The purpose of this study was to provide a detailed Thus,we predicted gender differences in sexual behavior examination of sexual behavior with different types of with casual acquaintances(H4a),but tendered no predic- partners.We first asked about sexual behavior with tions regarding gender differences with friends or friends romantic partners,friends,and casual acquaintances;we with benefits.Although not as well-documented as the 555

different partners. Grello et al. (2006), however, found that more affectionate sexual behavior (e.g., handhold￾ing, hugging, kissing, and massaging) occurred when they were friends than when they were acquaintances or strangers. Thus, the limited research suggests that sexual activity may vary across different kinds of non￾romantic partners. Not only have most investigators failed to differen￾tiate among categories of nonromantic partners, but they also have not typically distinguished among differ￾ent types of sexual behaviors. Intercourse does not occur in approximately 60% of hookups (Paul et al., 2000). Different sexual behaviors involve different levels of risk of sexually transmitted diseases. The type of sexual behavior that commonly occurs also varies as a function of the type of sexual partner (Grello et al., 2006). Finally, genital, heavy nongenital, and light nongenital sexual behaviors are differentially related to representa￾tions of romantic relationships (Jones & Furman, 2010). These findings suggest that it is important to distinguish among different types of sexual behaviors. Friends with Benefits Recently, the idea of ‘‘friends with benefits’’ has received considerable attention in the mass media (e.g., Denizet-Lewis, 2004). This relationship is commonly described by laypersons as friends engaging in sexual behavior without a monogamous relationship or any kind of commitment (see http://www.urbandictionary. com/define.php?term=friends+with+benefits). Social scientists have similarly described them as friends engag￾ing in sex or sexual activity (e.g., Bisson & Levine, 2009). What is less clear, however, is whether friends with benefits are typically seen as a distinct category of sexual partners—that is, it is not apparent if all friends one has engaged in sexual activity with are considered friends with benefits; for example, being a friend with benefits may imply some ongoing opportunities for sexual beha￾vior, rather than a single episode. Some types of sexual activity behavior may also be necessary to be considered a friend with benefits. In addition, it is unclear if it is even necessary to first be a friend in the traditional sense of a friend to be considered a friend with benefits. For example, it is not apparent if a casual acquaintance could be considered a friend with benefits or not. A clearer understanding of the nature of friends with benefits is needed. This Study The purpose of this study was to provide a detailed examination of sexual behavior with different types of partners. We first asked about sexual behavior with romantic partners, friends, and casual acquaintances; we then asked about sexual behavior with friends with benefits (see rationale in the Methods section). We distinguished among types of sexual behavior: 1. ‘‘Light’’ nongenital acts (kissing on the lips, cuddling, and ‘‘making out’’). 2. ‘‘Heavy’’ nongenital acts (light petting, heavy petting, and dry sex). 3. Genital acts (oral sex, vaginal intercourse, and anal intercourse). Based on the existing literature (e.g., Grello et al., 2006; Manning et al., 2006), we predicted that young adults would be more likely to engage in light nongeni￾tal, heavy nongenital, and genital sexual behaviors with romantic partners than with nonromantic partners of any type (H1a). Moreover, we expected that the fre￾quencies of all types of sexual behavior would be greater with romantic partners than with any type of nonro￾mantic partners because romantic relationships in early adulthood are more intimate in nature (Furman & Buhrmester, 1992; H1b). Based on prior research (Grello et al., 2006; Manning et al., 2006), we also pre￾dicted that a greater proportion of young adults would engage in sexual behaviors with friends than with casual acquaintances (H2a). The frequencies of sexual beha￾viors, especially light sexual behaviors such as kissing, cuddling, and making out, were also expected to be greater in friendships because of the affectionate nature of the relationships (H2b). The limited literature on friends with benefits provided little basis for predictions, but we expected fewer participants would report engag￾ing in sexual behavior with friends with benefits than with friends or casual acquaintances because a signifi- cant proportion of sexual activity with a nonromantic partner only occurs on one occasion, whereas being friends with benefits may require establishing a relation￾ship that involves some ongoing opportunities for sexual behavior (H3a). When young adults have friends with benefits, however, we expected the frequency of sexual behavior with friends with benefits to be higher than the frequencies with friends or casual acquaintances because of the ongoing opportunities with friends with benefits (H3b). Past work has consistently found that males have greater interest in sexual behavior with nonromantic partners (see Okami & Shackelford, 2001). To date, how￾ever, distinctions among different types of nonromantic partners have not been made. Gender differences may be less pronounced in friendships than in casual acquaint￾anceships, as friendships entail some level of intimacy that encounters with casual acquaintances may not. Thus, we predicted gender differences in sexual behavior with casual acquaintances (H4a), but tendered no predic￾tions regarding gender differences with friends or friends with benefits. Although not as well-documented as the SEXUAL PARTNERS 555

