Decision-making models, rigor and new puzzles

Comparative tests of predictive accuracy across models are exceptionally rare in political science. The collection of articles in this issue provides a rigorous, systematic evaluation of alternative models for explaining and predicting decision-making within the European Union. I examine how alternative models were evaluated and raise questions about the differences in the extent to which the operational definitions of variables match contending theoretical approaches. I also raise questions regarding the difference between models with regard to whether the issues are part of a repeated game or a single-shot game. Finally, I propose future tests to hone in on some of the puzzles raised by the findings reported here.