So far Google hasn't turned up anything.

I have a dataset that has up to 17% missing values for some variables. I'd like to be able to compare the values for all my variables between the complete cases and the cases with any missing values, to determine if my cases with missing data are different from the complete cases. So for continuous variables, I'd like to use a t-test, and for factor variables I'd like to use chi-square. 44% of my data are complete cases.

I will also perform multiple imputation with 20 imputations, to compare those results to my complete case analysis. But I'd really like to know if my complete cases differ from the cases with missing values on all my variables of interest. I can select just the complete cases, but haven't been able to figure out how to compare them to the cases with missing data.

Thanks!

More Christine Marie Mills's questions See All
Similar questions and discussions