Thanks a lot to Richard, Stat and Anders for their reply. I will look into the
references and ideas they gave about the problem and, when I got a asnwer I
will post something with my results and how I implemented. Futhermore, I am
facing -besides the measurement error in my dependent variable given by drifts
in the cutpoints- and endogeneity problem.........
Thanks again,
Marcos.
Quoting Anders Alexandersson <[email protected]>:
> If Richard 's comment "Who's to say that all respondents do this
> coding the same?" is important, then could an ordinal unfolding or
> ideal point model solve that problem? The unfolding model directly
> deals with the issue whether you "respond from below" or "respond from
> above". The latest gllamm article (which is available from
> http://www.gllamm.org/pub.html) has an example of a unidimensional
> ordinal unfolding model:
>
> Rabe-Hesketh, S. and Skrondal, A. (2007). Multilevel and latent
> variable modelling with composite links and exploded likelihoods.
> Psychometrika, in press.
>
> I sure would like to see the .do file for the example in their
> article, because theoretically I think unfolding models often make
> sense but a big problem is the implementation. Unfolding models also
> tend to require a fairly large sample size, so what sample size does
> Marco have?
>
> Anders Alexandersson
> [email protected]
>
>
> Marcos <[email protected]> wrote:
>
> >> Basically, I have an ordered categorical model (ordered probit) where,
> >> theoretically, people tend to overstate the values of the dependent
> >> variable for higher values. That is, y can take for values 1, 2, 3 or 4
> >> and my assumption is that certain people in my sample (i.e., with
> some >> particular values of covariates) tend to answer most often
> say, 3 or 4; >> so I am facing a measurement error problem in the
> dependent variable,
> >> and by modelling the cut-points as function of those covariates I
> would >> be able to take this into account. However, as mentioned by
> Stat
> >> Kolenikov's reply, I would face an identification problem if the
> >> covariates I use to measure these errors in the cut points are not
> >> different from the one in the main equation.
>
> Richard Williams <[email protected]> commented, in part:
>
> > In general, I think this is the kind of issue
> > that could make you want to give up quantitative
> > research altogether. :) You are counting on
> > respondents to code themselves as "above
> > average", "average", "below average" or
> > whatever. Who's to say that all respondents do
> > this coding the same? If we see differences by
> > gender, how do we know whether these differences
> > are real as opposed to just being differences due
> > to the fact that different people use different standards when coding?
> *
> * 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/
>
*
* 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/
