Let's say a student takes multiple different tests. I get a percentage for each of the test scores (the tests have a different amount of multiple choice questions, and the multiple choice questions are all different across tests).

What level of measurement can I use to compare these different percentages? Is it ordinal, interval or ratio?

This other post Measurement level of percentile scores suggests that it should be interval. However, my question is different as there are different "questionnaires".

Does having different sources for the percentages affect anything?

  • 1
    $\begingroup$ What would be the practical effect of choosing a "level of measurement"? What do you hope that would accomplish? $\endgroup$
    – whuber
    Aug 27, 2020 at 21:37
  • $\begingroup$ I'm hoping to justify what ways I can compare test scores by just looking at them. $\endgroup$ Aug 27, 2020 at 22:39


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.