0
$\begingroup$

I trying to run a one-way ANOVA to compare 3 groups (Northern, Center and Southern regions) but Levene test is significant. Kurtosis and Skewness are ok but the equal variances assumption is violated. So what do you think is the best way to proceed? Maybe using Welch and Brown - Forsythe test and Games Howell test as post hoc ?

Thank you for your attention and your time

$\endgroup$
1
$\begingroup$

Before you proceed to other tests - how large is the sampleset? In large samplesets Levene's test can be significant, although the Variance differences are very low. If you have a large sampleset test also the variance ratio (Hartley's Fmax).

If you still have a problem with the unequal variances "Maybe using Welch and Brown - Forsythe test and Games Howell test as post hoc ?" is the right way to handle this problem.

I can highly recommend the statistics book from andy field. ALthough it is about SPSS, it helps you with a lot of "basic" statistic problems and is written in an easy and understandable way. (This is no advertisment, but the source of which my answer is based on)

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.