if a researcher were studying the effects of a teaching method on patient learning outcomes, how must he or she word the research question (different from hypothesis) to use the t-test for statistical differences? What type of data must he or she collect (interval, ratio, ordinal, nominal)?

Answer 1

I don't see any need to reward research hypothesis, 9if it is adequately stated.

Assuming that your dependent variable is some test results, you would need at least ordinal data.

http://www.members.cox.net/dagershaw/lol/TwoPlusTwo.html