spssforum.com

Hello,

I am carrying out the Kruskal-Wallis test to determine whether there is any significant difference between some datasets.

As far as I know, this can be interpreted by the p-value (given as Sig.).

However SPSS also gives the Chi-Square value in the KW output and I am not sure what it implies, and whether it should it be included in the outcome interpretation.
I would really appreciate if you could shed some light on this matter.

Many thanks.

Thank you for your quick reply.

So taking the outcome in that link as an example:

H(2) = 8.520, P = 0.014

P-value is smaller than 0.05, meaning that the 3 drugs are not giving the same results.

My problem is whether the H-value is needed to interpret the outcome? As far as I am understanding the H-value does not have to be used, as the p-value is enough.

That's probably audience and format dependent. Check with others' work at your institute or target journal. I would rather be safe to attach it and then delete later if recommended by the supervisor or editor, that way you don't need to delve into the output again.