The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Are the respondents for varioud data sets the same group of people?
Are the respondents for varioud data sets the same group of people? [message #22664] Mon, 19 April 2021 07:01 Go to next message
kkkkkk is currently offline  kkkkkk
Messages: 12
Registered: November 2020
Member
I am currently working on the effects of maternal education on child mortality in Cambodia. I am planning to do a robustness check to see if the signs of the marginal effects are consistent (I assume this is the definition/aim of the check). I will merge the two data sets from 2014 and 2010 to increase the sample size and to add a year dummy variable and run the regressions that I originally had, plus the year dummy. However, I am concerned that if the respondents are the same group of people or they responded in both surveys, this will result in doublecounting. Therefore, I am wondering if this information is provided. Thank you!
Re: Are the respondents for varioud data sets the same group of people? [message #22673 is a reply to message #22664] Mon, 19 April 2021 15:05 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2267
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

All DHS surveys are statistically independent of one another, including successive surveys in the same country. You may find the same ID codes but they do not refer to the same individuals, households, or clusters.
Previous Topic: Still births-infant mortality
Next Topic: Confidence intervals for child mortality estimates
Goto Forum:
  


Current Time: Fri May 7 01:06:59 Coordinated Universal Time 2021