|P.Mean: Should I report the univariate or the multivariate logistic regression analysis? (created 2011-09-07).
News: Sign up for "The Monthly Mean," the newsletter that dares to call itself average, www.pmean.com/news.
Dear Professor Mean, I have results from four univariate logistic regression models and one multivariate logistic regression model with all four variables. In my univariate analysis, all the variables are significant. But, in multivariate analysis, only x1 is significant. Which results should I report?
There are several different approaches, but here is one I use frequently. The univariate analyses are labelled as "unadjusted" because they represent the effect without controlling for the other variables in the model. The multivariate model provides four "adjusted" effects because it represents the impact of a given variable after holding all the other variables constant. Your table of odds ratios and confidence intervals would look something like
odds ratio odds ratio
x1 2.4 (1.6,3.6) 2.3 (1.5,3.5)
x2 2.8 (2.0,4.2) 1.8 (0.9,2.1)
x3 3.3 (2.5,6.7) 1.5 (0.9,2.1)
x4 2.6 (1.2,4.0) 1.4 (0.7,2.2)
Notice in this example, that the effect of adjusting is to attenuate all of the odds ratio except for x1.
As a general guide on how to write up statistical analyses, look for a wonderful book: How to Report Statistics in Medicine: Annotated Guidelines for Authors, Editors, and Reviewers by Thomas A. Lang and Michelle Secic.
This page was written by Steve Simon and is licensed under the Creative Commons Attribution 3.0 United States License. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Logistic Regression.