Dear Tipsters,
John Kulig's post is excellent in distinguishing between different
kinds of significance. I would add that we could distingish three
kinds: statistical in the traditional sense, effect size (e.g., as
revealed by meta-analysis, where Cohens guidelines for d =( M1-M2)/s
are .2 small, .5 medium, .8 large) and then practical (clinical)
significance. There has been somerecent discussion of the latter
(see ref. below).
However,in a review of the debate about null hypothesis testing,
Nickerson (2000) argues that we should not state that a small p (<
alpha) means reliability (replicability). Indeed he labels this
belief a myth! He acknowledges that a small p makes it more likely
that a similar outcome would occur but he would not wish to replace
the term significance with reliability.
Sincerely,
Stuart
Kendall, P. C. (1999). Clinical significance. Journal of Consulting
and Clinical Psychology, 67, 283-284.
Nickerson. R. S. (2000). Null hypothesis significance testing.
Psychological Methods, 5, 241-301.
___________________________________________________
Stuart J. McKelvie, Ph.D., Phone: (819)822-9600
Department of Psychology, Extension 2402
Bishop's University, Fax: (819)822-9661
3 Route 108 East,
Lennoxville, e-mail: [EMAIL PROTECTED]
Quebec J1M 1Z7,
Canada.
Bishop's University Psychology Department Web Page:
http://www.ubishops.ca/ccc/div/soc/psy
___________________________________________________