I want to know how to perform a power analysis in the following case to know if my sample is good enough or not.

In my faculty, we have decided to apply a new lecturing methodology to a set of students, and we hypothesized that if we apply this methodology, then more than half of the students will have a drop in their marks.

However, we were only be able to test this methodology in a group of 30 students, because another classroom of approximately the same number of students was not able to follow this new methodology. We have also applied an h-test in our group, but we have received the critique that the sample size might be too small for such a text.

How to tackle this situation?


1 Answer 1


Now you have the data the concept of power becomes much less relevant as it is more of a planning issue. You can carry out your test and see what happens. It may well be that you cannot detect even quite a scientifically meaningful difference with what you have but that is life. You ask what you should do and I am afraid if the anonymous critic persists your only remedy may be to run the experiment on more classes.


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.