Re: st: Ologit question

["Anat (Manes) Tchetchik" <anatmanes@gmail.com>]

Thanks Jay, I really appreciate your help
Anat

On Wed, Oct 31, 2012 at 8:47 PM, JVerkuilen (Gmail)
<jvverkuilen@gmail.com> wrote:
> On Wed, Oct 31, 2012 at 1:44 PM, Anat (Manes) Tchetchik
> <anatmanes@gmail.com> wrote:
>>
>> Thanks,  I ran the Brant test of parallel regression assumption and it
>> turned out insignif. (p>chi2=0.877) does it supprt the use of ologit?
>
> Not if asymptotics themselves don't apply, though it's hard to say
> here. The real problem is that you have a lot of IVs for your sample
> size.
>
>
>
>> Also- what is the main difference between the ologit and rologit? The
>> data I have  actually refer to 3 groups of 50 observations each
>> (cardiology, oncology and orthopedic) I currently used dummy vars. to
>> control for this, maybe the ranked ordered logit is more appropriate?
>
> -rologit- is for responses that are ranked. It fits the Plackett-Luce
> model. This is used when a respondent has multiple objects and ranks
> them, and would thereby be giving multiple responses. So if each
> respondent were to be asked "what medical speciality do you intend to
> study for? rank order them" then yes -rologit- is more applicable.
>
> Jay
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/



-- 
Anat Tchetchik, PhD
Department of Hotel and Tourism Management
Guilford Glazer Faculty of Business and Management
Ben-Gurion University of the Negev
P.O.Box: 653
Beer-Sheva, Israel, 84105

E-mail:       anat@som.bgu.ac.il
Phone         972-(0)8-6479735
Fax:           972-(0)8-6472920
Web:          http://cmsprod.bgu.ac.il/Eng/som/hotelmanage/Staff/Academic/ChechikA.htm
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/