Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Nonparametric Statistics
Chapter 13
Nonparametric Statistics
LEARNING OBJECTIVES
This chapter presents several nonparametric statistics that can be used to analyze data
enabling you to:
1.
2.
3.
4.
CHAPTER OUTLINE
13.1
Runs Test
Small-Sample Runs Test
Large-Sample Runs Test
13.2
Mann-Whitney U Test
Small-Sample Case
Large-Sample Case
13.3
13.4
Kruskal-Wallis Test
13.5
Friedman Test
13.6
KEY TERMS
Friedman Test
Kruskal-Wallis Test
Mann-Whitney U Test
Nonparametric Statistics
Parametric Statistics
Runs Test
Spearmans Rank Correlation
Wilcoxon Matched-Pairs Signed Rank Test
SOLUTIONS TO CHAPTER 13
13.1
= .05, The lower tail critical value is 6 and the upper tail critical value is 16
n1 = 10 n2 = 10
R = 11
Since R = 11 is between the two critical values, the decision is to fail to reject the
null hypothesis.
The data are random.
13.2
R =
2n1n2
2(26)( 21)
+1 =
+ 1 = 24.234
n1 + n2
26 + 21
2n1n2 ( 2n1n2 n1 n2 )
=
( n1 + n2 ) 2 ( n1 + n2 1)
R=9
z=
R R 9 24.234
=
= -4.55
R
3.351
Since the observed value of z = -4.55 < z.025 = -1.96, the decision is to reject the
null hypothesis. The data are not randomly generated.
13.3
n1 = 8
n2 = 52
= .05
This is a two-tailed test and /2 = .025. The p-value from the printout is .
0264.
Since the p-value is the lowest value of alpha for which the null hypothesis can
be rejected, the decision is to fail to reject the null hypothesis
(p-value = .0264 > .025). There is not enough evidence to reject that the data are
randomly generated.
13.4
The observed number of runs is 18. The mean or expected number of runs
is 14.333. The p-value for this test is .1452. Thus, the test is not significant
at alpha of .05 or .025 for a two-tailed test. The decision is to fail to reject the
null hypothesis. There is not enough evidence to declare that the data are not
random. Therefore, we must conclude that the data a randomly generated.
13.5
R =
R =
z=
2n1n2
2(40)( 24)
+1 =
+ 1 = 31
n1 + n2
64
2n1n2 (2n1n2 n1 n2 )
=
(n1 + n2 ) 2 (n1 + n2 1)
R R 27 31
=
= -1.08
R
3.716
Since the observed z of -1.08 is greater than the critical lower tail z value
of -1.96, the decision is to fail to reject the null hypothesis. The data are
randomly generated.
13.6
13.7
Rank
1
2.5
2.5
4
5
6
7.5
7.5
9.5
9.5
11
12.5
12.5
14
15
16
Group
1
1
2
2
1
1
1
2
1
2
1
2
2
2
1
1
n1 = 8
n2 = 8
W1 = 1 + 2.5 + 5 + 6 + 7.5 + 9.5 + 15 + 16 = 62.5
U = n1 n2 +
n1 (n1 + 1)
(8)(9)
W1 = (8)(8) +
62 .5 = 37.5
2
2
13.8
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15.5
15.5
17
18
19
20
21
22
Group
2
2
2
2
2
1
1
2
2
1
2
1
2
2
2
2
1
2
1
1
1
1
256
283
23
24
1
1
n1 = 11
n2 = 13
W1 = 6 + 7 + 10 + 12 + 17 + 19 + 20 + 21 + 22 + 23 + 24 =
W1 = 181
=
=
n1 n2 (11)(13)
=
= 71.5
2
2
n1 n2 (n1 + n2 + 1)
=
12
U = n1 n2 +
z=
(11)(13)( 25)
= 17.26
12
n1 (n1 + 1)
(11)(12)
W1 = (11)(13 +
181 = 28
2
2
28 71 .5
= -2.52
17 .26
13.9
Contacts
6
8
9
9
10
11
11
12
12
13
13
13
14
Rank
1
2
3.5
3.5
5
6.5
6.5
8.5
8.5
11
11
11
13
Group
1
1
1
2
2
1
1
1
2
1
2
2
2
15
16
17
W1 = 39
U1 = n1 n2 +
14
15
16
2
2
2
n1 (n1 + 1)
(7)(8)
W1 = (7)(9) +
39 = 52
2
2
U 2 = n1 n2 U1 = (7)(9) 52 = 11
U = 11
From Table A.13, the p-value = .0156. Since this p-value is greater than = .01,
the decision is to fail to reject the null hypothesis.
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19.5
19.5
21
22.5
22.5
24
Group
U
R
R
U
U
U
U
R
U
R
U
U
R
U
U
R
R
R
U
U
R
R
R
R
n1 = 12
n2 = 12
W1 = 1 + 4 + 5 + 6 + 7 + 9 + 11 + 12 + 14 + 15 + 19.5 + 19.5 = 123
n1 n2 (12)(12)
=
= 72
2
2
n1 n2 (n1 + n2 + 1)
=
12
U = n1 n2 +
z=
n1 (n1 + 1)
(12 )(13)
W1 = (12 )(12) +
123 = 99
2
2
U 99 72
=
= 1.56
17 .32
= .05
/2 = .025
z.025 = +1.96
Since the observed z = 1.56 < z.025 = 1.96, the decision is to fail to reject the
null hypothesis.