FURMAN AND SHAFFER gender differences with nonromantic partners,women who did and did not participate in Wave 5 did not differ appear to be more likely to engage in intercourse and on any of 18 primary demographic.adjustment,and have higher frequencies of intercourse with romantic romantic and sexual variables collected at Wave 1.For partners than men(Carver,Joyner,Udry,2003;Prince the purpose of this study,we limited the sample to the Bernard,1998).We expected that we would replicate Wave 5 participants who were not married,engaged,or these gender differences with romantic partners and find cohabiting with someone (N=163;86 men and 77 similar gender differences in the occurrence and fre- women). quency of light nongenital and heavy nongenital behavior With regard to sexual orientation,87%said they were with romantic partners(H4b). heterosexual(straight),whereas the remaining partici- Another purpose of the study was to obtain a better pants said they were bisexual,gay,lesbian,or question- understanding of the nature of friends with benefits. ing.We chose to retain the sexual minorities in the As previously noted,it is not clear how similar friends sample to be inclusive and because the majority of them with benefits are to other friends.Because the focus of reported that they were either bisexual or questioning relationships with friends with benefits appears to be their sexual identity. on sexual activity,we hypothesized that young adults Participants were financially compensated for com- would engage in fewer nonsexual activities with friends pleting the questionnaires.The confidentiality of the with benefits than with typical friends;at the same time, participants'data was protected by a Certificate of we hypothesized that they would engage in more non- Confidentiality issued by the U.S.Department of Health sexual activities with friends with benefits than with and Human Services. casual acquaintances because friends with benefits appear to be ongoing relationships(H5). Finally,we interviewed young adults to obtain a Measures better understanding about their conceptualization of Sexual behavior questionnaire.Participants were friends with benefits.We hypothesized that most would first asked about their sexual behavior in the last 12 require friends with benefits to be friends,and would months with three types of partners:(a)romantic part- require that there be an ongoing opportunity for sexual ners,(b)friends,and (c)casual acquaintances or some- behavior (vs.a one-time experience;H6). one they just met.The participants were told they were going to be asked about all three types in advance, and the order of the questions concerning the three rela- Method tionships was fixed to eliminate potential confusion of categories (e.g.,romantic partners are often considered Participants friends as well). After they had answered the questions about the first The participants were part of a longitudinal study three types of sexual partners,we asked them to answer a investigating the role of relationships with parents,peers, parallel set of questions about friends with benefits. and romantic partners on psychosocial adjustment in Because it was unclear how friends with benefits would adolescence and young adulthood.Two hundred 10th- be categorized and how distinct they were from other grade high school students(100 boys and 100 girls;mean categories,we indicated that the term can be defined in age=15.88 years;range=14-16 years old)were orig- different ways and asked participants to use their own inally recruited from a diverse range of neighborhoods definition of friends with benefits,even if their partners and schools in a large,Western,metropolitan area by in this category overlapped with some of their partners distributing brochures and sending letters to families in the categories they had answered about already.This residing in various zip codes and to students enrolled in strategy allowed us to examine how a term was naturally various schools in ethnically diverse neighborhoods. used and provided a means of obtaining information Designed to be relatively representative of the eth- about whom young adults consider to be friends with nicity of the United States,the sample was 11.5%African benefits.We also believed that our strategy would be less American,12.5%Hispanic,1.5%Native American,1% confusing to the participants than initially asking them Asian American,4%biracial,and 69.5%White (non- about all four categories when we expected that the Hispanic).The sample was of average intelligence and friend with benefits category overlapped with the other did not differ from national norms on 11 of 12 measures categories,especially friends.We also thought it would of adjustment (see Furman,Low,Ho,2009).In the be inappropriate to force the four categories to not over- fifth wave of data collection.which was collected in lap with each other when they were likely to overlap in 2005 through 2007,we asked about sexual activity with actuality;moreover,we thought it would be confusing different types of partners.At that time,participants ran- to ask participants about sexual behavior with friends ged in age from 19.32 to 21.93 years old (M=20.51 who were not friends with benefits,or to ask them about years);186(94 men and 92 women)of the original 200 friends with benefits who were not friends or casual participants took part in the Wave 5 assessment.Those acquaintances. 556

