Omitted variable bias can be solved by including covariates that are omitted. However, can selection bias also be solved by including covariates?

  • $\begingroup$ By selection bias, do you mean the bias resulting from conditioning on consequences of the treatment and outcome? This term is used differently in different disciplines. $\endgroup$ – Noah Feb 19 '19 at 19:19
  • $\begingroup$ You can control only for measured variables, by definition. In addition, such adjustment is generally inefficient or with incomplete coverage. That is why randomized trials remain superior to observational studies or non-randomized trials.. $\endgroup$ – Joe_74 Feb 20 '19 at 8:21
  • $\begingroup$ @Noah yes I mean that kind of selection bias $\endgroup$ – Joep van der Plas Feb 28 '19 at 11:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.