13.11
10
Rank
1
2
3
4
5
6
7
8
9
10
11
12.5
12.5
14
15
16
17.5
17.5
19
20
21
22
23
n1 = 11
Gender
F
F
F
F
F
F
F
F
F
M
F
F
M
F
M
M
M
M
M
M
M
M
M
n2 = 12
n1 n2 (11)(12)
=
= 66 and =
2
2
n1 n2 (n1 + n2 + 1)
=
12
(11)(12 )( 24 )
=
12
16.25
U = n1 n2 +
z=
= .01,
n1 ( n1 + 1)
(11)(12 )
W1 = (11)(12 ) +
193 .5 = 4.5
2
2
4.5 66
= -3.78
16 .25
z.01 = 2.33
Since the observed z = 3.78 > z.01 = 2.33, the decision is to reject the null
hypothesis.
11
13.12 H0: There is no difference in the price of a single-family home in Denver and
Hartford
Ha: There is a difference in the price of a single-family home in Denver and
Hartford
Price
132,405
134,127
134,157
134,514
135,062
135,238
135,940
136,333
136,419
136,981
137,016
137,359
137,741
137,867
138,057
139,114
139,638
140,031
140,102
140,479
141,408
141,730
141,861
142,012
142,136
143,947
143,968
144,500
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
City
D
H
D
H
D
H
D
H
H
D
D
H
H
H
D
H
D
H
D
D
D
D
D
H
H
H
H
H
n1 = 13
n2 = 15
W1 = 1 + 3 + 5 + 7 + 10 + 11 + 15 + 17 + 19 +
20 + 21 + 22 + 23 = 174
U = n1 n2 +
n1 ( n1 + 1)
(13)(14 )
W1 = (13)(15) +
174 = 112
2
2
n1 n2 (13)(15)
=
= 97.5
2
2
=
=
z=
12
n1 n2 (n1 + n2 + 1)
=
12
(13)(15)( 29 )
= 21.708
12
112 97 .5
= 0.67
21 .708
2
179
184
213
167
189
200
212
221
223
217
208
215
187
198
189
201
d
33
50
6
32
5
6
22
4
-3
1
26
-3
32
-2
-11
12
n = 16
T- = 3.5 + 3.5 + 2 + 9 = 18
=
Rank
15
16
7.5
13.5
6
7.5
11
5
- 3.5
1
12
-3.5
13.5
-2
-9
10
=
z=
13
T 18 68
=
= -2.59
19 .34
= .10
/2 = .05
z.05 = 1.645
Since the observed z = -2.59 < z.05 = -1.645, the decision is to reject the null
hypothesis.
13.14 Ho: Md = 0
Ha: Md 0
Before
49
41
47
39
53
51
51
49
38
54
46
50
44
49
45
After
43
29
30
38
40
43
46
40
42
50
47
47
39
49
47
d
6
12
17
1
13
8
5
9
-4
4
-1
3
5
0
-2
Rank
+9
+12
+14
+ 1.5
+13
+10
+ 7.5
+11
- 5.5
+ 5.5
- 1.5
+4
+ 7.5
-3
14
After
12,600
10,660
11,890
14,630
8,580
10,115
14,320
6,900
8,890
16,540
11,300
13,330
9,990
14,050
9,500
12,450
13,450
d
-2,100
-1,790
410
-4,120
-3,010
-965
-2,370
-160
-1,550
-3,140
900
-2,760
-110
-1,950
-500
-650
-2,950
Rank
-11
-9
3
-17
-15
-7
-12
-2
-8
-16
6
-13
-1
-10
-4
-5
-14
=
z=
= .05
9 76 .5
= -3.20
21 .12
z.05 = -1.645
Since the observed z = -3.20 < z.05 = -1.645, the decision is to reject the null
hypothesis.
13.16 Ho:Md = 0
15
Ha:Md < 0
Manual
426
387
410
506
411
398
427
449
407
438
418
482
512
402
Scanner
473
446
421
510
465
409
414
459
502
439
456
499
517
437
d
-47
-59
-11
-4
-54
-11
13
-10
-95
-1
-38
-17
-5
-35
Rank
-11
-13
-5.5
-2
-12
-5.5
7
-4
-14
-1
-10
-8
-3
-9
n = 14
T+ = (+7)
T- = (11 + 13 + 5.5 + 3 + 12 + 5.5 + 4 + 14 + 1 + 10 + 8 + 3 + 9)= 98
T = min(T+,T-) = min(7, 98) = 7
from Table A.14 with = .05, n = 14, T.05,14 = 26
Since the observed T = 7 < T.05,14 = 26, the decision is to reject the null
hypothesis.
The differences are significantly less than zero and the after scores are
significantly higher.
