I realize the CPS doesn’t survey college residence halls (I tell my students that dorms are like prisons in the CPS). Thus, I assume the college student data in the CPS is for commuter students. But is there another Census product that does have college student data? I assume the 2020 Census does, but are there other products?
The CPS might not actually treat dorms like prisons. The CPS surveys both family and group quarters and identifies the type of quarters under the variable GQTYPE. The description there notes that “institutions such as college dorms or military barracks are examples of group quarters without families.” If you want to use CPS data you’ll want to be able to additionally distinguish between college dorms and other group quarters without families. SCHLCOLL indicates whether the respondent was enrolled in school or college in the previous week and whether they were enrolled full-time or part-time. Using this in combination with GQTYPE should let you restrict your analysis to college students living in college when surveyed by the CPS.
The ACS, accessed through IPUMS USA, also has data on college students. That survey classifies college dorms into non-institutional group quarters. Student data can be parsed from respondents in similar quarters using GRADEATT. An additional note is that data from the 2020 census is not yet available through IPUMS and we are likely years out from a public release.
Hope that helps!
Not sure that this works or why so few college students in dorms are in the CPS. For the year 2020, it gives zero people in the categories of “college or university” (full or part-time) X “group quarters without family.” Should other QGTYPE be considered? I really think they don’t survey dorms, but I don’t recall why I think or where I read that. Thanks.
Thank you for bringing this to our attention. After reviewing the documentation a bit closer, I realize you’re correct that the CPS does not survey college students living in dorms. While the CPS technical paper 77 states that group quarters “include college dormitories, retirement homes, and communes”, a few lines down it makes it clear that “as a cost-savings measure, student dormitories are not sampled, since the vast majority of students in dormitories either have a usual residence elsewhere, are not in the labor force, or both.” My apologies for providing incorrect information in my initial response.
My recommendation for using the ACS, on IPUMS USA, to find microdata on college students living in college dorms still stands. College/University student housing is one of the categories of group quarters that are explored in the ACS. According to the technical documentation, college dorms are not surveyed during summer months (May-August). The 2020 ACS gives 56,560 individuals who are either undergraduate, graduate, or professional students (GRADEATT = 6-7) residing in non-institutional group quarters (GQTYPE = 5).