statistics in health care
posted by Jean on .
If a researcher were studying the effects of a teaching method on patient learning outcomes, how would the research question need to be worded to use a ttest to test for statistical differences?

Ho: There is no difference in outcomes between patients who have taught "X" method and those who have not.
H1: There is a difference.
However, you need to define the specific problem(s) that the patients have, what your teaching method is, and what criterion you are using as a measure of change.
The comparison of the amount of change in your criterion from pretest to posttest (after the teaching or no teaching) would be used for your ttest. Since I assume you are looking only for more positive change with patients who were taught, you would be dealing with a onetailed test.
I hope this helps. Thanks for asking.