Match Couple in LFS

[Giulia Dalla Nora]

Hi there!

Can anyone help me with this question ?

I am using the LFS dataset.
I want conduct analyses on the couple level. For each person information about the presence of a partner in the household is available.
Where appropriate a partner identifier (HHSPOU) is available, which corresponds to the partner’s ID (HHSEQNUM).
I have to create a new variabile COUPLE, where I have to combine the information of both partners in one row of the dataset.

Can anyone help me?

Thank you.

Giulia

[Martin Bresslein]

Gulia,

your description is a bit superficial. What do you mean by combining? The answer possibly depends on this piece of information. E.g., if by combining you refer to adding incomes, collapse would help. If one item is number of children, you would possibly want the mean.
E.g.

collapse (sum) income (mean) children, by(hhid)

where hhid would be a variable identifying a household - if not already in the dataset, you could generate it with the information available through the identifiers you mention.

If you don't want to collapse your dataset, have a look at egen and egenmore, there are lots of functions which may help. That way you could check that Stata does what you want, and later keep only one observation per household

In terms of maximizing your chances of getting help, have a look at the forum's FAQ how to provide information here.

[Giulia Dalla Nora]

Thanks Martin,

I mean, for the moment I do not want to combine couple by number of children or income, I want only combine and find couple informations. In other words I want matching Husband to Wifes.

[Martin Bresslein]

Sorry, it's still not quite clear to me which information is supposed to be contained in the variable "couple". Is it the identifier? Btw, does LFS refer to the European Labor Force Survey?
Not sure about the makeup of the dataset, since you do not provide a snapshot...

It would help if you elaborate, else I can only speculate about what you want to achieve...

[Giulia Dalla Nora]

Yes, LFS refer to EU Labor Force Survey.
Data are organized according to households and so the reference person and the spouse/cohabitant to the reference person is really just information ABOUT the household.
I would match/combine spouse/cohabitant of the reference person using household-id as the matching identifier.

[Stephen Jenkins]

Giulia: Martin Bresslein wrote earlier:

In terms of maximizing your chances of getting help, have a look at the forum's FAQ how to provide information here.

So, Giulia, please re-read the Forum FAQ. Photos are not a recommended way of providing information. It's better to report, for example, the output from an appropriate list of selected variables and cases, and to do so using CODE delimiters. The FAQ explains how to do this. (It's easy.)

[Friedrich Huebler]

These two Stata FAQs may be useful.

• How do I create a variable recording whether any members of a group (or all members of a group) possess some characteristic?
• How do I create variables summarizing for each individual properties of the other members of a group?