In the book Business Research Methods, Donald R. Cooperand C. William Emory (1995) discuss a manager who wishes to comparethe effectiveness of two methods for training new salespeople. Theauthors describe the situation as follows:

The company selects 22 sales trainees who are randomly dividedinto two equal experimental groups—one receives type A and theother type B training. The salespeople are then assignedand managed without regard to the training they have received. Atthe year’s end, the manager reviews the performances of salespeoplein these groups and finds the following results:

	A Group	B Group
Average Weekly Sales	x??1x¯1 = $1,350	x??2x¯2 = $1,086
Standard Deviation	s1 = 233	s2 = 263

(a) Set up the null and alternative hypothesesneeded to attempt to establish that type A trainingresults in higher mean weekly sales than does type Btraining.

H₀: µ_A ?µ_B ?  versus H_a:µ_A ?µ_B  >

(b) Because different sales trainees areassigned to the two experimental groups, it is reasonable tobelieve that the two samples are independent. Assuming that thenormality assumption holds, and using the equal variancesprocedure, test the hypotheses you set up in part a atlevel of significance .10, .05, .01 and .001. How much evidence isthere that type A training produces results that aresuperior to those of type B? (Round your answer to3 decimal places.)

t =

(Click to select)RejectDo not reject H0with ? equal to .10.

(Click to select)Do not rejectReject H0with ? equal to .05

(Click to select)Do not rejectReject H0with ? equal to .01

(Click to select)Do not rejectReject H0with ? equal to .001

(Click to select)WeakVery strongNoStrongExtremelystrong  evidence that µA ? µB > 0

(c) Use the equal variances procedure tocalculate a 95 percent confidence interval for the differencebetween the mean weekly sales obtained when type Atraining is used and the mean weekly sales obtained when typeB training is used. Interpret this interval.(Round your answer to 2 decimal places.)

Confidence interval [, ]