Propensity score analysis — some critical remarks Our findings suggest that researchers need comprehensive knowledge of model assumptions and knowledge of plausible causal structure. From prior research, sources of selection bias must be understood. Substantive knowledge of plausible causal structure typically includes the theory of change of an intervention program being evaluated, which determines the covariates that should be used in the model predicting propensity scores and in the outcome analysis. Sample reduction after running a propensity score model is a key issue and should always be considered … Our findings suggest that it is of paramount importance to understand the assumptions of propensity score models and attend to potential violations of
Topics:
Lars Pålsson Syll considers the following as important: Statistics & Econometrics
This could be interesting, too:
Lars Pålsson Syll writes The history of econometrics
Lars Pålsson Syll writes What statistics teachers get wrong!
Lars Pålsson Syll writes Statistical uncertainty
Lars Pålsson Syll writes The dangers of using pernicious fictions in statistics
Propensity score analysis — some critical remarks
Our findings suggest that researchers need comprehensive knowledge of model assumptions and knowledge of plausible causal structure. From prior research, sources of selection bias must be understood. Substantive knowledge of plausible causal structure typically includes the theory of change of an intervention program being evaluated, which determines the covariates that should be used in the model predicting propensity scores and in the outcome analysis. Sample reduction after running a propensity score model is a key issue and should always be considered … Our findings suggest that it is of paramount importance to understand the assumptions of propensity score models and attend to potential violations of these assumptions. This requires both methodological and substantive knowledge …
Finally, this study supports a methodological caution made repeatedly by experienced observational researchers: OLS regression is a poor and ill-advised analytic approach in the presence of endogeneity or selection bias.