CHAPTER NINETEEN

NONPARAMETRIC METHODS

MULTIPLE CHOICE QUESTIONS

In the following multiple choice questions, circle the correct answer.

1. The level of measurement that allows for the rank ordering of data items is
a. nominal measurement
b. ratio measurement
c. interval measurement
d. ordinal measurement

2. The level of measurement that is simply a label for the purpose of identifying
an item is
a. ordinal measurement
b. ratio measurement
c. nominal measurement
d. internal measurement

3. The labeling of parts as "defective" or "non-defective" is an example of

a. ordinal data
b. ratio data
c. interval data
d. nominal data

4. In a questionnaire, respondents are asked to mark their marital status. Marital

status is an example of the
a. ordinal scale
b. nominal scale
c. ratio scale
d. interval scale

5. The scale of measurement that is used to rank order the observation for a
variable is called the
a. ratio scale
b. ordinal scale
c. nominal scale
d. interval scale

6. On a teacher evaluation form students are asked to rate their professor’s

performance as excellent, very good, good, and poor. This is an example of
the
a. ordinal scale
b. ratio scale

1
2 Chapter Nineteen

c. nominal scale
d. interval scale

7. Temperature is an example of a variable that uses

a. the ratio scale
b. the interval scale
c. the ordinal scale
d. either the ratio or the ordinal scale

8. The speed of an automobile is an example of a variable that uses the

a. ratio scale
b. interval scale
c. nominal scale
d. ordinal scale

9. Statistical methods that generally require very few, if any, assumptions about
the population distribution are known as
a. parametric
b. nonparametric
c. free methods
d. None of these alternatives is correct.

10. Which of the following tests would not be an example of nonparametric method?
a. Mann-Whitney-Wilcoxon test
b. Wilcoxon signed-rank test
c. sign test
d t test

11. A nonparametric version of the Parametric analysis of variance test is the

a. Kruskal-Wallis Test
b. Mann-Whitney-Wilcoxon Test
c. sign test
d. Wilcoxon Signed-rank test

12. A nonparametric method for determining the differences between two

populations based on two matched samples where only preference data is
required is the
a. Mann-Whitney-Wilcoxon test
b. Wilcoxon signed-rank test
c. sign test
d. Kruskal-Wallis Test

13. When ranking combined data in a Wilcoxon signed rank test, the data
that receives a rank of 1 is the
a. lowest value
b. highest value
c. middle value
 Nonparametric 3

d. average of the highest and the lowest of values

14. Statistical methods that require assumptions about the population are known as
a. distribution free
b. nonparametric
c. either distribution free of nonparametric
d. parametric

15. The Spearman rank-correlation coefficient is

a. a correlation measure based on the average of data items
b. a correlation measure based on rank-ordered data for two variables
c. a correlation measure based on the median of data items
d. None of these alternatives is correct.

16. A nonparametric test for the equivalence of two populations would be used
instead of a parametric test for the equivalence of the population parameters
if
a. the samples are very large
b. the samples are not independent
c. no information about the populations is available
d. The parametric test is always used in this situation.

17. A nonparametric test would be used if

a. nominal data is available
b. interval data is available
c. it is known that the population is normally distributed
d. None of these alternatives is correct.

18. If a null hypothesis that states that two populations are identical is rejected using
a nonparametric test, then it is safe to assume that
a. neither the means nor the variances are equal
b. the means of the populations are not the same
c. the variances of the populations are not the same
d. We cannot be sure of the way in which the populations differ from each other.

19. The Spearman rank-correlation coefficient for 20 pairs of data when di2 = 50
is. a. 0.0063
b. 0.0376
c. 0.9624
d. 0.9937

Exhibit 19-1
Fifteen people were given two types of cereal, Brand X and Brand Y. Two people
preferred Brand X and thirteen people preferred Brand Y. We want to determine whether
or not customers prefer one brand over the other.

