Journal of Bisexuality, 10:404–428, 2010 Copyright © Taylor & Francis Group, LLC ISSN: 1529-9716 print / 1529-9724 online DOI: 10.1080/15299716.2010.521056
Personal, Relational and Community Aspects of Bisexual Identity in Emerging, Early
and Middle Adult Cohorts
MELANIE E. BREWSTER and BONNIE MORADI University of Florida, Gainesville, Florida, USA
This study offered exploratory comparisons of personal, relational and community aspects of bisexual identity for women and men in emerging adulthood relative to those in early and middle adult cohorts. In this sample of 576 bisexual individuals, the pattern of findings for some aspects of bisexual identity (e.g., relationship com- mitment, bisexual community involvement) was consistent with the developmental perspective that characterizes emerging adulthood as a period of identity exploration and low commitment relative to increasing identity commitment across early and middle adult- hood. For other aspects of bisexual identity (e.g., self-described sex- ual orientation, outness and sexual behavior), gender appeared to play a role, either on its own or in conjunction with life stage cohort.
KEYWORDS bisexuality, bisexual youth, emerging adults, sexual orientation, sexual identity, romantic relationships, sexuality, life stage development, early adulthood, middle adulthood
Despite growing attention to identity issues in lesbian and gay populations, there remains a dearth of literature focused on bisexual identity (e.g., Dia- mond, 2008; Israel & Mohr, 2004; Phillips, Ingram, Smith, & Mindes, 2003). Specifically, the literature on lesbian and gay identity has delineated personal (e.g., self-identified orientation, outness), relational (e.g., romantic attractions and relationships) and community (e.g., connection with sexual minority
This research was supported in part by the APA Society for the Psychological Study of Lesbian, Gay and Bisexual Issues Bisexual Foundation Scholarship Award. This manuscript is based on data from the first author’s master’s thesis.
Address correspondence to Bonnie Moradi, Department of Psychology, University of Florida, P.O. Box 112250, Gainesville, FL 32611-2250, USA. E-mail: email@example.com
M. E. Brewster and B. Moradi 405
communities) aspects of sexual identity (e.g., Cass, 1979; Fassinger & Miller, 1996; McCarn & Fassinger, 1996; Morales, 1989; Troiden, 1989). For example, romantic attractions and relationships are thought to be important in sexual identity formation and, in turn, in coming out as a sexual minority person (McCarn & Fassinger, 1996). Similarly, awareness of and connection with sexual minority community groups have been conceptualized as important to identity development, as these communities can facilitate self-definition and self-acceptance and provide support and validation to sexual minority community members (Cass, 1979; Morales, 1989; Troiden, 1989).
Although personal, relational and community aspects of sexual identity have been discussed extensively for lesbian and gay people, attention is needed to these aspects of bisexual identity and how they may vary with developmental stages throughout life for bisexual individuals. Particularly, developmental conceptualizations suggest that emerging adulthood is a pe- riod of greater identity and relational exploration and lower commitment relative to early and middle adulthood (e.g., Arnett, 2000; Levinson, 1986; Nelson & Barry, 2005). The current study offers exploratory comparisons of bisexual individuals from emerging, early and middle adulthood cohorts on personal, relational and community aspects of bisexual identity.
To facilitate the present discussion of identity, we adopt the following conceptualization and terminology of sexual identity and its components (Moradi, Mohr, Worthington, & Fassinger, 2009; Worthington, Savoy, Dil- lon, & Vernaglia, 2002): Sexual identity is defined as the self-acknowledged collection of sexual behaviors, values, needs, preferences and so on that make up an individual’s sexuality. Sexual identity includes sexual orien- tation, which reflects sexual, affectional and relational orientations toward others as related to their gender or sex characteristics (e.g., same-, other-, pan-sexual orientation). Sexual orientation identity, in turn, reflects the in- ternal (e.g., self-identified orientation) and external (e.g., outness to others) claiming of sexual orientation at the individual or personal level and sense of group connection and membership at the collective or community level. Thus, in the present examination of bisexual identity across life stage co- horts, the relational aspects of sexual orientation are reflected in assessment of romantic attractions, behaviors and relationships; the personal and com- munity aspects of sexual orientation identity are reflected in assessment of self-identified sexual orientation, outness and community connection. These aspects of bisexual identity are compared across emerging, early and middle adult cohorts.
Emerging Adulthood and Distinctions from Early and Middle Adulthood
The ages of 18 to 25 are posited to represent a distinctly formative pe- riod of development termed emerging adulthood that is characterized by
406 Journal of Bisexuality
demographic flux (e.g., residential status, school attendance), role ambiva- lence (e.g., perceptions of self-sufficiency) and identity exploration (Arnett, 2000). Emerging adults are youth who have left their adolescent roles but are yet to commit to adult roles or to consider themselves adults (Arnett, 2000; Nelson & Barry, 2005). Emerging adulthood is marked by uncertainty and the sense that many different routes in life are available; individuals in this life stage may be inclined to experiment with social roles and identities in the course of establishing a life structure and considering various possible identity and role commitments (Erikson, 1968; Levinson et al., 1979).
Emerging adulthood is particularly important for identity formation in that this is “the period of life that offers the most opportunity for identity ex- ploration in the areas of love, work, and worldviews” (Arnett, 2000, p. 473). Of particular relevance to sexual identity is that emerging adulthood is a time of experimentation with dating and relationships, entering intimate and nonmarital relations, and beginning to treat romantic relationships as seri- ous quests for emotional and physical intimacy rather than as recreational activities (Arnett, 2000; Nelson & Barry, 2005). In emerging adulthood, in- dividuals begin to consider their own identities in the context of romantic relationships, asking: “Given the kind of person I am, what kind of person do I wish to have as a partner through life?” (Arnett, 2000, p. 473). Because romantic attraction and relationships are important aspects of sexual iden- tity, the task of self-definition through relationship definition may make the emerging adulthood stage a particularly salient time of intense exploration, interpersonal growth, and identity formation for bisexual people who expe- rience attraction to more than one gender. Within this context of personal exploration and experimentation, emerging adults are also likely to have more self- than other-oriented goals (Arnett, 2000; Nelson & Barry, 2005) that suggests potentially low commitment in community aspects of bisexual identity.
The identity exploration and experimentation that characterize emerging adulthood distinguish this stage from the increasing commitment and stabil- ity that characterize the subsequent developmental stages of early adulthood and middle adulthood (Berk, 2007; Levinson et al., 1979). Early adulthood is typically defined as an age range spanning from the mid-twenties to early forties, whereas middle adulthood encompasses ages from the early for- ties to mid-sixties. Early adulthood is marked by the peak years of the life cycle, in which individuals establish a niche in society and often form sta- ble relationships and occupations (Levinson, 1986). Beyond developing a more concrete sense of self, individuals in early adulthood may also incur stress related to relationship commitments, parenthood and financial obli- gations. Levinson (1986) described this period as a time when individuals make important choices regarding family, work and lifestyles without “the maturity or life experience to choose wisely” (p. 5). In contrast to the peak of early adulthood, middle adulthood is marked by a decrease in biological
M. E. Brewster and B. Moradi 407
capacities and increasing sense of personal responsibility for future genera- tions; individuals in this stage are thought to feel responsible for their own work and the work of others as they prepare to transition away from past social roles (i.e., parenthood, employee; Levinson, 1986). Thus, emerging adulthood is marked by tumultuous self-discovery, identity exploration and low identity commitment, whereas early and middle adulthood are concep- tualized as times of greater stability, personal identity and role commitment and collective responsibility and commitment.
