NOT collapsing race/ethnic categories to preserve power

I convinced my team to preserve race/ethnic categories with small sample sizes in our analyses. We are proposing this analysis plan in a grant proposal.
Can anyone help me identify a peer-reviewed article that I can use to support this decision in our proposal?

1 Like

Hi!

I’m on my phone trying to get my toddler up so don’t have access to my PDF library but one article (albeit an editorial) I find useful is https://jamanetwork.com/journals/jama/articlepdf/2783090/jama_flanagin_2021_ed_210061_1628523987.58592.pdf

Let me know if you’re looking for more.
Chris

This is something we recently had to weigh while curating a larger data set for public release. Because there was some potentially sensitive/stigma related data a decision was made to remove or aggregate smaller sample size categories (state, race, ethnicity, gender id, etc.). This was what we felt was safest - especially given the granularity of the data collected and some of the details reported. As a data curator/publisher I’m very interested to learn more about this topic and how to do this in a way that both honors privacy, consent, and the right to data about oneself.

Peers of mine have written some really good pieces that relate to this that may be of interest:

1 Like

Awesome resources and didn’t see the reply article! Have a great evening!