Question
a. If an automotive analyst wished to show that damage to SUVs with less than a three-star rating cost over $1500 more than SUVs with a three-star rating, how would you set up the null and alternative hypotheses?
b. If a random sample of 25 SUVs with a three-star rating that were involved in collisions showed an average damage amount of $5810 and a random sample of 25 SUVs with less than a three-star rating also involved in collisions showed an average damage amount of $8000, would these data support the alternative hypothesis in question a?
Use a significance level of 10%. Assume that the data are from normally distributed populations with known standard deviations of $1450 for the SUVs with three-star ratings and $1625 for the SUVs with less than three-star ratings.
c. Suppose that in question b, the standard deviations of $1450 and $1625 were computed from the sample. Using a t test, would the results of the test change? Explain. (Note: no calculations required).
Question
Problem 12-7:
Sample 1 Sample 2 Sample 3
14 17 17 13 16 14 12 16 15
15 18 16 16 14 16
A. Conduct a one way analysis of variance use alpha = .05
B. If warranted, use the tukey kramer to determine which populations have different means use an experiment wide error of .05