What if most political science studies are too weak to find the effects they’re looking for? In this episode, we dig into a new paper by Vincent Arel-Bundock and colleagues that reveals a striking truth: quantitative political science is greatly underpowered. With thousands of tests analyzed, the authors show that many studies have only a one-in-ten chance of detecting real effects — and that even experts vastly overestimate the field’s strength. Arel-Bundock is Professor of Political Science at the Université de Montréal. In his new paper, he concludes that methodologists greatly overestimate the statistical power of political science research.
What if most political science studies are too weak to find the effects they’re looking for? In this episode, we dig into a new paper by Vincent Arel-Bundock and colleagues that reveals a striking truth: quantitative political science is greatly underpowered. With thousands of tests analyzed, the authors show that many studies have only a one-in-ten chance of detecting real effects — and that even experts vastly overestimate the field’s strength.
Arel-Bundock is Professor of Political Science at the Université de Montréal. In his new paper, he concludes that methodologists greatly overestimate the statistical power of political science research.