gender differences with nonromantic partners, women appear to be more likely to engage in intercourse and have higher frequencies of intercourse with romantic partners than men (Carver, Joyner, & Udry, 2003; Prince & Bernard, 1998). We expected that we would replicate these gender differences with romantic partners and find similar gender differences in the occurrence and fre￾quency of light nongenital and heavy nongenital behavior with romantic partners (H4b). Another purpose of the study was to obtain a better understanding of the nature of friends with benefits. As previously noted, it is not clear how similar friends with benefits are to other friends. Because the focus of relationships with friends with benefits appears to be on sexual activity, we hypothesized that young adults would engage in fewer nonsexual activities with friends with benefits than with typical friends; at the same time, we hypothesized that they would engage in more non￾sexual activities with friends with benefits than with casual acquaintances because friends with benefits appear to be ongoing relationships (H5). Finally, we interviewed young adults to obtain a better understanding about their conceptualization of friends with benefits. We hypothesized that most would require friends with benefits to be friends, and would require that there be an ongoing opportunity for sexual behavior (vs. a one-time experience; H6). Method Participants The participants were part of a longitudinal study investigating the role of relationships with parents, peers, and romantic partners on psychosocial adjustment in adolescence and young adulthood. Two hundred 10th￾grade high school students (100 boys and 100 girls; mean age ¼ 15.88 years; range ¼ 14–16 years old) were orig￾inally recruited from a diverse range of neighborhoods and schools in a large, Western, metropolitan area by distributing brochures and sending letters to families residing in various zip codes and to students enrolled in various schools in ethnically diverse neighborhoods. Designed to be relatively representative of the eth￾nicity of the United States, the sample was 11.5% African American, 12.5% Hispanic, 1.5% Native American, 1% Asian American, 4% biracial, and 69.5% White (non– Hispanic). The sample was of average intelligence and did not differ from national norms on 11 of 12 measures of adjustment (see Furman, Low, & Ho, 2009). In the fifth wave of data collection, which was collected in 2005 through 2007, we asked about sexual activity with different types of partners. At that time, participants ran￾ged in age from 19.32 to 21.93 years old (M ¼ 20.51 years); 186 (94 men and 92 women) of the original 200 participants took part in the Wave 5 assessment. Those who did and did not participate in Wave 5 did not differ on any of 18 primary demographic, adjustment, and romantic and sexual variables collected at Wave 1. For the purpose of this study, we limited the sample to the Wave 5 participants who were not married, engaged, or cohabiting with someone (N ¼ 163; 86 men and 77 women). With regard to sexual orientation, 87% said they were heterosexual (straight), whereas the remaining partici￾pants said they were bisexual, gay, lesbian, or question￾ing. We chose to retain the sexual minorities in the sample to be inclusive and because the majority of them reported that they were either bisexual or questioning their sexual identity. Participants were financially compensated for com￾pleting the questionnaires. The confidentiality of the participants’ data was protected by a Certificate of Confidentiality issued by the U. S. Department of Health and Human Services. Measures Sexual behavior questionnaire. Participants were first asked about their sexual behavior in the last 12 months with three types of partners: (a) romantic part￾ners, (b) friends, and (c) casual acquaintances or some￾one they just met. The participants were told they were going to be asked about all three types in advance, and the order of the questions concerning the three rela￾tionships was fixed to eliminate potential confusion of categories (e.g., romantic partners are often considered friends as well). After they had answered the questions about the first three types of sexual partners, we asked them to answer a parallel set of questions about friends with benefits. Because it was unclear how friends with benefits would be categorized and how distinct they were from other categories, we indicated that the term can be defined in different ways and asked participants to use their own definition of friends with benefits, even if their partners in this category overlapped with some of their partners in the categories they had answered about already. This strategy allowed us to examine how a term was naturally used and provided a means of obtaining information about whom young adults consider to be friends with benefits. We also believed that our strategy would be less confusing to the participants than initially asking them about all four categories when we expected that the friend with benefits category overlapped with the other categories, especially friends. We also thought it would be inappropriate to force the four categories to not over￾lap with each other when they were likely to overlap in actuality; moreover, we thought it would be confusing to ask participants about sexual behavior with friends who were not friends with benefits, or to ask them about friends with benefits who were not friends or casual acquaintances. FURMAN AND SHAFFER 556