20. Refer to Exhibit 19-1. The null hypothesis that is being tested is
4 Chapter Nineteen

a. H0: =5
b. H0:  = 0.5
c. H0: P=5
d. H0: P = 0.5

21. Refer to Exhibit 19-1. To test the null hypothesis, the appropriate probability
distribution to use is
a. normal
b. chi-square
c. Poisson
d. binomial

22. Refer to Exhibit 19-1. The p-value for this test is

a. 0.0005
b. 0.001
c. 0.0037
d. 0.0074

23. Refer to Exhibit 19-1. At 95% confidence, the null hypothesis should
a. be rejected
b. not be rejected
c. be revised
d. None of these alternatives is correct.

Exhibit 19-2
Students in statistics classes were asked whether they preferred a 10-minute break or to
get out of class 10 minutes early. In a sample of 150 students, 40 preferred a 10-minute
break, 80 preferred to get out 10 minutes early, and 30 had no preference. We want to
determine if there is a difference in students' preferences.

24. Refer to Exhibit 19-2. The null hypothesis that is being tested is
a. H0:  = 5
b. H0:  = 0.5
c. H0: P = 5
d. H0: P = 0.5

25. Refer to Exhibit 19-2. The mean and the standard deviation of the sampling
distribution of the number of students who preferred to get out early are
a. 50 and 30
b. 60 and 30
c. 50 and 5.478
d. 60 and 5.478

26. Refer to Exhibit 19-2. To test the null hypothesis, the appropriate probability
distribution to use is the
a. normal
 Nonparametric 5

b. chi-square
c. t distribution
d. binomial

27. Refer to Exhibit 19-2. The test statistic based on the number of students who
preferred to get out early equals
a. 1.825
b. 0.67
c. 0.82
d. 3.65

28. Refer to Exhibit 19-2. The p-value for testing the hypotheses is
a. less than 0.002
b. between 0.002 and 0.05
c. between 0.05 and 0.10
d. greater than 0.10

29. Refer to Exhibit 19-2. The null hypothesis should be

a. rejected
b. not rejected
c. revised
d. None of these alternatives is correct.

Exhibit 19-3
It is believed that the median yearly income in a suburb of Atlanta is $70,000. A sample
of 67 residents was taken. Thirty-eight had yearly incomes above $70,000, 26 had
yearly incomes below $70,000, and 3 had yearly incomes equal to $70,000. The null
hypothesis to be tested is H0: median = $70,000.

30. Refer to Exhibit 19-3. To test the null hypothesis, the appropriate probability
distribution to use is
a. normal
b. chi-square
c. t distribution
d. binomial

31. Refer to Exhibit 19-3. The mean and the standard deviation (respectively) for this
test about the median are
a. 32 and 4
b. 32 and 16
c. 33.5 and 4
d. 33.5 and 16

32. Refer to Exhibit 19-3. The test statistic has a value of

a. 1.00
b. 1.50
6 Chapter Nineteen

c. 2.00
d. 2.50

33. Refer to Exhibit 19-3. The p-value for this test is

a. 0.4332
b. 0.8664
c. 0.0668
d. 0.1336

34. Refer to Exhibit 19-3. The null hypothesis should be

a. rejected
b. not rejected
c. revised
d. None of these alternatives is correct.

Exhibit 19-4
A company advertises that food preparation time can be significantly reduced with the
Handy Dandy Slicer. A sample of 12 individuals prepared the ingredients for a meal with
and without the slicer. You are given the preparation times below.

Preparation Times
Person With Slicer Without Slicer
1 20 22
2 12 18
3 20 18
4 14 22
5 19 19
6 20 21
7 19 18
8 15 12
9 22 18
10 19 25
11 21 26
12 23 20

35. Refer to Exhibit 19-4. To test the null hypothesis, the appropriate probability
distribution to use is
a. normal
b. chi-square
c. t distribution
d. binomial

36. Refer to Exhibit 19-4. The test statistic equals

a. -0.81 or 0.81
b. -0.89 or 0.89
c. -10 or 10
 Nonparametric 7

d. -20 or 20

37. Refer to Exhibit 19-4. The p-value for this test is

a. 0.3133
b. 0.6266
c. 0.3734
d. 0.8167

38. Refer to Exhibit 19-4. The null hypothesis should

a. be rejected
b. not be rejected
c. be revised
d. None of these alternatives is correct.

