Exploring the Q in LGBTQ: Demographic characteristic and sexuality of Queer people in a U.S. representative sample of sexual minorities

Abstract

Although queer identity has been used among sexual minorities for decades, little is known about the population of queer-identified people in the U.S. We compared people who identify as queer (unweighted n = 88; weighted % = 5.8%) with those who identify as lesbian/gay (n = 833; 46.9%), bisexual (n = 493; 40.6%) or other sexual minority identities (n = 93; 6.7%), in order to describe queer-identified people as a distinct sexual minority group. The study is the first to estimate demographic characteristics and sexuality of queer-identified people using a U.S. nationally representative sample. We found that queer people are overwhelmingly cisgender women and genderqueer/ nonbinary (GQNB), younger, and more highly educated than other groups. After stratifying by gender identity (cisgender women; cisgender men; GQNB), survey-weighted descriptive differences in attraction, sexual partnering, and relationship patterns show that queer individuals are more likely to report attraction to, and sexual relationships with, transgender and GQNB people, though differences by respondent gender identity were noted: The majority of queer women are attracted to and partnered with both women and men, and were more likely than other groups to be attracted to and partnered with cisgender and transgender people. In contrast, queer men are split in their attractions—about half were attracted exclusively to men, and half to men and women—but the majority partnered with men only. Of all groups, queer men are the most likely to partner with transgender men, but none had partnered with transgender women. GQNB people are more likely than cisgender people to identify as queer (25.9%) and are attracted to both cisgender and transgender women and men, yet predominantly partnered with cisgender people. The results provide support for queer as a distinct sexual identity.

Translational Abstract/Public Significance Statement:

We explored the demographics and sexuality (e.g. sexual partnering and attraction) of people who identify as queer, compared to those who identify as lesbian/gay, bisexual, or other sexual minority identities, to better understand if, and how, queer-identified people are distinct from other sexual minority groups. We found that queer people are overwhelmingly cisgender women and genderqueer/ nonbinary (GQNB), younger, and more highly educated than other groups. Sexuality wise, queer individuals are more likely to report attraction to, and sexual relationships with, transgender and GQNB people, though differences emerge by gender identity.

The modern gay liberation movement has provided an increasingly public social space for individuals to recognize and name their same-sex attractions, relationships, and identities. Following, the categories and labels that reflect distinctive identities have evolved: in addition to gay, lesbian, and bisexual, other non-heterosexual identity labels are increasingly being used, including asexual, pansexual, and, the identity group that is the focus of this article, queer. However, these identities remain largely understudied, and thus little is known about the characteristics and attractions, relationships, and sexual behaviors of people who are queer, how these differ among people who hold other sexual minority identities, and how overall patterns differ by gender identity. In this paper we aim to better understand queer identity, using a U.S. representative sample of sexual minorities.

The meaning of queer identity

People make meaning of their sexual orientation and identity such that any two people of the same sexual identity may differ substantially in whom they are attracted to, partner with sexually, and engage with in romantic relationships (Vrangalova & Savin-Williams, 2012).Yet typically, lesbian, gay, bisexual, and heterosexual identities have been defined by gendered attraction and partnering patterns. That is, heterosexual men and women are assumed to be exclusively attracted to, respectively, women and men; lesbians are attracted to women; gay men are attracted to men; and bisexual men and women are attracted to both men and women. Queer individuals, however, may or may not fall neatly within gendered patterns of attraction and partnering (Callis, 2014; Galupo, Lomash, & Mitchell, 2017).

The roots of queer as a personal identity have changed over time. In the late 19^th and early 20^th centuries, queer was a pejorative term used against persons with same-sex desires or relationships. By the late 20^th century, queer was reclaimed by both intellectuals and activists. In academic realms, queer was a term that, for some, offered a critique of the mainstream lesbian, gay, and bisexual (LGB) community (Howard, 2018). For example, the field of queer theory grew out of the academic realms of feminist and LGBT studies; specifically, queer theory saw sex and gender as non-essentialist, fluid, and non-binary (Sullivan, 2003). In the context of these community and academic discourses, queer emerged as a social and personal identity for some, defined by rejection of binary categories of gender and sexuality, and inclusion of any sexual or gender identity that is non-normative, disrupting categories such as man and woman, and gay and straight.

For others, queer became used in activist circles to describe a collective group, deliberately selecting the term in order to confront the prejudice implied in the pejorative use of the term. For them, queer has holds a political, rather than academic, meaning (Gray & Desmarais, 2014). With the rise of the AIDS epidemic in the 1980s and 1990s came a demand for visibility and respect, wherein the activist group Queer Nation popularized the chant “We’re here, we’re queer, get used to it” (“Queer Nation NY History,” n.d.). In this context, rather than a rejection of heteronormative binaries, queer became a rejection of heteronormative politics, and its use reflected a desire to reclaim a label, in order to make a political statement and demand change (Miller, Taylor, & Rupp, 2016).

Who is Queer?

In addition to its collective meaning in academic and activist circles, queer is often used as a catch-all umbrella term to include the group of all non-heterosexual and non-cisgender sexual and gender minorities (American Psychological Association, 2015; Zane, 2015). Yet, queer is also increasingly being recognized as an individual sexual identity within the broadening spectrum of non-traditional sexual and gender minority identities, that now includes asexual, pansexual and others, in addition to traditional identities (Russell, Clarke, & Clary, 2009; Smalley, Warren, & Barefoot, 2016; Watson, Wheldon, & Puhl, 2019). For those people who identify as queer in order to describe their individual sexual identity, rather than indicate affiliation with the ‘queer umbrella’ of sexual and gender minorities, queer reflects an identity label adopted in contrast to traditional identities such as lesbian, gay, or bisexual (Callis, 2014).

Qualitative researchers have conceptualized queer identity, among other identities, as plurisexual in contrast to monosexual (Galupo, Ramirez, & Pulice-Farrow, 2017), described as identities that make salient sexual and gender fluidity. However, to date, no study using a U.S. population sample has been conducted to assess demographics and sexuality of queer-identified people, limiting knowledge about who is queer in terms of demographic characteristics, attraction, and partnering.

Knowledge on the demographics of queer-identified adults has come mainly from community-based (non-representative) samples. These studies have shown that women and gender minorities are more likely to identify as queer (Bosse & Chiodo, 2016; Galupo, Mitchell, & Davis, 2018; Galupo, Ramirez, et al., 2017; Katz-Wise, Reisner, White, & Keo-Meier, 2016; Smalley et al., 2016). At the same time, a majority of research into queer identity has focused on women and gender-minority samples to begin with (Baldwin et al., 2017; Better, 2014; Katz-Wise et al., 2016; Kuper, Nussbaum, & Mustanski, 2012; Mereish, Katz-Wise, & Woulfe, 2017), limiting understanding of queer (cisgender) men specifically. One Australian study has explored how dimensions of sexuality differ between queer-identified women and men, finding that queer women were more likely than queer men to have had a non-cisgender sexual partner and that queer men were over twice as likely as queer women to report exclusively same-sex attraction or exclusively same-sex partnering (attraction to non-cisgender people was not reported; Morandini, Blaszczynski, & Dar-Nimrod, 2017).

