weighting cases in an analysis of variance

[Jürg Artho]

Hi everybody

My sample is not representative. Therefore I have a weighting variable w.

To calclutate tables with means I use the command WEIGHT BY w.

To calculate analysis of variance (two factors) I turn this off (WEIGHT OFF) and use the weighting variable w as WLS-variable.

Is that correct? Or how can I weight cases in a UNIANOVA-command?

If it is correct: The means given out by the UNIANOVA-Command are the same as in the tables. Then N however differ slightly. Why is that the case?

Thank you very much for your help.

Regards,
Jürg

--
Universität Zürich
Sozialforschungsstelle
Jürg Artho
Binzmühlestrasse 14 / Box 15
8050 Zürich

Tel: [phone]
Fax: [phone]

=====================
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