Full Statistics For Business and Economics 12Th Edition Mcclave Solutions Manual Online PDF All Chapter

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 72

Statistics for Business and Economics

12th Edition McClave Solutions Manual


Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/statistics-for-business-and-economics-12th-edition-mcclave-solutions-manual/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Statistics for Business and Economics 12th Edition


McClave Test Bank

https://testbankdeal.com/product/statistics-for-business-and-
economics-12th-edition-mcclave-test-bank/

Statistics for Business and Economics 13th Edition


McClave Solutions Manual

https://testbankdeal.com/product/statistics-for-business-and-
economics-13th-edition-mcclave-solutions-manual/

Statistics for Business and Economics 13th Edition


McClave Test Bank

https://testbankdeal.com/product/statistics-for-business-and-
economics-13th-edition-mcclave-test-bank/

Statistics for Business and Economics 12th Edition


Anderson Solutions Manual

https://testbankdeal.com/product/statistics-for-business-and-
economics-12th-edition-anderson-solutions-manual/
Statistics for Business and Economics Revised 12th
Edition Anderson Solutions Manual

https://testbankdeal.com/product/statistics-for-business-and-
economics-revised-12th-edition-anderson-solutions-manual/

Statistics 12th Edition McClave Test Bank

https://testbankdeal.com/product/statistics-12th-edition-mcclave-
test-bank/

First Course in Statistics 12th Edition McClave


Solutions Manual

https://testbankdeal.com/product/first-course-in-statistics-12th-
edition-mcclave-solutions-manual/

Statistics for Business and Economics 12th Edition


Anderson Test Bank

https://testbankdeal.com/product/statistics-for-business-and-
economics-12th-edition-anderson-test-bank/

Statistics for Business and Economics Revised 12th


Edition Anderson Test Bank

https://testbankdeal.com/product/statistics-for-business-and-
economics-revised-12th-edition-anderson-test-bank/
Chapter 10
Categorical Data Analysis
10.1 a. The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  3  1  2 .
From Table IV, Appendix D,  .05
2
 5.99147 . The rejection region is  2  5.99147 .

b. The rejection region requires   .10 in the upper tail of the  2 distribution with df  k  1  5  1  4 .
From Table IV, Appendix D,  .10
2
 7.77944 . The rejection region is  2  7.77944 .

c. The rejection region requires   .01 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .01
2
 11.3449 . The rejection region is  2  11.3449 .

10.2 The characteristics of the multinomial experiment are:

1. The experiment consists of n identical trials.


2. There are k possible outcomes to each trial.
3. The probabilities of the k outcomes, denoted p1, p2, ... , pk, remain the same from trial to trial, where
p1 + p2 +  + pk = 1.
4. The trials are independent.
5. The random variables of interest are the counts n1, n2, ... , nk in each of the k cells.

The characteristics of the binomial are the same as those for the multinomial with k  2 .

10.3 The sample size n will be large enough so that, for every cell, the expected cell count, Ei, will be equal to 5
or more.

10.4 The hypotheses of interest are:

H 0 : p1  .25, p2  .25, p3  .50


H a : At lease one of the probabilities differs from the hypothesized value

E1  np1,0  320(.25)  80 E2  np2,0  320(.25)  80 E3  np3,0  320(.50)  160

 ni  Ei 
2
(78  80) 2 (60  80) 2 (182  160) 2
The test statistic is  2       8.075
Ei 80 80 160

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  3  1  2 . From
Table IV, Appendix B,  .05
2
 5.99147 . The rejection region is  2  5.99147 .

Since the observed value of the test statistic falls in the rejection region (  2  8.075  5.99147) , H0 is
rejected. There is sufficient evidence to indicate that at least one of the probabilities differs from its
hypothesized value at   .05 .

564
Copyright © 2014 Pearson Education, Inc.
Categorical Data Analysis 565

10.5 Some preliminary calculations are:

If the probabilities are the same, p1,0  p2,0  p3,0  p4,0  .25

E1  np1,0  205(.25)  51.25  E2  E3  E4

a. To determine if the multinomial probabilities differ, we test:

H 0 : p1  p2  p3  p4  .25
H a : At lease one of the probabilities differs from .25

The test statistic is

[ni  Ei ]2 (43  51.25) 2 (56  51.25) 2 (59  51.25) 2 (47  51.25) 2


2        3.293
Ei 51.25 51.25 51.25 51.25

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is  2  7.81473 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  3.293  7.81473) , H0 is not rejected. There is insufficient evidence to indicate the multinomial
probabilities differ at   .05 .

b. The Type I error is concluding the multinomial probabilities differ when, in fact, they do not.