There is some evidence that identifying as queer may be associated with being of a younger birth cohort. This suggests that younger sexual minorities are more likely to view sexual orientation and gender identity as fluid and see queer as an preferred term, in contrast to older sexual minorities who may still continue to see queer as a slur, as it has been used historically (Laughlin, 2016; Miller et al., 2016; Russell et al., 2009). There is also evidence, that younger sexual minorities are increasingly adopting more plurisexual identity labels. An exploration of the LGBTQ National Teen Survey, a population-based sample of over 17,000 LGBTQ youth aged 13–17 in 2017, found that over 37% of the sample identified as something other than gay/lesbian or bisexual, including 4.1% who identified as queer and 13.2% who identified as pansexual (Watson et al., 2019). This is similar to an estimate from a 2003–2005 California-based sample of high school youth who participated in Gay-Straight Alliances (GSA), which found that 5.2% of the non-heterosexual sample identified as queer, with 8.5% identifying as something other than queer, straight, lesbian, gay, bisexual, or questioning (Russell et al., 2009). In contrast, other studies have found that queer respondents are older than other sexual minorities in their samples (Galupo, Ramirez, et al., 2017; Morandini et al., 2017). The present study comes to fill a gap in knowledge needed to understand cohort differences in queer identification.

A limited body of research suggests that queer people have unique patterns of sexuality characteristics, such as attraction, sexual partnering, and romantic relationships, though evidence is mixed. One consistent theme is that identifying as queer allows for the “flexibility [to be]...attracted to people of many different gender identities and presentations” (Galupo, Ramirez, et al., 2017, p. 117), addressing “the problem with fitting attraction to transpeople into labels like bisexual and homosexual” (Callis, 2014, p. 72). Galupo and colleagues (2017) conducted a qualitative study of bisexual, queer, and pansexual identified adults in the United States, and found that queer respondents used more inclusive language to describe their sexual attractions, and were less likely to make binary distinctions (e.g. “I’m interested in people”). Another study of bisexual and queer-identified adult women in the United States by Mereish and colleagues (2017) found that queer women were more likely than bisexual women to report attraction to non-cisgender people and sex with transgender and genderqueer or non-binary partners. In contrast, Morandini and colleagues (2017) in found that queer Australian adults, particularly queer women, reported patterns of sexual and romantic attraction and sexual partnering that are more similar to lesbian and gay individuals than bisexual or pansexual individuals. However, no study to date has extensively explored differences across identity in both cisgender and non-cisgender attractions and partnering, within a population-based sample.

We address these gaps in knowledge, using data from a nationally-representative sample of sexual minorities in the U.S., to describe demographic characteristics, and patterns of attraction and partnering which assess partners of multiple gender identities, in queer adults as compared with other sexual minorities, and how such patterns differ across gender identity.

Method

Respondents

Respondents were recruited in 2016 – 2017 for first wave of the Generations Study, a longitudinal study that compares three cohorts of sexual minority people that are distinct in the historical context and events that happened over their lifetime. The study investigators conceptualized three distinct cohorts of LGB people that correspond to significant social changes in the U.S. regarding LGB law, policy, and culture based on major events that characterize the social environment of LGB people since 1969 (available on the study website at www.generations-study.com). Cohorts were defined our cohorts based on significant events people who would have been 10 years old—considered a significant age for sexual development (Herdt & McClintock, 2000)—plus/minus 3 years, to approximate the ages when the social environment would make an impact on one’s socialization. Events examples include the Stonewall inn riots (1969), the formation of ACT UP (1987), and the Massachusetts Supreme Court ruling that it was unconstitutional to deny marriage to same sex couples (2003).

Three distinct cohorts were defined: The youngest, the Equality generation, were 18–25 years-old when recruited. They are respondents whose early life experiences were impacted by a national discourse about LGB (and, to some extent, T) equality, such as marriage equality, employment discrimination, and other forms of institutionalized LGB acceptance. The second cohort, the Visibility generation, were 34–41 years-old when recruited. They are respondents whose early life experiences were impacted by a period after the beginning of the AIDS epidemic, when LGBT institutions were strengthened, and LGBT people gained greater visibility than ever before. The oldest cohort, the Pride generation, were 52–59 years-old when recruited. The respondents’ early life experiences were immediately after the 1969 Stonewall riots and the start of the modern gay liberation movement, and thus were impacted by the emergence of a gay identity, discourse about gay pride, and coming out.

Generations Study respondents were recruited by the survey research company Gallup, using the Gallup Daily Tracking Survey as initial contact. Respondents were recruited via random-digit dialing of both landlines and cellphone users, first in March 2016 – March 2017, with an additional enhancement sample recruited in March 2017 – March 2018 with an oversample of Black and Latino respondents (for more details about the methods see Krueger, Lin, Kittle, & Meyer, 2015). The investigators used a two-phase recruitment procedure.

In the first phase, via a telephone interview of a representative sample of U.S. population of adults over age 18, LGBT individuals were identified utilizing a question asked of all Gallup respondents, “do you personally identify as lesbian, gay, bisexual or transgender?” Respondents who identified as LGBT were further screened to determine sexual and gender identity, and other eligibility criteria (e.g., age, race/ethnicity, and educational level). In this phase, sexual identity was assessed via the question “do you consider yourself to be...” with the following options provided: “straight or heterosexual; lesbian; gay; bisexual; queer; same-gender loving;, don’t know; and refuse.”¹ Respondents who selected an identity other than “straight or heterosexual”; were in one of the targeted birth cohorts; identified as Black, Latino, or White or had a mixed race/ethnicity that included one of these; had a sixth grade education or higher; and were able to speak English were eligible for inclusion. Respondents who had lower educational levels were not eligible due to the requirement to self-administer the survey questionnaire. Respondents who identified as other race/ethnic groups were not eligible because the low base rate of people in these groups (e.g., Asian, Native American) did not allow for sufficient numbers of respondents over the study recruitment period for meaningful statistical analyses of their responses.

In the second phase, eligible respondents were asked to participate in a self-administered online or mailed survey. In Phase 1, 366,640 respondents were screened in the brief telephone interview. Of them, 3.5% (n = 12,837) identified as LGBT and 3,525 were eligible to participate in the Generations study based on that study’s age group, race and ethnicity, and educational restrictions. Of those eligible, 82% (n = 2,882) agreed to participate in the Generations study (76% of them were sent the web version and 24% the mailed questionnaire). In that study, 49% of web surveys and 46% of mailed surveys were completed. The final cooperation rate was 39%

The baseline sample included 1,518 respondents (1,331 from original sample, 187 from the oversample of Black and Latino respondents). For the purpose of the current analysis, all baseline Generations respondents were included, with the exception of an additional 11 respondents who were removed due to identifying as “straight/heterosexual” on the baseline survey (not the screener), resulting in a sample for the present study of 1,507. Of these, 61.8% (n = 664) were from the Equality generation, 20.7% (n = 369) from the Visibility generation, and 17.5% (n = 474) from the Pride generation.

