Skip to main content
. Author manuscript; available in PMC: 2018 May 1.
Published in final edited form as: Epidemiology. 2017 May;28(3):387–395. doi: 10.1097/EDE.0000000000000627

Figure 2.

Figure 2

Comparison of weighted and matched sample sizes under different levels of covariate overlap. IPTW shows a weighted sample size identical to the original cohort. Matching weights and matching are similarly affected by exposure prevalence and poor covariate overlap, indicating shifts in the target population.

Abbreviations: U: Unmatched cohort, M: Matched cohort; Mw: matching weight cohort; Ip: Inverse probability of treatment weight cohort; pExpo: Exposure prevalence