Question

Use MegaStat, MINITAB, or another software package to perform Tukey’s test for significant pairwise differences. Perform the test using both the 5 percent and 1 percent levels of significance.

One particular morning, the length of time spent in the examination rooms is recorded for each patient seen by each physician at an orthopedic clinic.

Time in Examination Rooms (minutes)
Physician 1	Physician 2	Physician 3	Physician 4
34	32	19	26
22	34	28	35
25	36	30	32
32	30	25	29
24	40	26	30
37	31	30	35
19		24	40
		27

(a) Calculate the mean for each group and the Tukey test statistic T_calc for each pair. Provide the critical values for both α = .05 and α = .01. (Input the mean values within the input boxes of the first row and input boxes of the first column. Input T_calc in the appropriate boxes in the table. Round all answers to two decimal places.)

Post hoc analysis:

Tukey simultaneous comparison t-values (d.f. = 24)
		Physician 3	Physician 1	Physician 4	Physician 2
Physician 3
Physician 1
Physician 4
Physician 2
Critical values for experimentwise error rate:
		0.05
		0.01

(b) Use Tukey simultaneous comparison t-values and choose the correct answer.

• Physicians 1 and 4 differ

• Physicians 1 and 3 differ

• Physicians 2 and 4 differ

• Physicians 2 and 3 differ

(c) The Tukey test is used for simultaneously comparing means because it offers good power and maintains the desired overall probability of Type I error.

• True

• False

Answer 1

putting above data and from Megastat :Analysis of Variance -One way ANOVA:

Tukey simultaneous comparison t-values (d.f. = 24)
		Physician 3	Physician 1	Physician 4	Physician 2
	mean	26.1	27.6	32.4	33.8
Physician 3	26.1
Physician 1	27.6	0.58
Physician 4	32.4	2.51	1.88
Physician 2	33.8	2.95	2.32	0.52
critical values for experimentwise error rate:
		0.05	2.76
		0.01	3.47

b)

for alpha =0.05:

Physicians 2 and 3 differ

c)

true (since it adjust for pair wise error rate)