The study had IRB approval from multiple institutions whose researchers participated.

Measures

Sexual identity.

In addition to reporting their sexual identity in the Phase 1 screen, respondents reported their identity in the survey (Phase 2), there respondents indicated their “current sexual identity” as “straight/heterosexual, lesbian, gay, bisexual, queer, same-gender loving,” or “other” (write-in). Sexual identity was categorized into four groups: queer (5.8%; n = 88); bisexual (40.6%; n = 493), lesbian/gay (46.9%; n = 433), and other (6.7; n = 94) including same-gender loving, queer, asexual, pansexual, and anti-label.

Sex Assigned at Birth, Gender, and Gender Identity.

A four-category gender identity variable was created that combined information from self-reported “sex assigned at birth, on your original birth certificate” (sex assigned at birth [SAAB]; female vs. male) and “current gender” (woman, man, genderqueer/nonbinary [GQNB]). Gender identity (cisgender women; cisgender men; genderqueer/non-binary), was constructed based on self-reported SAAB and gender. Respondents assigned female at birth who identified their current gender as woman, were categorized as cisgender women; respondents assigned male at birth who identified as male for current gender identity were categorized as cisgender men. Respondents who reported a current genderqueer/nonbinary gender identity were first categorized based on their SAAB (GQNB, female SAAB; GQNB, male SAAB), then collapsed into a single GQNB group.

Demographics.

Respondents were categorized based on their birth cohort (Equality, age 18–25; Visibility, age 34–41; Pride, age 52–59); race/ethnicity (White /Black-African American/Latino-Hispanic); education (≤ High school/ Some college / Bachelor’s degree/ Post-graduate); living in poverty (yes/no, based on ratio of household income to Census-issued poverty thresholds; respondents with household incomes below the federal poverty limit were considered to be living in poverty); Census region (Northeast / Midwest/ South /West); and urbanity (Urban / non-Urban, based on zip code, following the 2010 US Department of Agriculture Rural-Urban Commuting Area (RUCA) coding system (2010 Rural-Urban Commuting Area Codes, 2016). Zip codes with RUCA urbanity score ≤ 3 are categorized as urban, those with scores 4–10 are categorized as non-urban.

Sexual attraction.

Two aspects of attraction were of interest to the present study. Firstly, given that sexual identities are often assessed (in survey research), and self-defined (among individuals) based on whether attractions are to the same, different, or multiple genders, relative to one’s own gender (Galupo et al., 2018; SMART (Sexual Minority Assessment Research Team), 2009), one variable (Gendered attraction patterns, and the parallel Gendered partnering patterns, described below) was included to assess this pattern among queer identified people, compared with other identities, and if this differed by respondent gender identity. In addition, separate variables for those attracted to women (and men), were constructed to assess differences in the gender identity (e.g. cisgender vs. transgender) of these attractions (and partners), following previous evidence that queer people may be more likely to be attracted to and/or partner with non-cisgender people (Callis, 2014; Galupo, Ramirez, et al., 2017; Mereish et al., 2017; Morandini et al., 2017)

Thus, the following four variables were constructed, based on responses to the prompt “how sexually attracted are you to the following types of people: Women, non-transgender; Men, non-transgender; Transgender Women/Male-to-Female (MTF); Transgender Men/Female-to- Male (FTM).” Responses were reported separately for each type of person using a 5-point Likert scale (“not at all, not very, somewhat, very, not sure”). Respondents who stated they were “somewhat: or “very” sexually attracted to a category of people were considered to be attracted to them, regardless of stated attraction to other genders; those who answered “not at all; not very; not sure” were considered not attracted to that category or people.

Any attraction:

Dichotomized yes (sexually attracted to at least one type of person) vs. no (not attracted to any of the four types of people).

Gendered attraction patterns:

Among those with any attraction, categorized as attracted to women only (cisgender and transgender women included); men only; or women and men.

Attraction to women:

Constructed as cisgender women only (exclusively attracted to cisgender women, and not attracted to transgender women); transgender women only; or cis- and transgender women.

Attraction to men:

Constructed as attraction to cisgender men only, transgender men only, cis- and transgender men.

Sexual Partnering.

Respondents were asked “in the last 5 years, who did you have sex with? By sex, we mean any activity you personally define as sexual activity” and could select multiple answers from the same categories presented for attraction (e.g. Women, non-transgender, Transgender Women/Male-to-Female (MTF); etc.). From these responses, four variables were constructed, similar to those constructed for attraction, above.

Current relationship status.

Respondents were asked whether or not they were “currently in a relationship or feel a special commitment to someone.”

Current partner gender.

Respondents who reported being in a relationship were asked to report their current partner’s gender, categorized as cisgender woman, cisgender man, and transgender/GQNB.

Data analyses

Respondents were stratified into one of three gender identity groups (cisgender women; cisgender men; GQNB). Bivariate differences in the distribution of demographic characteristics (including survey-weighted percentages and unweighted counts) across sexual identity were computed using Pearson’s chi² tests. When overall differences were detected across sexual identity, post-hoc adjusted Wald tests, incorporating Bonferroni corrections for multiple comparisons, assessed whether queer participants differed, specifically, from each of the non-queer groups, as well as whether non-queer groups differed from each other. A Bonferroni corrected adjusted Wald p-value <.05 was set as statistically significant. Analyses were conducted in STATA V14.0, and used survey weights to allow for generalization to the U.S. population of sexual minority adults, ages 18–25, 34–41, and 52–59. All proportions reported are weighted percent.

Results

Demographic characteristics

Demographic characteristics of respondents are presented in Table 1. Overall, 5.8% (n = 88) identified as queer. On average, queer respondents were younger than other groups (mean age 26.1). Queer respondents reported significantly higher educational attainment than others—38.6% had graduated from college or obtained a post-graduate degree compared with 31.9% of lesbians and gay men, and less than 30% of other groups. Less than ten percent of queer respondents were living in poverty, fewer than all other groups. No differences were seen across sexual identity in race/ethnicity or residential geographic region (both Census region and urbanity).

Table 1.

Survey-Weighted Demographic Characteristics Of Sample, By Sexual Identity (n=1,507)

