Political Science, Biometric Theory and Twin Studies: A Methodological Introduction

Abstract:     
As political scientists begin to incorporate biological influences as explanatory factors in political behavior the need to present a methodological road map for utilizing biometric genetic theory and twin data specifically, is apparent. The classical twin design (CTD) remains the most popular design for initial epidemiological examinations of the source of variance among social and political behaviors, and a vast majority of advanced variance components models as well as many molecular analyses are largely extensions of the CTD. Thus, it is appropriate to begin a series of works with the CTD and its most common variants. The CTD has strong roots in biometrical genetic theory, and provides estimates of the correlations between observed traits of monozygotic and dizygotic twins in terms of underlying genetic and environmental influences. The majority of these analyses utilize structural equation models (SEM) of observed covariances for both twin types to assess the relative importance of these 'latent' factors. SEM estimates model parameters by minimizing a goodness-of-fit function between observed and predicted covariance matrices, usually by maximum-likelihood criterion.

Working Paper Series