The DHS Program User Forum: India » Mapping Variables in Questionnaire to DHS V Dataset

Home » Countries » India » Mapping Variables in Questionnaire to DHS V Dataset (Mapping Variables in Questionnaire to DHS V Dataset)

Show: Today's Messages :: Show Polls :: Message Navigator

Mapping Variables in Questionnaire to DHS V Dataset [message #17636]

Wed, 01 May 2019 14:01

saveeg93
Messages: 2
Registered: May 2019

Member

Hi,

I am trying to map the variables in the DHS individual recode dataset to the women's questionnaire (NFHS 2005-06/DHS V for India). The recode manual which reflects the dataset variables (attached) and questionnaire (attached) have a different numbering system. Is there a codebook which provides a precise mapping? Some of the descriptions in the recode manual are not thorough enough to identify how they are developed from the questionnaire.

Specifically, I am wondering if someone can confirm the following:

1) v714, currently working, indicates if the respondent says yes to either 806, 807, or 808
2) v731, have worked in the past year, indicates if the respondent says yes to either 806, 807, 808, or 809
3) v741 comes directly from 816
4) v721 comes directly from 814

Thank you!

Attachment: DHS Recode Manual - DHS 5.pdf
(Size: 353.62KB, Downloaded 724 times)
Attachment: IND_DHS_2005_Questionnaire_Woman.pdf
(Size: 340.93KB, Downloaded 511 times)

Report message to a moderator

Re: Mapping Variables in Questionnaire to DHS V Dataset [message #17711 is a reply to message #17636]

Thu, 09 May 2019 07:56

Bridgette-DHS
Messages: 3199
Registered: February 2013

Senior Member

Following is a response from Senior Data Processing Specialist, Mianmian Yu:

Thank you for the questions.

(1) V714 was constructed using the following logic in CSPro

if Q806 = 1 | Q807 = 1 then
      V714 = 1
    else
      V714 = YesNo( Q808 );
endif;

As you interpreted correctly, it is "yes" when Q806, Q807 or Q808 is "yes".

(2) V714 was constructed using the following logic in CSPro

 if Q808 = 1 then
      V731 = 3               { have a job but didn't work last 7 days }
   elseif V714 = 1 then
      V731 = 2              { currently working }
   else
      V731 = YesNo( Q809 ); { whether worked last 12 months }
 endif;

This logic is different from what you guessed earlier.

(3) You are correct that V741 is from Q816 and V721 comes directly from Q814

There is no codebook that maps a question to a variable. If you have questions about additional variables, we will be happy to check them out for you.

[Updated on: Thu, 09 May 2019 07:57]

Report message to a moderator

Re: Mapping Variables in Questionnaire to DHS V Dataset [message #17844 is a reply to message #17636]

Mon, 24 June 2019 12:16

boyle014
Messages: 78
Registered: December 2015
Location: Minneapolis

Senior Member

You can find the specific survey text related to each variable at IPUMS DHS.

First, select that sample(s) that interest you.

Then do a search of the variable name, such as v741. You'll see the IPUMS DHS harmonized variable (it's called wkearntype). Click on the variable name, and you'll see a row of tabs.

SURVEY TEXT will show you precisely that. If you click on the Survey Form line on this page, you can jump into the complete survey right where the question was asked.

UNIVERSE will describe the individuals who were asked the question, which of course takes into account the survey skip pattern. For example, for v741/wkearntype, IPUMS DHS says the question was asked of: Women age 15-49 who are currently working or have worked in the last 12 months.

Professor Elizabeth Boyle
Sociology & Law, University of Minnesota, USA
Principal Investigator, IPUMS-DHS

Report message to a moderator

Previous Topic:	Mismatch in recode vs. stata file for DHS VII
Next Topic:	Village related questions (NFHS-3, NFHS-4)

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Tue Nov 26 23:26:46 Coordinated Universal Time 2024