Not that I doubt you Karl (you seem very educated on statistical issues) but 
all the textbooks I have used talk about HSD as a post hoc test that is only 
appropriate to use after finding significance with an ANOVA. Do you have 
something I could reference to support this? 
 
The other thing is if HSD is a replacement for ANOVA it seems even more unusual 
that they would come to different conclusions.
 
 
Thanks,
 
 
Rick
 
Dr. Rick Froman
Psychology Department
Box 3055
John Brown University
Siloam Springs, AR 72761
(479) 524-7295
[EMAIL PROTECTED]
"Pete, it's a fool that looks for logic in the chambers of the human heart"
- Ulysses Everett McGill

________________________________

From: Wuensch, Karl L [mailto:[EMAIL PROTECTED]
Sent: Tue 4/3/2007 6:18 PM
To: Teaching in the Psychological Sciences (TIPS)
Subject: [tips] ANOVA, HSD, and LSD



        I wouldn't, because Tukey's HSD was designed as a REPLACEMENT
for the ANOVA, not as something to do after an ANOVA.  Frankly, there is
not often any good reason to do an ANOVA other than that everybody else
does it and it is expected.  Better off to make a set of focused
contrasts that address the research questions you want to answer.  If
you just must make pairwise contrasts, use the REGWQ instead of the HSD
-- it has more power and holds familywise alpha at no more than its
nominal level.  If you have only three groups, use Fisher's procedure,
which, by the way, is the only one of the commonly employed pairwise
procedures which requires that you do an ANOVA first (and that the ANOVA
be significant).  With only three groups it holds familywise error at
the nominal level.  Although pooled error is commonly employed with
Fisher's procedure, one should use individual error terms if
heterogeneity of variance is evident.

Cheers,

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karl L. Wuensch, Chair, Faculty Information Technology Review Committee
East Carolina University, Greenville NC  27858-4353, USA, Earth
Voice:  252-328-9420     Fax:  252-328-6283
[EMAIL PROTECTED]
http://core.ecu.edu/psyc/wuenschk/klw.htm


-----Original Message-----
From: Rick Froman [mailto:[EMAIL PROTECTED]
Sent: Tuesday, April 03, 2007 2:50 PM
To: Teaching in the Psychological Sciences (TIPS)
Subject: [tips] ANOVA interpretation

How would you interpret an ANOVA result where the F-test was significant
but none of the multiple comparisons were significant in an HSD
comparison?

Rick


Dr. Rick Froman, Chair
Division of Humanities and Social Sciences
Professor of Psychology
John Brown University
2000 W. University
Siloam Springs, AR  72761
[EMAIL PROTECTED]
(479) 524-7295
http://www.jbu.edu/academics/hss/faculty/rfroman.asp



"Pete, it's a fool that looks for logic in the chambers of the human
heart."
- Ulysses Everett McGill




---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang= 
<http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0<=> 
english


---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english 
<http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0<=english> 



<<winmail.dat>>

---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english

Reply via email to