SEXUAL PARTNERS For each type of partner,participants were asked Results about the frequency of engaging in nine types of sexual behaviors during the last year:(a)three kinds of light Preliminary Analyses and Descriptive Information nongenital acts (kissing on the lips,cuddling,and All variables were examined to determine if the making out),(b)three kinds of heavy nongenital acts (light petting,heavy petting,and dry sex),and(c)three assumptions of univariate and multivariate analyses kinds of genital acts (oral sex,vaginal intercourse,and were met (Behrens,1997).Outliers were adjusted to fall anal intercourse).The distinctions among light non- 1.5 times the interquartile range below the 25th percen- genital,heavy nongenital,and genital acts were based tile or above the 75th percentile.All the resulting variables had acceptable levels of skew and kurtosis. on structural equation models examining the structure of sexual behavior (Jones Furman,2010).They rated the frequency of sexual activity using a scale ranging Occurrence of Sexual Behaviors with Four Types from 1 (not in the last 12 months)to 8 (almost every of Partners day or every day). Participants were also asked if they had commonly Table 2 presents the proportion of men and women engaged in each of 24 nonsexual activities with indivi- engaging in each level of sexual behavior with each of duals in each of the four relationship categories.Sample the four types of partners.These proportions include activities included drinking alcohol,watching TV,and both participants who engaged in additional forms of sharing something personal.The questions about friends sexual activity,as well as those who engaged in no more and casual acquaintances asked about all casual than that level of sexual activity;thus,the light nongeni- acquaintances and friends,not just sexual ones,so that tal proportions includes those who engaged in only light we could see if friends with benefits were similar to other nongenital activity and those who engaged in heavy casual acquaintances and friends.The sexual behavior nongenital or genital activity as well.(All participants questionnaire was administered by computer-assisted who reported engaging in genital sexual activity with a self-interviewing techniques to encourage participants particular type of partner had also reported engaging to respond honestly (Turner,Ku,Rogers,1998). in light nongenital and heavy nongenital sexual activity with that type of partner;similarly,all participants who reported engaging in heavy nongenital sexual activity Characteristics of friends with benefits.In light of with a particular type of partner had also reported the limited information regarding friends with benefits, engaging in light nongenital sexual activity with that we also asked participants a series of questions to clarify type of partner.) the nature of these relationships.Specifically,we asked We conducted the equivalent of two-way (Participant them whether friends with benefits are different from Gender x Partner Type)repeated-measures multivariate romantic partners,friends,and casual acquaintances analyses of variance (MANOVAs)for each level of (see the questions in Table 1).Responses were categor- sexual behavior using generalized hierarchical linear ized as yes,no,or qualified (e.g.,"it depends").We also modeling (for a description of the equivalence,see asked about the frequency of sexual encounters neces- Kenny,Bolger,Kashy,2002).Generalized hierarchi- sary to consider someone a friend with benefits.The cal linear modeling takes into account the nested nature questions regarding the characteristics of friends with of the data,including the dependency inherent in the benefits were not added until the first third of the data overlapping nature of the friends with benefits category. had been collected.The 109 participants who answered Unlike repeated-measures MANOVAs,generalized these questions did not differ from the other participants hierarchical linear modeling does not require ordinal on any of the primary variables of the study or in terms data and can be used to analyze proportional data by of gender and ethnicity. using a logit link function.In addition,it permits Table 1.Proportion (and n)of Participants'Answers to Questions about the Characteristics of Friends with Benefits Question Yes No Other 1.Are friends with benefits different from a romantic relationship? 97%(100) 3%(3) 0%(0) 3.Are they different from a casual sexual partner? 60%(66 37%(41) 3%(3) 7.Are friends with benefits different from other people you periodically hook up with? 58%(64) 41%(45) 0%(0) 2.Other than sexual behavior,are they different from friends? 27%(29) 73%(79) 0%(0) 4.Do they have to be a friend? 72%(79) 26%(28) 2%(2) 5.Could it be someone you don't know as well,such as a casual acquaintance or a stranger? 47%(51) 53%(58) 0%(0) 6.Would someone be a friend with benefits if you only engaged in sexual behavior with him or her once? 14%(15) 86%(94) 0%(0) Note.The number preceding the question refers to the order of the questions.The category of"other"answers refers to qualified ones that were not simple yes or no responses (e.g.,"it depends").Numbers slightly vary across questions because of technical problems 557

