Question

3. How does the price of a gallon of milk in California compare to the price in New York? The price of a gallon of milk was recorded at a random sample of stores in each state. Is there strong evidence that that average price of milk differs between the two states? a) What hypotheses would be tested? b) What kind of parametric (z, t, F, etc.) would be most appropriate for testing these hypotheses? c) What conditions need to be checked before using that test? d) If the conditions are not met, which non-parametric test(s) could be used instead?

Answer 1

a) Here we have to test, H0: = against H1: .

and are the prices of a gallon of milk in California and

New York respectively.

We test the difference of two means.

b) We have to go for unpaired t-test in this case.

c) The assumption for applying t-test of difference is that

the distribution of the two population should be

approximately normal i.e. the distribution of price of a

gallon of milk in California and New York should be normal.

d) If the normality condition is not satisfied, we should go

for Mann-Whitney U test, which is a non-parametric test.