Hello Melissa,

Your goal in model-building with your data set wasn't fully clear to me in your query, so please excuse the If-then approach below.

If your goal is to construct the model with fewest possible IVs to maximally account for differences in status on the outcome variable, there are far better ways to go about this than to apply the "10%" standard to each possible IV. The data mining community has a bunch of methods from which to choose.

If you wish to be able to make statements about how much or how little a given confounder/control variable adds to your ability to predict outcome variable status, then you should be comparing models with all of the relevant dummy variates (for that control variable) to models with none of the relevant dummy variates.

Finally, please note that applying a standard like the "10%" criterion is virtually a guarantee that you'll miss each and every instance of suppressor variables, should they exist.

Good luck with your work.

In binary logistic regression, the outcome is usually coded 1/0; then I am assuming you have an exposure of interest, and all other variables are potential confounders coded as indicator variables. Assuming a large sample size relative to the number of potential confounders, you can run a model with the exposure of interest and all confounders. The adjusted OR for the exposure-disease association (controlling for all confounders), is the most valid estimate of the exposure-disease association (assuming accurate measurements, etc.). You could stop there. Depending on the research question, you could consider simplifying the model by removing some of the variables which appear to be weak confounders. Remove one of these variables, then compared the adj OR for the exposure-disease association for this reduced model with the adj OR controlling for all variables (which is considered the "gold standard"). If the reduced model OR is similar to the gold std OR (within 10%), leave that variable out. Repeat this simplification approach always comparing to the "gold std" OR; if dropping one variable has an adj OR more than 10% away from the gold standard, place that variable back into the model and assess another variable. While not necessarily a perfect approach, this change-in-indicator approach is reasonable.