| Date: | Mon, 1 Aug 2005 12:37:53 +1000 |
| Reply-To: | Paul Ginns <p.ginns@itl.usyd.edu.au> |
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
| From: | Paul Ginns <p.ginns@itl.usyd.edu.au> |
| Subject: | Within-subjects analysis of ranked preferences |
| Content-Type: | text/plain; charset="iso-8859-1" |
Hello everyone,
a colleague has asked me for some advice about how to analyse some rank scores. He asked students to rank their preference for answering non-assessable in-class questions across 5 different possibilities. Thus, for each student, he would have 5 different variables:
Case 1: 3 4 2 1 5
Case 2: 1 2 4 5 3
Case 3: 1 2 4 3 5
etc.
He has a theory based expectation about the pattern of preferences, which initially made me think about using a within-subjects ANOVA contrast encoding the theory, but the fact that the scores are ranks would (I presume) invalidate this approach. Can anyone tell me what non-parametric test is appopriate, if it can be performed in SPSS, and if not, if any shareware is available for it?
Thanks in advance,
Paul
Dr. Paul Ginns
Survey Officer
Institute For Teaching and Learning
Room 385 Carslaw Building F07
The University of Sydney NSW 2006
Australia
Phone: (02) 93513607
Fax: (02) 9351 4331
Email: p.ginns@itl.usyd.edu.au
http://www.itl.usyd.edu.au
|