16
2006
54
38
38
80
53
68
43
67
73
55
58
57
63
49
39
34
66
90
57
58
d
-5
-11
1
-5
6
-1
-21
-6
-15
5
14
5
-14
-1
-20
-2
-6
-10
-2
10
Rank
-7.5
-15
2
-7.5
11
-2
-20
-11
-18
7.5
16.5
7.5
-16.5
-2
-19
-4.5
-11
-13.5
-4.5
13.5
n = 20
T+ = 2 + 11 + 7.5 + 16.5 + 7.5 + 13.5 = 58
T = 58
=
=
z=
( n)( n + 1) ( 20 )( 21)
=
= 105
4
4
n(n +1)( 2n +1)
=
24
20 (21)( 41)
= 26.79
24
T 58 105
=
= -1.75
26 .79
For = .10,
z.10 = -1.28
Since the observed z = -1.75 < z.10 = -1.28, the decision is to reject the null
hypothesis.
17
April
2006
57.1
66.4
61.8
65.3
63.5
66.4
64.9
65.2
65.1
62.2
60.3
57.4
58.2
65.3
67.2
64.1
d
5.7
0.7
3.7
2.7
3.1
-0.7
4.3
1.8
0.1
-1.5
3.1
1.8
4.7
4.1
0.1
2.7
Rank
16
3.5
12
8.5
10.5
-3.5
14
6.5
1.5
-5
10.5
6.5
15
13
1.5
8.5
n = 16
T- = 8.5
T = 8.5
=
=
z=
8.5 68
= -3.08
19 .339
3.08
18
2
165
197
204
214
183
1
1
6
4
3
9
10.5
_
Tj 33.5
nj
2
2
7.5
12
14
5
_
40.5
6
Tj
3
219
257
243
231
217
203
5
2
=
j
BY RANKS
3
18
26
23.5
19
16.5
10.5
__
113.5
6
4
286
243
259
250
279
5__
197
215
235
217
240
233
213
4
29
23.5
27
25
28
5__
7.5
15
21
16.5
22
20
13_
115
__
132.5
5
n = 29
2
K =
Tj
12
12
3( n + 1) =
(8,062 .67 ) 3(30 ) = 21.21
n(n + 1)
nj
29 (30 )
= .01
null
.01,4
df = c - 1 = 5 - 1 = 4
= 13.2767
.01,4
19
Group 2
30
38
35
24
29
Group 4
39
32
41
44
30
27
33
Group 2
8.5
14
12
4
6.5
_
45
Group 3
15
10
16
18
8.5
5
11_
83.5
By Ranks
Group 1
1
2
6.5
3
13
17
__
Tj
42.5
nj
Tj
=
j
n = 18
2
K =
Tj
12
12
3(n +1) =
(1,702 .08 ) 3(19 ) = 2.72
n(n +1)
nj
18 (19 )
= .05,
.05,2
df = c - 1 = 3 - 1 = 2
= 5.9915
.05,2
20
Region 2
$225
950
100
350
275
Region 3
$ 675
500
1,100
310
660
Region 4
$1,075
1,050
750
180
330
680
425
By Ranks
Region 1
23
12
2
18
10
4
_
Tj
69
nj
Tj
Region 2
5
19
1
9
6
Region 3
15
13
22
7
14
_
40
_
71
Region 4
21
20
17
3
8
16
11
96
n = 23
2
K =
Tj
12
12
3(n +1) =
(3,428 .27 ) 3(24 ) = 2.75
n(n +1)
nj
23 (24 )
= .05
.05,3
df = c - 1 = 4 - 1 = 3
= 7.8147
.05,3
21
City
$22,300
21,900
21,900
22,650
21,800
Suburb
$22,000
22,600
22,800
22,050
21,250
22,550
By Ranks
Small Town
4.5
12
3
10
2
__
Tj
31.5
nj
Tj
City
11
6.5
6.5
15
4.5
__
43.5
=
j
Suburb
8
14
16
9
1
13
61
n = 16
2
Tj
12
12
K =
3(n +1) =
(1,197 .07 ) 3(17 ) = 1.81
n(n +1)
nj
16 (17 )
= .05
.05,2
df = c - 1 = 3 - 1 = 2
= 5.9915
.05,2
22
Amusement Parks
2
5.5
5.5
2
11.5
5.5
2
Tj
nj
Tj
=
j
__
34
7
Lake Area
3
2
3
5
4
4
3
5
2
City
2
2
3
2
3
2
3
3
1
3
By Ranks
Lake Area
20.5
11.5
20.5
33
28.5
28.5
20.5
33
11.5
__
207.5
9
City
11.5
11.5
20.5
11.5
20.5
11.5
20.5
20.5
5.5
20.5
154.0
10
National Park
2
4
3
4
3
5
4
4
National Park
11.5
28.5
20.5
28.5
20.5
33
28.5
28.5
____
199.5
8
= 12,295.80
n = 34
2
Tj
12
12
K =
3( n +1) =
(12 ,295 .80 ) 3(35 ) = 18.99
n( n +1)
nj
34 (35 )
= .05
2.05,3 = 7.8147
df = c - 1 = 4 - 1 = 3
Since the observed K = 18.99 > 2.05,3 = 7.8147, the decision is to reject the
null hypothesis.
