Split Samples and Design Sensitivity in Observational Studies
- 1 September 2009
- journal article
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 104 (487), 1090-1101
- https://doi.org/10.1198/jasa.2009.tm08338
Abstract
An observational or nonrandomized study of treatment effects may be biased by failure to control for some relevant covariate that was not measured. The design of an observational study is known to ...Keywords
This publication has 22 references indexed in Scilit:
- War and WagesJournal of the American Statistical Association, 2008
- Sensitivity Analysis for m‐Estimates, Tests, and Confidence Intervals in Matched Observational StudiesBiometrics, 2007
- Preoperative Antibiotics and Mortality in the ElderlyAnnals of Surgery, 2005
- Heterogeneity and CausalityThe American Statistician, 2005
- Does a dose-response relationship reduce sensitivity to hidden bias?Biostatistics, 2003
- Bounding a Matching Estimator: The Case of a Norwegian Training ProgramOxford Bulletin of Economics and Statistics, 2001
- Choice as an Alternative to Control in Observational StudiesStatistical Science, 1999
- Using Omitted Variable Bias to Assess Uncertainty in the Estimation of an AIDS Education Treatment EffectJournal of Educational and Behavioral Statistics, 1997
- Detecting bias with confidence in observational studiesBiometrika, 1992
- A Multiplist Strategy for Strengthening Nonequivalent Control Group DesignsEvaluation Review, 1987