The Use Of Effect Size Estimates To Evaluate Covariate Selection, Group Separation, And Sensitivity To Hidden Bias In Propensity Score Matching.

Description:

Covariate quality has been primarily theory driven in propensity score matching with a general adversity to the interpretation of group prediction. However, effect sizes are well supported in the literature and may help to inform the method. Specifically, I index can be used as a measure of effect size in logistic regression to evaluate group prediction. As such, simulation was used to create 35 conditions of I, initial bias and sample size to examine statistical differences in (a) post-matching bias reduction and (b) treatment effect sensitivity. The results of this study suggest these conditions do not explain statistical differences in percent bias reduction of treatment likelihood after matching. However, I and sample size do explain statistical differences in treatment effect sensitivity. Treatment effect sensitivity was lower when sample sizes and I increased. However, this relationship was mitigated within smaller sample sizes as I increased above I = .50.

Creator(s): Lane, Forrest C.
Creation Date: December 2011
Partner(s):
UNT Libraries
Collection(s):
UNT Theses and Dissertations
Usage:
Total Uses: 275
Past 30 days: 16
Yesterday: 1
Creator (Author):
Publisher Info:
Publisher Name: University of North Texas
Publisher Info: www.unt.edu
Place of Publication: Denton, Texas
Date(s):
  • Creation: December 2011
Description:

Covariate quality has been primarily theory driven in propensity score matching with a general adversity to the interpretation of group prediction. However, effect sizes are well supported in the literature and may help to inform the method. Specifically, I index can be used as a measure of effect size in logistic regression to evaluate group prediction. As such, simulation was used to create 35 conditions of I, initial bias and sample size to examine statistical differences in (a) post-matching bias reduction and (b) treatment effect sensitivity. The results of this study suggest these conditions do not explain statistical differences in percent bias reduction of treatment likelihood after matching. However, I and sample size do explain statistical differences in treatment effect sensitivity. Treatment effect sensitivity was lower when sample sizes and I increased. However, this relationship was mitigated within smaller sample sizes as I increased above I = .50.

Degree:
Level: Doctoral
PublicationType: Disse
Language(s):
Subject(s):
Keyword(s): Simulation | effect sizes | propensity score matching
Contributor(s):
Partner:
UNT Libraries
Collection:
UNT Theses and Dissertations
Identifier:
  • ARK: ark:/67531/metadc103349
Resource Type: Thesis or Dissertation
Format: Text
Rights:
Access: Public
Holder: Lane, Forrest C.
License: Copyright
Statement: Copyright is held by the author, unless otherwise noted. All rights Reserved.