Hi Everyone,
I ran a logistic regression and identified several predictors of student satisfaction. However, some of the independent variables had low N's for some levels.
For example, geographic region was identified as significant ( In -State, N=558) (Out of State, N= 57).
Is it still safe to say the odds of In-State Students reporting a higher satisfaction score increase if they are in-state considering the sample size. I heard that the N for each level should be at least 1/3 of the total.
Thanks,
Keval
====================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
