I'm currently working with a dataset with a sample size of 174 from a questionnaire that measures 6 types of personality traits with 5 response options per item (Strongly Agree, Agree, Neutral, Disagree, Strongly Disagree)

I've generated a polychoric correlation matrix to inspect each item and I need some assistance with the interpretation.

Specifically I'm looking at the correlation between two items: Item 1 reads "Compares their successes to other's" and another reads "questions things learned."

During my reliability analysis, R recommended that item 1 be reverse scored, so that's the scoring reflected in the polychoric matrix:

enter image description here

Given the strong positive correlation between these two items, my question is, given the reverse scoring, could we interpret the correlation between items 1 and 3:

As people DO compare themselves to others (reverse scored), the more they feel like they are questioning things learned


As people DON'T compare themselves to others, the more they feel like they are contributing. I'm just not sure on how to interpret the direction of the relationship in this case.


1 Answer 1


Given that you recoded the item, the "original" (pre-recoding) correlation should have a negative sign, indicating a negative relationship. You can easily verify this by looking at a bivariate scatterplot with the original (unrecoded) variable.

  • $\begingroup$ Yes you are correct. The original pre-coded item does indeed indicate a negative relationship. Would it be fair to say (if we're looking at the pre-coded item) to say that the less often people compare themselves to others, the more they question things (or vice versa)? $\endgroup$
    – Darko
    Jun 27, 2023 at 14:48
  • 1
    $\begingroup$ From your description of your original variables, that interpretation sounds correct to me. Good luck with your work! $\endgroup$ Jun 27, 2023 at 14:50

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.