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 on April 23, and its replacement, is already up and running.

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

Re: st: continuous data measured as ordinal catogories

From   Shehzad Ali <>
Subject   Re: st: continuous data measured as ordinal catogories
Date   Fri, 9 Apr 2010 14:09:22 +0000 (GMT)

Thank you, Richard, Maaten and Clive. These are all very helpful suggestions. 

I want to use willingness-to-pay as the dependent variable. The category 'other' includes random amounts that are may be less or greater than the highest amount category (i.e. $10). These 'other' amounts account for around 10% of the data. I assume that using the response categories of $1, $5, $10 and 'other amounts' as a continuous variable would be inappropriate, or would it?

Many thanks,

----- Original Message ----
> From: Maarten buis <>
> To:
> Sent: Fri, 9 April, 2010 8:25:45
> Subject: Re: st: continuous data measured as ordinal catogories
> --------------------------
Maarten L. Buis
Institut fuer 
> Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
> Tuebingen

> >

--- On 
> Fri, 9/4/10, Richard Williams wrote:
> If you want it as an independent 
> variable, my first impulse
> is to break it into dummies.  

> somewhat uncommon but sometimes useful interpretation of such
a model is that 
> it simultaneously fits an optimal scale for 
your ordinal variable and an 
> effect of that optimally scaled 
variable. The scale is optimal in the sense 
> that it best 
predicts the dependent variable. The sheaf-coefficients, as 
implemented in -sheafcoef-, uses that approach. This allows 
you to 
> _both_ present a single effect of willingness to pay and 
use the more 
> flexible approach proposed Richard. You can 
download -sheafcoef- by typing 
> -ssc install sheafcoef-, and it
is discussed in this paper: <> href=""; target=_blank 
> >>

Hope this 
> helps,


> For searches and help try:
*  > href=""; target=_blank 
> >
*  > href=""; target=_blank 
> >
*  > href=""; target=_blank 
> >


*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index