(Table 1)	QUEER (5.8%; n=88)		LESBIAN/GAY (46.9%: n=833)		BISEXUAL (40.6%; n=493)		OTHER (6.7%; n=93)		Variable level Bonferroni- corrected p-value	Overall p-value
Measure	%	SE	%	SE	%	SE	%	SE
Birth Cohort
Equality (age 18–25)	76.1	4.6	48.0	2.2	73.1	2.2	77.2	4.2	a, d, e	<.001
Visibility (age 34–41)	21.6	4.5	22.2	1.7	20.5	2.0	11.1	3.0	e, f
Pride (age 52–59)	2.3	1.1	29.8	1.7	6.4	0.9	11.7	3.0	a, b, c, d, e
Mean Age	26.1	0.8	35.4	0.6	26.8	0.5	27.8	1.2	a, d, e	<.001
Race/ethnicity
White	55.3	6.3	61.0	2.1	65.1	2.5	61.5	6.0		0.419
Black/African American	15.0	4.1	17.5	1.7	14.6	1.8	21.2	4.8	N/A
Latinx/Hispanic	29.7	6.3	21.5	1.8	20.4	2.1	17.3	4.7
Education
≤ High School	16.5	6.0	39.3	2.4	47.8	2.8	45.2	6.5	a, b, c, d
Some College	44.9	6.2	28.8	1.9	35.1	2.4	29.8	5.3	a	<.001
Bachelor’s Degree	25.8	4.6	18.5	1.3	11.4	1.2	19.2	3.9	b, d
Post-Graduate	12.8	3.1	13.4	1.0	5.7	0.8	5.8	1.8	d, e
Living in Poverty	9.6	3.8	15.6	1.7	23.8	2.5	16.8	4.7	a, b	0.006
Living in Urban Area	96.5	1.9	87.6	1.5	85.5	2.0	87.9	4.2	N/A	0.118
Census Region
Northeast	17.1	4.4	18.2	1.6	19.9	2.2	22.6	5.2
Midwest	22.9	5.5	17.7	1.7	22.0	2.3	20.4	4.8	N/A	0.167
South	22.4	5.1	38.9	2.2	32.0	2.6	28.4	5.3
West	37.6	5.9	25.2	1.8	26.1	2.3	28.6	6.2
Sex at Birth
Female	83.3	3.9	39.0	2.2	76.8	2.1	78.8	4.8	a, d, e	<.001
Male	16.7	3.9	61.0	2.2	23.3	2.1	21.2	4.8
Current Gender
Woman	56.4	6.1	38.4	2.2	74.6	2.2	44.3	6.2	a, b, d, f
Man	10.0	2.8	59.2	2.2	21.8	2.1	11.5	3.2	a, b, d, f, e	<.001
Genderqueer/non-binary (GQNB)	33.5	5.9	2.4	0.8	3.6	1.0	44.2	6.4	a, b, e, f
Gender Identity
Cisgender women	56.4	6.1	38.4	2.2	74.6	2.2	44.3	6.2	a, b, d, f
Cisgender men	10.0	2.8	59.2	2.2	21.8	2.1	11.5	3.2	a b, d, e, f	<.001
GQNB, female sex at birth	26.9	5.7	0.6	0.3	2.2	0.9	34.4	6.3	a, b, e, f
GQNB, male sex at birth	6.7	2.9	1.8	0.7	1.4	0.5	9.8	3.9	f

Note. All percentages and standard errors (SE) are weighted using Generations survey weights; all n’s unweighted. Percentages may not add to 100.0% due to weighting and/or rounding

‘Variable level Bonferroni-corrected p-value’ indicates if distribution of specific variable level (e.g. “cisgender women only”) differs non-significantly (n.s.), or significantly (p<.05), using adjusted Wald tests, incorporating Bonferroni correction for multiple comparisons, between:

^aQueer and Lesbian/Gay

^bQueer and Bisexual

^cQueer and Other

^dLesbian/Gay and Bisexual

^eLesbian/Gay and Other

^fBisexual and Other

Variable-level analysis only computed for variables with statistically significant ‘overall p-value,’ which tests overall association (Pearson’s chi² test) between variable and sexual identity. If overall association is non-significant, variable-level analysis is not computed (hence ‘N/A,’ aka ‘not applicable’)

Over 83% of queer respondents were assigned female at birth—more than all other groups. Over 56% of queer respondents were categorized as cisgender women, significantly more than the proportion among lesbian/gay respondents (38.4%), but less than the proportion among bisexual respondents (74.6%). Only 10% of queer respondents were cisgender men, lower than all other identity groups (59.2% of lesbian/gay, 21.8% of bisexual, and 11.5% of other-identified were cisgender men). Over a third of queer respondents identified as GQNB, including 26.9% who were assigned female at birth, and 6.7% assigned male at birth. Queer respondents were significantly more likely than lesbian/gay and bisexual respondents to identify as GQNB, but did so at similar (albeit slightly less) frequency than other-identified respondents.

Attraction, Sexual Partnering, and Romantic Relationships By Sexual and Gender Identity

Differences in attraction, sexual partnering, and romantic relationships are shown separately in each of three groups: cisgender women, cisgender men, and GQNB.

Cisgender women (n = 741, Table 2).

Table 2.

Survey-Weighted Distribution of Attraction and Partnering Patterns Among Cisgender Women, by Sexual Identity (n=741)

(Table 2- Cisgender Women)	QUEER (6.0%; n=48)		LESBIAN/GAY (33.1%; n=301)		BISEXUAL (55.5%; n=347)		OTHER (5.5%; n=45)		Variable level Bonferroni- corrected p-value	Overall p-value
Measure	%	SE	%	SE	%	SE	%	SE
Any attraction (n=736)
No	6.7	6.3	0.6	0.5	0.2	0.2	7.1	4.2	n.s.	<.001
Yes	93.3	6.3	99.4	0.5	99.8	0.2	92.9	4.2
Gendered attraction patterns (n=729) †
Women only	15.3	6.5	75.8	3.0	1.5	0.8	7.4	3.4	a, d, e,	<.001
Men only	0	0	0	0	<.0.1	<.0.1	0	0	NA
Women and Men	84.7	6.5	24.2	3.0	98.4	0.8	92.6	3.4	a, d, e
Attracted to women (n=728) †
Cisgender women only	37.6	7.9	80.2	2.9	61.6	3.2	29.4	8.1	a, b, d, e, f
Transgender women only	0	0	0	0	0	0	0	0	N/A
Cis- and transgender women	62.4	7.9	19.8	2.9	38.5	3.2	70.6	8.1	a, b, d, e, f
Attracted to men (n=491) †
Cisgender men only	24.0	7.1	44.7	6.9	53.1	3.3	19.3	8.0	b, e, f	<.001
Transgender men only	8.7	4.3	37.2	6.8	1.8	1.0	12.1	5.8	a, d, e
Cis- and transgender men	67.3	7.8	18.1	5.6	45.2	3.3	68.6	9.1	a, b, d, e, f
Any sexual partners, past 5 years (n=741)
No	9.8	4.6	9.7	2.1	10.4	2.0	13.5	5.5	N/A	0.923
Yes	90.2	4.6	90.3	2.1	89.6	2.0	86.5	5.5
Gendered partnering patterns; n=652) †
Women only	18.8	6.6	83.3	3.3	5.0	1.4	23.0	7.3	a, d, e, f
Men only	15.6	7.9	2.1	1.2	34.9	3.3	32.6	9.2	d, e	<.001
Women and men	65.6	8.8	14.5	3.2	60.2	3.3	44.4	9.9	a, d, e
Sex Partners—Women (n=521) †										0.417
Cisgender women only	91.8	4.3	97.5	1.2	95.4	1.6	94.9	5.0	n.s.
Transgender women only	0	0	0.4	0.4	1.8	1.1	0	0	n.s.
Cis- and transgender women	8.3	4.3	2.1	1.7	2.8	1.2	5.1	5.0	n.s.
Sex Partners--Men (n=378) †
Cisgender men only	85.3	6.5	86.3	8.9	96.4	1.4	94.4	3.3	n.s.	0.036
Transgender men only	0	0	8.1	7.6	1.5	1.1	2.1	2.1	n.s.
Cis- and transgender men	14.7	6.5	5.7	5.5	2.1	1.0	3.5	2.5	n.s.
Currently in Relationship	65.9	7.8	65.5	3.5	66.4	3.1	66.7	8.1	N/A	0.997
Current Partner Gender (n=486)
Cisgender woman	40.1	10.2	97.5	1.6	9.4	2.0	41.8	10.9	a, b, d, e, f
Cisgender man	56.3	10.3	0	0	88.1	2.3	56.7	11.0	a, b, d, e, f	<.001
Trans/GQNB	3.6	2.6	2.5	1.6	2.5	1.1	1.6	1.6	n.s.

