cristina.solera--
Under the most generous assumptions, you can only identify the
"effect" of partner's occupational score for those with partners, i.e.
you can get a coefficient on the interaction married*partneroccscore
but not on the main effect partneroccscore, letting the interaction be
zero whenever someone has no partner (assuming haspartner==married in
this context). If you include both variables, the "marginal effect"
of moving from married==0 to married==1 will be much trickier to
estimate. You might prefer a set of indicators {notmarr, poorhusband,
richhusband}, or somesuch (with the excluded category the middle third
of partners' occupational scores, perhaps).
Note further that you would not want to give a causal interpretation
to any of the regressors you mentioned, since there are very good
reasons to suppose that unobserved factors predict each, and also
predict women's risk of exiting paid work. To be more concrete, a
woman's desire to exit paid work is probably a strong predictor of
marital status, potential partners' occ scores, and risk of exiting
paid work.
So you will not be estimating the effect of marriage, in any case.
Just my 0.016 Euro (poorly translated American idiom).
--Austin
On 10/10/06, solera <[email protected]> wrote:
Dear all,
I'm running a discrete time duration model on women's risk of exiting paid
work. In addition to covariates such as level of education, type of
contract, sector, age and number of children and so on, I would like to
include the marital status and, if married, the occupational score of the
partner.
However, which value should I give to the occupational score in the months
where the woman is not married? If I make occupational score missing, then
the model will drop the observations where the woman is not married, so that
I can not estimate the effect of marriage. If I give the value 0, STATA
reads it as a very low occupational score. A solution could be to make
occupational score a categorical varible, defing a certain number of
subgroups, with one of the subgroups being "not married". But for different
reasons I would prefer to keep occupational score as a continuous variable.
Any suggestion?
Thanks a lot in advance
Cristina
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