The Type II error is concluding the multinomial probabilities are equal, when, in fact, they are not.

c. For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 .

pˆ 3  59 / 205  .288

The confidence interval is:


ˆˆ
pq .288(.712)
pˆ 3  z.025  .288  1.96  .288  .062   .226, .350 
n 205

10.6 a. The data are categorical because they are measured using categories, not meaningful numbers. The
possible categories are legs only, wheels only, both legs and wheels, and neither legs nor wheels.

b. Let p1  proportion of social robots with legs only, p2  proportion of social robots with wheels only,
p3  proportion of social robots with both legs and wheels, and p4  proportion of social robots with
neither legs nor wheels. To determine if the design engineer’s claim is incorrect, we test:

H 0 : p1  .50, p2  .30, p3  .10, and p4  .10


H a : At least one of the probabilities differs from the hypothesized value

c. If the claim is true, E1  np1,0  106(.50)  53 , E2  np2,0  106 .30   31.8 ,


E3  np3,0  106 .10   10.6 , and E4  np4,0  106 .10   10.6 .

Copyright © 2014 Pearson Education, Inc.


566 Chapter 10

 ni  Ei 
2
(63  53) 2 (20  31.8) 2 (8  10.6) 2 (15  10.6) 2
d. The test statistic is  2        8.730
Ei 53 31.8 10.6 10.6

e. The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is  2  7.81473 .

Since the observed value of the test statistic falls in the rejection region (  2  8.730  7.81473) , H0 is
rejected. There is sufficient evidence to indicate that at least one of the probabilities differs from its
hypothesized value at.   .05 .

10.7 a. Let p1  proportion using total visitors, p2  proportion using paying visitors, p3  proportion using
big shows, p4  proportion using funds raised, and p5  proportion using members.

To determine if one performance measure is used more often than any of the others, we test:

H 0 : p1  p2  p3  p4  p5  .20
H a : At least one of the probabilities differs from the hypothesized value

From the printout, the test statistic is  2  1.66667 and the p-value is p  0.797 .

Since the p-value is not less than  ( p  .797  .10) , H0 is not rejected. There is insufficient evidence
to indicate that one performance measure is used more often than any of the others at   .10 .

b. For confidence coefficient .90,   .10 and  / 2  .10 / 2  .05 . From Table II, Appendix D,
z.05  1.645 .

pˆ 1  8 / 30  .267

The confidence interval is:


ˆˆ
pq .267(.733)
pˆ1  z.05  .267  1.645  .267  .133  (.134, .400)
n 30

We are 90% confident that the proportion of museums world-wide that use total visitors as their
performance measure is between .134 and .400.

10.8 a. The categorical variable is the rating of the student exposure to social and environmental issues. It
has 5 levels: 1-star, 2-stars, 3-stars, 4-stars, and 5-stars.

b. If there were no difference in the category proportions, then each proportion should be
pi  1 / 5  .20 . There were a total of n = 30 business schools sampled. The expected number would
be: E1  E2  E3  E4  E5  npi ,0  30(.20)  6

c. To determine if there are differences in the star rating category proportions of all MBA programs, we
test:

H 0 : p1  p2  p3  p4  p5  .20
H a : At lease one of the probabilities differs from the hypothesized value

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 567

 ni  Ei   2  6 9  6 14  6  5  6 0  6
2 2 2 2 2 2

d. The test statistic is  2         21


Ei 6 6 6 6 6

e. The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  5  1  4 .
From Table IV, Appendix D,  .05
2
 9.48773 . The rejection region is  2  9.48773 .

f. Since the observed value of the test statistic falls in the rejection region (  2  21  9.48773) , H0 is
rejected. There is sufficient evidence to indicate differences in the star rating category proportions of
all MBA programs at   .05 .

x3 14
g. Some preliminary calculations are: pˆ 3    .467
n 30

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ 3 qˆ3 .467(.533)
pˆ 3  z.025  .467  1.96  .467  .179  (.288, .646)
n 30

We are 95% confident that the proportion of all MBA programs that are ranked in the 3-star category
is between .288 and .646.

10.9 a. Since there are 10 income groups, we would expect 10% or 1, 072(.10)  107.2 givers in each of the
income categories.

b. The null hypothesis for testing whether the true proportions of charitable givers in each income group
are the same is:

H 0 : p1  p2    p10  .10

c. Some preliminary calculations are: E1  E2    E10  npi ,0  1, 072(.10)  107.2

[ ni  Ei ]2 (42  107.2) 2 (93  107.2) 2 (127  107.2) 2


2      ...   93.15
Ei 107.2 107.2 107.2

d. The rejection region requires   .10 in the upper tail of the  2 distribution with df  k  1  10  1  9 .
From Table IV, Appendix D,  .10
2
 14.6837 . The rejection region is  2  14.6837 .

e. Since the observed value of the test statistic falls in the rejection region (  2  93.15  14.6837) , H0 is
rejected. There is sufficient evidence to indicate that the true proportions of charitable givers in each
income group are not all the same at   .10 .

