Abstract
dc:description.abstractThe use of propensity score methods have increased tremendously in the last several years. Propensity scores are used as a tool to correct the effects of confounding that can result in biased estimates of treatment effects. However, there is more than one method that employs propensity scores, and there is no guide in choosing the optimal method, in terms of Type I error and statistical power. This research aims to assess these attributes through a complex simulation study of a five variable model. The results of the simulations indicate performance of a particular propensity score method is dependent
Degree
thesis:*- Level thesis:degree_level
- Dissertation
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Turley, Falynn Cartmill
- Contributors dc:contributor
-
- David T Redden
- Janice L Case
- Derek Dubay
- Charles R Katholi
- Jeff M Szychowski
Subjects
dc:subject × 4Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.library.uab.edu/etd-collection/3181