Although a developmental perspective conceptualizes emerging adult- hood as a period of low identity commitment relative to early and middle adulthood, attention to shifts in larger social contexts suggests po- tentially countervailing forces. For example, although sexual minority ori- entations and identities, including bisexuality, have become more visible over time (e.g., Grossman, D’Augelli, & Hershberger, 2000; Keppel, 2006; Savin-Williams, 2008), older bisexual cohorts may have experienced a vir- tual invisibility of bisexuality in culture, media and social contexts during most of their life. Keppel (2006) suggested that as a consequence of such experiences of invisibility, older bisexual cohorts may lack supportive social networks (e.g., sexual minority or bisexual communities), perceive greater risks associated with ‘outness’ (e.g., loss of employment, family issues and health care rights for partners) and struggle with coming out. From this per- spective, it follows that bisexual persons in early and middle adulthood may be less out and less likely to be involved with bisexual or sexual minority communities than are emerging adults. These countervailing developmental and social context expectations—that emerging adults would have lower or higher commitment to personal, relational and community aspects of bisex- ual identity relative to older cohorts—remain largely speculative and warrant empirical exploration.
Gender and Bisexuality
Attention to gender is also important in exploring bisexual identity across life stage cohorts because some differences in the experiences of women and men may have implications for personal, relational and community as- pects of sexual identity. For example, based on a review of the literature, Brown (2002) surmised that women may be more comfortable with identify- ing as bisexual and more reflective and comfortable with ambiguity in their coming-out process whereas men may be less comfortable with identifying as bisexual and come out more abruptly and with less reflection. Such dif- ferences in self-identification and coming out may be shaped by a view of same-sex attractions as a threat to masculinity and related social status, by societal perceptions that men are either gay or straight (in contrast to views that women’s sexuality is more flexible) and by societal stigmatization of
408 Journal of Bisexuality
bisexual men as AIDS vectors (Brown, 2002; Burleson, 2005; Dodge, Reece, & Gebhard, 2008). Indeed, there is some evidence to suggest more overall negative attitudes toward bisexual men than women (Eliason, 2001).
Some gender differences in relational identity variables—including con- texts for meeting same-sex partners, exploration of same-sex attractions, and same-sex sexual behaviors—also have been observed. In this regard, Baumeister (2000) suggested that women have greater “erotic plasticity” (p. 348) such that women’s sexual responses and behaviors are more mal- leable and shaped by cultural, social and situational factors than are men’s responses; such erotic plasticity may have implications for relational aspects of identity for bisexual women and men. For instance, Burleson (2005) found that many bisexual women, but not men, in other-sex relationships were allowed or encouraged by romantic partners to explore their same-sex at- tractions. Furthermore, many bisexual men met male partners through cruis- ing, sex parties or the Internet, whereas most women met female partners through friends, work or polyamory community events. Bisexual men also had significantly more same-gender sex than did bisexual women. In addi- tion, findings with sexual minority youth suggest gender differences in the centrality of sexual behavior to same-sex attractions and self-labeling. For instance, in their sample of emerging adults, Savin-Williams and Diamond (2000) found that women were more likely to describe their first same-sex attractions and self-labeling as emotionally oriented whereas men were more likely to provide sexually oriented descriptions. In this sample, men were also more likely to engage in same-sex sexual behavior prior to self-labeling as sexual minority whereas women were more likely to self-label first and then engage in same-sex sexual behavior.
With regard to community identity and connection, bisexual men and women may experience strain in finding support within the lesbian and gay community, but for different reasons. For example, bisexual men in Burleson’s (2005) sample reported that their sexual orientation was not taken seriously by some gay men who perceived bisexual men as gay but in de- nial (as reflected in the saying “bi now, gay later”). Women may experience disconnection from some lesbian communities because of the view that rela- tionships with men reinforce patriarchy and subjugation of women (Brown, 2002; Burleson, 2005; Israel & Mohr, 2004). Taken together, this literature suggests that gendered contexts might shape differences in personal, rela- tional, and community aspects of bisexual identity.
Based on the literature reviewed here, the current study provides exploratory comparisons of bisexual individuals in emerging, early and middle adult cohorts on personal identity variables (i.e., self-identified orientation and
M. E. Brewster and B. Moradi 409
level of outness), relational identity variables (i.e., relationship commitment status, polyamory, same- and other-sex attractions and sexual behaviors) and community identity variables (i.e., involvement in sexual minority and bisexual communities). Moreover, given prior literature suggesting potential variations in women and men’s experiences of aspects of bisexual identity (e.g., Brown, 2002; Burleson, 2005), gender and its interaction with life stage cohort will also be explored. Due to the limited literature on personal, re- lational and community aspects of bisexual identity across life stages, no directional hypotheses are made. Overall, developmental conceptualizations of the life stages suggest that emerging adulthood would be a period of low commitment to personal, relational and community aspects of bisexual identity, with a pattern of increasing commitment across early and middle adulthood; the social context perspective suggests the countervailing pattern that emerging adults might be more committed to these aspects of bisexual identity given greater visibility of bisexuality and bisexual communities in their social contexts.
Participants were recruited via online resources such as electronic listserves, discussion boards and virtual communities for bisexual or sexual minority individuals. The study was advertised as an examination of the life experi- ences of bisexual individuals. Participants were directed to an online survey that began with an informed consent page. To participate in the study, re- spondents had to first affirm that they (1) identified as bisexual, (2) were age 18 years older, and (3) resided in the United States. If respondents affirmed that they met these inclusion criteria and agreed to participate after reading the informed consent, they were prompted to complete the survey.
The Internet has been a useful tool for collecting data from lesbian, gay and bisexual (LGB) samples (Moradi et al., 2009). Specifically, sexual minority listserves and online message boards may be good recruitment venues because LGB people tend to view the Internet as a safe place to connect with other sexual minority individuals (Riggle, Rostosky, & Reedy, 2005). Furthermore, even if such persons are not ‘out’ broadly, they may feel comfortable being ‘out’ online because the Internet provides a shield of anonymity; thus, the Internet is a viable resource for recruiting these under- represented individuals (Mustanski, 2001). Online surveys also are shown to yield similar responses as traditional pen-and-paper methods while be- ing more cost-efficient (Gosling, Vazire, Srivastava, & John, 2004; Hiskey & Troop, 2002). In recent years, numerous studies have utilized online methods as their primary means to recruit sexual minority participants (e.g., Carballo- Dieguez, Miner, Dolezal, Rosser, & Jacoby, 2006; Fernández et al., 2005; Wang & Ross, 2002). In the current study, to reduce the risk of nonbisexual
410 Journal of Bisexuality
people participating, the survey link was distributed only to groups or net- works that included bisexual individuals or bisexuality issues. Additionally, four validity questions asking participants to mark a particular response (e.g., Please mark “strongly agree”) were included within the survey to ensure that participants were responding attentively.