Note. All proportions and standard errors (SE) weighted using Generations survey weights; all n’s unweighted. Percentages may not add to 100.0% due to weighting and/or rounding

‘Variable level Bonferroni-corrected p-value’ indicates if distribution of specific variable level (e.g. “cisgender women only”) differs non-significantly (n.s.), or significantly (p<.05), using adjusted Wald tests, incorporating Bonferroni correction for multiple comparisons, between:

^aQueer and Lesbian/Gay

^bQueer and Bisexual

^cQueer and Other

^dLesbian/Gay and Bisexual

^eLesbian/Gay and Other

^fBisexual and Other

Variable-level analysis only computed for variables with statistically significant ‘overall p-value,’ which tests overall association (Pearson’s chi² test) between variable and sexual identity. If overall association is non-significant, variable-level analysis is not computed (hence ‘N/A,’ aka ‘not applicable’)

^†Gendered attraction patterns are only reported for respondents with any stated attraction; for the purpose of this variable, women (and men) include both cis- and/or transgender people. Attracted to women /attracted to men are only reported for those who stated attraction to people of that gender. Sexual partnering variables follow similar patterns.

Approximately 6.0% of cisgender women (CW) identified as queer. The majority of CW (55.5%) identified as bisexual, with an additional 33.1% identifying as lesbian or gay, and 5.5% identifying as something else.

Attraction.

With the exception of a single bisexual respondent who reported exclusive attraction to men, every CW, regardless of identity, reported attraction to women, though largely in conjunction with attraction to men: 84.7% of queer CW and over 90% of bisexual and other-identified CW reported attraction to both men and women. Approximately 15.3% of queer CW were exclusively attracted to women, substantively more than both other-identified (7.4%) and bisexual (1.5%) CW, but significantly less than lesbian/gay CW (76%).

Queer CW were significantly more likely than lesbian/gay and bisexual CW to be attracted to transgender people, and transgender men in particular. Almost two-thirds (62.4%) of queer CW were attracted to both cisgender and transgender women, a significantly higher proportion than lesbians (19.8%) and bisexual women (38.4%). The remaining 37.6% of queer CW attracted to women were attracted exclusively to cisgender women, significantly less than both lesbians (80.2%) and bisexual women (61.6%). No queer cisgender women, nor CW of any identity, reported exclusive attraction to transgender women. However, among the almost 85% of queer CW attracted to men, over 76% reported attraction to transgender men, either exclusively (8.7%) or in conjunction with attraction to cisgender men (67.3%), significantly more than both lesbians (55.3%) and bisexual women (47%).

Sexual partners.

Over 90% of cisgender women reported at least one sexual partner in the five years prior to interview. Among sexually active CW, gendered partnering differed somewhat from gendered attraction. Whereas only one CW was exclusively attracted to men, exclusive partnering with men was common, though less so for queer and lesbian/gay CW. Over 15% of queer CW had partnered exclusively with men in the last five years, more than lesbian/gay CW (2.1%), but half that of both bisexual and other-identified CW. The majority of queer CW (65.6%) had partnered with both men and women, more than all other groups, though differences were significant only relative to lesbian/gay CW.

Among CW who had partnered with women, no significant differences were seen across sexual identity in the gender identity of women partners. Among CW who had partnered with men, queer CW were more likely than all other groups to have partnered with both cisgender and transgender men (reported by almost 15%), yet also were the only group not to have exclusively partnered with transgender men. Queer CW were also the least likely to report exclusively cisgender male partners.

Romantic relationships.

Among those in a relationship, queer CW paralleled other-identified women, and were more likely than lesbian/gay, but less likely than bisexual CW, to have a male partner (reported by 56.3% of queer women) and were more likely than all other groups to have a transgender or GQNB partner (3.6% queer CW, vs. 1.6%−2.5% of all other groups). The remaining 40.1% of queer CW were in a same-gender relationship with a cisgender woman.

Cisgender men (n = 672, Table 3).

Table 3.

Survey-Weighted Distribution of Attraction and Partnering Patterns Among Cisgender Men, by Sexual Identity (n=672)

(Table 3- Cisgender Men)	QUEER (1.5%; n=14)		GAY (73.1%; n=517)		BISEXUAL (23.2%; n-126)		OTHER (2.0%; n=15)		Variable level Bonferroni- corrected p-value	p-value
Measure	%	SE	%	SE	%	SE	%	SE
Any attraction (n=660)
No	0	--	1.4	0.6	0	--	0	--	N/A	0.505
Yes	100.0	--	98.6	0.6	100.0	--	100.0	--
Gendered attraction patterns (n=652) †
Women only	0	0	0	0	4.9	2.8	7.5	7.3	n.s.
Men only	53.2	13.9	87.9	1.8	2.3	1.2	39.3	14.0	a, b, d, e, f	<.001
Women and men	46.8	13.9	12.1	1.8	92.8	3.0	53.2	14.6	a, b, d, e, f
Attracted to women (n=194) †
Cisgender women only	19.4	13.7	54.4	7.9	43.8	5.4	42.7	18.6	n.s.	<.001
Transgender women only	10.4	10.2	32.2	7.5	1.3	0.9	0	0	d
Cis- and transgender women	79.2	16.5	13.4	5.5	55.0	5.4	57.3	18.6	a, d, e
Attracted to men (n=647) †
Cisgender men only	27.8	11.6	75.3	2.6	47.4	5.4	50.2	15.4	a, d	<.001
Transgender men only	0		0		2.0	1.7	0		N/A
Cis- and transgender men	72.2	11.6	24.7	2.6	50.6	5.4	49.8	15.4	a, d
Any sexual partners, past 5 years (n=672)
No	14.3	9.5	6.9	1.5	12.0	3.5	0		N/A	0.202
Yes	85.7	9.5	93.1	1.5	88.0	3.5	100.0
Gendered partnering patterns (n=611) †
Women only	7.9	7.7	0	0	16.6	3.9	16.1	10.5	n.s.
Men only	72.2	13.9	95.9	1.1	27.5	5.3	50.3	14.8	b, d, e	<.001
Women and men	19.9	12.8	4.1	1.1	55.9	5.7	33.6	15.2	b, d, e,
Sex Partners—Women (n=109) †										0.005
Cisgender women only	100.0		69.1	13.2	84.9	4.7	100.0		a, b, e, f
Transgender women only	0		30.9	13.2	2.1	2.1	0		e
Cis- and transgender women	0		0		13.0	4.4	0		n.s
Sex Partners--Men (n=586) †
Cisgender men only	70.6	14.7	98.0	1.0	89.7	3.7	100.0		f	< 001
Transgender men only	13.7	12.5	0.7	0.7	0.6	0.6	0		n.s.
Cis- and transgender men	15.7	10.5	1.4	0.7	9.6	3.7	0		f
Currently in Relationship	39.2	13.5	53.4	2.8	60.4	5.2	39.2	13.7	N/A	0.205
Current Partner Gender (n=367)
Cisgender woman	17.4	16.0	0.5	0.4	52.7	7.1	19.2	17.0	d
Cisgender man	26.8	18.2	97.2	1.4	45.8	7.1	80.9	17.0	a, d	<.001
Trans/GQNB	55.8	21.3	2.3	1.3	1.5	1.0	0		a, b, c

