16. To use the t-statistic to test for a difference between the means of
two populations, what assumptions must be made about the two
populations? About the two samples?
17.
18. Independent random samples are selected from two populations
and are used to test the hypothesis H 0 : (µ 1 - µ 2 ) = 0 against the
alternative H a : (µ 1 - µ 2 ) ≠ 0. An analysis of 234 observations from
population 1 and 310 from population 2 yielded a p-value of 0.113.
Complete parts a and b below.
19. A study was done to examine whether the perception of service
quality at hotels differd by gender. Hotel guests were randomly
selected to rate service items on a 5-point scale. The sum of the items
for each guest was determined and a summary of the guest scores are
provided in the table. Complete parts a and b.
20. To determine if winning a certain award leads to a challenge in
life expectancy, researches sampled 748 award winners and matched
each one with another person of the same sex who was in the same
profession and was born in the same era. The lifespan of each pair
was compared. Complete parts a through c below.
21. A new testing method was developed to reduce a certain ratio.
The data in the table show the ratios that resulted from testing six
components using the standard method and the new method. Compare
the two methods with a 90% confidence interval. Which method has
the smaller mean ratio?
22. Consider making an interference about p 1 – p 2 , where there are
x 1 successes in n 1 binomial trails and x 2 succeseses in n 2 binomial
trails.
23. Construct a 90% confidence interval for (p 1 – p 2 ) in each of the
following situations.
24. In auction bidding the “winner’s curse” is the phenomenon of the
winning (or highest) bid price being above the expected value of the
item being auctioned. A study was conducted to see if less-
experienced bidders were more likely to be impacted by the curse