Several years ago, Michael Bailey (Northwestern) and Richard Pillard (Boston School of Medicine) published sophisticated studies on the genetic contribution to the development of male and female homosexuality. These studies reported very high estimates of genetic influence and garnered much media attention. Newsweek had a cover story on the research, asking in the cover headline, "Is this child gay?"

Follow-up research that refutes earlier findings seldom receives the same level of coverage. It should, because Bailey's most recent study (forthcoming in the Journal of Personality and Social Psychology) suggests that genetic influence on homosexual development may be dramatically less than his earlier studies projected.

In their earlier studies, Bailey and Pillard searched for gays who were twins and investigated the sexual orientations of their siblings. Research subjects were solicited through advertisements in openly progay magazines and tabloids. (This method raised the concern that scientists call sample bias.) They found a "Probandwise Concordance" (PC) of 52 percent for male identical twins. For fraternal male twins, the PC was 22 percent. Similar concordances were reported for lesbian twins.

Other research had failed to produce estimates of genetic influence as strong as those of Bailey and Pillard. Now, Bailey himself (to his credit) has provided the crucial refutation of his earlier estimates. To avoid possible sample bias, Bailey sent a questionnaire on sexual preferences and experiences to the entire Australian Twin Registry, an exhaustive listing of all twins born in its population.

The influence of genetics on development of homosexual orientation would, on the basis of this superior research, appear to be half or ...