The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Bangladesh » Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence
Re: Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence [message #28325 is a reply to message #28320] Thu, 14 December 2023 07:37 Go to previous messageGo to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member

Following is a response from Senior DHS staff member, Tom Pullum:

Most of the biometric variables, including this one (sh284a), have two aspects of their coding that you must take into account. The first is that there are codes for "refused", etc., that must be bypassed in any analysis. The second is that the measurements have been multiplied by some power of 10 to get rid of a decimal point.

If you enter "describe sh284a" you will find that the label is SH284A. Then enter "label list SH284A" to identify the special codes. Here they are:

. label list SH284A
SH284A:
994 not present
995 refused
996 other

Then enter either "histogram sh284a if sh284a<994" or "summarize sh284a if sh284a<994" to find out what the multiplier is. The latter command shows that the (unweighted) mean is 55 and the range is 16 to 271. The cutoff values in table 15.5 are 3.9, 6.0, etc. Clearly sh284a is 10 times the scale of the glucose indicator in the table. Therefore you need to construct a new variable as follows: "gen glucose=sh284a/10 if sh284a<994". Please try again to construct the tables using "glucose".



 
Read Message
Read Message
Read Message
Previous Topic: Determining child's birth order
Next Topic: Merging DHS household Survey data with SPA data
Goto Forum:
  


Current Time: Sun Dec 1 20:12:41 Coordinated Universal Time 2024