Once upon a time a couple years B.C. (BC stands for before coronavirus), when I was a student, I asked my advisor about using two questionnaires in my research. One was some emotional intelligence questionnaire and the other one was Rosenberg Self-Esteem Scale. The goal was to check if high self-esteem people have higher emotional intelligence.

She said" Sure, you can, but you need to remember that 'self report measures always correlate with themselves'. She meant that even if you give two really different, conceptually unrelated questionnaires there will always be some correlation - the reason for that wasn't the correlational noise (or crud factor) but the fact, that there was the same people who answer both set of items. Q: Have you ever heard that? Is it a sort of another research legend or maybe it is true? It seems compelling, but I'm a bit sceptical about how it could work.

  • 1
    $\begingroup$ Self-report measures can be designed to be orthogonal (not correlate), as may be done for sub-scales for example. Here are the correlations between the sub-scales of the Big 5 personality inventory: researchgate.net/figure/… - I don't think this was done purposefully, but you can see that some correlate, and some don't. $\endgroup$
    – Arnon Weinberg
    Commented Aug 8, 2022 at 19:31
  • $\begingroup$ This is an interesting question and I will be eager to see what answers come out of it $\endgroup$ Commented Aug 10, 2022 at 6:45


Your Answer

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

Browse other questions tagged or ask your own question.