Support #508

BHPS identifying households or individuals across waves

Added by Christopher Booth about 8 years ago. Updated almost 8 years ago.

Data analysis
Start date:
% Done:




I am looking to analyse a number of variables such as income, wealth and consumption across waves and to see how they change. From what I understand I cannot link households across waves. But to link individuals across waves I have to use the person identifier (PID). I am specifically looking at the data files whhresp and windresp from waves J-R. Within these I can find the variable for household identification number but I cannot find the variable for PID. Please could you help identify where I can find PID so that I can analyse this data for individuals across waves.



Updated by Alita Nandi about 8 years ago

  • Assignee changed from Alita Nandi to Christopher Booth
  • % Done changed from 0 to 90

I am not sure why you cannot find pid but here are two possible reasons:

(1) In the BHPS all variables are lower case and Stata is case sensitive. So, if you look for PID Stata will say that the variable does not exist. The variable is pid

(2) The household file, whhresp, only include the household identifier, whid. The individual respondent file, windresp, includes both whid and pid. To merge these files by wave you should do a m:1 merge using the variable whid.

use windresp, clear
merge m:1 whid using whhresp

As you are interested in matching household level variables across waves, you may also look at a similar issue raised by another user: issue 509.

Hope this answers your question. If not, please let us know.



Updated by Alita Nandi about 8 years ago

Although issue 509 is about understanding society (ukhls) - the file structures for the two surveys are almost identical so the data management issues are the same.


Updated by Victoria Nolan almost 8 years ago

  • Status changed from New to Closed
  • % Done changed from 90 to 100

Also available in: Atom PDF