The DHS Program User Forum: Bangladesh » Number of respondents mismatch in dataset and published report

Home » Countries » Bangladesh » Number of respondents mismatch in dataset and published report

Show: Today's Messages :: Show Polls :: Message Navigator

Re: Number of respondents mismatch in dataset and published report [message #9684 is a reply to message #9665]

Fri, 06 May 2016 11:03

Bridgette-DHS
Messages: 3208
Registered: February 2013

Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

Two problems. First, you are not using weights. Virtually every number in DHS tables is weighted. Second, you are not using marital status. The table is restricted to currently married women. It is important to read the table titles carefully. If you run this line, "tab v501 v025 [iweight=v005/1000000]" in that file, the top line (rounded to the nearest integers) will give 4709 and 12149.

Report message to a moderator

[Message index]

		Number of respondents mismatch in dataset and published report By: gkibria1@jhu.edu on Mon, 02 May 2016 02:35
		Re: Number of respondents mismatch in dataset and published report By: Bridgette-DHS on Fri, 06 May 2016 11:03
		Re: Number of respondents mismatch in dataset and published report By: Soleil Von on Mon, 15 May 2017 00:19
		Re: Number of respondents mismatch in dataset and published report By: Bridgette-DHS on Mon, 15 May 2017 11:22

Previous Topic:	Height and Weight
Next Topic:	Mother's education

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Dec 13 04:20:50 Coordinated Universal Time 2024