Data from 576 participants were analyzed in the current study. These data were drawn from a larger data set (N = 699) on the life experiences of bisexual individuals (Brewster & Moradi, 2010); participants were selected for inclusion if they had data for most of the variables of interest. Participants ranged in age from 18 to 65 (M = 32.13, SD = 11.24, Mdn = 29.00) and were divided into three groups based on prior definitions of life stage ages (Arnett, 2000; Levinson et al., 1979; Nelson & Barry, 2005): emerging adults ranging from 18 to 25 (n = 207; M = 21.39, SD = 2.47, Mdn = 21.00), young adults ranging from 26 to 40 (n = 240, M = 32.28, SD = 4.45, Mdn = 32.00), and middle adults ranging from age 41 to 65 years (n = 129, M = 49.07, SD = 6.70, Mdn = 47.00). In terms of gender, about 59% of participants identified as female, 40% as male and 1% as transgender (two male-to-female and seven female-to-male participants). Approximately 78% of the sample identified as White, 10% as multiracial, 5% as Latino/a American, 4% as Black/African American, 2% as Asian American, 1% as other race/ethnicities and 1% as Native American. In terms of level of education, roughly 35% of participants reported having a college degree, 33% having some college education, 22% having a professional degree, 9% having a high school degree, 1% having some high school education and less than 1% having no high school education. With regard to social class, roughly 51% of participants described themselves as middle class, 25% as working class, 19% as upper middle class, 4% as lower class and 1% as upper class.
Personal identity variables. Participants were asked to rate their sexual orientation with the following item, “Please select your sexual orientation using the BEST descriptor.” Possible orientations were (1) exclusively les- bian/gay, (2) mostly lesbian/gay, (3) bisexual, (4) mostly heterosexual, and (5) exclusively heterosexual. Respondents who described their orientation as mostly lesbian or gay (10% of the sample) or mostly heterosexual (14% of the sample) were included in the current study because scholars have noted that bisexuality is a spectrum and that bisexual individuals may not experience equal attraction to both genders (e.g., Rust, 2000). None of the participants identified as exclusively lesbian/gay or heterosexual.
M. E. Brewster and B. Moradi 411
The 10-item Outness Inventory (OI; Mohr & Fassinger, 2000) was used to assess the degree to which respondents’ sexual orientation is known or talked about within different social spheres of their life. The measure asks participants to rate on a 7-point continuum (1 = person definitely does not know about your sexual orientation status to 7 = person definitely knows about your sexual orientation status, and it is openly talked about) how open they are about their sexual orientation to members of their social network (e.g., friends, coworkers, family). For the current study, we added the item “new lesbian/gay friends” to parallel an item examining outness with “new straight friends” resulting in a total of 11 OI items. Item ratings are averaged to yield an overall score, with higher scores indicating greater levels of outness. Balsam and Mohr (2007) found a Cronbach’s alpha of .89 for OI items with their sample of bisexual individuals. In terms of validity, prior studies have found that OI scores correlate positively with involvement in lesbian and gay communities for LGB individuals (Balsam & Szymanski, 2005; Mohr & Fassinger, 2000). Moreover, validity evidence specific to bisexual individuals was reported by Balsam and Mohr (2007) who found that outness was correlated negatively with a measure of desire to keep sexual orientation private. Cronbach’s alpha for OI items with the current sample was .87.
Relational identity variables. Participants were asked to rate their phys- ical and emotional attraction to members of the same and other sex (i.e., “How much are you physically attracted to members of your own [the other] sex? How much are you emotionally attracted to members of your own [the other] sex?”). Attraction to the target groups was rated on a 5-point rating scale: 1 (low), 3 = (moderate), and 5 (high).
Participants were also asked about the gender specificity of their sexual experiences (i.e., “Have you had sex with persons of your own gender, the other gender, or both genders?”). Response options included 0 (never had sex), 1 (my own gender only), 2 (my own gender mostly), 3 (both genders equally), 4 (other gender mostly), and 5 (other gender only).
Last, participants were asked about their romantic relationships. Specifi- cally, they described their current relationship status as (1) single, (2) casually dating, (3) long-term dating, or (4) married/partnered. They were also asked if they identified as polyamorous (yes or no).
Community identity variables. Participants were asked to describe how connected or involved they were with (1) LGBTQ and (2) bisexual communities using a 5-point continuum from 1 (not at all) to 5 (extremely).
Descriptive information and correlations among the variables of interest are reported in Tables 1, 2 and 3, respectively. To examine the relations of life stage cohort, gender and their interaction with the variables of interest,
412 Journal of Bisexuality
TABLE 1 Descriptive Data for Variables of Interest for All Participants
Variables N (%) M SD
Personal identity variables Self-identified sexual orientation 3.04 .49
1 = Exclusively lesbian/gay — — 2 = Mostly lesbian/gay 57 10 3 = Bisexual 437 76 4 = Mostly heterosexual 79 14 5 = Heterosexual — —
Outness (1 = not out, 7 = very out) 3.19 1.39 Relational identity variables
Same-sex physical attraction (1 = low, 5 = high) 4.11 .95 Same-sex emotional attraction (1 = low, 5 = high) 3.86 1.14 Other-sex physical attraction (1 = low, 5 = high) 3.92 1.02 Other-sex emotional attraction (1 = low, 5 = high) 3.89 1.06 Sexual behavior
Never had sex 44 8 (1 = own gender only, 5 = other gender only) 3.41 1.00
Polyamorous 1 = Yes 222 39 2 = No 354 61
Relationship status 1 = Single 162 28 2 = Casually dating 62 11 3 = Long-term dating 123 21 4 = Married/partnered 229 40
Community identity variables Connected/involved with LGBT community
(1 = not at all, 5 = extremely) 2.54 1.14 Connected/involved with bisexual community
(1 = not at all, 5 = extremely) 1.88 1.00 LGBT = lesbian, gay, bisexual, transgender.
log-linear analyses and follow-up chi-square tests were conducted when the categories of the dependent variable were of interest (e.g., sexual orientation category). Log-linear analysis is an extension of chi-square test that is used to examine relationships among more than two categorical variables; similar to Analysis of Variance (ANOVA), it fits a linear model to the data to predict the number of cases expected in a given category (Field, 2000). In the present log-linear analyses, second-order effects of interest are the relationship of (1) the criterion variable with gender and (2) the criterion variable with life stage cohort; highest-order effects reflected an interaction of gender by life stage cohort in relation to the criterion variable. Multivariate Analysis of Variance (MANOVA), ANOVAs, and follow-up pairwise comparisons were conducted when level (from low to high) of the dependent variable was of interest (e.g., level of same-sex attraction). The nine individuals who identified as transgender were grouped according to the gender with which they identified (e.g., male-to-female transgender grouped with women).
