ANCOVA versus change from baseline had more power in randomized studies and more bias in nonrandomized studies

Missing data: Our view of the state of the art.

Joseph L. Schafer, J. A. Graham · Psychological Methods · 2002 · 10.9K citations

Public Health, Biostatistics, Data Science +17

Estimating causal effects of treatments in randomized and nonrandomized studies.

Donald B. Rubin · Journal of Educational Psychology · 1974 · 9.5K citations · Full text

Public Health, Behavioral Sciences, Psychology +13

Applied Regression Analysis and Other Multivariable Methods

Esteban Walker, David G. Kleinbaum, Lawrence L. Kupper et al. · Technometrics · 1989 · 8.4K citations

Natural Sciences, Biostatistics, Causal Inference +16

Designing Experiments and Analysing Data: A Model Comparison Perspective.

G. H. Freeman, Scott E. Maxwell, Harold D. Delaney · Biometrics · 1991 · 1.2K citations

Public Health, Biostatistics, Experimental Design +12

Assignment to Treatment Group on the Basis of a Covariate

Donald B. Rubin · Journal of Educational Statistics · 1977 · 1.2K citations

Public Health, Biostatistics, Causal Inference +15