How large is your sample relative to your design?
If you only have a couple of subjects per cell you may not have enough df for an error term.
Michael
****************************************************
Michael Granaas
[hidden email]
Assoc. Prof. Phone: 605 677 5295
Dept. of Psychology FAX: 605 677 3195
University of South Dakota
414 E. Clark St.
Vermillion, SD 57069
*****************************************************
-----Original Message-----
From: SPSSX(r) Discussion on behalf of Coja
Sent: Mon 1/19/09 2:48 PM
To:
[hidden email]
Subject: Zero error sum of squares ?
Dear readers an fellow statisticians,
I am trying to create a model with random effects using PROC GLM. Using
this model, I am trying to get the residuals for a residual analysis.
However, even though the model looks is OK, I get a result with sum of
squares being zero (DF=0 as well). Therefore, no residuals. What do do ?
Can anyone advise me ?
Kind regards,
Coja.
=====================
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
====================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
Locked
