The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Countries » Kenya » Recreating published 2014 DHS data - long vs short survey version
Recreating published 2014 DHS data - long vs short survey version [message #13056] Tue, 12 September 2017 17:00 Go to next message
LMW1 is currently offline  LMW1
Messages: 1
Registered: September 2017
Member
We are having trouble re-creating the 2014 published Kenya mCPR data for married women. Our current approach is to look at those who received the short-form version based on variable HV027 while applying HV028 as a weight, using V502 code 1 to account for only married women. Those values are still slightly off the married mCPR published in Kenya's DHS report.

Has anyone else had success with a different approach when looking at the short vs long form survey?
Re: Recreating published 2014 DHS data - long vs short survey version [message #13097 is a reply to message #13056] Tue, 19 September 2017 14:05 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1156
Registered: February 2013
Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:

Hi Lauren--I suspect that you are using v501=1 to determine whether a woman is currently married. This table (and contraceptive prevalence in general) is based on v501=1 or 2, i. v502=1. Living with partner is included with currently married. Please try that change.

When trying to match numbers in a table, start with the totals row at the bottom of the table. First try to match the n (in this case 18,549) and then the overall percentage (in this case 53.2%).

Please let me know if you still do not get a match.
Previous Topic: Creating panel
Next Topic: KDHS 2003-04
Goto Forum:
  


Current Time: Fri Nov 17 21:26:54 Eastern Standard Time 2017