The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » General Data Questions » variable names -
variable names - [message #14476] Tue, 10 April 2018 18:41 Go to next message
annm__ann
Messages: 2
Registered: April 2018
Member
I want to find child weight in the women's file for Zambia.

In the recode manual, that is M19.

In the STATA file I downloaded for 2013-14 there is no variable with that name but there is:

"m19_1" "m19_2" "m19_3" "m19_4" "m19_5" "m19_6"

is this m19?

There also seem to be other variables that I cannot find in the recode.

Any help would be much appreciated. Thank you.
Re: variable names - [message #14479 is a reply to message #14476] Wed, 11 April 2018 07:46 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member
This is a multiple occurring variables, so each occurrence of every such variable must have its own name because statistical packages do not generally support the use of arrays or subscripts. For example, the third occurrence of the variable M19 would be named M19$03 in SPSS, or M19_03 in SAS and STATA. You might want to use the Child file (KR), because in the Child file, you will have just "M19" for each observation, because in this dataset an observation is a child.

For more information on other variables in the files, you will need the Guide to DHS Statistics, and the DHS Recode Manual. The Recode Manual contains a general description of the recode data file, including the rationale for recoding; a description of coding standards and recode variables, and a listing of the standard dictionary, with basic information relating to each variable.
Re: variable names - [message #14510 is a reply to message #14479] Sun, 15 April 2018 12:50 Go to previous messageGo to next message
annm__ann
Messages: 2
Registered: April 2018
Member
Thank you for your response. I am looking into the birth weight of a child as a dependent of the factors of the mother from Zambia. For this reason I am looking at m19 in the women's file.

I see that mmidx has _20 entries. The recode manual says, " It contains up to 20 entries containing information related to all of the sisters and
brothers of the respondents. "

How is this assessed? For every entry under maternal mortality there is potentially 20 recorded responses. Again, I don't see this in the recode manual. What do they mean?

Thanks
Re: variable names - [message #14514 is a reply to message #14510] Mon, 16 April 2018 07:52 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

The variables with names that begin with "mm" refer to maternal mortality and come from the sibling histories. If you look at the questionnaire for this Zambia survey. which is in an appendix to the main report, you will find a series of questions about the respondent's brothers and sisters and their survival. The responses are used to estimate adult and maternal mortality rates given in the body of the report. The underlying ASCII data files are in fixed format, with columns reserved for 20 siblings (defined as having the same mother as the respondent). The mm variables have suffixes or subscripts _01 through _20, completely analogous to the b variables for the respondent's children in her birth history. The index variable mmidx is helpful for some programming purposes. It just takes the value of the suffix, e.g. mmidx_01=1, mmidx_02=2, etc., out to the reported number of siblings, after which it is ".".
Previous Topic: Different calendar
Next Topic: Select Utility tool
Goto Forum:
  


Current Time: Fri Mar 29 09:59:16 Coordinated Universal Time 2024