Date:         Sun, 8 Mar 2009 09:49:55 -0700
Reply-To:     tanwanzang@YAHOO.COM
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         tanwanzang@YAHOO.COM
Organization: http://groups.google.com
Subject:      Re: ordinal data
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

QUOTE: "From our experience, treating ordinal data as continuous with
constant variance can provide a useful approximation when the number
of response categories is large, but may be inadequate when that
number is less than five." This is on page 2605 of  Agresti, Alan &
Chuang-Stein, Christy (1997) "A REVIEW OF TESTS FOR DETECTING A
MONOTONE DOSE-RESPONSE RELATIONSHIP WITH ORDINAL RESPONSE DATA",
Statistics in Medicine VOL. 16, 2599-2618.