multiple comparison procedures

Article View: bit.listserv.stat-l

Article #3555

Back to bit.listserv.stat-l Previous Next

#3555

From: W. Mark Coward
Date: Sat, 25 Jan 1992 00:33

12 lines
1114 bytes

Some people suggest that one is "allowed" k-1 contrasts in a one-way
ANOVA with k levels of the factor. They suggest that there is no
need to control for multiple comparisons when contrasts are independent
and take at most the number of degrees of freedom of the factor.

I know that Ryan (1959 or 1960) suggests that this is not appropriate,
however I have never encountered a paper that supports this idea.

Does anyone know of a reference that lays out the argument in favor of this?


Mark Coward

Technical Details

Message-ID: <STAT-L%92012419445003@VM1.MCGILL.CA>

Path: arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!paladin.american.edu!auvm!UICVM.BITNET!U56961

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: f5efb,83fe4d7f6b6ef828,start
X-Google-Attributes: gidf5efb,public
X-Google-ArrivalTime: 1992-01-25 01:10:48 PST
Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
Path: arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!paladin.american.edu!auvm!UICVM.BITNET!U56961
Message-ID: <STAT-L%92012419445003@VM1.MCGILL.CA>
Newsgroups: bit.listserv.stat-l
Date:         Fri, 24 Jan 1992 18:33:32 CST
Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
From:         "W. Mark Coward" <U56961@UICVM.BITNET>
Subject:      multiple comparison procedures