DeFries-Fulker and Pearson-Aitken model-fitting analyses of reading performance data from selected and unselected twin pairs

Abstract

Although a comparison of concordance rates for deviant scores in identical and fraternal twin pairs can provide prima facie evidence for a genetic etiology, information is not fully utilized when continuous measures are analyzed in a dichotomous manner. Thus, DeFries and Fulker (Behav Genet 15:467-473, 1985; Acta Genet Med Gemellol, 37:205-216, 1988) developed a regression-based methodology (DF analysis) to assess genetic etiology in both selected and unselected twin samples. While the DF analysis is a very versatile and relatively powerful statistical approach, it is not easily extended to the multivariate case. In contrast, structural equation models may be readily extended to analyze multivariate data sets (Neale and Cardon, Methodology for genetic studies of twins and families, 1992). However, such methodologies may yield biased estimates of additive genetic, shared environmental, and non-shared environmental influences when multivariate models are fitted to selected twin data. Therefore, the Pearson-Aitken (PA) selection formula (Aitken, Proc Edinburgh Math Soc B, 4:106-110, 1934) was used to analyze reading performance data from twins with reading difficulties (selected sample) and a population of normally-achieving twin pairs (control sample). As a comparison, DF models were also fitted to these same data sets. In general, resulting estimates of additive genetic, shared environmental, and non-shared environmental influences were similar when the DF and PA models were fitted to the data. However, the PA selection formula may be more readily generalized to the multivariate case.