Thursday, 8 October 2020

Social Science and the Replication Crisis

Keynote by Andrew Gelman, Department of Statistics and Department of Political Science, Columbia University

The replication crisis is typically discussed in the context of particular silly claims, or in terms of the sociology of science, or with regard to controversies in statistical practice. Here we discuss the content of unreplicated or otherwise shaky empirical claims in social science, which often seem to be associated with a model in which important attitudes and behavior can be easily manipulated using irrelevant stimuli. This set of theories, if true, would have important implications for politics and society, supporting certain views held on the left, right, and technocratic center of the political spectrum.  Conversely, the lack of empirical support for the manipulable-human model has social and political implications which are worth considering: If people are not so easily swayed in this way, this suggests that we should try to more carefully understand their direct motivations.