Improving interpretability: gamma as an alternative to R(2) as a measure of effect size.
A traditional measure of effect size associated with tests for difference between two groups is the variance explained by group membership (R(2)). If exposure to a disease causes a small but long term deficit in performance, however, R(2) does not capture that cumulating effect. We propose an altern...
Main Authors: | , , , |
---|---|
Format: | Journal article |
Language: | English |
Published: |
2000
|
_version_ | 1797050956391645184 |
---|---|
author | Reidpath, D Diamond, MR Hartel, G Glasziou, P |
author_facet | Reidpath, D Diamond, MR Hartel, G Glasziou, P |
author_sort | Reidpath, D |
collection | OXFORD |
description | A traditional measure of effect size associated with tests for difference between two groups is the variance explained by group membership (R(2)). If exposure to a disease causes a small but long term deficit in performance, however, R(2) does not capture that cumulating effect. We propose an alternative statistic, gamma, based on the probability of an unexposed person outperforming an exposed person. Although gamma is also a point estimate, it more easily conveys what the cumulating effect of a deficit would be. We discuss some of the advantages of this measure. |
first_indexed | 2024-03-06T18:12:54Z |
format | Journal article |
id | oxford-uuid:03a0076a-c770-4055-b27d-6dca6fb6a703 |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-06T18:12:54Z |
publishDate | 2000 |
record_format | dspace |
spelling | oxford-uuid:03a0076a-c770-4055-b27d-6dca6fb6a7032022-03-26T08:47:18ZImproving interpretability: gamma as an alternative to R(2) as a measure of effect size.Journal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:03a0076a-c770-4055-b27d-6dca6fb6a703EnglishSymplectic Elements at Oxford2000Reidpath, DDiamond, MRHartel, GGlasziou, PA traditional measure of effect size associated with tests for difference between two groups is the variance explained by group membership (R(2)). If exposure to a disease causes a small but long term deficit in performance, however, R(2) does not capture that cumulating effect. We propose an alternative statistic, gamma, based on the probability of an unexposed person outperforming an exposed person. Although gamma is also a point estimate, it more easily conveys what the cumulating effect of a deficit would be. We discuss some of the advantages of this measure. |
spellingShingle | Reidpath, D Diamond, MR Hartel, G Glasziou, P Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title | Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title_full | Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title_fullStr | Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title_full_unstemmed | Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title_short | Improving interpretability: gamma as an alternative to R(2) as a measure of effect size. |
title_sort | improving interpretability gamma as an alternative to r 2 as a measure of effect size |
work_keys_str_mv | AT reidpathd improvinginterpretabilitygammaasanalternativetor2asameasureofeffectsize AT diamondmr improvinginterpretabilitygammaasanalternativetor2asameasureofeffectsize AT hartelg improvinginterpretabilitygammaasanalternativetor2asameasureofeffectsize AT glaszioup improvinginterpretabilitygammaasanalternativetor2asameasureofeffectsize |