Social epidemiology of gender diversity in early adolescents

Background: Few large U.S. cohort studies have examined multidimensional measures of gender diversity; therefore, this study investigates their associations with sociodemographic factors in a national sample of 12- to 13-year-old adolescents. Methods: We conducted a cross-sectional analysis of Adolescent Brain Cognitive Development (ABCD) Study data (2019–2021, Year 3 follow-up, N = 10,089). Associations between sociodemographic characteristics and gender were evaluated using ordinal logistic regression for ordinal measures (gender non-contentedness and gender expression), multinomial regression models for categorical measures (transgender identity and felt gender), and linear regression models for continuous measures (felt gender and gender spectrum). Results: Gender spectrum measures revealed the most diverse responses: 25.9% of sex-assigned females self-describing masculinity and 9.9% of sex-assigned males self-describing femininity. Female sex assigned at birth was associated with greater gender-diverse responses across all continuous and categorical gender measures compared to male sex assigned at birth. Higher household income was associated with less gender diversity relative to lower household income. Conclusion: Assessment of gender across multiple measures beyond binary gender identity (e.g., transgender vs. cisgender) may yield more diversity of responses in early adolescents. Future research could explore explanations for the sociodemographic factors associated with greater gender diversity in early adolescents. Impact: Our study uses a large and diverse cohort to analyze sociodemographic associations with gender diversity using multidimensional measures. Across gender measures, female sex assigned at birth and lower household income were associated with greater gender diversity. Future research on gender diversity in early adolescents should use multidimensional gender measures to better understand the nuances of gender development.