The construct of empathy was examined in 20 boys with gender identity disorder (GID), 20 clinical control boys with externalizing disorders (ECC), 20 community control boys (NCB), and 20 community control girls (NCG). The mean age of the children was 6.86 years (range = 4-8 years). It was hypothesized that boys with GID would show similar levels of empathy to those shown by NC girls and higher levels of empathy than the NC and ECC boys. Three measures of empathy were administered: a maternal-report questionnaire, a self-report questionnaire, and an in vivo evaluation in which children's reactions to pain simulations to two adult actors (mother, experimenter) were coded for empathy levels. On the maternal report and in-vivo measures, the NC girls had significantly higher empathy levels than the NC boys, but not on the self-report measure. By maternal report, the NC girls were rated as significantly more empathic than were the GID boys, with a "large" effect size. There were no significant differences between the GID boys and the NC girls on the self-report and in-vivo measures and the effect size differences were "small." No significant differences were observed between the GID and NC boys; however, there were "medium" and "small" effect size differences with boys with GID showing more empathy on the in-vivo and self-report measures, respectively. On the maternal-report measure, the GID boys were rated as significantly more empathic than the ECC boys and there was a trend for the GID boys to show greater levels of empathy than the ECC boys on both the self-report and in-vivo measures. The effect size differences on all three empathy measures were "medium" to "large," with GID boys showing more empathy than ECC boys. Empathy as a dispositional characteristic in the genesis and perpetuation of GID in boys is discussed.