ia b le
s o f In
st b y
e C o h o rt
A d u lth
o o d
ly A d u lth
o o d
A d u lth
o o d
ia b le
so n al
va ri ab
Se lf -i d en
al o ri en
= E xc
ia n /g
M o st
ly h et
er o se
er o se
O u tn
va ri ab
x p h ys
at tr ac
ig h )
at tr ac
ig h )
ex p h ys
at tr ac
ig h )
at tr ac
ig h )
al b eh
4 P o ly
am o ro
at u s
Si n gl
= C as
ly d at
Lo n g-
C o m
m u n ity
va ri ab
C o n n ec
te d /i
u n ity
a ll , 5
C o n n ec
te d /i
ex u al
m u n ity
a ll , 5
ia n , ga
y, b is
ex u al
, tr an
io n s
am o n g
ia b le
s o f In
ia b le
co h o rt
so n al
va ri ab
al o ri en
va ri ab
5. R el
at u s
6. P o ly
am o ro
p h ys
at tr ac
p h ys
at tr ac
em o tio
at tr ac
10 . O
em o tio
at tr ac
11 . Se
— C o m
m u n ity
va ri ab
12 . In
u n ity
13 . In
ex u al
m u n ity
n , ga
y, b is
ex u al
, tr an
. P ar
w h o
re p o rt ed
er h ad
cl u d ed
fr o m
ra n ge
6 d u e
a. P le
va ri ab
y. ∗ p
∗∗ ∗ p
M. E. Brewster and B. Moradi 415
Personal Identity Variables
A three-way log-linear analysis was conducted with Gender, Life Stage Cohort and Sexual Orientation (mostly lesbian/gay, bisexual, mostly heterosexual). The likelihood ratio of this model was χ2(0) = 0, p = 1, and the analysis indicated that a second-order effect was significant, χ2(8) = 84.45, p < .001, but the highest-order interaction of Life Stage Cohort × Gender × Sexual Orientation was not significant. To further decompose this finding, follow- up Pearson chi-square tests were conducted on the second-order effects of interest (i.e., Gender × Sexual Orientation, Life Stage Cohort × Sexual Orientation). These tests revealed that Gender had a small association with Sexual Orientation, χ2(2) = 14.47, p < .01, V = .16; a greater proportion of men than women identified their orientation as mostly lesbian/gay (men = 14%, women = 7%) and mostly heterosexual (men = 18%, women = 11%), whereas a greater proportion of women than men identified as bisexual (men = 68%, women = 82%). Life Stage Cohort was not related significantly to Sexual Orientation. Thus, Gender but not Life Stage Cohort was associated with self-identified Sexual Orientation, with a greater proportion of women than men describing their sexual orientation as bisexual.
An ANOVA was conducted in which Life Stage Cohort, Gender, and their interaction were examined with level of outness as the dependent variable. There was a small main effect of Gender, F(1, 570) = 22.53, p < .001, η2 = .04, which was qualified by a small Gender × Life Stage Cohort interaction, F(2, 570) = 8.66, p < .001, η2 = .03; however, there was no significant main effect for Life Stage Cohort. Pairwise comparisons to interpret the significant Gender × Life Stage Cohort interaction effect indicated that emerging adult women (M = 3.07, SD = 1.10) reported lower levels of outness than women in early adulthood (M = 3.52, SD = 1.44, t = −2.88, p < .01) and women in middle adulthood (M = 4.05, SD = 1.38, t = −4.13, p < .001); the latter two cohorts also differed significantly, t(570) = −2.20, p < .05. Men’s level of outness did not differ significantly across emerging adulthood (M = 3.13, SD = 1.32), early adulthood (M = 3.00, SD = 1.52) and middle adulthood (M = 2.70, SD = 1.43). Women also reported greater outness than men in early (women: M = 3.52, SD = 1.43; men: M = 3.00, SD = 1.52; t = 2.94, p < .01), and middle (women: M = 4.05, SD = 1.38; men: M = 2.70, SD = 1.43; t = 5.27, p < .001) adult cohorts, but women’s and men’s outness did not differ significantly in emerging adulthood (see Figure 1). Thus, emerging adult women in this sample reported lower outness than did women in early and middle adult cohorts, whereas men across life stage cohorts reported similar levels of outness.
416 Journal of Bisexuality
FIGURE 1 Outness interaction effect.
Relational Identity Variables
A three-way log-linear analysis was conducted with Gender, Life Stage Cohort, and Relationship Status (single, casually dating, long-term dating, married/partnered). The likelihood ratio of this model was χ2(0) = 0, p = 1, and the analysis indicated that a second-order effect was significant, χ2  = 202.81, p < .001, but the highest-order interaction of Life Stage Cohort × Gender × Relationship Status was not significant. To further decompose this finding, follow-up Pearson chi-square tests were conducted on the second- order effects of interest (i.e., Gender × Relationship Status, Life Stage Cohort × Relationship Status). These tests indicated that there was a medium associ- ation between Life Stage Cohort and Relationship Status, χ2(6) = 114.78, p < .001, V = .32, with a pattern of increasing relationship commitment across the life stage cohorts such that most emerging adults were single or casually dating whereas most early and middle adults were married/partnered (see Table 2). There was also a small association between Gender and Relation- ship Status, χ2(3) = 22.23, p < .001, V = .20. Specifically, among women, 28% were single, 12% were casually dating, 27% were in a long-term rela- tionship and 33% were married/partnered; among men, 29% were single, 9% were casually dating, 14% were in a long-term relationship and 49% were married/partnered. Thus, a higher proportion of women than men described
M. E. Brewster and B. Moradi 417
themselves to be in long-term relationships whereas a higher proportion of men than women described themselves as married/partnered, and there was a pattern of increasing relationship commitment across life stage cohorts.
IDENTIFICATION AS POLYAMOROUS
A three-way log-linear analysis was conducted with Gender, Life Stage Co- hort, and identification as Polyamorous (yes or no). The likelihood ratio of this model was χ2(0) = 0, p = 1, and the analysis indicated that a second-order effect was significant, χ2(5) = 102.37, p < .001. To further decompose this finding, follow-up Pearson chi-square tests were conducted on the second-order effects of interest (i.e., Gender × Polyamory, Life Stage Cohort × Polyamory). Follow-up Pearson chi-square tests indicated no sig- nificant association between Gender and Polyamory, but there was a medium association between Life Stage Cohort and Polyamory, χ2(2) = 35.17, p < .001, V = .25. As indicated in Table 2, there was a pattern of increasing iden- tification as polyamorous across emerging, early and middle adult cohorts.
PHYSICAL AND EMOTIONAL ATTRACTION
To examine same- and other-sex attractions by participant gender and life stage cohort, MANOVAs were conducted because same-sex physical and emotional attraction were correlated (r = .36, p < .001) and other-sex phys- ical and emotional attraction were correlated (r = .45, p < .001).
