Date:         Thu, 22 Sep 2005 17:38:52 +0200
Reply-To:     Marta García-Granero
              <biostatistics@terra.es>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Marta García-Granero
              <biostatistics@terra.es>
Organization: Asesoría Bioestadística
Subject:      Re: Standard Error
In-Reply-To:  <s332763d.004@gwdom2-med.med.utah.edu>
Content-Type: text/plain; charset=ISO-8859-15

Hi Adrian,

AM> I did a one-way ANOVA in SPSS 13 with a Tukey HSD Post-Hoc test.  In
AM> the output I noticed that the Standard Error in the Multiple Comparisons
AM> table was 17.30069 for each and every comparison.  I am comparing the
AM> amount of foraging in 4 different types of grass, each with 5
AM> observations.  My understanding is that the Standard Error is calculated
AM> by dividing the Standard Deviation by the square root of the number of
AM> observations -  SD/sqrt(N)

Not in this case. The Standard Error the output refers to is the
SE(mean difference):

SE(diff)=SQRT(MSwithin)/(SQRT(1/n1+1/n2)).

You say your sample size is 5 for every group. The SE(diff) is in this
case:

SE(diff)=SQRT(MSwithin)/SQRT(2/5)

--
Regards,
 Marta                            mailto:biostatistics@terra.es