home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pipex!unipalm!uknet!comlab.ox.ac.uk!ph
- From: ph@physiology.oxford.ac.uk (Patrick Haggard)
- Newsgroups: sci.math.stat
- Subject: Variances significantly different ? (independent samples)
- Message-ID: <1992Dec23.114358.7198@galen.physiol.ox.ac.uk>
- Date: 23 Dec 92 11:43:58 GMT
- Sender: ph@physiol.ox.ac.uk (Patrick Haggard)
- Organization: Physiology Department, Oxford University, Oxford, UK.
- Lines: 22
- Originator: ph@galen.physiol
-
- Hello. There was a thread a while back about the Pitman procedure
- for testing whether the variances of two related samples are
- significantly different. Could anyone fill me in on how to do the
- equivalent for unrelated samples. I have approx. 25 measurements in test
- A, and approx. 25 measurements taken in test B. The units of measurement
- are the same in the two tests. I want to see whether performance
- is more variable on test A.
- Hays' book ("Statistics"), and most other stats books,
- just give the simple F test between the
- two variances, but I seem to remember people being rather unhappy
- about using that method when it was last discussed. Can anyone
- suggest any alternative test, please?
- Thankyou
- Patrick
- --------------------------------------------------------------------------
- Patrick Haggard Email (WORLD): ph@physiol.ox.ac.uk
- University Laboratory of Physiology Email (JANET): ph@uk.ac.ox.physiol
- Parks Road, Tel. (0865) 272116
- Oxford, OX1 3PT Fax. (0865) 272469
- England
- --------------------------------------------------------------------------
-