13.24 Ho: The 3 populations are identical
23
Day Shift
16.5
22
18
21
20
13
14.5
_
125
7
Tj
nj
Tj
=
j
Swing Shift
45
48
44
51
48
54
49
43
By Ranks
Swing Shift
7
9.5
6
14.5
9.5
19
11.5
5 _
82
8
Graveyard Shift
41
46
39
49
42
35
52
Graveyard Shift
3
8
2
11.5
4
1
16.5
___
46
7
n = 22
2
Tj
12
12
K =
3(n +1) =
(3,374 .93 ) 3(23 ) = 11.04
n( n +1)
nj
22 (23)
= .05
.05,2
df = c - 1 = 3 - 1 = 2
= 5.9915
.05,2
24
.05,4
= 9.4877
If the observed value of 2 > 9.4877, then the decision will be to reject the null
hypothesis.
Shown below are the data ranked by blocks:
1
1
1
2.5
3
4
4
3
1
2
2
3
4
4
4
3
5
5
5
5
5
2
2
2.5
1
1
Rj
11.5
12
18
25
8.5
Rj2
132.25
144
324
625
72.25
1
2
3
4
5
Rj2 = 1,297.5
r 2 =
12
2
12
R j 3b(c +1) = (5)( 5)( 6) (1,297 .5) 3(5)( 6) = 13.8
bc (c +1)
Since the observed value of r2 = 13.8 > 4,.052 = 9.4877, the decision is to
reject the null hypothesis. At least one treatment population yields larger values
than at least one other treatment population.
25
.05,5
= 11.0705
If the observed value of 2 > 11.0705, then the decision will be to reject the null
hypothesis.
Shown below are the data ranked by blocks:
1
1
3
1
1
3
1
1
3
1
2
3
5
3
3
1
3
2
1
2
3
2
1
2
4
2
2
4
2
3
4
6
6
6
6
4
6
6
6
6
5
5
4
5
5
6
5
5
5
5
6
4
2
4
2
5
4
3
4
4
15
25
25
56
50
38
Rj2 225
625
625
3136
2500
1444
1
2
3
4
5
6
7
8
9
Rj
Rj2 = 8,555.5
r 2 =
12
2
12
R j 3b(c +1) = (9)( 6)( 7) (8,555 ) 3(9)( 7) = 82.59
bc (c +1)
Since the observed value of r2 = 82.59 > 5,.052 = 11.0705, the decision is to
reject the null hypothesis. At least one treatment population yields larger values
than at least one other treatment population.
26
.01,3
= 11.3449
If the observed value of 2 > 11.3449, then the decision will be to reject the null
hypothesis.
Shown below are the data ranked by blocks:
1
2
3
4
5
6
1
1
2
1
1
1
2
2
4
3
4
3
3
3
3
3
4
3
4
4
4
4
2
1
2
2
2
1
Rj
20
22
10
Rj2
64
400
484
100
Rj2 = 1,048
r 2 =
12
R j
bc (c +1)
3b(c +1) =
12
(1,048 ) 3(6)( 5) = 14.8
(6)( 4)( 5)
Since the observed value of r2 = 14.8 > 23,.01 = 11.3449, the decision is to
reject the null hypothesis. At least one treatment population yields larger values
than at least one other treatment population.
27
.05,2
= 5.9915
If the observed value of 2 > 5.9915, then the decision will be to reject the null
hypothesis.
Shown below are the data ranked by blocks:
Worker
1
2
3
4
5
6
7
8
9
10
5-day
3
3
3
3
2
3
3
3
3
3
4-day
2
2
1
2
3
2
1
2
2
1
3.5 day
1
1
2
1
1
1
2
1
1
2
Rj
29
18
13
Rj2
841
324
169
Rj2 = 1,334
r 2 =
12
R j
bc (c +1)
3b(c +1) =
12
(1,334 ) 3(10 )( 4) = 13.4
(10 )( 3)( 4)
Since the observed value of r2 = 13.4 > 2.05,2 = 5.9915, the decision is to
reject the null hypothesis. At least one treatment population yields larger values
than at least one other treatment population.
13.29 c = 4 treatments
S=
2
r
28
b = 5 blocks
Since the p-value of .564 > = .10, .05, or .01, the decision is to fail to reject
the null hypothesis. There is no significant difference in treatments.
13.30 The experimental design is a random block design that has been analyzed using a
Friedman test. There are five treatment levels and seven blocks. Thus, the
degrees of freedom are four. The observed value of S = 13.71 is the equivalent of
r2. The p-value is .009 indicating that this test is significant at alpha .01. The
null hypothesis is rejected. That is, at least one population yields larger values
than at least one other population. An examination of estimated medians shows
that treatment 1 has the lowest value and treatment 3 has the highest value.
