Date:         Thu, 19 Aug 1999 12:37:53 -0700
Reply-To:     glaser@pacific-science.com
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Dale Glaser <glaser@PACIFIC-SCIENCE.COM>
Subject:      Re: SPSS: t-test: two independent sample
Comments: To: Shail Dobson <dobson_shail@HOTMAIL.COM>
In-Reply-To:  <19990819190220.98186.qmail@hotmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Assuming you mean n = 100 per group, the answer, in my opinion, is yes; even
though I have seen some intro. texts divide hypothesis testing for two
groups as follows: (1) if n is small (e.g., n < 30) and the population
standard deviation is unknown then use the t distribution; (2) if n is large
and the population standard deviation is known then use z distribution;
however, a cursory glance at the research journals will find many examples
of t-tests being employed when n is well over 30 per group; further as df
increases for the t-distribution, it approximates the z-distribution (i.e.,
critical value for df > 120 = 1.96, when alpha = .05 for two-tailed test);
however, I will be interested if other researchers differ in their
opinions...............dale glaser

-----Original Message-----
From:   SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
Shail Dobson
Sent:   Thursday, August 19, 1999 12:02 PM
To:     SPSSX-L@LISTSERV.UGA.EDU
Subject:        SPSS: t-test: two independent sample

Can independent t-test for means of two independent samples be performed for
sample size over 100? If not, is there any other equivalent test available?

Thank you very much,

Shail


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com