Date: Tue, 5 Oct 1999 14:14:43 -0500
Reply-To: "Nichols, David" <nichols@SPSS.COM>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: "Nichols, David" <nichols@SPSS.COM>
Subject: Re: Peto's Test
What we have in SPSS that might be called a Peto test is for comparing
survival distributions in two or more groups. If this is something
different, then it's likely that you would indeed have to program it.
Principal Support Statistician and
Manager of Statistical Support
From: Frederic Villamayor [mailto:villagri@MX3.REDESTB.ES]
Sent: Monday, October 04, 1999 8:57 AM
Subject: Re: [SPSSX-L] Peto's Test
Peto's test in fact are three type of procedures that combine to analyse
tumoral frequencies from long-term carcinogenesis assays. Each procedure is
adequate to some kind of tumours, defined by its "observation context", i.e.
(i) if the tumour caused the death of the animal (direct or indirectly),
(ii) if it did not cause the death of the animal,
(iii) if the animal was found independently of the death of the animal
(tumours found in vivo).
The main reference is
Peto, R. Pike, MC, Day, NE, et al. (1980). Guidelines for simple, sensitive
significance tests for carcinogenic effects in long-term animal experiments.
In: Long-term and short-term screening assays for carcinogens: A critical
appraisal (IARC Monographs Supplement 2), Lyon, International Agency for
Research on Cancer, pp. 311-426.
The whole procedure is in fact simple to understand, but I'd appreciate some
help if I can avoid to write any script or macro.
Thanks in advance