Ariel Linden
Linden Consulting Group, LLC
In contrast to randomized studies in which individuals have no control over their treatment assignment, participants in observational studies self-select into the treatment arm and are therefore likely to differ in their characteristics compared to those who elect not to participate. Analytic approaches using the propensity score to adjust for differences between study groups are thus popular among investigators of observational data. In this paper, I describe how the new Stata package for implementing CTA can be used to generate propensity score weights.
odajournal.com 