10.10 a. The qualitative variable is firm position on off-shoring. There are four levels: “currently off-shoring,”
“not currently off-shoring, but plan to do so,” “off-shored in the past, but no more,” and “off-shoring is
not applicable.”

b. Let p1  proportion of firms currently off-shoring, p2  proportion of firms not currently off-shoring,
but plan to do so, p3  proportion of firms off-shored in the past, but no more, and p4  proportion of
firms where off-shoring is not applicable.
Copyright © 2014 Pearson Education, Inc.
568 Chapter 10

Some preliminary calculations are: E1  E2  E3  E4  npi,0  600 .25  150

To determine if the proportions of U.S. firms in the four off-shoring position categories is significantly
different, we test:

H 0 : p1  p2  p3  p4  .25
H a : At least one of the probabilities differs from the hypothesized value

The test statistic is


 ni  Ei 
2
(126  150) 2 (72  150) 2 (30  150) 2 (372  150) 2
 
2
     468.96
Ei 150 150 150 150

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is  2  7.81473 .

Since the observed value of the test statistic falls in the rejection region (  2  468.96  7.81473) , H0
is rejected. There is sufficient evidence to indicate that at least one of the proportions of U.S. firms in
the four off-shoring position categories is significantly different at   .05 .

c. For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 .

pˆ 1  126 / 600  .21

The confidence interval is:


ˆˆ
pq .21(.79)
pˆ1  z.025  .21  1.96  .21  .033  (.177, .243)
n 600

We are 95% confident that the proportion of U.S. firms who are currently off-shoring is between .177
and .243.

10.11 Let p1  proportion users using both hands/both thumbs, p2  proportion of users using right hand/right
thumb, p3  proportion of users using left hand/left thumb, p4  proportion of users using both hands/right
index finger, p5  proportion of users using left hand/right index finger and p6  proportion of users using
other. Some preliminary calculations: E1  E2  E3  E4  E5  E6  npi,0  859 1/ 6   143.167 .

To determine if the proportions of mobile device users in the six texting style categories differ, we test:

H 0 : p1  p2  p3  p4  p5  p6  1 / 6
H a : At least one of the probabilities differs from the hypothesized value

The test statistic is


 n  Ei   (396  143.167) 2  (311  143.167)2  (70  143.167) 2  (39  143.167)2
2

2   i
Ei 143.167 143.167 143.167 143.167
(18  143.167) 2 (25  143.167) 2
   756.436
143.167 143.167

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 569

The rejection region requires   .10 in the upper tail of the  2 distribution with df  k  1  6  1  5 . From
Table IV, Appendix D,  .10
2
 9.23635 . The rejection region is  2  9.23635 .

Since the observed value of the test statistic falls in the rejection region (  2  756.436  9.23635) , H0 is
rejected. There is sufficient evidence to indicate that the proportions of mobile device users in the six
texting style categories differ at   .10 .

10.12 Let p1  proportion of anchor tenants, p2  proportion of major space users, p3  proportion of large
standard tenants, p4  proportion of small standard tenants, and p5  proportion of small tenants. Some
preliminary calculations:

E1  np1,0  1,821.01  18.21 E2  np2,0  1,821.05  91.05 E3  np3,0  1,821.10   182.1

E4  np4,0  1,821.40   728.4 E5  np5,0  1,821.44   801.24

To determine if the mall developer’s belief is correct, we test:

H 0 : p1  .01, p2  .05, p3  .10, p4  .40, p5  .44


H a : At least one of the probabilities differs from the hypothesized value

The test statistic is

 ni  Ei 
2
(14  18.21) 2 (61  91.05) 2 (216  182.1) 2 (711  728.4) 2
2      
Ei 18.21 91.05 182.1 728.4
(819  801.24) 2
  18.011
801.24

The rejection region requires   .01 in the upper tail of the  2 distribution with df  k  1  5  1  4 . From
Table IV, Appendix D,  .01
2
 13.2767 . The rejection region is  2  13.2767 .

Since the observed value of the test statistic falls in the rejection region (  2  18.011  13.2767) , H0 is
rejected. There is sufficient evidence to indicate that the proportions of tenants in the five categories differ
from the developer’s belief at   .01 .

Copyright © 2014 Pearson Education, Inc.