Same-sex attraction. In the first MANOVA, Life Stage Cohort, Gender, and their interaction were examined with same-sex physical and emotional attraction as the dependent variables. Results from the overall MANOVA in- dicated that the Gender, F(2, 569) = 28.66, p < .001, η2 = .09, and Gender × Life Stage Cohort interaction, F(4, 1140) = 3.25, p < .05, η2 = .01, effects were significant, but the main effect of Life Stage Cohort was not.1 Follow- up univariate analyses for same-sex physical attraction yielded a significant Gender main effect, F(1, 570) = 14.04, p < .001, η2 = .02, but no significant interaction effect; women reported higher levels of same-sex physical attrac- tion than men (women: M = 4.26, SD = .84; men: M = 3.89, SD = 1.06). For same-sex emotional attraction, the Gender main effect, F(1, 570) = 55.68, p < .001, η2 = .09, and the Gender × Life Stage Cohort interaction effects, F(2, 570) = 4.08, p < .05, η2 = .01, were significant. Pairwise comparisons were used to interpret the significant Gender × Life Stage Cohort interaction effect for same-sex emotional attraction indicated that among men, those in the emerging adult cohort (M = 3.69, SD = 1.28) reported higher same- sex emotional attraction than did those in the middle adult cohort, M = 3.24, SD = 1.18), t(570) = 2.37, p < .05; the early adult cohort (M = 3.46, SD = 1.21) did not differ from the two other cohorts. Among women, level
418 Journal of Bisexuality
of same-sex emotional attraction did not differ significantly across emerging adulthood (M = 4.11,SD = 1.01), early adulthood (M = 4.13, SD = .95) and middle adulthood (M = 4.41, SD = .84). Women also reported significantly greater same-sex attraction than men in emerging (women: M = 4.11, SD = 1.01; men: M = 3.69, SD = 1.28; t = 2.35, p < .05), early (women: M = 4.13, SD = .95; men: M = 3.46, SD = 1.21; t = 4.79, p < .001), and middle (women: M = 4.42, SD = .84; men: M = 3.24, SD = 1.18; t = 5.79, p < .001) adult cohorts. Thus, women generally reported greater same- sex physical and emotional attraction than men. Furthermore, the youngest cohort of men reported greater same-sex emotional attraction than did the oldest cohort of men, whereas women across life stage cohorts reported similar levels of same-sex emotional attraction (see Figure 2).
Other-sex attraction. A second MANOVA was conducted in which Life Stage Cohort, Gender, and their interaction were examined with other-sex physical and emotional attraction as dependent variables. Results from the overall MANOVA indicated that the Gender, F(2, 569) = 4.70, p < .01, η2 = .02, and Life Stage Cohort, F(4, 1138) = 2.80, p < .05, η2 = .01, main effects were significant, but their interaction was not. Follow-up univariate analyses for other-sex physical attraction indicated that the Life Stage Cohort main effect was not significant, but there was a significant Gender effect, F(1, 570) = 4.02, p < .05, η2 = .01; men reported greater other-sex phys- ical attraction than women (men: M = 4.02, SD = .98; women: M = 3.85,
FIGURE 2 Interaction effect for same-sex emotional attraction.
M. E. Brewster and B. Moradi 419
SD = 1.05). For other-sex emotional attraction, the Gender, F(1, 570) = 8.85, p < .01, η2 = .02, and Life Stage Cohort, F(2, 570) = 3.06, p < .05, η2 = .01, main effects both were significant. Men reported greater other-sex emotional attraction than women (men: M = 4.04, SD = .99; women: M = 3.79, SD = 1.09). Furthermore, pairwise comparisons indicated that relative to the early adult cohort (M = 3.74, SD = 1.12), level of other-sex emotional attraction was higher in the emerging adult cohort (M = 4.00, SD = .99, t = 2.66, p < .01) and in the middle adult cohort (M = 3.99, SD = 1.02, t = 2.20, p < .05). Thus, other-sex emotional attraction was higher in the emerging and middle adult cohorts than in the early adult cohort. Moreover, men generally reported greater other-sex physical and emotional attraction than women.
An ANOVA was conducted in which Life Stage Cohort, Gender, and their interaction were examined with sexual behavior as the dependent variable (where 1 = sex with my gender only to 5 = sex with other gender only; participants who reported that they had never had sex were excluded from these analyses). Gender had a small significant main effect, F(1, 525) = 9.93, p < .01, η2 = .02, with women (M = 3.52, SD = 1.01) reporting more other- gender sexual behavior than men (M = 3.24, SD = .96); but Life Stage Cohort and the Gender × Life Stage Cohort interaction effects were not significant.
Community Identity Variables
A MANOVA was conducted to examine involvement in sexual minority and bisexual communities (r = .53, p < .001) by Life Stage Cohort, Gender, and their interaction. Results from the overall MANOVA indicated that the Gender, F(2, 564) = 9.73, p < .001, η2 = .03, and Life Stage Cohort, F(4, 1130) = 4.94, p < .01, η2 = .02, main effects were significant, but their interaction was not.2
Follow-up univariate analyses for sexual minority community involve- ment indicated small but significant main effects for Gender, F(1, 565) = 19.03, p < .001, η2 = .03, and Life Stage Cohort, F(2, 565) = 3.20, p < .05, η2 = .01. Women (M = 2.68, SD = 1.16) were more involved with sexual minority communities than were men (M = 2.33, SD = 1.09). Furthermore, the overall Life Stage Cohort effect was significant, but the pairwise compar- isons did not yield specific significant differences among life stage cohorts. Follow-up univariate analyses yielded similar results for bisexual community involvement, such that small but significant main effects emerged for Gen- der, F(1, 565) = 8.09, p < .01, η2 = .01, and Life Stage Cohort, F(2, 565) = 10.02, p < .001, η2 = .03. Women (M = 1.93, SD = 1.07) were more involved
420 Journal of Bisexuality
with the bisexual community than men (M = 1.82, SD = .88). Furthermore, pairwise comparisons indicated that emerging adults (M = 1.71, SD = .86) were less involved in the bisexual community than early adults (M = 1.90, SD = 1.04, t = −2.12, p < .05) and middle adults (M = 2.13, SD = 1.06, t = −3.86, p < .001); the latter two cohorts also differed significantly from each other, t(565) = −2.13, p < .05. Thus, women reported more in- volvement in sexual minority and bisexual communities than men, and there was an overall pattern of increasing involvement in bisexual communities from the youngest to the oldest cohort.
The current study offers exploratory comparisons of personal, relational and community aspects of bisexual identity for women and men in emerging adulthood relative to those in early and middle adult cohorts. Overall, the pattern of findings for some aspects of bisexual identity was consistent with the general developmental perspective that characterizes emerging adult- hood as a period of identity exploration and low commitment and posits increasing identity commitment across early and middle adulthood. For other aspects of bisexual identity, gender appeared to play a role, either on its own or in conjunction with life stage cohort. Thus, the data offered a picture of developmentally consistent life stage cohort differences on some aspects of bisexual identity and gender variability on most aspects of bisexual identity.
