I'm attempting to use the PUMS data to compute earnings percentiles for each undergraduate degree, for various subsets of the population.
For instance, "what are the 10th, 25th, 50th, 75th and 90th percentile incomes for non-Hispanic white men aged 35-40 with a physics degree?"
Is there a straightforward way to compute percentile values over the weighted person records?
It feels very wrong, but the first thing that comes to mind is to create a new dataset with each record duplicated N times, where N = the record's PWGTP value.
IIRC this was the issue that first made me turn to R years ago (never turned back) but I believe this is the package I used rdrr.io/.../weighted_ntile.html