For each type of partner, participants were asked about the frequency of engaging in nine types of sexual behaviors during the last year: (a) three kinds of light nongenital acts (kissing on the lips, cuddling, and making out), (b) three kinds of heavy nongenital acts (light petting, heavy petting, and dry sex), and (c) three kinds of genital acts (oral sex, vaginal intercourse, and anal intercourse). The distinctions among light non￾genital, heavy nongenital, and genital acts were based on structural equation models examining the structure of sexual behavior (Jones & Furman, 2010). They rated the frequency of sexual activity using a scale ranging from 1 (not in the last 12 months) to 8 (almost every day or every day). Participants were also asked if they had commonly engaged in each of 24 nonsexual activities with indivi￾duals in each of the four relationship categories. Sample activities included drinking alcohol, watching TV, and sharing something personal. The questions about friends and casual acquaintances asked about all casual acquaintances and friends, not just sexual ones, so that we could see if friends with benefits were similar to other casual acquaintances and friends. The sexual behavior questionnaire was administered by computer-assisted self-interviewing techniques to encourage participants to respond honestly (Turner, Ku, & Rogers, 1998). Characteristics of friends with benefits. In light of the limited information regarding friends with benefits, we also asked participants a series of questions to clarify the nature of these relationships. Specifically, we asked them whether friends with benefits are different from romantic partners, friends, and casual acquaintances (see the questions in Table 1). Responses were categor￾ized as yes, no, or qualified (e.g., ‘‘it depends’’). We also asked about the frequency of sexual encounters neces￾sary to consider someone a friend with benefits. The questions regarding the characteristics of friends with benefits were not added until the first third of the data had been collected. The 109 participants who answered these questions did not differ from the other participants on any of the primary variables of the study or in terms of gender and ethnicity. Results Preliminary Analyses and Descriptive Information All variables were examined to determine if the assumptions of univariate and multivariate analyses were met (Behrens, 1997). Outliers were adjusted to fall 1.5 times the interquartile range below the 25th percen￾tile or above the 75th percentile. All the resulting variables had acceptable levels of skew and kurtosis. Occurrence of Sexual Behaviors with Four Types of Partners Table 2 presents the proportion of men and women engaging in each level of sexual behavior with each of the four types of partners. These proportions include both participants who engaged in additional forms of sexual activity, as well as those who engaged in no more than that level of sexual activity; thus, the light nongeni￾tal proportions includes those who engaged in only light nongenital activity and those who engaged in heavy nongenital or genital activity as well. (All participants who reported engaging in genital sexual activity with a particular type of partner had also reported engaging in light nongenital and heavy nongenital sexual activity with that type of partner; similarly, all participants who reported engaging in heavy nongenital sexual activity with a particular type of partner had also reported engaging in light nongenital sexual activity with that type of partner.) We conducted the equivalent of two-way (Participant Gender
Partner Type) repeated-measures multivariate analyses of variance (MANOVAs) for each level of sexual behavior using generalized hierarchical linear modeling (for a description of the equivalence, see Kenny, Bolger, & Kashy, 2002). Generalized hierarchi￾cal linear modeling takes into account the nested nature of the data, including the dependency inherent in the overlapping nature of the friends with benefits category. Unlike repeated-measures MANOVAs, generalized hierarchical linear modeling does not require ordinal data and can be used to analyze proportional data by using a logit link function. In addition, it permits Table 1. Proportion (and n) of Participants’ Answers to Questions about the Characteristics of Friends with Benefits Question Yes No Other 1. Are friends with benefits different from a romantic relationship? 97% (100) 3% (3) 0% (0) 3. Are they different from a casual sexual partner? 60% (66) 37% (41) 3% (3) 7. Are friends with benefits different from other people you periodically hook up with? 58% (64) 41% (45) 0% (0) 2. Other than sexual behavior, are they different from friends? 27% (29) 73% (79) 0% (0) 4. Do they have to be a friend? 72% (79) 26% (28) 2% (2) 5. Could it be someone you don’t know as well, such as a casual acquaintance or a stranger? 47% (51) 53% (58) 0% (0) 6. Would someone be a friend with benefits if you only engaged in sexual behavior with him or her once? 14% (15) 86% (94) 0% (0) Note. The number preceding the question refers to the order of the questions. The category of ‘‘other’’ answers refers to qualified ones that were not simple yes or no responses (e.g., ‘‘it depends’’). Numbers slightly vary across questions because of technical problems. SEXUAL PARTNERS 557