Note. All proportions and standard errors (SE) weighted using Generations survey weights; all n’s unweighted. Percentages may not add to 100.0% due to weighting and/or rounding

‘Variable level Bonferroni-corrected p-value’ indicates if distribution of specific variable level (e.g. “cisgender women only”) differs non-significantly (n.s.), or significantly (p<.05), using adjusted Wald tests, incorporating Bonferroni correction for multiple comparisons, between:

^aQueer and Lesbian/Gay

^bQueer and Bisexual

^cQueer and Other

^dLesbian/Gay and Bisexual

^eLesbian/Gay and Other

^fBisexual and Other

Variable-level analysis only computed for variables with statistically significant ‘overall p-value,’ which tests overall association (Pearson’s chi² test) between variable and sexual identity. If overall association is non-significant, variable-level analysis is not computed (hence ‘N/A,’ aka ‘not applicable’)

^†Gendered attraction patterns are only reported for respondents with any stated attraction; for the purpose of this variable, women (and men) include both cis- and/or transgender people. Attracted to women /attracted to men are only reported for those who stated attraction to people of that gender. Sexual partnering variables follow similar patterns.

Only 1.5% of cisgender men (CM) (n = 14) identified as queer. The majority of CM identified as gay (73.1%), with about a quarter (23.3%) identifying as bisexual, and 2% identifying as something other. In contrast to CW, few significant differences were seen between queer CM and other sexual identity groups in terms of both attraction and sexual partnering, and differences were confined to queer versus gay men.

Attraction.

Among CM with stated attraction, queer CM were relatively split in the gendered patterns of their attractions—53.2% reported attraction exclusively to men; 46.8% to both men and women. Queer CM were significantly more likely than gay CM (12.1%) to report attraction to both women and men, and were significantly less likely to do so than bisexuals (92.8%). No queer (or gay) CM were attracted exclusively to women, though almost 5% of bisexual CM, and 7.5% of other-identified CM, were.

Among queer CM attracted to women, the vast majority were attracted to transgender women. Over 79% of queer CM stated attraction to both cis- and transgender women, substantively more than any other group, and an additional 10.4% were exclusively attracted to transgender women. Similarly, among CM attracted to men, 72.2% of queer CM were attracted to both cisgender and transgender men, more so than all other groups.

Sexual partners.

Among those who were sexually active, queer CM partnering patterns were unique, relative to both queer CW, and CM with other sexual identities. Whereas the majority of queer CW had partnered with both men and women, the majority of queer CM (over 72%) had partnered exclusively with men. An additional 20% of queer CM had partnered with both men and women, significantly less than bisexual CM (55.9%). The remaining 7.9% of queer CM had partnered exclusively with women, half that of bisexual and other-identified CM, but more than gay CM, none of whom had done so.

Queer CM were also unique in the gender identity of their partners, particularly with regard to transgender (men) partners. Whereas 100% of queer CM who had partnered with women, did so exclusively with cisgender women—significantly more so than gay (69.1%) and bisexual (84.9%) CM—queer CM were the least likely group to have partnered exclusively with cisgender men. Almost 14% of queer CM had exclusively partnered with transgender men (compared with less than 1% of each other identity group), and an additional 15.7% had partnered with both cisgender and transgender men (vs. 1.4% of gay, 9.6% of bisexual, and none of the other-identified CM), though results were not statistically significant.

Romantic relationships.

Among the 45.5% of CM who were in a relationship, the majority of queer CM were in a relationship with a transgender/GQNB partner (55.8%), significantly more than gay (2.3%) and bisexual (1.5%) men. Less than one-fifth of queer CM (17.4%) were in an other-gender relationship with a CW, and the remaining 26.8% were in a same-gender relationship with a CM, less than all other groups

Genderqueer/non-binary individuals (n = 94; Table 4).

Table 4.

Survey-Weighted Distribution of Attraction and Partnering Patterns Among Genderqueer/Nonbinary (GQNB) respondents, by Sexual Identity (n=94)

(Table 4- GQNB)	QUEER (24.9%; n=26)		LESBIAN/GAY (14.9%; n=15)		BISEXUAL (19.5%; n=20)		OTHER (39.8%; n=33)		Variable level Bonferroni- corrected p-value	p-value
Measure	%	SE	%	SE	%	SE	%	SE
Any attraction (n=94)
No	0		18.2	12.9	0		30.0	10.2		0.024
Yes	100.0		81.8	12.9	100.0		70.0	10.2	c, f
Gendered attraction patterns (n=84) †
Women only	16.2	7.4	11.0	6.1	0		3.9	2.8	n.s
Men only	5.6	4.1	68.4	14.8	0		7.2	5.3	a, d, e	<.001
Women and men	78.2	8.3	20.6	13.1	100.0		88.9	6.0	a, b, d, e
Attracted to women (n=75) †
Cisgender women only	40.1	10.8	43.0	21.2	33.1	13.0	5.0	3.0
Transgender women only	2.6	2.6	0		0		0		N/A	0.095
Cis- and transgender women	57.3	11.0	57.0	21.2	66.9	13.0	95.1	3.0
Attracted to men (n=72) †
Cisgender men only	12.8	7.6	63.6	19.2	20.5	11.0	10.6	6.0	a, e	0.038
Transgender men only	9.2	6.8	20.2	14.5	23.1	14.2	9.2	6.3	n.s.
Cis- and transgender men	78.0	9.7	16.1	15.0	56.5	14.4	80.3	8.5	a, d
Any sexual partners, past 5 years (n=94)
No	86.2	9.1	96.3	2.9	81.8	13.4	62.0	10.0	N/A	0.104
Yes	13.8	9.1	3.7	2.9	18.2	13.4	38.0	10.0
Gendered partnering patterns (n=74) †
Women only	17.4	8.4	23.0	12.1	22.9	9.8	50.3	13.7	n.s
Men only	32.0	10.8	77.0	12.1	40.0	15.2	10.0	5.4	b, d	0.005
Women and men	50.6	11.5	0		37.0	13.0	39.7	13.3	n.s.
Sex Partners—Women (n=51) †
Cisgender women only	95.9	4.1	78.7	19.9	91.1	8.5	78.0	11.6		0.627
Transgender women only	3.1	4.1	21.3	19.9	0		10.8	10.1	N/A
Cis- and transgender women	0		0		8.9	8.5	11.3	6.7
Sex Partners--Men (n=49) †
Cisgender men only	73.6	11.1	100.0		73.3	14.0	58.6	17.2		0.396
Transgender men only	3.1	3.2	0		7.4	5.5	3.9	4.1	N/A
Cis- and transgender men	23.2	10.9	0		19.4	13.1	37.5	17.0
Currently in Relationship	70.5	11.2	35.7	15.2	73.9	13.4	54.9	10.6	N/A	0.249
Current Partner Gender (n=64)
Cisgender woman	16.9	7.6	14.2	9.8	27.8	11.5	44.6	13.4
Cisgender man	57.9	12.1	72.6	17.0	63.5	13.2	45.0	13.2	N/A	0.346
Trans/GQNB	25.2	11.4	13.2	13.1	8.7	6.5	10.4	6.2

