William C. Duncan at NRO's The Corner:
The journal Demography has just published a very interesting article that reexamines the claims of a 2010 study that suggested (and was widely reported) as showing that children raised by same-sex couples experienced no academic disadvantages. The catch of the earlier study was that it was significantly different from previous studies on same-sex children and their parents since it used a large sample from the Census rather than a small self-selected one which is more typical of this body of research.
The 2010 study had excluded children who were not biologically related to the head of household and who were not in the same home for at least five years. This reduced “the sample size by more than one-half.” The 2012 study explains that putting the children who had been in unstable households (lived at the same address less than five years) back into the sample increases the sample “by more than 80 percent.” This fact alone seems important. The new study’s conclusion is that “children being raised by same-sex couples are 35 percent less likely to make normal progress through school.”