Testing the Assumption of Similar Distribution Shape for the Kruskal Wallis H Test using SPSS

[@lucyperry8075]

Great video, super clear and thorough!

[@joelaffoon6656]

what do you do if the 4th assumption is violated? As in, the shape of the distributions are different.

[@kelvinyong7555]

You continue with the Kruskal Wallis test. However, when you interpret it later, you do no compare the medians of the independent variables, but instead only compare the rank means.

[@Lozzalo66]

I am in the same boat. I have violated assumption number 4, and doing a Kruskall-Wallis test it has retained the null hypotheses. Do i then need to interpret the mean ranks in my output? They don't quite make sense to me in their values.

[@nicolastabchouri6985]

Thank you for this video. It was very helpful. I have one question though, how do you include the post test (mann whitney analysis between groups) in your box plots ? other statistics programs such as prism graph pad usually show an asterisk when there is a statistically significant difference found between two groups.