FURMAN AND SHAFFER Table 2.Proportions of Participants Engaging in Sexual To determine if there was a significant omnibus effect Behaviors with Different Types of Partners of the interaction between gender and type of partner, Friend we compared the full model with a two main effects Romantic Casual with model,which did not contain the terms that reflect an Variable Partner Friend Acquaintance Benefits interaction in the Level 2 equations:711 (Gender),721 (Gender),and 731(Gender).If the deviance of the full Light nongenital:Women 861 .51, .452 293 model was significantly smaller than the deviance of Light nongenital:Men .711 .432 .641 292 Heavy nongenital:Women 811 .202 242 272 the two main effects model (i.e.,the fit was better),it Heavy nongenital:Men .691 153 332 2623 would indicate a significant interaction between gender Genital:Women .771 182 212 242 and type of partner existed.If the deviance of the full Genital:Men .621 .142 .302 232 model was not significantly smaller than the two main Note.Different subscripts for different relationships in the same row effects models,it would indicate there was not a signifi- indicate that the proportions for that type of sexual behavior signifi- cant interaction between gender and type of partner. cantly differ between the two relationships. To determine if there was a significant effect of gender,we compared the deviance of the two main effects model with the deviance of a partner type only missing data;subsequent analyses examine the frequen- cies of sexual behavior with different types of partners. model,which only contained the partner effects terms. If the deviance of the two main effects model was signifi- If a participant did not have a particular type of sexual cantly smaller than the partner type only models,it partner,the participant's scores for that type of partner would indicate there was a significant gender effect. were treated as missing scores.Less than 15%of the To determine if there was a significant main effect of participants had engaged in sexual behavior with all four types of partners;thus,the analyses of the fre- partner,we compared the deviance of the two main quencies would not be possible if complete data were effects models with the deviance of a gender only model, which only contained the gender term.If the deviance of required. An example of a full model in generalized hierarchical the two main effects model was significantly smaller than the gender only model,it would indicate there linear modeling was as follows: was a significant partner effect. We found significant main effects of partner type for Level 1 Model: all levels of sexual behavior (all differences in deviances >104.14,ps<.001).The interaction between Logit(Y)=Boi BuiCr-a+B2iCr-b partner type and gender was significant for light nonge- +B3iCr-a+B2iCra-b+ nital behavior (difference in deviance=16.33,p<.001) and genital behavior (difference in deviance=8.89, Level 2 Model: p=.03),and approached significance for heavy non- genital behavior (difference in deviance=6.60,p=.09). Boi =700+701 (Gender)+uoi To understand the nature of the interactions,we con- Bui=710+711(Gender) ducted the hierarchical linear modeling equivalent of B2i=720+721 (Gender) tests of simple main effects in an analysis of variance. B3i=730+731(Gender) To determine the effect of partner type for each gender, we compared the deviance of the partner type only model for a gender with the deviance of a random intercept This model contained three orthogonal dummy vari- model for that gender,which did not include the terms able contrasts:Cr-a represents a contrast between reflecting a partner effect.The simple main effects of romantic partners and casual acquaintances;Cr-b repre- partner were significant for all three levels of sexual beha- sents a contrast between friends and friends with bene- vior for both genders (all differences in deviances> fits;finally,Cp reflects a contrast between romantic 48.90,ps<.001).We then examined the specific partners and casual acquaintances,on the one hand, dummy-variable contrasts of pairs of means.Consistent and friends and friends with benefits.on the other hand. with Hla,these analyses revealed that both men and The outcome Y is whether a type of sexual behavior women were almost always more likely to engage in each occurred or not. level of sexual behavior with romantic partners than with In traditional MANOVAs,the significance of main friends,casual acquaintances,or friends with benefits. effects and interactions are obtained as part of the stan- The one noteworthy exception is that men were as likely dard output.To determine if an interaction or main to engage in light nongenital sexual behavior with casual effect is significant in generalized hierarchical linear acquaintances as with romantic partners.Contrary to modeling,however,it is necessary to compare the fit H2a,men were also significantly more likely to engage (deviance)of pairs of models that contain or do not in light nongenital and heavy nongenital sexual behavior contain the terms of interest. with casual acquaintances than with friends.Consistent 558

