<p>Sexual orientation may be shaped, in part, by organizing effects of gonadal steroids during development. However, the interpretation of associations between sexual orientation and markers of such organizing effects may be hampered by the use of Kinsey-type scales of hetero-/homosexuality, because they assess sexual orientation from the individual’s (i.e., egocentric) perspective and inherently represent a difference score between attraction to women versus attraction to men, with unclear meaning of medium scores. Using illustrative data from three studies, we show that a clearer and more consistent pattern emerges once one uses measures of sexual orientation that assess sexual attraction from the (allocentric) perspective of one’s object of attraction; that is, as attraction to women (gynephilia) and to men (androphilia). We find that, in both men and women, gynephilia is negatively, and androphilia positively, associated with digit ratio, a measure of the relative prenatal exposure to testosterone versus estradiol. This suggests that relatively more exposure to testosterone predisposes individuals to be more attracted to women and less to men and vice versa for relatively more exposure to estradiol. We also find that ulna-to-fibula ratio, a putative marker of pubertal steroid effects, is positively associated with gynephilia, particularly in men, suggesting that relatively more testosterone exposure during puberty may be another source of individuals’ sexual attraction to women. When we constructed a Kinsey-type measure of homo-/heterosexuality from androphilia and gynephilia scales, our findings for digit ratio replicated those reported in the literature for generic Kinsey-type scales. But we argue that the findings we obtained with allocentric gynephilia and androphilia measures are more straightforward to interpret.</p>