Exhibit 19-5
It has been hypothesized that there is no difference in the mathematical ability of men and
women. A sample of men and women were given math tests. The scores on the tests are
given below.

Women Men
Person Score Person Score
1 95 1 80
2 86 2 87
3 100 3 93
4 100 4 95
5 99 5 97
6 98 6 82
7 88 7 89
8 92 8 86
9 94 9 75
10 89 10 82
11 79

39. Refer to Exhibit 19-5. To test the null hypothesis, the appropriate probability
distribution to use is
a. normal
b. chi-square
c. t distribution
d. binomial

40. Refer to Exhibit 19-5. The test statistic equals (using the women as population 1)
a. -5.246
b. 0.176
c. 0.722
d. 2.5
8 Chapter Nineteen

41. Refer to Exhibit 19-5. The p-value for this test is

a. 0.4938
b. 0.0062
c. 0.9876
d. 0.0124

Exhibit 19-6
Forty-one individuals from a sample of 60 indicated they oppose legalized abortion.
We are interested in determining whether or not there is a significant difference
between the proportions of opponents and proponents of legalized abortion.

42. Refer to Exhibit 19-6. The null hypothesis that is being tested is
a. H0:  = 5
b. H0:  = 0.5
c. H0: P = 5
d. H0: P = 0.5

43. Refer to Exhibit 19-6.  in this situation is

a. 60
b. 30
c. 41
d. 2

44. Refer to Exhibit 19-6.  in this problem is

a. 15
b. 5.47
c. 3.87
d. 7.45

45. Refer to Exhibit 19-6. The test statistics is

a. 3.87
b. 2.84
c. 60
d. 0.5

46. Refer to Exhibit 19-6. p-value is

a. 0.0023
b. 0.0046
c. 0.4954
d. 0.4977

47. Refer to Exhibit 19-6. The null hypothesis should be

a. rejected
b. not rejected
c. Not enough information is given to answer this question.
d. None of these alternatives.
 Nonparametric 9

48. Refer to Exhibit 19-6. The conclusion is that there

a. is no significant difference between the proportions
b. is a significant difference between the proportions
c. could be a difference in proportions, depending on the sample size
d. None of these alternatives is correct.
10 Chapter Nineteen

PROBLEMS

1. From the courthouse records, it is found that in 60 divorce cases, the filing
for divorce was initiated by the wife 41 times. Using the sign test, test for a
difference in filing between husband and wives. Let  = 0.05

2. Two employers (A and B) ranked five candidates for a new position.

Their rankings of the candidates are shown below.

Candidate Rank by A Rank by B

Nancy 2 1
Mary 1 3
John 3 4
Lynda 5 5
Steve 4 2

Compute the Spearman rank-correlation and test it for significance. Let  = 0.05.

3. The following data show the test scores of six individuals on a standardized
test before and after attending a preparation seminar for the test.

Person Before After

A 108 110
B 102 118
C 107 105
D 97 97
E 112 116
F 108 106

Use the Wilcoxon Signed-Rank test in order to determine whether or nor the
seminar has been effective. Hint: This is a one tailed test. Let  = 0.05.

4. Fifteen people were asked to indicate their preference for domestic

versus imported cars. The following data showed their preferences.

Individual Domestic vs. Imported

1 +
2 +
3 -
4 +
5 +
6 +
7 -
8 +
9 +
10 +
 Nonparametric 1

11 +
12 +
13 +
14 +
15 -

With  = 0.05, test for a significant difference in the preferences for cars. A "+"
indicates a preference for imported cars.

5. The following data show the preference of 20 people for a candidate to a

public office. A "+" indicates a preference for the Democratic candidate, and a
"-" indicates a preference for the Republican candidate.

