In the final run, we were cheap and only collected a sample of size n = 5. However, the other sample statistics happened to turn out the same. We assume was not known.

Our new sample statistics were:

n = 5, = 110 mg, s = 27 mg

And the resulting 95% confidence interval was: (76.47, 143.53)

a) Based on this scenario, define an appropriate null hypothesis and an appropriate

alternative hypothesis for a statistical test to determine whether the mean caffeine

content is 100 mg.

b) State the appropriate conclusion of this hypothesis test using the data from this

sample. (e.g. Do you reject the null hypothesis?). What did you base this conclusion

on?

c) Would you expect the p-value of this test to be above or below 0.05? Justify your

answer.)