Merge household survey sections with no unique identifier

[Elise Space]

Hi everyone

I am working on a university project and I have household survey for a country "X" , the survey has different sections. In particular I am interested in working on section 12 which has 3 different subsections 12.a , 12.b and 12.c . I would like to merge these sections togheter the problem is that i do not really have a "unique identifier", or I seem not to find it. This section is related to agriculture and as possible identifier i have household_id, number of the plot etc. Do you have any clues on how i can proced ?

[Clyde Schechter]

Clearly, in order to meaningfully combine these data sets you have to have some way of identifying which observation in each of the subsections corresponds to which observation in the others. If household_id alone does not do it, then probably you need to combine household_id with some other variable(s) to get to a unique identifier. What that (those) might be should be found in the documentation provided by the purveyor of the data sets. You will have to scour the documentation to get that.

Alternatively, you might post back giving the title of the survey, identifying the country, and perhaps providing a link to the data source (if it is publicly available) and somebody might recognize it and, having worked with it before, might know off the top of his or her head how to do this.