Republicans
Individual vs. Democrats
1 +
2 -
3 +
4 +
5 +
6 +
7 -
8 +
9 +
10 +
11 +
12 -
13 -
14 +
15 +
16 -
17 -
18 +
19 +
20 +

With  = 0.05, test for a significant difference in the preference for the candidates.

6. In a sample of 120 people, 50 indicated that they prefer domestic automobiles,

60 said they prefer foreign-made cars, and 10 indicated no difference in their
preference. At a 0.05 level of significance, determine if there is evidence of a
significant difference in the preferences for the two makes of automobiles.

7. In a sample of 300 shoppers, 160 indicated they prefer fluoride toothpaste, 120
favored nonfluoride, and 20 were indifferent. At a 0.05 level of significance, test
for a difference in the preference for the two kinds of toothpaste.
12 Chapter Nineteen

8. Ten secretaries were sent to take a typing efficiency course. The following
data show the typing speeds of the secretaries before and after the course.

Typing Speed Typing Speed

Before After
Secretary the Course the Course
1 59 57
2 57 62
3 60 60
4 66 63
5 68 69
6 59 63
7 72 74
8 52 56
9 58 64
10 63 64

At  = 0.05, what can be concluded about the effectiveness of the course?

9. Ten drivers were asked to drive two models of a car. Each car was given one
gallon of gasoline. The distance that each automobile traveled on a gallon of
gasoline is shown below.

Distance Traveled (Miles)

Driver Model A Model B
1 27.7 27.1
2 28.4 28.0
3 28.9 28.7
4 27.9 27.6
5 26.5 26.0
6 29.1 29.0
7 28.9 28.2
8 28.9 28.0
9 28.8 28.0
10 28.0 27.0

At  = 0.05, what can be concluded about the performance of the two models?

10. The sales records of two branches of a department store over the last 12
months are shown below. (Sales figures are in thousands of dollars.)

Month Branch A Branch B

1 257 210
2 280 230
3 200 250
 Nonparametric 1

4 250 260
5 284 275
6 295 300
7 297 320
8 265 290
9 330 310
10 350 325
11 340 329
12 272 335

Use  = 0.05 and test to determine if there is a significant difference in the

populations of the sales of the two branches.

11. Independent random samples of ten day students and ten evening students at
a university showed the following age distributions.

Day Evening
26 32
18 24
25 23
27 30
19 40
30 41
34 42
21 39
33 45
31 35

Use  = 0.05 and test for any significant differences in the age distribution of the
two populations.

12. A PTA group wishes to determine whether a barrage of letters sent to the local
station has reduced the amount of violence broadcast between the hours of 4
P.M. and 9 P.M. The results of a survey of viewers are given here.

Response Number of Respondents

More Violence 4
Less Violence 11
No Change 6

Carry out a sign test to determine whether or not the letters were effective in
reducing the amount of violence during the 4 to 9 p.m. period. Use a .05 level of
significance. Be sure to state the null and alternative hypotheses.

13. A clothing manufacturer purchased some newly designed sewing machines in

the hopes that production would be increased. The production records of a
random
14 Chapter Nineteen

sample of workers are shown below.

Worker Old Machine New Machine

1 28 36
2 36 40
3 27 25
4 25 32
5 38 30
6 36 32
7 40 40
8 29 28
9 32 35
10 28 33
11 20 26
12 32 31
13 32 23
14 32 34
15 36 36

Use the Wilcoxon signed-rank test to determine whether the new machines have
significantly increased production. Use a .05 level of significance.

14. The president of a company wants to see if the new anti-smoking campaign is
having any influence on his employees. A sample of 100 employees who
smoked prior to the campaign is taken. Thirty-six employees said they smoked
less, 15 employees said they smoked more, and 49 employees said there was no
change.

a. State the null and alternative hypotheses.

b. Test the null hypothesis at the 1% level of significance.

