Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Nominal or ordinal?


From   Richard Williams <richardwilliams.ndu@gmail.com>
To   statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu
Subject   Re: st: Nominal or ordinal?
Date   Thu, 12 Aug 2010 17:55:49 -0500

I am not so trusting, partly because I don't think ordinal regression is all that much harder than ols regression. And, if you do things like a plot of residuals versus fitted, it is pretty clear that a 4 category ordinal dependent variable is not behaving the way you would like it to for OLS regression, e.g. OLS assumptions about error terms being homoskedastic and normally distributed are clearly violated. Example:

use "http://www.indiana.edu/~jslsoc/stata/spex_data/ordwarm2.dta";, clear
tab1 warm
reg  warm yr89 male white age ed prst
rvfplot

That plot certainly doesn't look like a random scatter of points.

Having said that, if the ordinal variable is independent rather than dependent, it is pretty easy to test whether it is ok to treat it as continuous. Long and Freese (in the 2006 book available from Stata Press that I referenced earlier) show how in Ch. 9 of their book. Basically, you run a model where the variable is treated as continuous and another model where you break the variable up into dummies. If breaking up into dummies does not significantly improve fit you can treat the variable as continuous.

At 04:26 PM 8/12/2010, Michael N. Mitchell wrote:
Dear Dave (and all others)

I know I am personally rather trusting of treating such scales as interval data... do you or any others have suggestions on references justifying the treatment of scales like this as interval?

Many thanks,

Michael N. Mitchell
Data Management Using Stata      - http://www.stata.com/bookstore/dmus.html
A Visual Guide to Stata Graphics - http://www.stata.com/bookstore/vgsg.html
Stata tidbit of the week         - http://www.MichaelNormanMitchell.com



On 2010-08-12 1.29 PM, David Bell wrote:
--
Chelsea,

Most of the world is willing to treat scales like this as interval data. Sure it isn't "exactly" interval. Be sure to consider whether your audience will be familiar with interpretations of ordinal logit regressions.

Dave
====================================
David C. Bell
Professor of Sociology
Indiana University Purdue University Indianapolis (IUPUI)
(317) 278-1336
====================================




On Aug 12, 2010, at 2:59 PM, Polis, Chelsea B. wrote:

Dear Statalisters,

I am working with a dependent variable that has the following four potential responses: (1) Not Likely, (2) Slightly Likely,
(3) Quite Likely, (4) Extremely Likely.

A colleague thinks this is an ordinal variable which should be analyzed using ordered logit regression. My sense was that this is a nominal variable, and should be analyzed using multinomial regression - since we cannot know if the levels are
equally spaced in people's minds.

My apologies for what is probably a very simplistic question, but I've searched Statalist and online, and I still am not
entirely certain.  I would greatly appreciate input on this question.

Thanks,
Chelsea


*
*   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/


*
*   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/
*
*   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/

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index