13.31
x
23
41
37
29
25
17
33
41
40
28
19
y
201
259
234
240
231
209
229
246
248
227
200
x Ranked
3
10.5
8
6
4
1
7
10.5
9
5
2
y Ranked
2
11
7
8
6
3
5
9
10
4
1
n = 11
rs = 1
6d 2
n(n 1)
2
=1
6( 23 .5)
= .893
11(120 )
d
d2
1
1
-.5
0.25
1
1
-2
4
-2
4
-2
4
2
4
1.5
2.25
-1
1
1
1
1
1
2
d = 23.5
13.32
x
4
5
8
11
10
7
3
1
2
9
6
y
6
8
7
10
9
5
2
3
1
11
4
d
-2
-3
1
1
1
2
1
-2
1
-2
2
29
d2
4
9
1
1
1
4
1
4
1
4
4
2
d = 34
n = 11
rs = 1
13.33
x
99
67
82
46
80
57
49
91
6d 2
n(n 1)
2
y
108
139
117
168
124
162
145
102
=1
6(34 )
= .845
11(120 )
x Ranked
8
4
6
1
5
3
2
7
y Ranked
2
5
3
8
4
7
6
1
n =8
rs = 1
6d 2
n( n 1)
2
=1
6(164 )
= -.95
8(63 )
d
d2
6
36
-1
1
3
9
-7
49
1
1
-4
16
-4
16
6
36
d2 = 164
13.34
x
92
96
91
89
91
88
84
81
83
y
x Ranked
9.3
8
9.0
9
8.5
6.5
8.0
5
8.3
6.5
8.4
4
8.1
3
7.9
1
7.2
2
n =9
rs = 1
13.35 Bank
Credit
Card
2.51
2.86
2.33
2.54
2.54
2.18
3.34
2.86
2.74
2.54
3.18
3.53
3.51
3.11
6d 2
n(n 1)
Home
Equity
Loan
2.07
1.95
1.66
1.77
1.51
1.47
1.75
1.73
1.48
1.51
1.25
1.44
1.38
1.30
y Ranked
9
8
7
3
5
6
4
2
1
=1
Bank
Cr. Cd.
Rank
12
6.5
13
10
10
14
3
6.5
8
10
4
1
2
5
d
d2
-1
1
1
1
-.5
.25
2
4
1.5
2.25
-2
4
-1
1
-1
1
1
1
2
d = 15.5
6(15 .5)
= .871
9(80 )
Home
Eq. Loan
Rank
d
1
11
2
4.5
6
7
3
7
7.5
2.5
10
4
4
-1
5
1.5
9
-1
7.5
2.5
14
-10
11
-10
12
-10
13
-8
d2
121
20.25
49
49
6.25
16
1
2.25
1
6.25
100
100
100
64
2
d = 636
n = 14
rs = 1
6d 2
n(n 2 1)
30
=1
6(636 )
14 (14 2 1)
= -.398
There is a very modest negative correlation between overdue payments for bank
credit cards and home equity loans.
13.36
Iron
Rank
12
11
3
2
4
10
9
8
7
1
6
5
Year
1
2
3
4
5
6
7
8
9
10
11
12
rs = 1
6d 2
n( n 1)
Steel
Rank
12
10
5
7
6
11
9
8
4
3
2
1
=1
d
0
1
-2
-5
-2
-1
0
0
3
-2
4
4
d2
0
1
4
25
4
1
0
0
9
4
16
16
d2 = 80
6(80 )
= 0.72
12 (144 1)
Rank NYSE
Rank AMEX
11
1
10
2
9
5
8
3
7
4
6
6
4
7
2
8
1
9
3
10
5
11
2
d = 162
n = 11
rs = 1
6d 2
n(n 1)
2
31
=1
d2
100
64
16
25
9
0
9
36
64
49
36
d
10
8
4
5
3
0
-3
-6
-8
-7
-6
= 408
6( 408 )
= -0.855
11(11 2 1)
32
13.38 = .05
H0: The observations in the sample are randomly generated
Ha: The observations in the sample are not randomly generated
n1 = 13, n2 = 21
R = 10
Since this is a two-tailed test, use /2 = .025. The critical value is: z.025 = + 1.96
R =
2n1n2
2(13)( 21)
+1 =
+ 1 = 17.06
n1 + n2
13 + 21
R =
z=
2n1n2 (2n1n2 n1 n2 )
=
(n1 + n2 ) 2 (n1 + n2 1)
R R 10 17.06
=
= -2.61
R
2.707
Since the observed z = - 2.61 < z.025 = - 1.96, the decision is to reject the null
hypothesis. The observations in the sample are not randomly generated.
13.39
Sample 1
573
532
544
565
540
548
536
523
Sample 2
547
566
551
538
557
560
557
547
= .01 Since n1 = 8, n2 = 8 < 10, use the small sample Mann-Whitney U test.
