The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » India » Comparing cluster-level outcomes between NFHS-4 and NFHS-5 (How to match DHS clusters across different survey years)
Comparing cluster-level outcomes between NFHS-4 and NFHS-5 [message #26918] Thu, 25 May 2023 06:06 Go to next message
cgt9
Messages: 1
Registered: May 2023
Member
Hello,

I am trying to leverage the data collection pause due to COVID-19 in the NFHS-5 to look at the impact of COVID-19 on various outcomes. I have a set of clusters that were surveyed before COVID-19 lockdowns and a set that were interviewed after lockdowns. I am specifically looking at Uttar Pradesh and Madhya Pradesh since these states have interviews in both time periods.

The respondents interviewed after lockdowns appear to be in lower wealth quintiles. This could be because of COVID-19 economic stress, or it could be because the post-lockdown respondents live in poorer areas. To investigate this further, I would like to compare NFHS-4 wealth outcomes between the pre- and post-lockdown clusters. If wealth outcomes are similar in the NFHS-4 data, I may be more confident that economic differences in NFHS-5 are due to COVID.

However, I am struggling to identify the clusters in my NFHS-5 data in the NFHS-4 data because the cluster IDs appear to have changed between the two surveys. For example, one of the Uttar Pradesh clusters I am looking at in NFHS-5 is cluster ID 13203. However, in NFHS-4 that cluster ID does not exist. Because of the GPS displacement process, I cannot match on latitude and longitude. Is there another way I can match the cluster areas?

Thanks in advance.
Re: Comparing cluster-level outcomes between NFHS-4 and NFHS-5 [message #26920 is a reply to message #26918] Thu, 25 May 2023 10:11 Go to previous message
fred.arnold@icf.com is currently offline  fred.arnold@icf.com
Messages: 83
Registered: May 2021
Senior Member
The NFHS-4 and NFHS-5 samples are separate independent nationally representative samples. They are not related to one another in any way. The cluster numbers are also independent. Therefore, there is no way to match the clusters between NFHS-4 and NFHS-5.
Previous Topic: Tetanus injections before birth description
Next Topic: Implausible waist and hip circumference values in NFHS-5 data
Goto Forum:
  


Current Time: Thu Oct 3 09:16:51 Coordinated Universal Time 2024