Developmentally consistent life stage cohort differences emerged for relationship commitment and community connection, which we conceptu- alized as relational and community aspects of bisexual identity, respectively. Specifically, there was a pattern of increasing relationship commitment across life stage cohorts, with a greater proportion of the youngest cohort describ- ing themselves as single versus married/partnered and a greater proportion of the two older cohorts describing themselves as married/partnered versus single. This pattern may be associated with the parallel finding of increas- ing identification as polyamorous across the three cohorts. Specifically, a smaller proportion of emerging adults compared to the early and middle adult cohorts identified as polyamorous. This pattern may reflect the notion that polyamory implies relationship commitment (although to more than one person; e.g., Barker, 2005) which is less characteristic for emerging adults, the majority of whom were single or dating casually. Thus, the lower proportion of emerging adults in married/partnered relationships and in polyamorous relationships may be consistent with the developmental perspective that emerging adulthood is a period of relationship and relational identity explo- ration, with a pattern of increasing commitment in early and middle adult life stages (Arnett, 2000; Levinson, 1986; Levinson, Darrow, Klein, Levinson, & Mckee, 1976; Ornstein, Cron, & Slocum, 1989).
M. E. Brewster and B. Moradi 421
Also consistent with the developmental perspective were findings re- garding community identity. Specifically, there was a pattern of increasing involvement in sexual minority and bisexual communities from the youngest to the oldest cohorts. For sexual minority community connection, the spe- cific life stage cohort comparisons did not reach significance, although the overall pattern was significant. For bisexual community involvement, emerg- ing adults reported significantly lower involvement, followed by early and middle adult cohorts. Thus, the overall pattern of findings for community identity was consistent with the developmental perspective of movement from self-focused identity and goal orientation in emerging adulthood to more other orientation in later life stages (Levinson, 1986; Nelson & Barry, 2005; Ornstein et al., 1989; Super, 1980).
In addition to these developmentally consistent life stage cohort patterns for relationship commitment and community connection, gender played a role—alone, in conjunction, or in interaction with life stage cohort—in most of the identity variables examined in the current study (all but polyamory). Specifically, with regard to personal aspects of bisexual identity, the data suggested similarities across life stage cohorts in self-identified sexual orien- tation, but women were more likely than men to describe their orientation as bisexual. As theorized by many sexuality scholars, there may be soci- etal expectations for men to fall cleanly into a gay or straight orientation whereas women’s sexuality is afforded more flexibility (e.g., Baumeister, 2000; Burleson, 2005; Diamond, 2000). Some men may also be wary of de- scribing their orientation as bisexual due to negative societal associations of male bisexuality with the spread of AIDS (Eliason, 2001). Thus, women may be more comfortable than men in describing their orientation as bisex- ual. Indeed, longitudinal data point to the stability of self-identified bisexual orientation in women (Diamond, 2008).
For level of outness (conceptualized as external expression of personal identity), a gender by life stage chort interaction yielded a developmentally consistent pattern for women such that women in the emerging adult cohort reported the lowest level of outness, followed by women in early and middle adult cohorts. By contrast, for men, level of outness was similar across life stage cohorts. This pattern contrasts with speculation that bisexual adults in older life stage cohorts would be more closeted than younger bisexual adults due to changing social contexts and visibility of bisexuality over time (Kep- pel, 2006). One possible reason for emerging adult women being ‘less out’ than women in older cohorts is that bisexual women of older cohorts may experience less dismissal of their bisexual orientation as ‘just a phase’ if they have identified as such for many years. Furthermore, popular caricatures of young bisexual women as immature, attention seeking and hypersex- ual might be particularly salient for the emerging adult cohort (Ault, 1996). Indeed, young bisexual women are frequently objectified in the media as promiscuous and emotionally unstable (e.g., the reality show A Shot at Love
422 Journal of Bisexuality
with Tila Tequila, Sharon Stone in Basic Instinct). Emerging adult bisexual women may avoid coming out to deidentify with these negative societal stereotypes (Ault, 1996; O’Connor, 1997). Admittedly, popular depictions of bisexual women of older cohorts and bisexual men in general are nearly nonexistent, but the explicitly negative depictions of young bisexual women may seem less salient for these other groups.
There were also some gender differences in the relational aspects of identity examined in the current study. Women and men did not differ in identification as polyamorous, but a higher proportion of women than men described themselves as dating long term whereas a higher proportion of men than women described themselves as married/partnered. However, this finding should be interpreted with the caution that the distinction be- tween dating long term and married/partnered may have been ambiguous for some participants. For instance, some participants may have interpreted married/partnered to imply legal status, which is restricted to very few lo- cations in the United States. Thus, in future research, it might be useful to specify whether researchers are interested in legal or personally defined commitment status. We were interested in the latter in light of the restricted access to legal commitment for same-sex couples.
Some interesting gender variability also emerged in same-sex and other- sex attraction, and some of these effects were accompanied by Life Stage Co- hort main effects or interactions. Women reported greater same-sex physical and emotional attraction than men whereas men reported greater other-sex physical and emotional attraction than women. Thus, there was an overall pattern of women and men reporting more physical and emotional attraction toward women. This pattern may reflect actual differences in women and men’s attraction. Indeed, in a sample of undergraduate students, Lippa and Arad (1997) found that men’s reports of same- and other attraction were negatively correlated whereas women’s reports were uncorrelated. This uni- dimensionality versus bidimensionality in men and women’s same and other- sex attraction may reflect differences in private and public acknowledgement of these attractions which could be shaped by differential societal tolerance of same-sex attraction and intimacy between women and between men (Baumeister, 2000). Specifically, there might be some societal leniency to- ward women’s same-sex attraction and even fetishizing of sexuality between young and feminine women. By contrast, expression of same-sex attraction between men conflicts with traditional masculinity norms of restricted emo- tionality and heterosexual self-presentation (Burleson, 2005; Mahalik et al., 2003; Parent & Moradi, 2009). Moreover, a man’s declaration of intimate at- tractions (physical or emotional) to other men often sparks assumptions that he is gay, such that, bisexual identity is frequently either ignored or denied for men (Brown, 2002; Burleson, 2005).
In addition, in terms of sexual behavior, the sample’s average was near the midpoint of the response continuum, suggesting that participants had
M. E. Brewster and B. Moradi 423
had sex with both genders equally, but women’s responses were signifi- cantly closer to the other gender mostly anchor than were men’s responses. This finding is consistent with Burleson’s (2005) finding that bisexual men reported more same-gender sexual behavior than bisexual women. Such a pattern might be surprising given the present finding that women reported greater levels of same-sex physical and emotional attraction than men. How- ever, personal definitions of sex seem important to consider in reconciling these two patterns. Based on prior literature that sexual minority women’s sexuality includes a broad range of behaviors (e.g., Fassinger & Arseneau, 2008; Moradi et al., 2009; Rose, 2000), we did not specify a definition of sex and were interested in participants’ personally defined sexual behavior. However, sexuality research and popular conceptualization of sex tend to construe penile intercourse as its defining feature (e.g., see Moradi et al., 2009; Sanders & Reinisch, 1999). Thus, it is possible that women and men in the current study were more likely to view intercourse with men (rather than sexual behaviors with women) as sex. This possibility may underlie women’s higher reports of other-gender sexual behavior than men despite their higher reports of same-sex emotional and physical attraction than men.
