<- file stat 97bonf.html ->

Bonferroni REFs

=======================Ronan Conroy, 20 Mar 1997==========ssc Message-ID: <199703201029.KAA04996@gate.rcsi.ie> From: Ronan Conroy <rconroy@RCSI.IE> Subject: Re: Reference for Bonferroni method >Hello, > >I am looking for the paper where the Bonferroni method for multiple >testing was reported. Other references related with this issue will be >also apreciated. Thanks, > > > Ananias Escalante Try Hancock GR and Klockars AJ. The quest for alpha; developments in multiple comparison procedures in the quarter century since Games (1971). Review of Educational Research 1996;66(3):269-306 Includes good discussion of sequential methods and the general philosophy of control of type 1 error.

Bonferroni REFs (variations)

=======================Paul Bernhardt, 08 Feb 1997==========sse Message-ID: <199702081712.KAA29328@cor.oz.cc.utah.edu> From: Paul Bernhardt <Paul.Bernhardt@m.cc.utah.edu> Subject: Re: Bonferroni correction You might want to consider the modified Bonferroni described in Keppel 1991, page 169. Basically, you multiply your nominal alpha (usually .05) by the degrees of freedom for the effect over which your comparisons will be made. Then you divide this number by the number of comparisons you will actually make. So, if you have 4 groups, using only 5 comparisons. You multiply .05 by 3 (DF for the factor) and divide by 5. This yields an alpha for each comparison of .03. This is only used when the number of comparisons will exceed the number of orthogonal comparisons for the factor under examination. Keppel provides much more clarity than I do, so go to that source: Keppel, G (1991). Design and Anlaysis: A Researcher's Handbook, Prentice-Hall.

Bonferroni, *original* REFs

=======================Greg Hancock, 21 Mar 1997==========ssc Message-ID: <Pine.SOL.3.95.970321074530.2550J-100000@rac4.wam.umd.edu> From: Greg Hancock <ghancock@WAM.UMD.EDU> Subject: Re: Bonferroni After years of thinking of Bonferroni as a nice pasta dish (delicious with a white clam sauce, by the way), I did in fact run into a reference to an actual human being. Bonferroni, C. E. (1937). Teoria statistica delle classi e calcolo delle probabilita. In "Volume in Onore di Ricarrdo dalla Volta," Universita di Firenza, 1-62. I've seen a second reference to a Bonferroni paper on the same topic, but do not have that reference available (it appeared in some encyclopedia of statistical or research something...). I have not read either paper, nor do I imagine being able to even if they were right in front of me. It is my understanding, though, that while Bonferroni expressed the initial inequality, it was Olive Dunn in the late 1950's and early 1960's who used such a relationship in the context of multiple comparisons. She did not, however, cite Bonferroni (to the best of my knowledge), nor is Bonferroni cited in multiple comparison texts such as Hochberg & Tamhane (1987). * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Document by Rich Ulrich. E-mail to wpilib+@pitt.edu

FAQ top.

Ulrich home page.

Ulrich FAQ. http://www.pitt.edu/~wpilib/stats99.html