Preregistration of experimental research on democratic innovations

Preregistration is one of the crucial practices advanced by the open science movement and it is gaining prominence within the social and behavioral sciences. It involves defining the research questions and outlining the analysis plan before seeing the research outcomes (i.e. collecting the data). The main goal of preregistration is transparency. The idea is that by registering the studies beforehand, we can prevent some of the common biases within academic research, such as publication bias and p-hacking. Sceptics, however, have argued that preregistration can take away the curiosity and discovery which lie at the heart of scientific endeavor and can put off the scientists from doing a high-risk research. Preregistration is very new to democratic innovations literature. In this paper, I discuss the benefits and challenges associated with preregistration of studies on democratic innovations. I showcase my arguments with two large experimental studies (N=5000) both of which investigate the macro effects of one type of democratic innovations, citizens’ assemblies. I conclude by underscoring the lessons learned from my first-hand experience with preregistration and making a set of practical recommendations for preregistering experimental studies on deliberative minipublics.