卡方檢定 CHI-SQUARE TEST

1. Use the sample data below to test the hypotheses

H0: p1 = p2 = p3
Ha: Not all population proportions are equal
where pi is the population proportion of Yes responses for population i. Using a .05 level of significance,
what is the p-value and what is your conclusion?

[sol] H0:  
p1  p2  p3
Ha:  Not all population proportions are equal
Observed Frequencies (fij)
1 2 3 Total
Yes 150 150 96 396
No 100 150 104 354
Total 250 300 200 750

m
Expected Frequencies (eij)

er as
1 2 3 Total

co
eH w
Yes 132.0 158.4 105.6 396
No 118.0 141.6 94.4 354

o.
Total 250 300 200 750
rs e
Chi Square Calculations (fij – eij)2 / eij
ou urc
1 2 3 Total
Yes 2.45 .45 .87 3.77
o

No 2.75 .50   .98 4.22

                    2
aC s

  7.99
v i y re

Degrees of freedom = k – 1 = (3 – 1) = 2
Using the   table with df = 2, = 7.99 shows the p–value is between .025 and .01  
2 2
Using Excel or Minitab, the p–value corresponding to  = 7.99 is .0184
ed d

2
ar stu

p–value  .05, reject H0.  Conclude not all population proportions are equal.


2. Kate Sanders, a researcher in the department of biology at IPFW University, studied the effect of
sh is

agriculture contaminants on the stream fish population in Northeastern Indiana (April 2012). Specially
Th

designed traps collected samples of fish at each of four stream locations. A research question was, Did
the differences in agricultural contaminants found at the four locations alter the proportion of the fish
population by gender? Observed frequencies were as follows.

a. Focusing on the proportion of male fish at each location, test the hypothesis that the population
proportions are equal for all four locations. Use a .05 level of significance. What is the p-value and
what is your conclusion?
b. Does it appear that differences in agricultural contaminants found at the four locations altered the fish
1

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 population Gender A B C D Total by gender?
[sol] a. H0: Male 49 44 49 39 181  
p1  p2  p3  p4
Female 41 46 36 44 167
               Ha: 90 90 85 83        348 Not all population 
proportions are equal
Observed Frequencies (fij)

Expected Frequencies (eij)
Gender A B C D Total
Male 46.81 46.81 44.21 43.17 181
Female 43.19 43.19 40.79 39.83 167
90 90 85 83  348

Chi Square Calculations (fij – eij)2 / eij
Gender A B C D Total

m
er as
Male .10 .17 .52 .40 1.19
Female .11 .18 .56 .44 1.29

co
eH w
 2  2.49
Degrees of freedom = k – 1 = (4 – 1) = 3 

o.
2
rs e
Using the   table with df = 3, = 2.49 shows the p–value is greater than .10
2
ou urc
Using Excel or Minitab, the p–value corresponding to  = 2.49 is .4771
2
p–value > .05, do not reject H0.  Conclude that we are unable to reject the hypothesis that the 
o

population proportion of male fish are equal in all four locations.
aC s
v i y re

b. No.     There   is   no   evidence   that   differences   in   agricultural   contaminants   found   at   the   four
locations have altered the gender proportions of the fish populations.
ed d
ar stu

3. [Test of Independence] Health insurance benefits vary by the size of the company (Atlanta Business
sh is

Chronicle, December 31, 2010). The sample data below show the number of companies providing health
insurance for small, medium, and large companies. For purposes of this study, small companies are
Th

companies that have fewer than 100 employees. Medium-sized companies have 100 to 999 employees,
and large companies have 1000 or more employees. The questionnaire sent to 225 employees asked
whether or not the employee had health insurance and then asked the employee to indicate the size of the
company.

a. Conduct a test of independence to determine whether health insurance coverage is independent of

the size of the company. What is the p-value? Using a .05 level of significance, what is your
conclusion?
2

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 b. A newspaper article indicated employees of small companies are more likely to lack health insurance
coverage. Use percentages based on the above data to support this conclusion.
[sol] a. H0: Having health insurance is independent of the size of the company
      Ha:  Having health insurance is not independent of the size of the company
Observed Frequencies (fij)
Health
Insurance Small Medium Large Total
Yes 36 65 88 189
No 14 10 12 36
Total 50 75 100 225
Expected Frequencies (eij)
Health
Insurance Small Medium Large Total
Yes 42 63 84 189
No 8 12 16 36
Total 50 75 100 225
Chi–Square Calculations (fij – eij)2 / eij
Health

m
Insurance Small Medium Large Total

er as
Yes .86 .06 .19 1.11

co
No 4.50 .33 1.00 5.83

eH w
       = 6.94

o.
rs e 2
Degrees of freedom = (r – 1)(c – 1)= (2 – 1)(3 – 1) = 2
ou urc
Using the   table with df = 2, = 6.94 shows the p–value is between .025 and .05.  
2 2
Using Excel or Minitab, the p–value corresponding to  = 6.94 is .0311.
o

