All Else Equal in Theory and Data ( Big or Small ) ∗

The forms of explanation that dominate political science research in the formal theory and causal inference traditions are closely connected. Specifically, each makes essential use of different, but related, kinds of all-else-equal claims. The emergence of “big data” has already begun to alter the landscape of empirical social science by making new sources of information (and, thus, new phenomena) amenable to quantitative analysis. But neither the centrality of all-else-equal explanations, nor the challenges associated with providing them, will be altered in the least by big data.