homogeneous subsets vs multiple comparison

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

homogeneous subsets vs multiple comparison

E. Bernardo
Hi all,

  One-way ANOVA showed that the five group means (G1=3.53, G2=3.79, G3= 3.82, G4=4.02, and G5=4.22) are significantly different (F=4.286, p<0.05).  The multiple comparison via Scheffe test revealed as follows:

               * G1 and G4 are significantly different (p=0.024)
               * G1 and G5 are significantly different (p=0.016).
               * The other pairs are nonsignificant (p>..05)

  While the Scheffe homogeneous subset for alpha=0.05 revealed as follows:

              * G1,G2, G3 and G4 are homogeneous groups (fall on subset 1with
                 p-value of .117)
              * G2,G3, G4, and G5 are also homogeneous (fall on subset 2 with
                 p-value of .231).

  I am puzzled with the results of multiple comparison and homogeneous subsets.  Why are the results of the two different?

  Thank you.
  Eins


---------------------------------
  Happy Holidays from Yahoo! Messenger
 Spread holiday cheers to your friends and loved ones today!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD