Can you reject the null hypothesis at a=.05?

Question

Ms Z has two groups of band students. She asks Group 1 to use her new embouchure strengthening cream before practice and asks Group 2 to practice as usual. Can you reject the null hypothesis at a=.05? why?

Answer 1

You don't have the information to answer that question.

The information you're missing is:

1. the sample size of the 2 groups
2. the endpoint (how is she measuring the strength of the embouchure before & after using the cream): is she using the difference between the two, or just comparing the measurement made at the end of the study between the 2 groups.
3. some information regarding the variability of the endpoint
4. what test statistic is she using? the appropriateness of that statistic depends on what the endpoint is.

I hope that's helpful for you to defend the answer of "I need more information to answer that question."

Good Luck, Marla

Answer 2

In your question, you asked if one can reject given alpha of 0.05. That probably is meant to read p=0.05. The way the alpha/p-value dynamic is designed is such that p must be lower than alpha to reject the null.

If you're being a purist, don't reject the null if p=0.05. In practice, I would bet that most researchers would write about this finding as if it were significant (assuming appropriate power).

https://www.youtube.com/watch?v=3SEy6nGTpqo