Matching PSUs across samples

Hi; I’m using some IDHS data across two DHS surveys. We believe there may be significant non-response in some regions of the (one) country we’re looking at in the second survey relative to the first. If this were the case (which we’d like to investigate) we would like to restrict our analysis to those areas without non-response problems.

For that to happen, we’d have to have some geographic identifiers lower than the state level (which is in the current geographic shapefiles), preferably at the PSU. I’m certain that’s not possible, but thought I’d ask.


You are correct that identifying observations at geographic levels below the state or region level in IPUMS DHS is not possible. For this level of detail, you’ll need to use the un-harmonized data, available directly from the DHS Program. The DHS Program allows users to access GPS coordinates for each of the households (within a cluster) in the data.