Date:         Mon, 8 Mar 1999 18:17:33 -0500
Reply-To:     Dale Pietrzak <dpietrzak@EDUC.KENT.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Dale Pietrzak <dpietrzak@EDUC.KENT.EDU>
Subject:      Re: Stat Help
Comments: To: "Simon, Steve, PhD" <ssimon@CMH.EDU>
Content-Type: text/plain; charset="iso-8859-1"

One of my major dislikes of SPSS for some time has been the
output format.  I truly like ASCII text only output, but have not
found the way to get it in many places.
 
Dale
 
-----Original Message-----
From: Simon, Steve, PhD <ssimon@CMH.EDU>
Newsgroups: bit.listserv.spssx-l
To: SPSSX-L@UGA.CC.UGA.EDU <SPSSX-L@UGA.CC.UGA.EDU>
Date: Monday, March 08, 1999 11:23 AM
Subject: Re: Stat Help
 
 
>Jim Pudilo wrote about some difficulty in getting p-values for
t-tests out
>of SPSS and that the results contradicted common sense and
disagreed with a
>previous analysis done in Excel.
>
>It's impossible to solve these problems via e-mail but I do have
a practical
>suggestion. Invest in an introductory statistics textbook that
has a lot of
>worked out examples. Try running these examples in SPSS and then
match up
>the SPSS output with the book.
>
>With a t-test, it is pretty easy to specify the wrong type of
test (one
>sample versus paired samples versus independent samples). It is
also easy to
>input the data in a format that makes analysis more difficult.
>
>Finally, it is easy to misread your output. The SPSS output for
the
>independent samples t-test is (in my humble opinion) needlessly
confusing
>because it includes two different variants of the t-test plus an
(again in
>my opinion) totally unnecessary test of equality of variances.
What's worse,
>there is no way to turn off some of these tests. With three
different
>p-values in the span of a half page, it's no wonder that a
student would get
>confused. <grin>
>
>Seriously, I have a Ph.D. in Statistics and even I have problems
from time
>to time with the output from SPSS and other programs. I find
that running a
>textbook example when I am using new software or a new procedure
helps me
>get my bearings.
>
>Steve Simon, ssimon@cmh.edu, Standard Disclaimer.
>STATS - Steve's Attempt to Teach Statistics:
http://www.cmh.edu/stats