Re: Newbie: How do I...
- From: Bruce Weaver <bweaver@xxxxxxxxxxxx>
- Date: Sat, 24 Dec 2005 09:38:14 -0500
MegaWhizz wrote:
Hey people,
I'm new to this group, so please hear me out. I work in healthcare and am trying to use SPSS to help analyze a study I recently completed.
What I am trying to do is to prove that there is no difference between two groups. It's pretty easy to use the null hypothesis that mean(a) = mean(b), use the t-test and I get a p-value of 0.74 - which means I was unable to detect any significant difference between the means, and cannot reject the null hypotheses. (Using alpha =0.05) - unfortunately this means I can only say "we were unable to detect a significant difference between the two groups" and not "we were able to detect no significant difference between the two grups"
stop me if I'm wrong...
So if I understand correctly, I need to use a null hypothesis that the absolute difference between the two means is greater than a predefined clinical threshold, (called tolerance?) i.e. H0 is mean(a) - mean(b) > delta, where delta is the tolerance. if I can run a test on this hypothesis and reject it, I will prove that the two means are similar.
Anybody know how to do this using SPSS?
MW
You're describing a test of equivalence. See the following link for more info:
www.graphpad.com/index.cfm?cmd=library.page&pageID=36&categoryID=4
-- Bruce Weaver bweaver@xxxxxxxxxxxx www.angelfire.com/wv/bwhomedir .
- References:
- Newbie: How do I...
- From: MegaWhizz
- Newbie: How do I...
- Prev by Date: Merry Christmas and a Happy Good Year
- Next by Date: Linearity of Independent varialbles in Multiple Regression.
- Previous by thread: Re: Newbie: How do I...
- Next by thread: Merry Christmas and a Happy Good Year
- Index(es):
Relevant Pages
|
