I have two likert scale questions:

1: not well at all, a little likely, somewhat likely, likely, very likely

2: not well at all, a little, somewhat, quite well, very well 

I am trying to see if there is a relationship/correlation between the responses to the two questions. How can I test this?

  • $\begingroup$ The short answer is that you can probably just use a chisq test of independence and then assess the contingency table for direction. $\endgroup$ – RickyB Apr 7 '18 at 14:44

For correlation, you can use rank correlation measures such as Spearman's rank correlation or Kendall's Rank Correlation. Depending on the relationships you are looking for, there are other means of looking for a relationship such as comparing means using ANOVA methods.

| cite | improve this answer | |

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