At various points, demographics data are joined alongside task data. Make sure all appropriate columns are matched in the join call and that there are no ".x" and ".y"-suffixed columns!
The implementation may need to be handled separately for Nexus vs legacy Explorer data, bc the "demographic" columns that are bundled in with the raw module-level data aren't consistent between the two versions.
Re: testthat testing for this, I think there already is a test for this, but you may just need to update the test for Nexus data.
At various points, demographics data are joined alongside task data. Make sure all appropriate columns are matched in the join call and that there are no ".x" and ".y"-suffixed columns!
The implementation may need to be handled separately for Nexus vs legacy Explorer data, bc the "demographic" columns that are bundled in with the raw module-level data aren't consistent between the two versions.
Re: testthat testing for this, I think there already is a test for this, but you may just need to update the test for Nexus data.