15. It is believed that the median age of college students is 21 years. A sample of 80
college students is taken. Thirty of the students were under 21, 45 of the students
were over 21, and 10 were 21 years old.

a. State the null and alternative hypotheses.

b. Test the null hypothesis at the 1% level of significance.

16. We want to see if the workers on the day and night shifts differ significantly
in their productivity levels. A sample of workers from both shifts was taken.

Day Shift Night Shift

Worker Output Worker Output
1 31 1 31
2 32 2 25
3 26 3 29
4 34 4 30
5 24 5 27
 Nonparametric 1

6 35 6 33
7 39 7 37
8 28 8 39
9 44 9 38
10 42 10 36
11 40

a. State the null and alternative hypotheses.

b. Test the null hypothesis at the 5% level of significance.

17. The manager of a company believes that differences in sales performance

depends upon the salesperson's age. Independent samples of salespeople were
taken and their weekly sales record is reported below.

Below 30 Years Between 30 & 45 Years Over 45 Years

No. of Sales No. of Sales No. of Sales
24 23 30
16 17 20
21 22 23
15 25 25
19 18 34
26 29 36
27 28

a. State the null and alternative hypotheses.

b. At 95% confidence, test the hypotheses.

18. Two faculty members ranked 12 candidates for scholarships. Calculate the
Spearman rank-correlation coefficient and test it for significance. Use a .02 level
of significance.

Rank by Rank by
Candidate Professor A Professor B
1 6 5
2 10 11
3 2 6
4 1 3
5 5 4
6 11 12
7 4 2
8 3 1
9 7 7
10 12 10
11 9 8
12 8 9

19. A comprehensive statistics examination is given to 16 students in order to

determine whether or not there is a significant difference in the performance
of
16 Chapter Nineteen

students majoring in the various disciplines of Business Administration. The

following data show the scores of the 16 students (5 majoring in accounting, 6
majoring in management, and 5 majoring in marketing).

Accounting Management Marketing

91 63 95
80 92 80
70 86 70
60 75 60
85 70 90
99

At  = 0.05 level of significance, test to see if there is a significant difference in

the performance of the students in the three majors.

20. A survey of male and female students showed the following ranking of
12 professors in the management department:

Ranking By Ranking By
Professor Female Students Male Students
1 7 8
2 8 7
3 1 2
4 2 3
5 9 1
6 3 10
7 10 9
8 11 4
9 4 6
10 6 11
11 12 5
12 5 12

Do the rankings given by the female students agree with the rankings given by the
male students? Use  = 0.05.

21. Two individuals were asked to rank the performances of eight

different automobiles. The following show their rankings.

Ranking by Ranking by
Automobile First Person Second Person
1 3 2
2 5 1
3 1 4
4 6 7
5 2 5
6 4 8
7 7 6
8 8 3
 Nonparametric 1

Determine the Spearman rank-correlation coefficient; and at 95% confidence, test

for its significance.

22. Two groups of students were asked to rank the activities sponsored by the
Student Government Association on campus. The following show their rankings.

Resident Nonresident
Activity Student Ranking Student Ranking
1 3 6
2 1 2
3 8 5
4 2 1
5 5 7
6 7 8
7 4 3
8 6 4

Determine the Spearman rank-correlation coefficient and test for a significant

correlation with  = 0.05.

23. Two faculty members (X and Y) ranked five candidates for scholarships.
The rankings are shown below.

Candidate Rank By X Rank By Y

Peter 5 1
Nancy 2 2
Michael 1 3
Mary 3 5
Judy 4 4

Compute the Spearman rank-correlation and test it for significance. Let  = 0.05.

24. Three universities in your state have decided to administer the same
comprehensive examination to the recipients of MBA degrees. From each
institution, a random sample of MBA recipients has been selected and given
the test. The following table shows the scores of the students from each
university.

Northern Central Southern

University University University
56 62 94
85 97 72
65 91 93
86 82 78
93 54
77

Use the Kruskal-Wallis test to determine if there is a significant difference in the

average scores of the students from the three universities. Let  = 0.01.
18 Chapter Nineteen