Using the data from 3.16. Suppose that dosage of 20g is the default dosage and ...

Using the data from 3.16. Suppose that dosage of 20g is the default dosage and the company wants to compare the effect of the other two dosages 30g and 40g with 20g, respectively. How would you perform the test? What would be your answer to the company’s question, if you want to control the overall type I error at α=0.05? Construct 95% confidence interval for the corresponding comparisons.

Data in 3.16 is here.