𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Improving interpretability: γ as an alternative to R2 as a measure of effect size

✍ Scribed by Daniel D. Reidpath; Mark R. Diamond; Gunter Hartel; Paul Glasziou


Publisher
John Wiley and Sons
Year
2000
Tongue
English
Weight
70 KB
Volume
19
Category
Article
ISSN
0277-6715

No coin nor oath required. For personal study only.

✦ Synopsis


A traditional measure of e ect size associated with tests for di erence between two groups is the variance explained by group membership (R 2 ). If exposure to a disease causes a small but long term deÿcit in performance, however, R 2 does not capture that cumulating e ect. We propose an alternative statistic, , based on the probability of an unexposed person outperforming an exposed person. Although is also a point estimate, it more easily conveys what the cumulating e ect of a deÿcit would be. We discuss some of the advantages of this measure.