I have a large dataset looking at how race is recorded. I want to see if there is an association between race being recorded in a specific dataset and not being recorded in another. For example, if someone is recorded as East Asian in the first dataset but their race is missing in the second dataset, I would like to test the association between being East Asian and having missing data in the second dataset. I am using R statistical package and I have very little experience using R so please let me know the best approach.

  • $\begingroup$ My first step would be to see if the two databases are equally likely to report (any) race. Maybe start with a $2\times 2$ table of counts: Rows DB1/DB2, Columns Race Given/Not Given. If they are similar as to diligence of reporting race in general, then it might be worthwhile looking at diligence for various races of interest. // It might be that the survey for one DB has race as a mandatory item and the survey for the other DF has race as an optional item. $\endgroup$
    – BruceET
    Aug 21 at 15:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.