| Re: NFHS 1998-1999 Diarrhoea Questions [message #2917 is a reply to message #2751] |
Tue, 09 September 2014 14:01  |
lee7573
Messages: 3
Registered: August 2014
|
Member |
|
|
Hi, thanks for answering my questions quickly. Your comments were especially helpful in analyzing the dataset.
I do have one more question regarding the NFHS 1998-1999 dataset. I am unable to get the prevalence rate of fever to match the NFHS-2 Report on Table 6.17 for Uttar Pradesh. I got an estimate of 27.7%, whereas the report says 27.8%. I was wondering if this could be due to the differences in the weighted denominator used to form the estimate. One interesting note is that the weighted denominator is not matching for me throughout the prevalence rate of ARI, diarrhea and fever for all youths aged months 0-35. I got a weighted denominator of 4050.05255 for ARI and diarrhea, whereas 4045.0725 for fever. I was wondering if you happen to know how to account for the 5 youths that were missing between the two estimates? This is something that my client is interested in.
FYI, this is how I set up my analysis:
The dataset I used is: UPKR42FL.DTA
The variables I used to set up my survey are v022 v001 to set up the strata and cluster. The weight I used is weight = v005/1000000 and defined the kids aged in month as age_month = v008 - b3.
To define fever:
if h22 in (0,8,9) then fever = 0; *no fever or don't know;
else if h22 in (1) then fever = 100; /*had fever within two weeks */
Thanks again for all your assistance!
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
