Levene's test - SPSS (part1)

I demonstrate how to perform and interpret *three* different Levene's tests within SPSS. The "regular" Levene's test available through the ANOVA procedure, as well as two versions of Levene's that are robust to non-normal distributions and unequal sample sizes. The first robust Levene's test is known as the Brown-Forsythe test. The second robust test is known as the nonparametric Levene's test, which is based on ranked data.

[howstats]

Technically, Kruskal Wallis does assume homogeneity of variance (or homogeneity of distributional shapes, which is implied by homogeneity of variance). Kruskal Wallis is good when you have non-normally distributed data, but the distributions across groups have to be distributed non-normally in the same way. If this is not true, then you should consider doing the Median test. I have a video on that too.

[deviation]

I found this really helpful. Thanks.

[MrSchmouz]

Very well explained! Thanks a lot man!

[oTBePka]

what do I do when I get significant Levene's on mixed ANOVA :(((

[Non-disjunction]

My p-values are all well below 5%. So that means the variance is statistically different, and therefore I cannot use anova?

[manodemetate]

hi, thanks for your vids, so is this first test you perform ideal for normally distributed data?

[jangelgg]

Estimated. that reference books recommend to identify the types of post hoc for nonparametric analysis. Or if this can recommend others as Nemenyi test for post hoc analysis. Thanks for your attention

[KuldeepSingh-xtku]

I have not find any option next to descriptive analysis ?

[thoughtclaw]

This is the second video I've watched showing how to do a Levene's test on SPSS, and the second one that didn't tell me what the threshold is for statistical significance. It would have been so easy to say that. I'm so frustrated!

[craftychinurse]

This one was too fast. I loved all the others. Thank you.

[TheShipz]

Hi thanks for your video, had a quick question i need help with. what happens if my lavene test is actually significant? I had to perform the kruskal wallis test and my lavene stats have come up significant, does that mean its pointless then reporting my results or do i just continue as normal and make a note that my lavene test is sig. Thank you

[mytwocents]

Go to 3:09 Homogeneity of variance test.

[edelynreonal]

anyone here knows how to perform homogeneity of variance for paired t test?

[godwinnnaemeka]

Your interpretation of your result is vague, its not clear and therefore not explicit