Publications of Zoltán Fazekas

The importance of context in the genetic transmission of US party identification

In this study, we discuss one example where behavior genetic findings vary greatly across political contexts. We present original findings on how party identification is heritable around the 2008 election on a sample of twins from Minnesota. As this is in contrast with findings from the late 1980s and with how a mid‐2000 study interpreted their results, we explain how the increasing partisan ideological polarization could be responsible for these seemingly contradictory findings. In the Minnesota sample, we show a genetic correlation between party identification and ideology, a finding consistent in the political science literature. We highlight how heritability of political characteristics, like all others, is population specific and highly context dependent stressing its nondeterministic nature.

Validity of Survey Response Propensity Indicators: A Behavior Genetics Approach

Objectives This study explains how behavior genetic analysis using a twin design can help us assess the validity of our measures. Methods We test multiple indicators of response propensity, a measure used by survey researchers to better understand the similarities and differences between survey respondents and nonrespondents. The response propensity indicators evaluated include response to follow‐up surveys and subsequent waves of a panel and the completion of a sensitive recontact information sheet to aid subsequent recontact efforts. Results A classical and the newly proposed method of validation all point to insufficient validity of our response propensity measures. Construct validation using data from the National Survey of Midlife Development in the United States exhibited little correlation between indicators. Genetic analysis suggests that the success of subsequent data‐collection efforts is predominantly driven by additive genetic effects, while nonresponse to inquiries for recontact information is influenced predominantly by familial environmental predictors. Conclusion Our results indicate that different underlying constructs drive the response propensity indicators, suggesting that nonresponse is, at minimum, multidimensional.