| Couple file from male and female dataset [message #13106] |
Thu, 21 September 2017 02:22  |
dnbhatta
Messages: 9
Registered: September 2017
Location: Nepal
|
Member |
|
|
HI, I am trying to make couple file from male and female dataset. I created new file using V001+V002+V034 (with female dataset) and MV001+MV002+MV034$1 (with male dataset), I renamed the variables name in male dataset similar to female dataset. How we can match the husband and wife where V034 and MV034 values are too different (such as in female dataset mostly have value 1 but in male dataset mostly have value 2)
What is the command to merge only to keep couple dataset, STATA command is okey, if R command is not available. Thank you.
DNB
[Updated on: Thu, 21 September 2017 13:38] Report message to a moderator |
|
|
|
|
|
| Re: Couple file from male and female dataset [message #13109 is a reply to message #13107] |
Thu, 21 September 2017 14:02  |
dnbhatta
Messages: 9
Registered: September 2017
Location: Nepal
|
Member |
|
|
Hi Tom,
Thank you for your reply. I am using IR and MR files. I would like to combine both files to make couple file for further analysis. Shall I do this or not to make couple file. If yes, how could I combine and If not, What should I do.
I could not find the CR file for Pakistan (2012-13) dataset. For this, how could we make CR file.
Thank you,
DNB
[Updated on: Thu, 21 September 2017 16:28] Report message to a moderator |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Re: Couple file from male and female dataset [message #24206 is a reply to message #24201] |
Tue, 15 March 2022 06:28 |
Bridgette-DHS
Messages: 3189
Registered: February 2013
|
Senior Member |
|
|
Following is response from DHS Research & Data Analysis Director, Tom Pullum:
It is not true that all surveys have a couples file. Some surveys do not include men. The surveys you mentioned, in fact, did not include men and therefore do not have MR or CR files.
If a woman is asked a question about her husband, for example about his age or occupation, and the man is also asked the same thing about himself, then the man's self-response would have priority. However, the two responses can be compared. Similarly, the woman's responses in the women's survey about her age, education, etc. are given priority over the corresponding responses from the household respondent during the household survey. Self-responses are always preferable but you can learn something about data quality (and communication within the household) by comparing self-responses with what someone else said.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
