home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!IUBACS.BITNET!RTITZER
- Original_To: JNET%"stat-l@mcgill1"
- Message-ID: <STAT-L%93012220380655@VM1.MCGILL.CA>
- Newsgroups: bit.listserv.stat-l
- Date: Fri, 22 Jan 1993 16:34:00 EST
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: RTITZER@IUBACS.BITNET
- Subject: Cluster Analysis Problem
- Lines: 20
-
- We would like to use a clustering procedure to group similar
- practice schedules, but we're not sure what to use as our proximity measure. In
- the experiment subjects controlled the order in which they practiced three
- tasks. Each subject performed 54 trials. Subjects were not restricted to
- practice each task the same number of times. Here is part of an order that was
- chosen by one subject (Trial Number, Task Number):
-
- 1 2
- 2 3
- 3 3
- 4 1
- ... ...
- 54 3
-
- Each subject chose a different order. We would like to form groups
- of subjects based on the practice schedules they selected. We are most
- interested in differentiating the subjects' practice schedules based on how
- many times they practiced a task before they switched to a different task.
- We would appreciate suggestions on how to group these practice schedules.
- Bob & Jennifer
-