The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Geographic Data » Two Way Fixed Effects with Year and Cluster FEs (Using DHS Clusters across Surveys)
Two Way Fixed Effects with Year and Cluster FEs [message #28798] Sat, 09 March 2024 21:51 Go to next message
Joseph is currently offline  Joseph
Messages: 1
Registered: March 2024
Member
Hello Everyone,

The idea is to examine the effect of the distance of the centroids of DHS clusters to nearby facilities on child health. We pull the independent cross-sections together to run two-way fixed effects. However, we just read that DHS clusters are not the same across surveys, which has serious consequences for our analysis. We will be glad if anyone is willing to give any suggestions on how best to tackle this issue?

We are running high-dimensional fixed effects (reghdfe) in STATA.
Re: Two Way Fixed Effects with Year and Cluster FEs [message #30406 is a reply to message #28798] Sun, 24 November 2024 22:47 Go to previous message
BrandynKunde is currently offline  BrandynKunde
Messages: 1
Registered: November 2024
Member
Instead of using DHS clusters, you can switch to more stable geographic units such as districts, provinces, raft wars, or administrative regions. Combine cluster data with these administrative units for consistency.

[Updated on: Sun, 24 November 2024 22:47]

Report message to a moderator

Previous Topic: Rwanda District identifier: 2005 and 2007-08 DHS
Next Topic: How to Access GPS Data
Goto Forum:
  


Current Time: Sun Dec 8 03:20:02 Coordinated Universal Time 2024