What is it about?
Recommendations on what relative measure of association (e.g., risk ratio or odds ratio) to use and when to use them. This article contains a review of the methods, along with code for fitting various statistical models to obtain risk ratios in four software packages: R, SAS, Stata, and SPSS.
Photo by Jonathan Petersson on Unsplash
Why is it important?
There are still many areas of research (including global health) where logistic regression is used by default, even when reporting odds ratios may be not be most appropriate. Some of this is because of convention, or because the logistic regression approach is what many people initially learn and default to using. Our paper provides rationale for when and why the risk ratio may be preferred, and we provide code to facilitate the use of these methods by researchers.
Read the Original
This page is a summary of: Relative Measures of Association for Binary Outcomes: Challenges and Recommendations for the Global Health Researcher, Annals of Global Health, January 2019, Ubiquity Press, Ltd., DOI: 10.5334/aogh.2581.
You can read the full text:
The following have contributed to this page