The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » General Data Questions » total population factors for all women
total population factors for all women [message #12357] Fri, 05 May 2017 05:43 Go to next message
Anabb is currently offline  Anabb
Messages: 3
Registered: May 2017
Member
Dear Sir/Madam,

For my master's thesis, I am using a few DHS datasets where only ever-married women have been sampled (i.e. Egypt: DHS 1992 and 2014, Indonesia: DHS 1991, Pakistan: DHS 1990-91 and 2012-13). I would like to know how I can calculate the percent distribution of all women age 15-49 by education level attended based on these datasets (and, if possible, how I can calculate the distribution of all women age 15-49 by education in single years). I think I have to use the all women factor, for total population (AWFACTT) and all women factor, for education (AWFACTU) in order to make these calculations, but I am not sure which steps I should take next. Could you perhaps guide me in the right direction?

Thank you very much in advance,

Kind regards,

Anne

Re: total population factors for all women [message #12362 is a reply to message #12357] Fri, 05 May 2017 08:51 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3013
Registered: February 2013
Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:

You are apparently using the IR files from these surveys. It could be easier to use the PR files. For example, if you go to the PR file for the Egypt 1992-93 survey, EGPR21FL.dta, you have sex (hv104), age in single years (hv105), level of schooling (hv106), and single years of schooling (hv108). This file includes all women, men, and children in the household sample. You can use the weight variable (hv005) to get the distributions you want. The restriction to ever-married women only affects the IR file. You could use that file, with weights v005*awfacte (awfacte, not awfactu), but that would not be as accurate or as easy as using the PR file.
Re: total population factors for all women [message #12369 is a reply to message #12362] Sat, 06 May 2017 06:46 Go to previous messageGo to next message
Anabb is currently offline  Anabb
Messages: 3
Registered: May 2017
Member
Dear Tom Pollum,

Many thanks for your response on my question. I am using the PR files now. I have one remaining question. In the DHS final reports, a section about educational attainment is usually included (e.g. in the final DHS report of Indonesia 2012, paragraph 3.2; http://dhsprogram.com/pubs/pdf/FR275/FR275.pdf). Are the tables in this section based on the de jure or the de facto population?

Thank you in advance,

Kind regards,

Anne Abbing
Re: total population factors for all women [message #12376 is a reply to message #12369] Mon, 08 May 2017 09:07 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3013
Registered: February 2013
Senior Member
Following is a response from DHS Senior Research Associate, Shireen Assaf:

This is a de facto table. I was able to match the numbers in the report with a weighted tabulation of v106 in the IR file. I could not get an exact match using the PR file.

Re: total population factors for all women [message #12395 is a reply to message #12362] Tue, 09 May 2017 09:15 Go to previous message
Anabb is currently offline  Anabb
Messages: 3
Registered: May 2017
Member
Dear Shireen,

Thank you for your answer and for checking the numbers in the report with the IR and PR file. I can use this information in order to do my own calculations.

Kind regards, Anne
Previous Topic: Dropping observations/weights
Next Topic: Liberia - Not possible to replicate education indicators
Goto Forum:
  


Current Time: Tue Mar 19 04:38:13 Coordinated Universal Time 2024