Expected frequency within CSTABULATE command

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

Expected frequency within CSTABULATE command

Shirin Nuesslein

I am working with the /TEST INDEPENDENCE sub-command within the CSTABULATE command.

Is it possible to have SPSS show me the expected frequencies in the output? I need to check if the chi-square test is valid and meets the assumption of expected counts n > 5 per cell before I proceed with my analysis. The default output only shows me unweighted counts and estimated population size. 

Alternatively, I could calculate the expected frequencies manually. Would I need to use the unweighted counts or estimated population size for my observed data? 

Any pointers appreciated! Thank-you!

Shirin Nuesslein
Research Assistant - Health, Environment and Indigenous Communities Research Group
Master's Student - Sustainability Studies Program at Trent University

Trent University - Indigenous Studies Department
1600 West Bank Drive - Peterborough, ON, K9J 7B8
Phone: 705-748-1011 ext 7242
===================== 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