| Surveying same HH in BD DHS [message #29683] |
Sun, 21 July 2024 23:14  |
Thasin
Messages: 3
Registered: April 2024
|
Member |
|
|
|
DHS collects information for all children under 5 years. However, Bangladesh DHS waves are conducted every three years. Since clusters are not same in every wave, is it possible to resurvey same HH? If I want to work with all waves, should I only consider children under 3 years of age to rule out the possibility of over-estimation? Please suggest! Thanks!
|
|
|
|
| Re: Surveying same HH in BD DHS [message #29689 is a reply to message #29683] |
Mon, 22 July 2024 11:02  |
Bridgette-DHS
Messages: 3201
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
DHS surveys, including closely-spaced surveys in Bangladesh, are completely independent of one another. Two successive surveys may use the same sampling frame from the most recent census, but it is very unlikely that they will repeat a cluster (enumeration area) and even less likely that they will repeat a household.
All the clusters and households are renumbered (that is, the numbers on the sampling frame and cluster listing are changed) so there is no way to check whether there has been a repeat, but it's very very unlikely, even though theoretically it could happen.
A good thing about these closely spaced surveys, that is hardly ever utilized (to my knowledge), is that you have repeated observations of the same national-level birth cohort of children. If two surveys are 3 years apart, then you can identify a calendar year birth cohort for which you have observations (1) when the cohort was age 0-1 and then (2) when it was age 3-4. It would be possible to make some aggregate-level comparisons in this way, but not individual-level comparisons.
|
|
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
