ANOVA vs t-tests: Fundamentally, what can ANOVA do that t-tests can’t?
Work with more than 2 means
Cook steak fuckin’ perfectly
Compare means directionally
Homogeneity of variance: Levene’s and Cochran’s are different tests for this homogeneity/heterogeneity of variance. If they are found to be significant (P<0.05), this means:
Need to transform first
Data is good, go ahead
Homogeneity of variance: Levene’s and Cochran’s are different tests for this homogeneity/heterogeneity of variance. If they are found to be not significant (P>0.05), this means:
Need to transform data first
ANOVA is good an all that, but: Data must be independent isn’t important must be homogeneous need not be independent need not be homogeneous is important( must be independent, isn’t important, must be homogeneous, need not be independent, need not be homogeneous, is important ); Normality must be independent isn’t important must be homogeneous need not be independent need not be homogeneous is important( must be independent, isn’t important, must be homogeneous, need not be independent, need not be homogeneous, is important ); Variance must be independent isn’t important must be homogeneous need not be independent need not be homogeneous is important( must be independent, isn’t important, must be homogeneous, need not be independent, need not be homogeneous, is important ).
What conclusion(s) can be drawn from this 2-way ANOVA?
Vegetation and Disturbance have a significant interaction
Vegetation has a significant effect
Disturbance has a significant effect
No significance can be drawn from this ANOVA
Bitte warte - es wird geladen…