Kruskal-Wallis test with very small p-values
7 views (last 30 days)
The results from the Kruskall-Wallis test I applied in my data retrives some p-values like e-218, can anyone tell me if this can be related with some kind of issue? Or it is an acceptable value? I used the command:
p = kruskalwallis(x,group)
Jeff Miller on 17 Feb 2021
p values can be anywhere between 0 and 1, depending on your data. a tiny p value like that is telling you that you would very rarely obtain such large group differences by chance if there were no real differences. if you have done everything correctly, you should be able to see that the groups are obviously different just by eye--by looking at the scores from the different groups.