Measurement and causal identification in constitutional law: A reply to Niels Petersen and Konstantin Chatziathanasiou

The past few decades have seen a trend toward empirical constitutional law scholarship. In a new article, Niels Petersen and Konstantin Chatziathanasiou take stock of this emerging field and suggest that this body of research may be all “smoke and mirrors.” They focus on explaining two well-documented potential problems with empirical research using observational data—omitted variable bias and measurement error—and argue that these problems are severe enough to call into question the growing body of cross-national research on the spread and effect of constitutional provisions. But while Petersen and Chatziathanasiou’s article points out genuine obstacles for empirical research in constitutional law, it fails to accurately document the way that these issues have been explained and addressed by the existing literature. In this reply, we make four points about the state of the empirical literature in comparative constitutional studies. First, we explain that the problems identified by Petersen and Chatziathanasiou have already been extensively documented and are the subject of ongoing conversation among those working in the field. Second, we describe the work that researchers have done to address these problems. Third, we argue that the methodological pluralism that Petersen and Chatziathanasiou advocate for as a way forward for the field is already well underway. Finally, we suggest that evaluating an entire literature, like Petersen and Chatziathanasiou seek to do in their article, is best done through the use of a “systematic review,” to ensure that the evaluation will offer a fair characterization of the field.