Have a personal or library account? Click to login
Corrigendum: Relative Measures of Association for Binary Outcomes: Challenges and Recommendations for the Global Health Researcher Cover

Corrigendum: Relative Measures of Association for Binary Outcomes: Challenges and Recommendations for the Global Health Researcher

Open Access
|Jul 2020

Full Article

Corrigendum

Shortly after pubication of Gallis and Turner [1] the authors informed us that they accidentally omitted an important option in the Stata code of Table 4.

In the clustered setting, the generalized estimating equations (GEE) Stata code must include the option “vce(robust)” in order to obtain robust standard errors. In the modified log-Poisson model, this option is essential in order to obtain valid statistical inference. The “vce(robust)” option may also be added for the log-binomial model when clustering is present, and should be added in order to match the output from the R, SAS, and SPSS code, which report the robust standard errors.

The updated modified log-Poisson code in the clustered setting is:

xtgee binaryoutcome exposure, family(poisson) link(log) corr(exchangeable) vce(robust) eform

This omission does not affect any data published in the article.

Competing Interests

The authors have no competing interests to declare.

DOI: https://doi.org/10.5334/aogh.2998 | Journal eISSN: 2214-9996
Language: English
Published on: Jul 6, 2020
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2020 John A. Gallis, Elizabeth L. Turner, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.