For tests such as chi square, the significance level is usually 0.05, but I've seen some people interpret p-values for Somers'd such as .000 as significant, displaying the results similar to this: (d= .603, p<.0005)

So is the significance level for Somers'd .0005? If so, is this the same with other similar ordinal measures of association such as gamma and Kendall? Thanks!


The significance level does not depend on the type of test. It's arbitrary and set by the investigator.

And p = 0.000 is below both 0.05 and 0.005.

Displaying it as < 0.005 is a bit unusual but is not related to the type of test, either. You could do the same with any p value.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.