Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Cohen d
Replies: 1   Last Post: Nov 8, 2012 12:05 PM

 Messages: [ Previous | Next ]
 Luis A. Afonso Posts: 4,758 From: LIsbon (Portugal) Registered: 2/16/05
Cohen d
Posted: Nov 1, 2012 4:50 PM

Hodges g, Cohen d, GOOD FOR WHAT?

Karl L. Wuensch (2008)
__This document is a guide for my students in Experimental Psychology . . .
core.ecu.edu/psyc/wuenschk/docs2210/Strength_of_Effect.doc
____________

An example on significance difference between two groups:
__Group 1 : N=6, Mean=22.67, St.dev.=4.274
__Group 2 :___8,______ 13.25,_______4.083

_______g = (22.67-13.25)/4.15 = 9.42/4.15 = 2.27
very large effect following Hedges.
______________
Applying the standard NHST, with the current d, Student t, 6 + 8 - 2 = 12 degrees of freedom, one tail, 5% significance level, critical value= 1.782
_______T = (9.42 - d)/ 4.15 = 1.782
_______d = 9.42 - (1.782*4.15) = 2.02
And the difference on Population mean values is at least 2.02 with 95% confidence. Simply as this . . . better than Cohen-Hodges procedure because provides a numeric alpha and uses the conventional NNHST (my acronym for no-null hypotheses statistical test . . .)

Luis A. Afonso[b][/b][i][/i][u][/u]

Message was edited by: Luis A. Afonso

Date Subject Author
11/2/12 Luis A. Afonso
11/8/12 Luis A. Afonso