This study examined the birth order of girls with gender identity diso
rder (N=22). Each proband was matched to 3-7 clinical control girls fo
r age at assessment and number of siblings (the mode number of control
s per proband was 7) (total N=147). The number of older brothers, olde
r sisters, younger brothers, and younger sisters was recorded. Slater'
s birth order index showed that the probands were significantly more l
ikely to be early born than were the controls. A modified Slater's ind
ex also compared the birth order of the probands and the controls only
to their brothers (when they had one or more) and only to their siste
rs (when they had one or more). Compared to the controls, the probands
were born early compared to their sisters, but not to their brothers.
These findings are the inverse of two previous studies of boys with g
ender identity disorder, who were later born relative to clinical cont
rol boys (11), an effect that appeared to be accounted for primarily b
y being born later relative to older brothers, but nor to older sister
s (46).