Hi

On Mon, 11 Nov 2002, Rob Flint wrote:

> One of my students doing her senior thesis ran her stats and got results of
> .056 and .08 for two different ANOVAs. In the past I have seen published
> studies indicating that these are "marginally significant." How do you deal
> with results of this nature? More importantly, do you have any citations
> (journals or books) that discuss the value of including/discussing results
> that seem to "approach significance"?

Unlike some others who have responded, I do not see significance
as an either-or proposition.  Indeed, APA style now calls for the
reporting of p-values, rather than dichotomous judgments,
presumably on the grounds that a p of .056 or even .08 might or
might not be meaningful to readers.

Nor do I read "approaching" as meaning moving in the direction of
significance, although that is certainly one interpretation.  I
always took it as synomymous with marginal (i.e., close to some
arbitrary cut-off).

None of the responses that I saw raised the issue of directional
vs. non-directional tests (i.e., one- vs. two-tailed ps).  Are
these ps for non-directional tests in a study where a more
knowledgeable researcher might have predicted the direction of
difference on the basis of theory or past research.

A second issue is how many df were associated with the numerator
of the F ratios?  If the df are >1 then I would view them more
positively, especially if the "predicted" effect could readily be
translated into a single df contrast.

Another important (and related) question is whether the ps are
for interactions.  ANOVAs are quite insensitive when it comes to
interactions (i.e., lack power) and I would certainly look
favorably on a (perhaps-expected) interaction with a p of .056 or
.08 (e.g., doing simple effects analyses or partitioning the
interaction).

So before answering this question, I would like to see the actual
variables and levels involved, the theoretical relationships
predicted, and (related to people's concern with power) the n's
for the conditions/cells.

Best wishes
Jim

============================================================================
James M. Clark                          (204) 786-9757
Department of Psychology                (204) 774-4134 Fax
University of Winnipeg                  4L05D
Winnipeg, Manitoba  R3B 2E9             [EMAIL PROTECTED]
CANADA                                  http://www.uwinnipeg.ca/~clark
============================================================================


---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]

Reply via email to