Note. All proportions and standard errors (SE) weighted using Generations survey weights; all n’s unweighted. Percentages may not add to 100.0% due to weighting and/or rounding

‘Variable level Bonferroni-corrected p-value’ indicates if distribution of specific variable level (e.g. “cisgender women only”) differs non-significantly (n.s.), or significantly (p<.05), using adjusted Wald tests, incorporating Bonferroni correction for multiple comparisons, between:

^aQueer and Lesbian/Gay

^bQueer and Bisexual

^cQueer and Other

^dLesbian/Gay and Bisexual

^eLesbian/Gay and Other

^fBisexual and Other

Variable-level analysis only computed for variables with statistically significant ‘overall p-value,’ which tests overall association (Pearson’s chi² test) between variable and sexual identity. If overall association is non-significant, variable-level analysis is not computed (hence ‘N/A,’ aka ‘not applicable’)

^†Gendered attraction patterns are only reported for respondents with any stated attraction; for the purpose of this variable, women (and men) include both cis- and/or transgender people. Attracted to women /attracted to men are only reported for those who stated attraction to people of that gender. Sexual partnering variables follow similar patterns.

The majority (67.3%) of genderqueer/non-binary (GQNB) people were assigned female at birth, with the remaining 33.7% assigned male at birth. Over one-quarter (25.9%) of GQNB respondents identified as queer, substantively more than both CM and CW. The majority of GQNB, almost 40%, identified as some other identity; only 19.5% identified as bisexual, and 14.9% identified as lesbian or gay.

Attraction.

The vast majority of GQNB respondents, including 100% of queer (and bisexual) respondents, reported some attraction. Similar to CW, over 78% of queer GQNB people were attraction to both women and men, significantly more than lesbian/gay respondents (20.6%), and significantly less than bisexuals (100%). Queer GQNB respondents were more likely (but not significantly) than all groups to be attracted exclusively to women (16.2%), and were less likely than gay and other-identified respondents to be attracted exclusively to men (5.2%).

Among GQNB respondents attracted to women, no differences were seen across sexual identity groups in gender identity patterns of this attraction. Some significant differences emerged among those attracted to men, and where these occurred, queer GQNB respondents closely resembled other-identified respondents, and differed significantly only from lesbians/gays: Queer and other-identified respondents were tied for being the least likely group to be attracted exclusively to transgender men (9.2%), with similar majorities of both groups attracted to both cisgender and transgender men (78% of queer and 89.3% of other-identified).

Sexual partners.

There were no differences in the proportion of GQNB people who were sexually active in the last year. Among those who were sexually active, over half of queer respondents (50.6%) had partnered with both women and men, more than all other identity groups (though not significantly so), and almost one-fifth (17.4%) had partnered exclusively with women, less than all other identity groups. There were no significant differences across sexual identity in the gender identity of either women or men sex partners.

Romantic relationships.

No significant differences were seen among GQNB across sexual identity groups in terms of proportion being in a romantic relationship, or, among those in a relationship, in the gender identity of their partner.

Discussion

Our study is the first to use data from a population-based, nationally-representative sample to describe the demographic characteristics and sexuality of queer people compared with other sexual minorities, and to examine how patterns differ by gender. We found that, when given the opportunity to do so, an estimated 5.8% of sexual minority people identified as queer, with almost a quarter of GQNB identifying as such (24.9%), along with 6% of CW, and 1.5% of CM. Several characteristics emerged that distinguished queer respondents from other sexual minority groups, suggesting that queer identity is a unique sexual identity.

In terms of sexuality, queer individuals have sexual attraction and relationship patterns that distinguishes them from lesbian/gay and bisexual identities, though not necessarily from other identified sexual minorities (e.g., pansexual). In general, queer people are more likely than lesbian/gay, and less likely than bisexual people, to report sexual attraction to, and sexual relationships with people of multiple genders, and transgender and GQNB people, in particular. Additional differences emerged by gender identity: Queer CW are distinguished from lesbian and bisexual women in their attraction to, and partnering with, both men and women (in contrast to lesbian/gay women who almost exclusively were attracted to/partnered with women), and attraction to, and partnering with, both transgender and cisgender people (in contrast to bisexual women, who largely reported cisgender partners). This suggests that for CW, identifying as queer may reflect preferences for partners of all gender identities (rather than only cisgender partners of either gender). Among CM, queer respondents report attraction to both cisgender and transgender men and women, yet had partnered only with transgender men, suggesting queer identity among CM may reflect experience with or at least openness to transgender male partners. Queer CW were generally more likely than queer CM to be attracted to and partnered with both men and women, whereas CM were more likely than CW to be attracted to and partnered with transgender people (particularly of the same gender). Both findings largely parallel those reported in Morandini et al. (2017), wherein queer women and men were more likely than lesbian/gay, but less likely than bisexual, same-gender peers, to report sexual and romantic attraction to same-sex and other-sex partners, with multi-gender attractions and partnering more common among queer women than queer men. However, in the Australian study, none of the queer men (or any men at all) had a current non-cisgender partner; in our study almost 56% of queer CM reported a current transgender/GQNB partner, with queer CM additionally more likely than queer CW to have previous transgender partners (Morandini et al., 2017). We are unable to assess why non-cisgender partnering was more common among queer men in our sample than the Australian sample. This discrepancy points to the need for cross-cultural studies, in order to better understand how geopolitical context intersects with sexuality.

Queer GQNB individuals’ attraction patterns typically resembled those of queer CW (e.g. attraction to both women and men and to both cis- and transgender people of either gender). This is perhaps related to GQNB respondents largely being female assigned at birth. However, GQNB queer individuals are largely not distinguished in sexual or relationship partnering from GQNB of other identities, though the relatively smaller sample of GQNB respondents limited statistical power for comparisons.

