Family Configuration and Intelligence

Abstract

A survey conducted in New York [USA], beginning with children born in 1963, showed that there was a definite rise in test scores coincident with the sharp reversal in birth trends. Several other states have also reported rising test scores in lower elementary grades. A variety of findings reveal the impact of family configuration [age, gap, family size, birth rank, parental absence, etc.] on intelligence. The data pattern is consistent with the analysis of intellectual development based on the confluence model. Not all variation in intelligence is accounted for by variation in family configuration, e.g., in the USA the large decline in SAT [scholarship aptitude test] scores cannot be a function of changes in family configuration alone because it is considerably larger than can be expected on the basis of a simple extrapolation from the samples. The confluence model does not generate the only interpretation of these facts; for each is possible to supply another reasonable explanation. The drop in SAT scores may be due to a general decline in intellectual interests, and the lower intelligence scores of children living in 1 parent homes may be due to a history of conflict or stress. The confluence model has the advantage of economy, and since it makes rather specific predictions it can be readily verified. On the basis of the empirical evidence now available it is not possible to evaluate the relative importance of genes and environment on intelligence. Some specific derivations with implications for the analysis of genetic effects on intelligence [twins and non-twin siblings] follow directly from the confluence model. Judging from the consistency and magnitude of some of the effects reviewed, the variables specified by the confluence model seem to account only for the small fraction allowed by heritability analysis to environmental factors and to error.