st: RE: Breadwinner variable

[Nick Cox <n.j.cox@durham.ac.uk>]

gen coupleid = ///
string(min(individual, spouse)) + " " + string(max(individual, spouse))  

bysort coupleid wave : gen difference = earnings - earnings[3 - _n] if _N == 2 

Notes: 

1. You need an identifier for each couple. 

2. If and only if there are two observations for each couple and each wave, the other person's earnings are in 3 - _n. That is because if _n is 1 then 3 - _n is 2, and vice versa. Note that 

bysort coupleid wave : gen difference = earnings[1] - earnings[2] if _N == 2

is not what you want. 

See also 

SJ-8-4  dm0043  . Tip 71: The problem of split identity, or how to group dyads
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
        Q4/08   SJ 8(4):588--591                                 (no commands)
        tip on how to handle dyadic identifiers

Nick 
n.j.cox@durham.ac.uk 

Peter Gambrill

I have data like this

Wave   Individual  Spouse  Earnings  Difference
1          101           102       x1          x1-y1
2          101           102       x2          x2-y2
3          101           102       x3          x3-y3
1          102           101       y1         y1-x1
2          102           101       y2         y2-x2
3          102           101       y3         y3-x3

I need to create a variable that shows how much more or less one
individual (101) in the couple earns compared to the other individual
(102). Is there anyway of doing this? Something like the variable
"Difference" above

Also for each couple they don't always have observations for the same
number of waves as each other or as the next couple in the data.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/