x
523
532
536
538
540
544
547
547
548
551
557
557
560
565
566
573
Rank
1
2
3
4
5
6
7.5
7.5
9
10
11.5
11.5
13
14
15
16
33
Group
1
1
1
2
1
1
2
2
1
2
2
2
2
1
2
1
W1 = 1 + 2 + 3 + 5 + 6 + 9 + 14 + 16 = 56
U1 = n1 n2 +
n1 (n1 + 1)
(8)(9)
W1 = (8)(8) +
56 = 44
2
2
U 2 = n1 n2 U1 = 8(8) - 44 = 20
34
13.40 = .05, n = 9
H0: Md = 0
Ha: Md 0
Group 1
Group 2
Rank
5.6
1.3
4.7
3.8
2.4
5.5
5.1
4.6
3.7
6.4
1.5
4.6
4.3
2.1
6.0
5.2
4.5
4.5
-0.8
-0.2
0.1
-0.5
0.3
-0.5
-0.1
0.1
-0.8
-8.5
-4.0
2.0
-6.5
5.0
-6.5
-2.0
2.0
-8.5
35
13.41 nj = 7, n = 28, c = 4, df = 3
Group 1
6
11
8
10
13
7
10
Group 2
4
13
6
8
12
9
8
Group 3
3
7
7
5
10
8
5
Group 4
1
4
5
6
9
6
7
By Ranks:
Tj
Group 1
9.5
25
17.5
23
27.5
13.5
23
139
Tj
Group 2
3.5
27.5
9.5
17.5
26
20.5
17.5
122
Group 3
2
13.5
13.5
6
23
17.5
6
81.5
Group 4
1
3.5
6
9.5
20.5
9.5
13.5
63.5
Tj
12
12
K =
3(n + 1) =
(6411 .36 ) 3(29 ) = 7.75
n(n + 1)
nj
28 (29 )
3,.01
3,.01
= 11.3449.
36
13.42 = .05, b = 7, c = 4, df = 3
.05,3
= 7.8147
Group 1
16
8
19
24
13
19
21
Group 2
14
6
17
26
10
11
16
Group 3
15
5
13
25
9
18
14
Group 4
17
9
9
21
11
13
15
By Ranks:
Blocks
1
2
3
4
5
6
7
Rj
Rj2
Group 1
3
3
4
2
4
4
4
24
Group 2
1
2
3
4
2
1
3
16
Group 3
2
1
2
3
1
3
1
13
Group 4
4
4
1
1
3
2
2
17
576
256
169
289
12
12
2
R j 3b(C +1) = (7)( 4)( 5) (1,290 ) 3(7)( 5) = 5.57
bC (C +1)
Since r2 = 5.57 < 2.05,3 = 7.8147, the decision is to fail to reject the null
hypothesis. The treatment population means are equal.
13.43
Ranks
1
101
129
133
147
156
179
183
190
2
87
89
84
79
70
64
67
71
1
1
2
3
4
5
6
7
8
2
7
8
6
5
3
1
2
4
d
d2
-6
36
-6
36
-3
9
-1
1
2
4
5
25
5
25
4
16
2
d = 152
n=8
rs = 1
6d 2
n( n 1)
2
=1
6(152 )
= -.81
8(63 )
5 Gal.
2.9
2.5
2.6
2.2
2.1
2.0
2.7
10 Gal.
3.1
2.4
3.0
2.3
2.9
1.9
5 Gal.
17.5
14
15
11
9.5
8
16
91
10 Gal.
20
13
19
12
17.5
6.5
By Ranks
1 Gal.
1
3.5
5
2
6.5
3.5
9.5
Tj 31
nj 7
88
37
Tj
=
j
38
n = 20
2
K =
Tj
12
12
3( n +1) =
( 2,610 .95 ) 3(21) = 11.60
n(n +1)
nj
20 (21)
= .01
df = c - 1 = 3 - 1 = 2
= 9.2104
.01,2
.01,2
13.45 N = 40 n1 = 24 n2 = 16 = .05
Use the large sample runs test since both n1, n2 are not less than 20.
H0: The observations are randomly generated
Ha: The observations are not randomly generated
With a two-tailed test, /2 = .025, z.025 = + 1.96. If the observed z > .196
or < -1.96, the decision will be to reject the null hypothesis.
R = 19
R =
R =
z=
2n1n2
2( 24)(16)
+1 =
+ 1 = 20.2
n1 + n2
24 + 16
2n1n2 ( 2n1n2 n1 n2 )
=
(n1 + n2 ) 2 ( n1 + n2 1)
R R 19 20 .2
=
= -0.40
R
2.993
Since z = -0.40 > z.025 = -1.96, the decision is to fail to reject the null hypothesis.
39
Machine 1
231
233
229
232
235
234
236
Machine 2
229
232
233
235
228
237
233
Machine 3
234
231
230
231
232
231
230
Machine 1
2
1
2
3
2
3
Machine 2
1
3
3
1
3
2
Machine 3
1
2
1
2
1
1
By ranks:
Operator
1
3
4
5
6
7
Rj
Rj2
16
256
If the observed
df = c - 1 = 2
.05,2
15
225
11
121
= 5.99147.
12
2
12
R j 3b(c +1) =
(602 ) 3(7)( 4) = 2
bc (c +1)
(7)( 3)( 4)
Since 2r = 2 <
hypothesis.
.05,2
40
Rank
1
2
3
4
6
6
6
9
9
9
11
13
13
13
15
16.5
16.5
18.5
18.5
20
21
22
n1 = 10
Group
1
1
1
1
1
2
2
1
2
2
2
2
2
2
1
2
2
1
2
1
2
1
n2 = 12
W1 = 1 + 2 + 3 + 4 + 6 + 9 + 15 + 18.5 + 20 + 22 = 100.5
n1 n2 (10)(12)
=
= 60
2
2
n1 n2 (n1 + n2 + 1)
=
12
U = n1 n2 +
z=
(10)(12 )( 23)
= 15.17
12
n1 (n1 + 1)
(10 )(11)
W1 = (10 )(12 ) +
100 .5 = 74.5
2
2
74 .5 60
= 0.96
15 .17
1.645
41
null hypothesis.
