# Homegeneity Tests

#### Nee

##### New Member
Hie,

This question is regarding the assumption for ANOVA on having the variances equal for all populations. Have read that if this assumption is violated and the sample sizes differ among groups, the p-value and the F test is untrustworthy. It is preferable to use statistics that do not assume equality of population variances such as the Browne-Forsythe or the Welch statistics.

1) How does the siginificant value differs in the Browne-Forsythe or Welch test compared to the Levene test?
2) Do we need to run the Levene test first to see whether the homogeneity assumption holds? And if it doesnt, then only we run the Browne-Forsythe/Welch test?

Thanks so much!

#### JohnM

##### TS Contributor
1) I believe they are less powerful (more conservative) - in other words, a larger statistic is required before a difference is called significant

2) yes