missing data; subsequent analyses examine the frequen￾cies of sexual behavior with different types of partners. If a participant did not have a particular type of sexual partner, the participant’s scores for that type of partner were treated as missing scores. Less than 15% of the participants had engaged in sexual behavior with all four types of partners; thus, the analyses of the fre￾quencies would not be possible if complete data were required. An example of a full model in generalized hierarchical linear modeling was as follows: Level 1 Model: LogitðYÞ ¼ b0i þ b1iCra þ b2iCfb þ b3iCra þ b2iCrafb þ ej Level 2 Model: b0i ¼ c00 þ c01 ðGenderÞ þ u0i b1i ¼ c10 þ c11 ðGenderÞ b2i ¼ c20 þ c21 ðGenderÞ b3i ¼ c30 þ c31 ðGenderÞ This model contained three orthogonal dummy vari￾able contrasts: Cr–a represents a contrast between romantic partners and casual acquaintances; Cf–b repre￾sents a contrast between friends and friends with bene- fits; finally, Crp–fb reflects a contrast between romantic partners and casual acquaintances, on the one hand, and friends and friends with benefits, on the other hand. The outcome Y is whether a type of sexual behavior occurred or not. In traditional MANOVAs, the significance of main effects and interactions are obtained as part of the stan￾dard output. To determine if an interaction or main effect is significant in generalized hierarchical linear modeling, however, it is necessary to compare the fit (deviance) of pairs of models that contain or do not contain the terms of interest. To determine if there was a significant omnibus effect of the interaction between gender and type of partner, we compared the full model with a two main effects model, which did not contain the terms that reflect an interaction in the Level 2 equations: c11 (Gender), c21 (Gender), and c31 (Gender). If the deviance of the full model was significantly smaller than the deviance of the two main effects model (i.e., the fit was better), it would indicate a significant interaction between gender and type of partner existed. If the deviance of the full model was not significantly smaller than the two main effects models, it would indicate there was not a signifi- cant interaction between gender and type of partner. To determine if there was a significant effect of gender, we compared the deviance of the two main effects model with the deviance of a partner type only model, which only contained the partner effects terms. If the deviance of the two main effects model was signifi- cantly smaller than the partner type only models, it would indicate there was a significant gender effect. To determine if there was a significant main effect of partner, we compared the deviance of the two main effects models with the deviance of a gender only model, which only contained the gender term. If the deviance of the two main effects model was significantly smaller than the gender only model, it would indicate there was a significant partner effect. We found significant main effects of partner type for all levels of sexual behavior (all differences in deviances > 104.14, ps < .001). The interaction between partner type and gender was significant for light nonge￾nital behavior (difference in deviance ¼ 16.33, p < .001) and genital behavior (difference in deviance ¼ 8.89, p ¼ .03), and approached significance for heavy non￾genital behavior (difference in deviance ¼ 6.60, p ¼ .09). To understand the nature of the interactions, we con￾ducted the hierarchical linear modeling equivalent of tests of simple main effects in an analysis of variance. To determine the effect of partner type for each gender, we compared the deviance of the partner type only model for a gender with the deviance of a random intercept model for that gender, which did not include the terms reflecting a partner effect. The simple main effects of partner were significant for all three levels of sexual beha￾vior for both genders (all differences in deviances > 48.90, ps < .001). We then examined the specific dummy-variable contrasts of pairs of means. Consistent with H1a, these analyses revealed that both men and women were almost always more likely to engage in each level of sexual behavior with romantic partners than with friends, casual acquaintances, or friends with benefits. The one noteworthy exception is that men were as likely to engage in light nongenital sexual behavior with casual acquaintances as with romantic partners. Contrary to H2a, men were also significantly more likely to engage in light nongenital and heavy nongenital sexual behavior with casual acquaintances than with friends. Consistent Table 2. Proportions of Participants Engaging in Sexual Behaviors with Different Types of Partners Variable Romantic Partner Friend Casual Acquaintance Friend with Benefits Light nongenital: Women .861 .512 .452 .293 Light nongenital: Men .711 .432 .641 .292 Heavy nongenital: Women .811 .202 .242 .272 Heavy nongenital: Men .691 .153 .332 .2623 Genital: Women .771 .182 .212 .242 Genital: Men .621 .142 .302 .232 Note. Different subscripts for different relationships in the same row indicate that the proportions for that type of sexual behavior signifi- cantly differ between the two relationships. FURMAN AND SHAFFER 558