Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: query pooling data on dependent variable logit mnl
From
Richard Williams <[email protected]>
To
[email protected], [email protected]
Subject
Re: st: query pooling data on dependent variable logit mnl
Date
Fri, 16 Mar 2012 09:15:30 -0500
At 07:59 AM 3/16/2012, Christine Zozula wrote:
Hi statalist,
Our dependent variable is dichotomous: whether or not someone received
a sanction for misconduct. But there are actually two kinds of
sanctions (we can call them A and B), and we're interested in knowing
whether the same variables predict both sanction A and B in the same
way. In other words, are we justified in combining the two sanction
types into a single (logistic) regression, or must we treat them
separately in something like a multinomial logit regression? There are
no a priori exclusion restrictions that suggest themselves, so what we
really want to do is test whether the same set of explanatory
variables influence getting sanction A (vs no sanction) in the same
way they influence getting sanction B (vs no sanction). Is there a
way to test whether all the coefficients in one panel of a multinomial
regression are the same as all the coefficients in another panel?
Are the two sanctions mutually exclusive, or is it possible to
receive both? Or is it the case that one subgroup can receive A while
another subgroup can receive B? The structure of your data isn't clear to me.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/