Fmax
I am a little confused on the fmax test... if it is < 10, do we use equal variances assumed or not?
posted by bfullmer @ 4:23 PM
This blog is for the use of students enrolled in either section of PG851. It is intended to be a forum for asking, discussing, clarifying, and helping.
posted by bfullmer @ 4:23 PM
2 Comments:
At 4:26 PM,
Ashli said…
If the Levene's test is significant you have to use equal variance NOT assumed 'cause that means you violated ANOVAs assumption of equal variance. The Fmax of less than 10 (if there is a 4:1 ratio or less of cells) lets you know ANOVA is still robust to the violation of equal variance.
At 5:34 PM,
David said…
yup, great answer Ashli!
Post a Comment
<< Home