The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Understanding the variables
Understanding the variables [message #9525] Fri, 08 April 2016 00:44 Go to next message
mef15505@grips.ac.jp is currently offline  mef15505@grips.ac.jp
Messages: 1
Registered: April 2016
Location: Japan
Member
I am using DHS data for Zimbabwe 2010-2011. I have been naming some variables in ZWHR62FL. Now i want to work on file ZWIR62FL. The issue is I would like to merge the two files. I prefer renaming for easy understanding. Is it possible to combine data in household file to household members file and then to women file? I noted that to merge household data to women`s data requires renaming eg HV001 to V001. Does it mean that the I have to rename HV001 and HV002 only or other variables of interest also? Is it correct to rename "source of drinking water" in the women`s data and that in the household file data the same? When the survey is done, is one questionnaire used to prepare the various data files, for instance, questions asked a household member (female), is it the same member who is also asked for the women`s data.

Am looking at the impact of women empowerment on reproductive health. With education being the proxy for women empowerment, which variables from the DHS maybe the best to use?

Am a new learner and Stata is also new, so please bear with me.
Re: Understanding the variables [message #9549 is a reply to message #9525] Tue, 12 April 2016 13:12 Go to previous message
user-rhs is currently offline  user-rhs
Messages: 132
Registered: December 2013
Senior Member
I would only change the names of the variables that you are going to merge on, i.e. the unique identifiers. Otherwise, if two variables are the same name in both datasets, Stata will keep only the master dataset's values, and that may not necessarily be what you want:

Quote:
By default, merge treats values from the master as inviolable. When observations match, it is the
master's values of the overlapping variables that are recorded in the merged result.

(From the Stata Data Management Reference Manual, section "Treatment of overlapping variables")


Since you are a beginner Stata learner, I encourage you to visit IDRE at UCLA's Stata resources page and also by looking at the help files of the commands you want to use by typing "help [command]" into the command window, for example help merge .

rhs
Previous Topic: Unable to reproduce results of age at first marriage table
Next Topic: Within country narrowest location identifier
Goto Forum:
  


Current Time: Fri Apr 19 14:42:55 Coordinated Universal Time 2024