In addition to these gender effects, levels of same- and other-sex physi- cal attraction were similar across life stage cohorts in the sample. But, some life stage cohort differences emerged in same- and other-sex emotional at- traction. Specifically, there was a pattern of dipping other-sex emotional attraction in the early adult cohort relative to the emerging and middle adult cohorts. Furthermore, among women, level of same-sex emotional attraction was similar across life stage cohorts; however, among men the youngest cohort reported greater same-sex emotional attraction than did the oldest cohort. These life stage cohort differences in same- and other-sex emotional attraction suggest the need for longitudinal research to explore potential un- derlying developmental processes. Furthermore, the distinctive patterns of life stage cohort findings for physical and emotional attraction are consistent with the need to conceptualize and assess these dimensions of attraction sep- arately in future research (Brewster & Moradi, 2010; Worthington & Navarro, 2003).
Limitations and Future Directions
The findings of the current study must be interpreted in light of a num- ber of limitations and directions for future research. We have noted some limitations and directions for future exploration throughout the discussion; however, some of these issues warrant further consideration. As noted previ- ously (e.g., Moradi et al., 2009; Mustanski, 2001; Riggle et al., 2005), Internet recruitment offers a number of advantages in research with sexual minority populations (e.g., access to large numbers of potential participants, reducing oversaturation of local venues, facilitating participation of individuals who
424 Journal of Bisexuality
are not comfortable “coming out” to researchers in-person), but Internet samples limit participation to individuals who have computer and Inter- net access. Thus, findings of Internet samples may not be generalizable to broader populations.
Relatedly, most of the current study’s participants were college educated and identified as middle class and White. Thus, the current findings must be interpreted with caution when considering their applicability to populations outside of these generalizability limits. Cultural variability in conceptualiza- tions of sexual orientation and bisexuality are important to consider, and research is needed to evaluate the replicability of the current findings with racially, ethnically and socioeconomically diverse populations. Such efforts can also inform future research and theory about bisexuality and its in- tersections with other sociodemographic identities. To this end, consistent assessment of sexual orientation, including bisexuality, in large-scale popu- lation studies can offer descriptive information about demographic diversity in the broader bisexual population.
Moreover, it is important to consider potential measurement issues in bisexuality research in general and in the current study. As noted previ- ously, personal definitions of relationship status (e.g., dating long term, married/partnered) and sexual behavior vary. When personally defined con- structs are of interest, assessments such as those used in the current study might be appropriate. But, in cases where predetermined conceptualizations of constructs are of interest (e.g., legal marriage or civil union, specific sexual behaviors), then more specific assessments are necessary to capture the di- mensions of interest. Similarly, bisexual individuals’ reasons for community involvement may vary and include social (e.g., dating) and political (e.g., civil right activism) motivations (Stearns & Sabini, 1997). This distinction be- tween social and political aspects of community involvement may be useful to explore in future research. Given that many of the variables of interest in the current study were basic personal descriptors (e.g., age, gender, rela- tionship status), we used single face valid items to assess these participant characteristics. However, the use of single items with limited prior psycho- metric evaluation is a limitation to consider in interpreting results involving some of the more complex variables, such as community involvement.
Another measurement issue is that many instruments designed for use in sexual minority research did not originate in bisexual populations’ ex- periences. For instance, the Outness Inventory is a frequently used mea- sure in sexual minority research and contains items that ask about the respondent’s “sexual orientation status.” This terminology is intended to be inclusive of bisexual individuals but may unintentionally collapse out- ness as sexual minority in general (i.e., not heterosexual) with outness as bisexual in particular (Balsam & Mohr, 2007). Thus, exploration of patterns of relations involving outness about one’s sexual minority status in general versus outness as bisexual in particular seems important. Although this may
M. E. Brewster and B. Moradi 425
be less of a concern in the current study given that the entire survey focused on bisexuality, this distinction may be important to consider in studies that sample the broader sexual minority population. We underscore the need for further conceptual and empirical advancements in measurement of key constructs with bisexual populations.
The current data also point to interesting directions for future research on bisexual women and men’s experiences of marginalization and inclusion in sexual minority communities. Overall, the data suggested a pattern that relative to men, women were more likely to describe their sexual orienta- tion as bisexual, were more out and were more involved in sexual minority and bisexual communities. Further research is needed to explore underly- ing reasons for these patterns. It is possible that these gender differences reflect greater social acceptance of bisexuality in women than in men (Elia- son, 2001). These differences may also reflect the distinctive manifestations of stigma to which bisexual women and men are exposed (e.g., bisexual men perceived as gay but in denial, stigmatized as AIDS/HIV vectors; bisex- ual women perceived as colluders with patriarchal subjugation of women; Brown, 2002; Burleson, 2005; Dodge et al., 2008). Although the current study provides some information about personal, relational and community aspects of bisexual identity in women and men across life stage cohorts, more re- search is clearly needed to elucidate bisexual women and men’s experiences and the social contexts that shape these experiences.
1. Pillai’s Criterion was used as a more conservative F-approximation because Box’s M was signif- icant (Tabachnick & Fidell, 2006).
2. Pillai’s Criterion was used as a more conservative F-approximation because Box’s M was signif- icant (Tabachnick & Fidell, 2006).
Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–490.
Ault, A. (1996). Ambiguous identity in an unambiguous sex/gender structure: The case of bisexual women. Sociological Quarterly, 37, 449–463.
Balsam, K. F., & Mohr, J. J. (2007). Adaptation to sexual orientation stigma: A com- parison of bisexual and lesbian/gay adults. Journal of Counseling Psychology, 54, 306–319.
Balsam, K., & Szymanski, D. (2005). Relationship quality and domestic violence in women’s same-sex relationships: The role of minority stress. Psychology of Women Quarterly, 29, 258–269.
Barker, M. (2005). This is my partner and this is my . . . partner’s partner: Construct- ing polyamorous identity in a monogamous world. Journal of Constructivist Psychology, 18, 78–88.
426 Journal of Bisexuality
Baumeister, R. F. (2000). Gender differences in erotic plasticity: The female sex drive as socially flexible and responsive. Psychological Bulletin, 126, 347–374.
Berk, L. E. (2007). Development through the lifespan (4th ed.). Boston: Pearson Division, Allyn & Bacon.
Brewster, M. E., & Moradi, B. (2010). Perceived experiences of anti-bisexual preju- dice: Instrument development and evaluation. Journal of Counseling Psychology, 57(4), 451–468.
Brown, T. (2002). A proposed model of bisexual identity development that elaborates on experiential differences of women and men. Journal of Bisexuality, 1, 69–91.
Burleson, W. E. (2005). Bi America: Myths, truths, and struggles of an invisible community. New York: Harrington Park Press.
Carballo-Dieguez, A., Miner, M., Dolezal, C., Rosser, S., & Jacoby, S. (2006). Sexual negotiation, HIV-status disclosure, and sexual risk behavior among Latino men who use the internet to seek sex with other men. Archives of Sexual Behavior, 35, 473–481.
Cass, V. (1979). Homosexual identity formation: A theoretical model. Journal of Homosexuality, 4, 219–235.
Diamond, L. M. (2000). Sexual identity, attractions, and behavior among young sexual-minority women over a 2-year period. Developmental Psychology, 36, 241–250.
