Paula, IVEware will select the highest coded group as the reference and automatically create dummy variables for the remaining groups. It will be used in regressions for imputing other variables if you keep it in the analysis. I usually code the highest populated category with the highest value to use as the reference. The caveat here is that you specify the default value in the analysis to be categorical, then define limits for all continuous variables.
Craig Craig D. Newgard, MD, MPH Assistant Professor Department of Emergency Medicine Department of Public Health & Preventive Medicine Oregon Health & Science University 3181 Sam Jackson Park Road Mail Code CR-114 Portland, OR 97239-3098 (503) 494-1668 (Office) (503) 494-4640 (Fax) [email protected] >>> Paula Andreea Tufis <[email protected]> 02/17/05 3:45 PM >>> Hi, I have recently begun working with IVEware and I have a question about how this software handles categorical variables when it uses them as predictors. For example, I have a list of 15 variables for which I want to impute missing values, among which a variable "race" with values 1 Black non-Hispanic, 2 Hispanic, 3 Asian, 4 Other and 5 White non-Hispanic. I want to have the missing values for this 5 category variable imputed, but I would also like to use the race variable as a predictor in the regressions used to impute missing values for the other 14 variables in my data set. If I leave the race variable in its 5 category format, will the software automatically dummy it out and use the dummies as predictors, or do I need to construct the dummies myself and instruct the software to use these dummies instead of the 5 category race variable as predictors when imputing missing values for the other 14 variables? I would appreciate your help with this, Thanks, Paula. -- ~~~~~~~~~~~~~~~~~~~~~~~~~ Paula Andreea Tufis Doctoral Student 211 Oswald Tower Department of Sociology Penn State University ~~~~~~~~~~~~~~~~~~~~~~~~~ _______________________________________________ Impute mailing list [email protected] http://lists.utsouthwestern.edu/mailman/listinfo/impute
