The Interpretation of Interaction in Contingency Tables

By A. Simpson. 1951

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Causal Reasoning for Algorithmic Fairness (2018, methodology): “The Berkeley admissions example [3] is often used to explain Simpson’s paradox [38] and highlight the importance of adjusting for covariates.”
Averaging Gone Wrong: Using Time-Aware Analyses to Better Understand Behavior (2016, background): “This is puzzling: when each of the cohorts exhibits a steady increase in their average comment length, how can the overall mean comment length decrease? This anomaly is an instance of the Simpson’s paradox [63], and occurs because we fail to properly condition on different cohorts when computing mea”
Simpson's Paradox in COVID-19 Case Fatality Rates: A Mediation Analysis of Age-Related Causal Effects (2020, background): “It constitutes a textbook example of a statistical phenomenon known as Simpson’s paradox (or reversal), which refers to the observation that aggregating data across subpopulations (here, age groups) may yield opposite trends (and, thus, lead to reversed conclusions) from considering subpopulations s”
Counterfactual reasoning and learning systems: the example of computational advertising (2012, background): “Such an inversion of the conclusion is called Simpson’s paradox (Simpson, 1951).”
Causal inference in statistics: An overview (2009, background): “The phenomenon was discovered by statisticians a century ago (Pearson et al., 1899; Yule, 1903) analyzed by statisticians for half a century (Simpson, 1951; Blyth, 1972; Cox and Wermuth, 2003) lamented by statisticians (Good and Mittal, 1987; Bishop et al., 1975) and wrestled with by statisticians…”
Identifying Causal Effects With Proxy Variables of an Unmeasured Confounder (2016, background): “Simpson’s paradox (Simpson, 1951) is an elegant illustration of the type of bias that may arise in causal inference subject to unmeasured confounding.”
UvA-DARE ( Digital Academic Repository ) Simpson ' s paradox in psychological science : a practical guide (2017, methodology): “A new consequence of Simpson’s paradox: stable cooperation in one-shot prisoner’s dilemma from populations of individualistic learners. J. Exp. Psychol. Gen. 137, 403–421. doi: 10.1037/00963445.137.3.403 Chuang, J. S., Rivoire, O., and Leibler, S. (2009). Simpson’s paradox in a synthetic microbial s”
ActiveClean: Interactive Data Cleaning While Learning Convex Loss Models (2016, background): “Aggregates over mixtures of different populations of data can result in spurious relationships due to the well-known phenomenon called Simpson’s paradox [32].”
Foundations of Linear and Generalized Linear Models (2015, background): “9The name refers to Simpson (1951), but the result had been shown by Yule (1903).”
TSCAN: Pseudo-time reconstruction and evaluation in single-cell RNA-seq analysis (2016, background): “Sometimes, using the population average to study cell type specific behavior can also be misleading due to Simpson’s paradox (7,8).”
ActiveClean: Interactive Data Cleaning For Statistical Modeling (2016, methodology): “It is known that aggregates over mixtures of different populations of data can result in spurious relationships due to the well-known phenomenon called Simpson’s paradox [32].”
Defining cell types and states with single-cell genomics (2015, background): “The misleading effects of Simpson’s Paradox are likely to be pervasive in modern experiments using bulk assays.”
Forecasting: theory and practice (2020, unknown): “This is closely related to the well-known Simpson’s Paradox (Simpson, 1951).”