“We had stuff like, here's the result, and it's statistically significant. And I would say our book was quite moderate in how we used it. We always emphasize that statistical significance doesn't mean it's true.”
Why: Emphasizes that statistical significance does not mean a result is true, and models integrating theory with data more openly.
Data Analysis Using Regression and Multilevel/Hierarchical Models by Andrew Gelman
Guest on EconTalkAndrew Gelman on Social Science, Small Samples, and the Garden of the Forking Paths