2
aC s
v i y re

p–value  .05, reject H0.  Conclude health insurance coverage is not independent of the size of

  the company.  Health coverage is expected to vary depending on the size of the company.
b. Percentage of no coverage by company size
ed d

Small 14/50 = 28%

ar stu

Medium 10/75 = 13%

Large 12/100 = 12%
More than twice as many small companies do not provide health insurance coverage when compared
to medium and large companies.
sh is
Th

4. [Test of Independence] The Wall Street Journal Corporate Perceptions Study 2011 surveyed readers and
asked how each rated the quality of management and the reputation of the company for over 250 world-
wide corporations. Both the quality of management and the reputation of the company were rated on an
excellent, good, and fair categorical scale. Assume the sample data for 200 respondents below applies to
this study.

a. Use a .05 level of significance and test for independence of the quality of management and the
reputation of the company. What is the p-value and what is your conclusion?
3

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 b. If there is a dependence or association between the two ratings, discuss and use prob- abilities to justify
your answer.
[sol] a. H0:  Quality of Management is independent of the Reputation of the Company 
 
      Ha:  Quality of Management is not independent of the Reputation of the Company 
Observed Frequencies (fij)
Quality of
Management Excellent Good Fair Total
Excellent 40 25 5 70
Good 35 35 10 80
Fair 25 10 15 50
Total 100 70 30 200
Expected Frequencies (eij)
Quality of
Management Excellent Good Fair Total
Excellent 35.0 24.5 10.5 70
Good 40.0 28.0 12.0 80
Fair 25.0 17.5 7.5 50
Total 100 70 30 200

m
er as
Chi Square Calculations (fij – eij)2 / eij

co
Quality of

eH w
Management Excellent Good Fair Total

o.
Excellent .71 .01 2.88 3.61
Good
rs e
.63 1.75 .33 2.71
ou urc
Fair .00 3.21 7.50
            10.71
 2  17.03
o

Degrees of freedom =  (r – 1)(c – 1) =  (3 – 1)(3 – 1) = 4
Using the   table with df = 4, = 17.03 shows the p–value is less than .005  
aC s

2 2
v i y re

Using Excel or Minitab, the p–value corresponding to  = 17.03 is .0019
2
p–value < .05, reject H0.  Conclude that the rating for the quality of management is not independent 
ed d

of the rating for the reputation of the company.
ar stu

      b. Using the highest column probabilities, if the reputation of the company is 
Excellent:  There is a 40/100 = .40 chance the quality of management will also be excellent.
Good:  There is a 35/70 = .50 chance the quality of management will also be good.
Fair:  There is a 15/30 = .50 chance the quality of management will also be fair.
sh is

The highest probabilities are that the two variables will have the same ratings.  Thus, the two rating are 
Th

associated.

5. [Test of Independence]The National Sleep Foundation used a survey to determine whether hours of sleep
per night are independent of age (Newsweek, January 19, 2004). A sample of individuals was asked to
indicate the number of hours of sleep per night with categorical options: fewer than 6 hours, 6 to 6.9
hours, 7 to 7.9 hours, and 8 hours or more. Later in the survey, the individuals were asked to indicate
their age with categorical options: age 39 or younger and age 40 or older. Sample data follow.

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 a. Conduct a test of independence to determine whether hours of sleep are independent of age. Using a .05
level of significance, what is the p-value and what is your conclusion?
b. What is your estimate of the percentages of individuals who sleep fewer than 6 hours, 6 to 6.9 hours, 7
to 7.9 hours, and 8 hours or more per night?
[sol] a. H0: Hours of sleep per night is independent of age
           Ha: Hours of sleep per night is not independent of age

Observed Frequencies (fij)
Hours of 39 or
Sleep younger 40 or older Total

m
er as
Fewer than

co
6 38 36 74

eH w
6 to 6.9 60 57 117

o.
7 to 7.9 77 rs e 75 152
8 or more 65 92 157
ou urc
Total 240 260 500
Expected Frequencies (eij)
Hours of 39 or
o

Sleep younger 40 or older Total

aC s

Fewer than
v i y re

6 35.52 38.48 74
6 to 6.9 56.16 60.84 117
7 to 7.9 72.96 79.04 152
ed d

8 or more 75.36 81.64 157

ar stu

Total 240 260 500

Chi Square Calculations (fij – eij)2 / eij
Hours of 39 or
sh is

Sleep younger 40 or older Total

Fewer than
Th

6 .17 .16 .33

6 to 6.9 .26 .24 .50
7 to 7.9 .22 .21 .43
8 or more 1.42 1.31 2.74
       = 4.01
2
Degrees of freedom = (r – 1)(c – 1) = (4 – 1)(2 – 1) = 3
Using the   table with df = 3, = 4.01 shows the p–value is greater than .10.  
2 2
Using Excel or Minitab, the p–value corresponding to  = 4.01 is .2604.
2