13.48 Ho: The population differences = 0
Ha: The population differences 0
With
1180
874
1071
668
889
724
880
482
796
1207
968
1027
1158
670
849
559
449
992
1046
852
Without
1209
902
862
503
974
675
821
567
602
1097
962
1045
896
708
642
327
483
978
973
841
d
-29
-28
209
165
-85
49
59
-85
194
110
6
-18
262
-38
207
232
-34
14
73
11
Rank
-6
-5
18
15
-12.5
9
10
-12.5
16
14
1
-4
20
-8
17
19
-7
3
11
2
n = 20
T- = 6 + 5 + 12.5 + 12.5 + 4 + 8 + 7 = 55
T = 55
=
=
z=
( n)( n + 1) ( 20 )( 21)
=
= 105
4
4
n(n +1)( 2n +1)
=
24
20 (21)( 41)
= 26.79
24
T 55 105
=
= -1.87
26 .79
= .01, /2 = .005
z.005 = 2.575
Since the observed z = -1.87 > z.005 = -2.575, the decision is to fail to reject the
null hypothesis.
42
Rank
1
2
3
4
5
6
7
8
9.5
9.5
11
12.5
12.5
14
15.5
15.5
17
18
19
20
Month
J
M
J
J
M
M
M
J
M
J
M
M
J
M
M
J
J
M
J
J
n2 = 10
n1 (n1 + 1)
(10 )(11)
W1 = (10 )(10 ) +
109 .5 = 45.5
2
2
From Table A.13, the p-value for U = 45 is .3980 and for 44 is .3697. For a
two-tailed test, double the p-value to at least .739. Using = .10, the decision is
to fail to reject the null hypothesis.
43
1
176
156
203
183
147
190
.05,3
= 7.8147
Location
2
3
58
111
62
98
89
117
73
118
46
101
83
113
4
120
117
105
113
114
115
By ranks:
Location
Brand
A
B
C
D
E
F
Rj
Rj2
1
4
4
4
4
4
4
24
576
2
1
1
1
1
1
1
3
2
2
3
3
2
2
4
3
3
2
2
3
3
6
36
14
196
16
256
Rj2 = 1,064
r 2 =
12
2
12
R j 3b(c +1) =
(1,064 ) 3(6)( 5) = 16.4
bc (c +1)
(6)( 4)( 5)
Since r2 = 16.4 > 2.05,3 = 7.8147, the decision is to reject the null hypothesis.
At least one treatment population yields larger values than at least one other
treatment population. An examination of the data shows that location one
produced the highest sales for all brands and location two produced the lowest
sales of gum for all brands.
44
No Box
170
112
90
87
51
49
11
40
165
82
94
139
21
55
60
52
d
15
-3
2
18
9
-4
14
18
-4
26
-5
-16
13
13
-1
26
Rank
11
-3
2
13.5
7
-4.5
10
13.5
-4.5
15.5
-6
-12
8.5
8.5
-1
15.5
n = 16
T- = 3 + 4.5 + 4.5 + 6 + 12 + 1 = 31
T = 31
=
=
z=
= 19.34
T 31 68
=
= -1.91
19 .34
= .05, /2 = .025
z.025 = 1.96
Since the observed z = -1.91 > z.025 = -1.96, the decision is to fail to reject the
null hypothesis.
13.52
Cups
25
41
16
0
11
28
34
18
5
Stress
80
85
35
45
30
50
65
40
20
Ranked
Cups
6
9
4
1
3
7
8
5
2
Ranked
Stress
8
9
3
5
2
6
7
4
1
45
d
-2
0
1
-4
1
1
1
1
1
d2 =
d2
4
0
1
16
1
1
1
1
1
26
n=9
rs = 1
6d 2
n(n 1)
2
=1
6(26 )
= .783
9(80 )
= .05, /.025
H0: The observations in the sample were randomly generated.
Ha: The observations in the sample were not randomly generated
From Table A.11, lower tail critical value = 10
From Table A.12, upper tail critical value = 22
R = 21
Since R = 21 between the two critical values, the decision is to fail to reject the
null hypothesis. The observations were randomly generated.
46
After
465
475
535
410
425
505
450
480
520
430
460
500
530
d
-35
10
-15
-50
15
-5
-25
-10
-5
0
-10
-5
10
Rank
-11
5.5
- 8.5
-12
8.5
-2
-10
-5.5
-2
OMIT
-5.5
-2
5.5
n = 12
T+ = 5.5 + 8.5 + 5.5 = 19.5
T = 19.5
From Table A.14, using n = 12, the critical T for = .01, one-tailed, is 10.
Since T = 19.5 is not less than or equal to the critical T = 10, the decision is to fail
to reject the null hypothesis.
