1
$\begingroup$

I have one variable that is categorical (with 5 levels), and another variable that consists of either mean or sum scores (0 to 28) due to the way you have to calculate the scores of this measure.

Can I perform a chi-squared test with my categorical data and sum scores?

$\endgroup$
2
  • 1
    $\begingroup$ Do you have the original data that the means were calculated from? Do you just want to test for association, or do you think of 1 of the variables as a response? $\endgroup$ Mar 10, 2015 at 22:17
  • $\begingroup$ I do have the original data, which are responses in three areas of X that are computated together to give one score of X. I want to test for association. $\endgroup$
    – UserXbar
    Mar 11, 2015 at 11:34

1 Answer 1

1
$\begingroup$

You cannot use a chi-squared test in this way. The simplest approach would be to use a one-way ANOVA, with the categorical variable as your independent variable and the combined X as your response. You could also do a MANOVA, using the three areas of X as distinct multiple responses. However, that would be more complicated, and if there is an a-priori theoretical reason to see the combined X as a single variable, you are OK to skip it.

$\endgroup$

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.