Binary contrasts for unordered polytomous regressors
Year: 2020
Type: preprint
Abstract: In observational studies, regression coefficients for categorical regressors are overwhelmingly presented in terms of contrasts with a reference category. For unordered regressors with many categories, however, this approach often focuses on contrasting different pairs of categories with one another with little substantive rationale for foregrounding those comparisons. Mean contrasts, which compare categories to the overall mean, provide an alternative to the reference category, but the magnitude of mean contrasts is conflated with the relative sizes of the categories. Instead, binary contrasts compare a category to all the other categories, allowing the familiar interpretation for dichotomous regressors. We provide a Stata package that computes binary contrasts.
Authors: Jeremy Freese, Sasha Johfre
Institution: Stanford University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4230660471