Table 2.
Descriptive statistics for preference, similarity, and implicit gender identity measures (Studies 1a and 1b)
| Study 1a | Study 1b | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Preferences1 | Similarity2 | IAT3 | Preferences | Similarity | IAT | |||||||
| M | SD | M | SD | M | SD | M | SD | M | SD | M | SD | |
| Unrelated cisgender children | 75.77 | 14.23 | 2.03 | 1.20 | 0.35 | 0.44 | 76.48 | 15.05 | 1.94 | 1.19 | 0.46 | 0.47 |
| Siblings | 78.12 | 15.40 | 1.97 | 1.52 | 0.37 | 0.51 | 75.22 | 16.88 | 2.04 | 1.27 | 0.41 | 0.50 |
| Transgender | 76.38 | 14.48 | 1.97 | 1.32 | 0.27 | 0.51 | 72.42 | 17.26 | 1.94 | 1.35 | 0.34 | 0.47 |
| Gender nonconforming | 69.70 | 22.52 | 0.93 | 1.72 | 0.16 | 0.53 | 60.11 | 22.93 | 0.63 | 1.86 | −0.16 | 0.41 |
| Group comparisons (ANOVA) | F(3,380) = 2.43, p = .065, ηp2 = .02 | F(3,380) = 6.24, p < .001, ηp2 = .05 4 | F(3,273) = 1.43, p = .234, ηp2 = .02 | F(3,528) = 12.90, p < .001, ηp2 = .07 5 | F(3,538) = 15.70, p <.001, ηp2 = .08 6 | F(3,338) = 14.50, p < .001, ηp2 = .11 7 | ||||||
Notes.
Preference scores ranged from 0 to 100. For transgender and cisgender participants, 0 indicated preferences aligned with the other gender, 100 indicated preferences aligned with their current gender. For gender nonconforming participants, 0 aligned with their gender assigned at birth, 100 aligned with the other gender.
Similarity scores ranged from 0 to 4. For transgender and cisgender participants, 0 indicated perceived similarity to the other gender, 4 indicated similarity to their current gender. For gender nonconforming participants, 0 aligned with their gender assigned at birth, 4 aligned with the other gender.
For transgender and cisgender participants, negative scores indicated identification with the other gender, and greater positive scores indicated identification with their current gender. For gender nonconforming participants, negative scores showed identification with gender assigned at birth, positive scores showed identification with the other gender.
Post-hoc Tukey comparisons showed that only gender nonconforming children’s scores differed from all other groups (ps ≤ .001); all other groups were not significantly different (ps > .980).
Post-hoc Tukey comparisons showed that only gender nonconforming children’s scores differed from all other groups (ps < .001); all other groups were not significantly different (ps > .100).
Post-hoc Tukey comparisons showed that only gender nonconforming children’s scores differed from all other groups (ps < .001); all other groups were not significantly different (ps > .900).
Post-hoc Tukey comparisons showed that only gender nonconforming children’s scores differed from all other groups (ps < .001); all other groups were not significantly different (ps > .200).