Arch Sex Behav (2007) 36:864–867 DOI 10.1007/s10508-007-9169-0 ORIGINAL PAPER Reply to Blanchard’s (2007) “Older-Sibling and Younger-Sibling Sex Ratios in Frisch and Hviid’s (2006) National Cohort Study of Two Million Danes” Morten Frisch · Anders Hviid Published online: 17 February 2007 C  Springer Science+Business Media, LLC 2007 As correctly pointed out by Blanchard (2007), the study de- sign and statistical procedures used in our article (Frisch & Hviid, 2006) were quite different from those of prior studies on the possible association between older brothers and male homosexuality, the so-called fraternal birth order (FBO) hy- pothesis. Our investigation was the first nationwide cohort study to address the issue, a study design known from basic textbooks of epidemiology to be less prone to the influence of various biases than the case-control design used in prior studies. We agree with Blanchard that the childhood family corre- lates of heterosexual marriage in our study should not be viewed as correlates of heterosexual orientation. By fol- lowing a nationwide cohort of (predominantly heterosex- ual) unmarried persons until their date of first heterosexual marriage, the observed associations reflect childhood dif- ferences between heterosexuals who stay formally unmar- ried (whether remaining single or living with a partner) and heterosexuals who marry. The relatively small proportion of non-heterosexual persons in the population is unlikely to have influenced the observed associations to any notice- able extent. Therefore, rate ratios (RRs) in the heterosexual marriage analyses reflect childhood differences between two (predominantly) heterosexual subsets of the Danish popu- lation so, clearly, these are not correlates of heterosexual orientation. In our parallel analysis of correlates of homosexual marriage, we followed the same cohort of (predominantly M. Frisch () · A. Hviid Department of Epidemiology Research, Division of Epidemiology, Statens Serum Institut, 5 Artilerivej, DK-2300 Copenhagen S, Denmark e-mail: mfr@ssi.dk heterosexual) unmarried persons until their date of first homosexual marriage, censoring periods when they were heterosexually married. Consequently, the associations we observed for homosexual marriage reflect childhood dif- ferences between (predominantly heterosexual) unmarried cohort members and homosexual persons who entered same-sex marriage. The extent to which the significant childhood correlates of homosexual marriage in our study can be interpreted as correlates of homosexual orientation will depend on how similar childhoods of homosexuals in same-sex marriages are to those of other homosexual persons. If homosexually married persons have childhood experiences that are not substantially different from those of other homosexual persons, our analyses for homosexual marriage may actually provide some reasonable suggestions as to possible childhood differences between broader groups of homosexual and heterosexual persons. Older-Sibling Sex Ratios The older-sibling sex ratios (ratios of older brothers to older sisters) calculated by Blanchard suggest that the way we cut our observation period in the study cohort, using restric- tions in birth year (1952 to 1983), calendar period (1970 or 1989 to 2001), and age (18 to 49 years) may have led to a relative deficit of older sisters in the sibships of some cohort members born early in the birth year inclusion pe- riod. The establishment of parent-offspring links for Danish citizens born before the initiation of the Civil Registration System in 1968 depended to some extent on the household compositions in 1968. Therefore, because women generally establish their own families and move away from the parental household at a younger age than men do, sibships of some early-born cohort members may have been incomplete, and Springer