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 p–value > .05, do not reject H0.  Cannot reject the assumption that age and hours of sleep are 
independent.
b. Since age does not appear to have an association on hours of sleep, use the overall row percentages.
Fewer than 6   74/500  = .148 14.8%
6 to 6.9 117/500 = .234 23.4%
7 to 7.9 152/500 = .304 30.4%
8 or more 157/500 = .314 31.4%
30.4% + 31.4% = 61.8% of individuals get seven or more hours of sleep a night.

6. During the first 13 weeks of the television season, the Saturday evening 8:00 P.M. to 9:00 P.M. audience
proportions were recorded as ABC 29%, CBS 28%, NBC 25%, and in- dependents 18%. A sample of 300 homes
two weeks after a Saturday night schedule revision yielded the following viewing audience data: ABC 95
homes, CBS 70 homes, NBC 89 homes, and independents 46 homes. Test with α = .05 to determine
whether the viewing audience proportions changed.
[sol] H0: pABC = .29, pCBS = .28, pNBC = .25, pIND = .18

m
Ha: The proportions are not pABC = .29, pCBS = .28, pNBC = .25, pIND = .18

er as
Expected frequencies: 300(.29) = 87, 300(.28) = 84

co
eH w
300(.25) = 75, 300(.18) = 54
e1 = 87, e2 = 84, e3 = 75, e4 = 54

o.
Observed frequencies: f1 = 95, f2 = 70, f3 = 89, f4 = 46
rs e
ou urc
(95  87) 2 (70  84) 2 (89  75) 2 (46  54) 2
2    
87 84 75 54
6.87
k – 1 = 3 degrees of freedom
o

Using the table with df = 3, = 6.87 shows the p–value is between .05 and .10.
aC s

2 2
v i y re

Using Excel or Minitab, the p–value corresponding to = 6.87 is .0762.

2
p–value > .05, do not reject H0. There has not been a significant change in the viewing audience
ed d

proportions.
ar stu

7. Mars, Inc. manufactures M&M’s, one of the most popular candy treats in the world. The milk chocolate
candies come in a variety of colors including blue, brown, green, orange, red, and yellow (M&M website,
March 2012). The overall proportions for the colors are .24 blue, .13 brown, .20 green, .16 orange, .13 red,
sh is

and .14 yellow. In a sampling study, several bags of M&M milk chocolates were opened and the following
Th

color counts were obtained.

Use a .05 level of significance and the sample data to test the hypothesis that the overall proportions for
the colors are as stated above. What is your conclusion?
[sol] 
Observed Expected
Hypothesize Frequency Frequency Chi Square
d
Category Proportion (fi) (ei) (fi – ei)2 / ei
Blue .24 105 120 1.88

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 Brown .13 72   65   .75
Green .20  89 100 1.21
Orange .16  84   80    .20  
Red .13  70   65    .38
Yellow .14  80   70  1.43
Total: 500   2 = 5.85

k – 1 = 6 – 1 = 5 degrees of freedom
Using the   table with df = 5, = 5.85 shows the p–value is greater than .10
2 2
Using Excel or Minitab, the p–value corresponding to  = 5.85 is .3211
2
p–value  > .05, do not reject H0. We cannot reject the hypothesis that the overall percentages of 
colors in the population of M&M milk chocolate candies are .24 blue, .13 brown, .20 green, .16 orange, .13 red 
and .14 yellow.

8. The National Highway Traffic Safety Administration reported the percentage of traffic accidents occurring
each day of the week (Time, March 12, 2012). Assume that a sample of 420 accidents provided the
following data.

m
er as
co
a. Conduct a hypothesis test to determine if the proportion of traffic accidents is the same for each day of

eH w
the week. What is the p-value? Using a .05 level of significance, what is your conclusion?

o.
b. Compute the percentage of traffic accidents occurring on each day of the week. What day has the
rs e
highest percentage of traffic accidents? Does this seem reasonable? Discuss.
ou urc
[sol] 
o
aC s
v i y re
ed d
ar stu
sh is
Th

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
 m
er as
co
eH w
o.
rs e
ou urc
o
aC s
v i y re
ed d
ar stu
sh is
Th

This study source was downloaded by 100000826599679 from CourseHero.com on 10-07-2021 19:00:52 GMT -05:00

https://www.coursehero.com/file/26487093/%E5%8D%A1%E6%96%B9%E6%AA%A2%E5%AE%9Adocx/
Powered by TCPDF (www.tcpdf.org)