0
$\begingroup$

I find some references and discussion state that Chi-square test used for comparison analysis, but some said for correlation also. I am confused, so Chi-square is test for correlation or comparison? or both? If both, when we used it for correlation or for comparison beside of based on hypothesis (association or comparison)?

Also, if we run Chi-square test in SPSS, it found at crosstabs menu and non-parametric test which are different way to settings the analysis. Which menu should used for correlation or comparison?

$\endgroup$

1 Answer 1

0
$\begingroup$

I personally have not heard the chi square test being used to measure correlation. That said, if you want to test if the population correlation is 0, the sample correlation is asymptotically a centered Normal under the null. So its square will be a scaled chi square and hence I can see that the test statistic will be chi squared under the null.

The Pearson chi square test can be used to test if a sample comes from a specified distribution, which is I guess what you mean by comparison?

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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