Stability and Change in Gender Identity and Sexual Orientation Across Childhood and Adolescence

Abstract

As increasing numbers of transgender, gender diverse, and queer youths come out to their friends, families and communities, their rights to express their identities in public life have become the subject of intense media scrutiny and political debate. But for all the attention transgender, gender diverse, and queer youth have received from politicians, journalists, and public intellectuals, basic science research on how these youth actually experience their identities over time remains scarce. In this monograph, we contribute to the emerging knowledge base on this topic by presenting a detailed quantitative description of gender identity and sexual orientation in a sample of over 900 North American transgender, gender diverse, and cisgender youths in the Trans Youth Project (M_age = 8.1 years at first visit; M_age = 14.3 at latest visit; 99% living in the United States, 1% in Canada; 69% non-Hispanic white; 73% household income >$75,000). Youths are in one of three groups: (1) a group of early identifying transgender youths, who were supported by their parents in a social gender transition (changing their name, pronouns, hairstyle, and clothing) by age 12 (M_{age at transition} = 6.5; N = 317); (2) a group of their siblings, who were cisgender at the beginning of their participation in the study (N = 218); and (3) a group of cisgender youths who were age- and gender-matched to, but not family members of, the early identifying transgender youths (N = 377). Data on the youths' identities have been collected from the youths themselves and their parents between 2013 and 2024. We had two primary research goals. First, we described stability or change in youths' gender identity (Chapter 4) and sexual orientation (Chapter 6). We asked whether transgender youths' rates of change were or were not different from those of cisgender youths. Second, we examined whether measures of gender development earlier in development were related to youths' later gender identity (Chapter 5) or sexual orientation (Chapter 6) trajectories into adolescence. Stability in gender identity was by far the most common pathway for youths in all three groups, with over 80% of youths showing stability throughout their participation in the study. We saw similarity between the three groups of youths, such that the early identifying transgender youths were no more or less likely to show gender change than their siblings or youths in the unrelated comparison sample. Nevertheless, 11.9% of youths who started as cisgender were not so at their most recent report-a much higher proportion than would be predicted based on assumptions held in classic developmental psychology research about gender since the 1950s. When gender change did occur in all three groups, it overwhelmingly involved change to (and, to a lesser extent, from) a nonbinary gender identity. Results were similar regardless of whether youth- or parent-report data were considered, and we found no evidence that youths were more or less likely to change at particular ages. We observed some evidence that more gender nonconformity in childhood (e.g., more femininity in childhood among children living as boys) was related to later gender change, but results were somewhat inconsistent across measures and gender identities. Youths showed diverse sexual orientations, with 60% of binary transgender and 33% of cisgender adolescents expressing queer (i.e., not straight) romantic or sexual interest. A high percentage of youths overall (37%) indicated interest in both boys and girls-a pattern particularly common among nonbinary youths. Finally, more than a third of youths have shown change in their sexual orientation, and childhood gender nonconformity was associated with whether currently binary transgender or cisgender teenagers most recently reported a queer identity. Our results accord with recent evidence indicating that today's youth are defying assumptions about gender and sexual orientation from decades of developmental research, considering gender and sexual orientation to be relatively flexible social identities rather than ones that are fixed, and view gender as having more than two categories. Early identifying transgender children's sense of their own gender was no more or less stable than cisgender children's, suggesting that children who are supported in their transgender identities tend to show developmental patterns that mirror their cisgender peers. Finally, in Chapter 7, we discuss how our findings exemplify and respond to this unique historical moment, the ways in which our findings do and do not align with past work about gender- nonconforming children, and how future research can continue to make strides toward better understanding a wider swath of gender development trajectories.

Figure 1

Key legal, historical, cultural, and linguistic LGBTQ milestones.

Figure 2

Relevant gender‐related milestones among Recruited as Transgender youths. Note: “Blocker use” refers to gonadotropin‐releasing hormone (GnRH) agonists, medications that suppress the physiological changes associated with endogenous puberty (e.g., breast growth in assigned females, voice deepening in assigned males). “Hormone use” refers to hormonal treatments that induce physical changes associated with exogenous puberty (e.g., testosterone causing voice‐deepening and facial hair growth in transgender boys, estrogens causing breast development and body fat redistribution in transgender girls). Distributions of First “Cross‐gender” Behavior and First Social Transition include all 317 youths in the Recruited as Transgender group, whereas distributions for Starting Blocker Use and Starting Hormone Use do not, because some youths will never access these treatments, and others will (likely) begin to access them after the time period reported in this monograph.

Figure 3

Example of Step 1 and Step 2 gender‐at‐visit coding.

Figure 4

Density curves of youth‐report Continuum scores in all three recruitment groups, broken down by Step 1 coding on measures 4, 5, and 6. Dashed lines represent means. Note: All three recruitment groups are included in every panel of this figure due to small N's in some categories.

Figure 5

Density curves of youth‐report Continuum scores in all three recruitment groups, broken down by Step 2 gender‐at‐visit coding. Dashed lines represent means. Note: In 7.4% of visits represented in this figure, human coders had access to Continuum scores when deciding visit‐level gender codes; all three recruitment groups are included together in this figure due to small N's in some categories.

Figure 6

Between youth‐ and parent‐reported gender‐at‐visit codes. Note: Youths with multiple visits at which both parent and youth provided relevant answers are included multiple times in this figure.

Figure 7

Relation between recruitment gender and youths' current gender identities, per youth report.

Figure 8

Most recent gender identity according to youth report, among youth with at least one codable follow‐up report.

Figure 9

Relation between recruitment gender and youth's current gender identity, per parent‐report. Note: Darker yellow and blue overhanging portions on the top and bottom of the left side of each panel represent youths whose parents did not provide a codable report of the youth's gender on a follow‐up visit.

Figure 10

Most recent gender identity according to parent‐report, among youths with at least one codable follow‐up report from parents.

Figure 11

Early continuum scores by gender at time of report and recruitment modality. Note: Black diamonds represent means; points are jittered for visual clarity.

Figure 12

Early preference scores by gender at time of report and recruitment modality. Note: Black diamonds represent means; points are jittered for visual clarity.

Figure 13

Early parent questionnaire scores by gender at time of report and recruitment modality. Note: Black diamonds represent means; points are jittered for visual clarity.

Figure 14

Relationship between target of romantic attraction and asexual, interest in nonbinary people, and questioning.

Figure 15

Earliest continuum scores broken down by target of attraction, current gender, and current modality. Note: Includes only youths who are currently boys or girls (not gender diverse); “Gay” as it applies to current girls is equivalent to “Lesbian.” Black diamonds represent group means. Points are jittered for visual clarity.

Figure 16

Earliest preference scores broken down by target of attraction, current gender, and current modality. Note: Includes only youths who were most recently boys or girls (not gender diverse); “Gay” as it applies to current girls is equivalent to “Lesbian.” Black diamonds represent group means. Points are jittered for visual clarity.

Figure 17

Earliest parent questionnaire scores broken down by target of attraction, current gender, and current modality. Note: Includes only youths who were most recently boys or girls (not gender diverse); “Gay” as it applies to current girls is equivalent to “Lesbian.” Black diamonds represent group means. Points are jittered for visual clarity.

Figure 18

Relationship between earliest and most recent report of attraction.