Diamond, L. M. (2008). Female bisexuality from adolescence to adulthood: Results from a 10-year longitudinal study. Developmental Psychology, 44, 5–14.
Dodge, B., Reece, M., & Gebhard, P. H. (2008). Kinsey and beyond: Past, present, and future considerations for research on male bisexuality. Journal of Bisexuality, 8, 177–191.
Eliason, M. J. (2001). Bi-negativity: The stigma facing bisexual men. Journal of Bi- sexuality, 1, 137–154.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton. Fassinger, R. E., & Arseneau, J. R. (2008). Diverse women’s sexualities. In F. L.
Denmark & M. Paludi (Eds.), Psychology of women: A handbook of issues and theories (2nd ed., pp. 484–508). Westport, CT: Praeger.
Fassinger, R. E., & Miller, B. A. (1996). Validation of an inclusive model of sexual minority identity formation on a sample of gay men. Journal of Homosexuality, 32, 53–78.
Fernández, M. I., Perrino, T., Collazo, J. B., Varga, L. M., Marsh, D., Hernandez, N., et al. (2005). Surfing new territory: Club-drug use and risky sex among Hispanic men who have sex with men recruited on the internet. Journal of Urban Health, 82, 179–188.
Field, A. (2000). Discovering statistics using SPSS. London: Sage. Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust
web-based studies? A comparative analysis of six preconceptions about internet questionnaires. American Psychologist, 59, 93–104.
Grossman, A. H., D’Augelli, R. D., & Hershberger, S. L. (2000). Social support net- works of lesbian, gay, and bisexual adults 60 years of age and older. Journal of Gerontology, 55, 171–179.
Hiskey, S., & Troop, N. A. (2002). Online longitudinal survey research: Viability and participation. Social Science Computer Review, 20, 250–259.
M. E. Brewster and B. Moradi 427
Israel, T., & Mohr, J. J. (2004). Attitudes toward bisexual women and men: Current research, future directions. Journal of Bisexuality, 4, 117–134.
Keppel, B. (2006). Affirmative psychotherapy with older bisexual women and men. Journal of Bisexuality, 6, 85–104.
Levinson, D. J. (1986). A conception of adult development. American Psychologist, 41, 3–13.
Levinson, D. J., Darrow, C. M., Klein, E. B., Levinson, M. H., & McKee, B. (1976). Pe- riods in the adult development of men: Ages 18 to 45. Counseling Psychologist, 6, 21–25.
Levinson, D. J., Darrow, C. M., Klein, E. B., Levinson, M. H., & McKee, B. (1979). The seasons of a man’s life. New York: Knopf.
Lippa, R., & Arad, S. (1997). The structure of sexual orientation and its relation to masculinity, femininity, and gender diagnosticity: Different for men and women. Sex Roles, 37, 187–207.
Mahalik, J. R., Locke, B. D., Ludlow, L. H., Diemer, M. A., Scott, P. J. S., Gottfried, M., et al. (2003). Development of the conformity to masculine norms inventory. Men and Masculinity, 4, 3–25.
McCarn, S. R., & Fassinger, R. E. (1996). Revisioning sexual minority identity forma- tion: A new model of lesbian identity and its implications for counseling and research. Counseling Psychologist, 24, 508–534.
Mohr, J. J., & Fassinger, R. (2000). Measuring dimensions of lesbian and gay male experience. Measurement and Evaluation in Counseling and Development, 33, 66–90.
Moradi, B., Mohr, J. J., Worthington, R. L., & Fassinger, R. E. (2009). Counseling psychology research on sexual (orientation) minority issues: Conceptual and methodological challenges and opportunities. Journal of Counseling Psychology, 56, 5–22.
Morales, E. S. (1989). Ethnic minority families and minority gays and lesbians. Mar- riage and Family Review, 14, 217–239.
Mustanski, B. S. (2001). Getting wired: Exploiting the Internet for the collection of valid sexuality data. Journal of Sex Research, 38, 292–301.
Nelson, L. J., & Barry, C. M. (2005). Distinguishing features of emerging adulthood: The role of self-classification as an adult. Journal of Adolescent Research, 20, 242–262.
O’Connor, P. (1997). Warning! Heterosexual contents under pressure. Hypatia, 12, 183–188.
Ornstein, S., Cron, W. L., & Slocum, J. W. (1989). Life stage versus career stage: A comparative test of the theories of Levinson and Super. Journal of Organiza- tional Behavior, 10, 117–133.
Parent, M., & Moradi, B. (2009). Confirmatory factor analysis of the Conformity to Masculine Norms Inventory and development of the CMNI-46. Psychology of Men and Masculinity, 10, 175–189.
Phillips, J. C., Ingram, K. M., Smith, N. G., & Mindes, E. J. (2003). Methodological and content review of lesbian-, gay-, and bisexual-related articles in counseling journals: 1990–1999. Counseling Psychologist, 31, 25–62.
Riggle, E. D. B., Rostosky, S. S., & Reedy, C. S. (2005). Online surveys for BGLT research issues and techniques. Journal of Homosexuality, 49, 1–21.
428 Journal of Bisexuality
Rose, S. (2000). Heterosexism and the study of women’s romantic and friend rela- tionships. Journal of Social Issues, 56, 315–328.
Rust, P. C. (2000). Bisexuality: A contemporary paradox for women. Journal of Social Issues, 56, 205–221.
Sanders, S. A., & Reinisch, J. M. (1999). Would you say you “had sex” if. . . . Journal of the American Medical Association, 281, 275–277.
Savin-Williams, R. C. (2008). Then and now: Recruitment, definition, diversity, and positive attributes of same-sex populations. Developmental Psychology, 44, 135–138.
Savin-Williams, R. C., & Diamond, L. M. (2000). Sexual identity trajectories among sexual-minority youth: Gender comparisons. Archives of Sexual Behavior, 29, 607–627.
Stearns, D. C., & Sabini, J. (1997). Dyadic adjustment and community involvement in same-sex couples. Journal of Gay, Lesbian, and Bisexual Identity, 2, 265–283.
Super, D. E. (1980). A life-span, life-space approach to career development. Journal of Vocational Behavior, 16, 282–298.
Tabachnick, B. G., & Fidell, L. S. (2006). Using multivariate statistics (5th ed.). New York: HarperCollins.
Troiden, R. R. (1989). The formation of homosexual identities. Journal of Homosex- uality, 17, 43–73.
Wang, Q., & Ross, M. W. (2002). Differences between chat room and e-mail sampling approaches in Chinese men who have sex with men. AIDS Education and Prevention, 14, 361–366.
Worthington, R. L., & Navarro, R. L. (2003). Pathways to the future: Analyzing the contents of a content analysis. Counseling Psychologist, 31, 85–92.
Worthington, R. L., Savoy, H. B., Dillon, F. R., & Vernaglia, E. R. (2002). Heterosex- ual identity development: A multidimensional model of individual and group identity. Counseling Psychologist, 30, 496–531.
Copyright of Journal of Bisexuality is the property of Taylor & Francis Ltd and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission.
However, users may print, download, or email articles for individual use.