The Detection of Environmental Influences on Academic Achievement Appears to Depend on the Analytic Approach.

ABSTRACT

One long-standing analytic approach in adoption studies is to examine correlations between features of adoptive homes and outcomes of adopted children (hereafter termed 'measured environment correlations') to illuminate environmental influences on those associations. Although results from such studies have almost uniformly suggested modest environmental influences on adopted children's academic achievement, other work has indicated that adopted children's achievement is routinely higher than that of their reared-apart family members, often substantially so. We sought to understand this discrepancy. We examined academic achievement and literacy-promotive features of the home in 424 yoked adoptive/biological families participating in the Early Growth and Development Study (EGDS; i.e., adopted children, adoptive mothers, birth mothers, and biological siblings of the adopted children remaining in the birth homes) using an exhaustive modeling approach. Results indicated that, as anticipated, adopted children scored up to a full standard deviation higher on standardized achievement tests relative to their birth mothers and reared-apart biological siblings. Moreover, these achievement differences were associated with differences in the literacy-promotive features of the adoptive and birth family homes, despite minimal measured environment correlations within adoptive families. A subsequent simulation study highlighted noise in measured environmental variables as an explanation for the decreased utility of measured environment correlations. We conclude that the field's heavy focus on measured environment correlations within adoptive families may have obscured detection of specific environmental effects on youth outcomes, and that future adoption studies should supplement their measured environment analyses with mean differences between reared-apart relatives.