The Simpson's paradox unraveled

作者: Miguel A Hernán , David Clayton , Niels Keiding

DOI: 10.1093/IJE/DYR041

关键词:

摘要: Background In a famous article, Simpson described hypothetical data example that led to apparently paradoxical results. Methods We make the causal structure of Simpson's explicit. Results show how paradox disappears when statistical analysis is appropriately guided by subject-matter knowledge. also review previous explanations attributed it two distinct phenomena: confounding and non-collapsibility. Conclusion Analytical errors may occur problem stripped its context analyzed merely in terms.

参考文章(30)
P. R. Rosenbaum, D. B. Rubin, Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome Journal of the royal statistical society series b-methodological. ,vol. 45, pp. 212- 218 ,(1983) , 10.1111/J.2517-6161.1983.TB01242.X
Alvan R. Feinstein, Principles of medical statistics Chapman and Hall/CRC. ,(2001) , 10.1201/9781420035681
Clark N. Glymour, Peter Spirtes, Richard Scheines, Causation, prediction, and search ,(1993)
Morris Raphael Cohen, Ernest Nagel, An Introduction to Logic and Scientific Method ,(1934)
SANDER GREENLAND, Interpretation and choice of effect measures in epidemiologic analyses American Journal of Epidemiology. ,vol. 125, pp. 761- 768 ,(1987) , 10.1093/OXFORDJOURNALS.AJE.A114593
Donald R. Hedeker, Robert D. Gibbons, Longitudinal Data Analysis ,(2006)
Colin R. Blyth, On Simpson's Paradox and the Sure-Thing Principle Journal of the American Statistical Association. ,vol. 67, pp. 364- 366 ,(1972) , 10.1080/01621459.1972.10482387
Laurence D. Robinson, Nicholas P. Jewell, Some surprising results about covariate adjustment in logistic regression models International Statistical Review. ,vol. 59, pp. 227- 240 ,(1991) , 10.2307/1403444
Miguel A. Hernán, Sonia Hernández-Díaz, James M. Robins, A structural approach to selection bias. Epidemiology. ,vol. 15, pp. 615- 625 ,(2004) , 10.1097/01.EDE.0000135174.63482.43
I. J. Good, Y. Mittal, The Amalgamation and Geometry of Two-by-Two Contingency Tables Annals of Statistics. ,vol. 15, pp. 694- 711 ,(1987) , 10.1214/AOS/1176350369