| Date: | Thu, 13 May 2004 08:18:07 -0400 |
|---|
| Reply-To: | Art@DrKendall.org |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Art Kendall <Arthur.Kendall@verizon.net> |
|---|
| Organization: | Social Research Consultants |
|---|
| Subject: | Re: Only 30 responses for T-test? |
|---|
|
| In-Reply-To: | <BAY15-F10TLYYJAeh9b0004f90a@hotmail.com> |
|---|
| Content-Type: | text/plain; charset=ISO-8859-1; format=flowed |
|---|
It seems that someone was pulling your leg.
Art
Art@DrKendall.org
Social Research Consultants
University Park, MD USA
(301) 864-5570
Elsa-Sofia Morote wrote:
> Dear colleagues,
> I heard in a discussion that SPSS has some deficiences when applying
> F-tes
> and T-test to large samples. The person said "when you compare two
> groups,
> and you have lets say 100 individuals in one group and 120 in the
> other, you
> have to take a randon sample of 30 to run a T-test, because SSPSS has
> deficiences"..."you must only have 27-30 responses for run T-test and
> F-test
> in SPSS to overcome SPSS deficiences in samples...."
> Could somebody tell me if this is correct? Which deficiences SPSS has in
> F-test, and T-test? Why we have to take randon sample of our sample?
> Why we
> can only test with 27-30 responses? Any thoughts?
>
> _________________________________________________________________
> Getting married? Find tips, tools and the latest trends at MSN Life
> Events.
> http://lifeevents.msn.com/category.aspx?cid=married
>
|
