Re: Regarding the relation between individual recode and children's recode [message #9706 is a reply to message #9651] |
Mon, 09 May 2016 10:39 |
Trevor-DHS
Messages: 805 Registered: January 2013
|
Senior Member |
|
|
Yes, you are basically correct. In the earlier surveys women were only measured if they had a living child born in the past 3 or 5 years (survey specific). Thus in the KR file you will find v438 is set for all mothers, but in the IR file you will not find data in v438 for those women without births in the 3 or 5 years. In the KR file women will be duplicated if they had two or more children in the past 3 or 5 years. If you are analyzing women's height and weight, I recommend using the IR file.
|
|
|