Demographically, queer respondents were largely cisgender women (56.4%) or GQNB (33.6%) who were mostly female assigned at birth. This result corresponds to earlier studies based on community-based samples that have found women (Bosse & Chiodo, 2016; Katz-Wise & Hyde, 2015; Morandini et al., 2017; Smalley et al., 2016) and gender minorities (Bosse & Chiodo, 2016; Katz-Wise et al., 2016; Kuper et al., 2012; Watson et al., 2019) more likely to identify as queer. Evidence from the broader sexual orientation literature suggests some clues. One explanation is that women are more likely to experience non-exclusive heterosexual attractions, increasing the likelihood of adopting a non-monosexual sexual identity such as queer, to account for attractions and partners of multiple genders, as well as those partners who themselves are sexually and/or gender fluid. Biomedical studies, such as a study of sexual arousal (both genital and subjective) by Chivers, Rieger, Latty, and Bailey (2004), partially confirmed this, finding that regardless of sexual identity (heterosexual vs. homosexual), women experienced sexual arousal to both female and male sexual content, whereas men only experienced arousal to stimuli that corresponded to preferred gender (Chivers, Rieger, Latty, & Bailey, 2004). An alternate explanation may be that women hold stronger beliefs than men about the fluidity of their sexual identity over time, leading them to prefer labels that are inclusive and allow for such flexibility. In her longitudinal study of female gender fluidity, Diamond (2008) has described changes in sexual attraction and orientation among women, noting that women are often attracted to specific people, not their gender. Similarly, in a study by Katz-Wise and Hyde (2015), non-heterosexual women had significantly higher ratings than men on the Sexual Fluidity Beliefs Scale, which included items such as “romantic love depends on the person, not the gender.” Our findings are consistent with these observations.

Queer respondents are also younger. Our findings suggest that younger birth cohorts may also be more likely to utilize different identity labels, including, in particular, queer, echoing what has been reported elsewhere (Jones, 2018; Russell et al., 2009). For example, 7.1% of the Equality cohort, compared with less than 1% of the Pride cohort, identified as queer.

Queer identity was also associated with higher levels of education and they are least likely to be living in poverty. One explanation may be that attending college increases likelihood of identifying as queer, because college students have greater opportunities to learn discourse involving queer theory, as well as learn about emerging identities and terminologies---notably, early discourses of queer theory and identity began in the academy, as did more radical, activist approaches that intersect with social movements (Miller et al., 2016; Sullivan, 2003), and it may be that this heritage led to the association with higher levels of education among queer respondents. College is also a place where young people are living independently from their families of origin for the first time, and thus have the freedom to try out new identities, as well as potentially be exposed to, and thus adopt, a wider range of identities, including queer (Wagaman, 2016).

Alternatively, adopting an openly queer identity may be a result of economic privilege, conferred by attending college. Previous studies have noted that higher SES is associated with greater sexual fluidity among women (Diamond, 2008) and transgender people (Katz-Wise et al., 2016), with the latter study suggesting higher SES may afford people the ability to be more sexually fluid, including identifying as queer and/or GQNB, and partnering with non-cisgender people, with economic privilege offering a protective buffer against stigma. However, at the same time, bisexuals (the more typical plurisexual identity group) report worse socioeconomic status (SES) than other sexual minorities, including lower educational attainment and higher rates of poverty, both in our sample, and in the broader literature(Mirza, 2018; Movement Advancement Project, BiNet USA, & Bisexual Resource Center, 2014). Mereish and colleagues (2017) further emphasize the queer-bisexual disparity, noting that queer women were similar in age to bisexual women in their community sample, yet were significantly more likely to have graduated college or obtained a graduate degree, though the two groups did not differ in employment status or income; (Mereish et al., 2017). Taken together, findings suggest this association is more nuanced than simply a protective effect for plurisexual identities—it instead is only for certain plurisexual identities. Future research is needed to better disentangle the relationship between economic privilege, and social privilege to openly and visibly self-identify with one’s gender and/or sexual identity—particularly with regard to why queer plurisexual identities appear to be advantaged, while bisexuals are disadvantaged.

Strengths and Limitations

This study has several strengths worthy of note. Our analysis represents the first exploration of queer identity in a US nationally-representative sample; currently, no other population-based, representative surveys include queer as a selectable identity option. Another strength is the inclusion of unique measures of the gendered nature of attractions and sexual behaviors, involving transgender as well as cisgender people, in contrast to typical approaches, which ask about only men, only women, or both (SMART (Sexual Minority Assessment Research Team), 2009). As it was attractions to/partnering with transgender people that largely distinguished queer respondents from other identities, access to these measures was a particularly salient strength. Inclusion of these measures may also have proved empowering to queer respondents, as queer people, similar to other plurisexual identities, have reported elsewhere that typical sexuality questions are often unrepresentative of their experiences (Carrotte et al., 2016; Galupo, Lomash, et al., 2017; Galupo et al., 2018; Levy & Johnson, 2012). The study is further strengthened by the inclusion of GQNB respondents. Whereas previous studies have found high amounts of queer identification among transgender samples (e.g. Katz-Wise et al., 2016; Kuper et al., 2012; Morandini et al., 2017), no studies to date have explored dimensions of sexuality with this level of detail among a non-cisgender sample, rendering the sexuality of this group largely unknown. Given that GQNB respondents were also substantially more likely to identify as queer, inclusion of this group, along with our diverse set of sexuality measures, is particularly beneficial for explorations of how queer identity intersects with non-cisgender gender identities.

In the context of these strengths, an important limitation of the study, which potentially limited the scope of the populations we describe, is that the foundational screening question identified respondents based on identifying as “LGBT”—it did not include queer as an option. Thus, we cannot know whether or how many people who identify as queer were excluded because they do not share affinity with a broader “LGBT” identity. Only future population-based studies that include queer as a specific identity category in recruitment could answer whether the demographic, attraction, and relationship differences we see here for queer people (overall, and compared to other sexual minorities) would apply to queer people who identify separately from LGBT.

Analyses were further limited by sample size: despite this being a population-based sample with a large overall sample size, the small number of queer (and other-identified) CM, and GQNB respondents, suggests results should be interpreted with some caution, and preclude more rigorous statistical analyses that account for potential demographic differences (e.g. age, race/ethnicity, SES) in some of the smaller subgroups. For example, sample size precluded us from disaggregating GQNB assigned male at birth and those assigned female at birth. Future population based studies are needed that oversample gender minorities and offer expanded response options for gender identity, to explore how gender identity intersects with queer identity in more depth.

Conclusion

Sexual minorities, in particular young sexual minorities, are increasingly using a growing number of identity labels that were largely unavailable to older generations of sexual minorities. The results of the present study indicate that queer-identified individuals are a sizable group, reflecting almost 6% of the broader sexual minority population (and over a quarter of GQNB people), who are distinct in a number of ways from other sexual minority people both in terms of demographic characteristics and sexuality. Differences in queer sexuality patterns further emerged across gender identity, suggesting that queer identity intersects with gender identity in important ways, particularly with regards to inclusion of transgender/GQNB partners and attractions, as well as GQNB people themselves Our results encourage both survey and qualitative researchers to continue to expand knowledge of the lives of queer individuals, both by assessing measures which offer queer as a response item when assessing sexual identity, and which capture attractions and partnering with non-cisgender people, and by further exploring the personal and demographic characteristics, and sexual lives and relationships, of this growing population.