Missing values [message #18208] |
Tue, 15 October 2019 05:17 ![Go to next message Go to next message](/theme/default/images/down.png) |
dasanchezp
Messages: 11 Registered: February 2017 Location: San Sebastian
|
Member |
|
|
Dear DHS,
I am working with file KEIR72FL and I see calendar data (vcal_1 and vcal_2) was only collected for 47% of women. Why is that? This means not every birth is recorded on vcal_1 so I cannot compute TFR from it?
Also, 52% of births in the previous 36 months to the date of interview do not include information on M10 nor V367. If I want to compute the proportion of unwanted pregnancies, should I reweight the sample? Please, help me out!
Thanks in advance.
Best regards,
David A. Sanchez P.
[Updated on: Tue, 15 October 2019 08:28] Report message to a moderator
|
|
|
Re: Missing values [message #18320 is a reply to message #18208] |
Fri, 08 November 2019 12:53 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Bridgette-DHS
Messages: 3151 Registered: February 2013
|
Senior Member |
|
|
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
This survey had both "long form" and "short form" samples. Half of the households were assigned to each. Only the "long form" subsample included the calendar. Households with the long form included interviews with men, and for the women, included the calendar, some additional maternal health questions, and the domestic violence module. When you are working with the IR file, the easiest way to determine that the woman was given the short form is that vcal_1 and vcal_2 are blank and d005 is 0.
Fertility rates are calculated from birth histories, not from the calendar. Birth histories were collected for all women with v201>0 (that is, women who had at least one child.
|
|
|