search

jcoin-maarc / stigma-surveys

Tools for generating a data package from the MAARC Survey Core's Amerispeak Stigma Surveys
0 stars 1 forks source link

Do overlapping caseids across waves represent the same individuals? #8

Open pschumm opened 6 months ago

pschumm commented 6 months ago

There are a lot of instances in which a caseid from one wave appears in the data for another wave. Are these the same individuals, and if so, can the caseids be relied on to establish cross-wave participation in all cases? If so, we should document this, in part so that people make sure to account for possible within-person correlation when analyzing data from multiple waves. If not, we should replace the caseids with another ID that is unique across waves, to avoid any confusion.

mbkranz commented 6 months ago
  1. My understanding is the caseids are generated independently across datasets (eg 1445 in wave1 may not be the same person as 1445 in wave 2).
  2. Amerispeak generated a caseid crosswalk excel file that can be used which maps individual caseids to a Cross-wave case id. That is, one unique id across participants. We can map this in next update of package. Thanks for getting this Phoebe!