47
Rank
1
2
3.5
3.5
5
6.5
6.5
9
9
9
11.5
11.5
13
15.5
15.5
15.5
15.5
19
19
19
21
22
Group
2
2
2
2
2
2
1
1
1
2
1
2
2
1
1
1
2
1
1
2
1
1
n2 = 11
=
=
n1 n2 ( n1 + n2 + 1)
=
12
U = n1 n2 +
z=
(11)(11)( 23)
= 15.23
12
n1 (n1 + 1)
(11)(12 )
W1 = (11)(11) +
163 .5 = 23.5
2
2
23 .5 60 .5
= -2.43
15 .23
2.43
48
Rank
1
2
3
4
5
6
7.5
7.5
9
10
11
12
13
14
15
16
17
18
Type of Dispenser
M
M
M
A
M
M
A
M
M
M
A
M
A
A
A
A
A
A
n2 = 9
W1 = 4 + 7.5 + 11 + 13 + 14 + 15 + 16 + 17 + 18 = 115.5
U1 = n1 n2 +
n1 ( n1 + 1)
(9)(10 )
W1 = (9)(9) +
115 .5 = 10.5
2
2
U 2 = n1 n2 U1 = 81 10.5 = 70.5
= .01
From Table A.13, the p-value = .0039. The decision is to reject the null
hypothesis since the p-value is less than .01.
49
55
228
224
225
222
226
225
70
219
220
221
223
224
85
218
216
217
221
218
217
55
23
18.5
20.5
16
22
20.5
120.5
70
11.5
13
14.5
17
18.5
74.5
85
9
4
6.5
14.5
9
6.5
49.5
By Ranks:
Tj
45
4
2
9
4
11.5
1
31.5
nj
Tj
=
j
= 4,103.84
n = 23
2
Tj
12
12
K =
3(n +1) =
(4,103 .84 ) 3( 24 ) = 17.21
n(n +1)
nj
23( 24 )
= .01
.01,3
df = c - 1 = 4 - 1 = 3
= 11.3449
.01,3
50
13.58
Sales
150,000
210,000
285,000
301,000
335,000
390,000
400,000
425,000
440,000
Miles
1,500
2,100
3,200
2,400
2,200
2,500
3,300
3,100
3,600
Ranks
Sales
1
2
3
4
5
6
7
8
9
n=9
rs = 1
6d 2
n(n 1)
2
=1
6(26 )
= .783
9(80 )
Ranks
Miles
1
2
7
4
3
5
8
6
9
d
0
0
-4
0
2
1
-1
2
0
d2
0
0
16
0
4
1
1
4
0
d2 = 26
51
Quality
27
38
25
40
31
19
35
Mgmt. Inv.
16
21
18
28
29
20
31
By Ranks:
3-day
2
4
7
3
12
5
1
Tj
34
nj
Tj
7
2
Quality
14
20
13
21
17.5
9
19
113.5
Mgmt. Inv.
6
11
8
15
16
10
17.5
83.5
n = 21
2
K =
Tj
12
12
3( n +1) =
(3,001 .5) 3(22 ) = 11.96
n(n +1)
nj
21(22 )
= .10
.10,2
df = c - 1 = 3 - 1 = 2
= 4.6052
.10,2
52
Wives
35
29
30
20
27
31
22
19
29
28
21
22
33
38
34
31
36
31
d
-8
-7
-2
-1
1
-2
-4
2
-4
-10
-1
2
-10
-13
-12
-15
-13
-1
Rank
-12
-11
-6.5
-2.5
2.5
-6.5
-9.5
6.5
-9.5
-13.5
-2.5
6.5
-13.5
-16.5
-15
-18
-16.5
-2.5
n = 18
T+ = 2.5 + 6.5 + 6.5 = 15.5
T = 15.51
=
=
z=
= .01
15 .5 85 .5
= -3.05
22 .96
z.01 = -2.33
Since the observed z = -3.05 < z.01 = -2.33, the decision is to reject the null
hypothesis.
53
13.61 This problem uses a random block design, which is analyzed by the Friedman
nonparametric test. There are 4 treatments and 10 blocks. The value of the
observed r2 (shown as S) is 12.16 (adjusted for ties) and has an associated
p-value of .007 that is significant at = .01. At least one treatment population
yields larger values than at least one other treatment population. Examining the
treatment medians, treatment one has an estimated median of 20.125 and
treatment two has a treatment median of 25.875. These two are the farthest apart.
13.62 This is a Runs test for randomness. n1 = 21, n2 = 29. Because of the size of the
ns, this is a large sample Runs test. There are 28 runs, R = 28.
R = 25.36
z=
R = 3.34
28 25 .36
= 0.79
3.34
The p-value for this statistic is .4387 for a two-tailed test. The decision is to fail
to reject the null hypothesis at = .05.
are
54
13.64 A Kruskal-Wallis test has been used to analyze the data. The null hypothesis is
that the four populations are identical; and the alternate hypothesis is that at least
one of the four populations is different. The H statistic (same as the K statistic) is
11.28 when adjusted for ties. The p-value for this H value is .010 which indicates
that there is a significant difference in the four groups at = .05 and marginally
so for = .01. An examination of the medians reveals that all group medians are
the same (35) except for group 2 that has a median of 25.50. It is likely that it is
group 2 that differs from the other groups.