Primary schoolchildren in married heterosexual households are 35 percent more likely to make typical school progress than peers in same-sex households, according to a new study published in the respected academic journal Demography.

The finding is based on data from 1.6 million children in the 2000 Census, which included 8,632 children who lived in same-sex households.

The new study also re-examines findings from a 2010 study that used the same data source but concluded that children raised in sex-same households progressed just as well as children in married heterosexual households when differences in the socio-economic status (i.e., household income and parental education) are taken into consideration.

Why the different conclusions regarding children’s grade retention using the same data source?

The Census provides only a single-year picture of children’s living arrangements, so its data do not reflect their full family history. For example, married households include both first-marriage and remarried couples. The 2010 study tried to address this issue by limiting its sample in two (as it turns out, significant) ways.

First, it included only biological children (or “own children,” in Census terms) of the heads of the household in the Census. Second, the 2010 study examined only biological children who had lived in the same location for five or more years. Effectively, the 2010 study examined a subset of residentially stable children living with at least one biological parent.

Noting that these two restrictions may not represent the full family experiences of all children, the new study re-examines the data without these two limitations. When both restrictions are lifted, the sample size increased by nearly 125 percent, from 716,740 children to 1,610,880 children.

What happens when more children are included in the analysis? The new study finds that:

  • When the sample consisted of only biological children, regardless of residential stability, children in married heterosexual households were 25.8 percent more likely to make typical school progress than peers raised in same-sex households;
  • When the sample consisted only of children who stayed in the same residence for five or more years, regardless of their biological status, children in married heterosexual households were 29.5 percent more likely to make typical school progress than peers in same-sex households;
  • Even when adopted children are excluded from the residentially stable sample (taking into consideration that children adopted by married heterosexual couples may be different from those adopted by same-sex couples), children in married heterosexual households were still 24 percent more likely to make typical progress in school; and
  • When the sample consisted of all children, regardless of their biological status or residential stability, children in married heterosexual households were 35.4 percent more likely to make normal progress in school than peers in same-sex households.

The new study accounts for households’ disability status, race, income, education, birthplace, metropolitan status, private-school attendance, and state residence. In other words, it compares children who are nearly identical on these characteristics.

Consistent with previous research, these findings suggest that when considering how children’s family environment influences their outcomes, it is important to look at both family structure and stability.

Together, the pair of studies underlines the complex dynamics between children’s family situations and their well-being, as well as the difficulty of analyzing that relationship even with sophisticated research methods and data. The studies also underscore the necessity for policymakers to weigh the full accumulating research evidence in their decision-making.