Assessing causal inference problems with Bayesian process tracing: the economic effects of proportional representation and the problem of endogeneity

ABSTRACT

In recent years, causal process tracing is becoming increasingly formalised as scholars have begun to follow Bayesian logic and thereby manage to combine the interpretative and contextual nuance of older forms of process tracing with the inferential rigours of Bayesian analysis. This paper illustrates the basic logic of Bayesian process tracing by drawing on the political economy literature that studies the social policy and economic effects of electoral systems. It compares and contrasts how each approach deals with the problem of endogeneity.

KEYWORDS:

• Process tracing
• Bayesianism
• endogeneity
• electoral system choice

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Marcus Kreuzer is Professor in the Political Science Department of Villanova University, Villanova, PA. He works on the formation of party systems in interwar and post-communist Europe. He also has a side-interest in qualitative methodology.

Notes

1. Some scholars prefer the term certainty over specificity. I follow Bennett who draws on medical literature differentiation of the specificity and sensitivity of medical tests. Sensitivity and specificity capture two qualities of medical testing instruments that will determine the probability of producing false-positive and false-negative results. (Bennett 2014) In analogy to such medical testing instruments, hypotheses can also be viewed of having distinct qualities that affect their false-positive and negative results. There still is some debate on what those specific qualities are. I propose to judge a hypothesis' specificity in terms of the explicitness of its causal mechanisms, historical boundary conditions, as well as attention paid to potential confounder, endogeneity problems and omitted variables.