Exploratory analysis and the expected value of Experimentation

Abstract It is increasingly easy to acquire a large amount of data about problem before formulating hypothesis. The idea exploratory analysis (EDA) predates this situation, but many researchers find themselves appealing EDA as an explanation what they are doing with these new resources. Yet there has been relatively little explicit work on or why it might be important. I canvass several positions in the literature, them wanting, and suggest alternative: analysis, when done well, shows expected value experimentation for particular