- Understanding the Ecological Fallacy and Simpson’s Paradox can help you identify and reduce potential instances of false conclusions
- Accounting for incorrect inferences in your analysis
- Taking analytics beyond simple correlations
In this week’s a.university session, Piotr Urbanski, Ph.D., Ayzenberg associate director of marketing science, explains how to scan your data for false conclusions produced by a phenomenon called the Simpson’s Paradox—where trends reverse when a dataset is separated into groups. Urbanski shares how to think more critically about data analysis to analysts elevate their causal inference and help leadership better understand the performance of marketing campaigns.