570 Chapter 10

10.13 a. The data come from a multinomial experiment because there are several possible categorical
responses to the question.

b. To determine if the multinomial probabilities agree with the theory, we test:

H 0 : p1 =.50, p2  p3  p4  p5  .10, p6  p7  .05

c. Using MINITAB, the results are:

Chi-Square Goodness-of-Fit Test for Observed Counts in Variable: C1

Test Contribution
Category Observed Proportion Expected to Chi-Sq
1 869 0.50 1059.50 34.252
2 339 0.10 211.90 76.236
3 338 0.10 211.90 75.041
4 127 0.10 211.90 34.016
5 85 0.10 211.90 75.996
6 128 0.05 105.95 4.589
7 233 0.05 105.95 152.352

N DF Chi-Sq P-Value
2119 6 452.483 0.000

To determine if the multinomial probabilities agree with the theory, we test:

H 0 : p1 =.50, p2  p3  p4  p5  .10, p6  p7  .05


H a : At least one of the probabilities differs from its hypothesized value

The test statistic is  2  452.843 and the p-value is p  0.000 . Since the p-value is less that   .01 ,
H0 is rejected. There is sufficient evidence to indicate that at least one of the proportions differs from
its hypothesized value at   .01 .

10.14 Some preliminary calculations are:

E1  np1,0  2, 023(.45)  910.35 E2  np2,0  2, 023(.35)  708.05

E3  np3,0  2, 023(.15)  303.45 E4  np4,0  2, 023(.05)  101.15

To determine if the percentages of all adults falling into the four response categories changed after the
Enron scandal, we test:

H 0 : p1  .45, p2  .35, p3  .15, and p4  .05


H a : At lease one of the probabilities differs from the hypothesized value

The test statistic is


 ni  Ei  1,173  910.35  587  708.05 182  303.45 81  101.15
2 2 2 2 2

2      
Ei 910.35 708.05 303.45 101.15
 149.096

The rejection region requires   .01 in the upper tail of the  2 distribution with df  k  1  4  1  3 . From
Table IV, Appendix D,  .01
2
 11.3449 . The rejection region is  2  11.3449 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 571

Since the observed value of the test statistic falls in the rejection region (  2  149.096  11.3449) , H0 is
rejected. There is sufficient evidence to indicate the percentages of all adults falling into the four response
categories changed after the Enron scandal at   .01 .

10.15 Let p1  proportion of mail only users, p2  proportion of Internet only users, and p3  proportion of both
mail and Internet. Some preliminary calculations:

E1  E2  E3  npi,0  440 1 / 3  146.667

To determine if the professor’s beliefs are correct, we test:

H 0 : p1  p2  p3  1 / 3
H a : At least one of the probabilities differs from the hypothesized value

The test statistic is

 ni  Ei 
2
(262  146.667) 2 (43  146.667) 2 (135  146.667) 2
2       164.895
Ei 146.667 146.667 146.667

The rejection region requires   .01 in the upper tail of the  2 distribution with df  k  1  3  1  2 . From
Table IV, Appendix D,  .01
2
 9.21034 . The rejection region is  2  9.21034 .

Since the observed value of the test statistic falls in the rejection region (  2  164.895  9.21034) , H0 is
rejected. There is sufficient evidence to indicate that the proportions mail only, Internet only, and both
mail and Internet users differ at   .01 .

10.16 Some preliminary calculations are:

E1  np1,0  943(.51)  480.93 E2  np2,0  943(.37)  348.91


E3  np3,0  943(.09)  84.87 E4  np4,0  943(.03)  28.29

To determine if the data from the independent survey contradict the percentages reported by the CPS Cell
Phone Supplement, we test:

H 0 : p1  .51, p2  .37, p3  .09 and p4  .03


H a : At least one of the probabilities differs from the hypothesized value

The test statistic is


 ni  Ei  (473  480.93)2 (334  348.91)2 (106  84.87) 2 (30  29.29) 2
2

 
2
     6.132
Ei 480.93 348.91 84.87 29.29

The rejections region requires   .10 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .10
2
 6.25139. The rejection region is  2  6.25139 .

Since the test statistic does not fall in the rejection region (  2  6.132  6.25139) , H0 is not rejected.
There is insufficient evidence to indicate the data from the independent survey contradict the percentages
reported by the CPS Cell Phone Supplement at   .10 .

Copyright © 2014 Pearson Education, Inc.


572 Chapter 10

10.17 To determine if the number of overweight trucks per week is distributed over the 7 days of the week in
direct proportion to the volume of truck traffic, we test:

H0: p1 = .191, p2 = .198, p3 = .187, p4 = .180, p5 = .155, p6 = .043, p7 = .046


Ha: At least one of the probabilities differs from the hypothesized value

E1  np1,0  414(.191)  79.074 E2  np2,0  414(.198)  81.972 E3  np3,0  414(.187)  77.418

E4  np4,0  414(.180)  74.520 E5  np2,0  414(.155)  64.170 E6  np3,0  414(.043)  17.802

E7  np3,0  414(.046)  19.044

The test statistic is


[ni  Ei ]2  90  79.074   82  81.972   72  77.418   70  74.520 
2 2 2 2

2      
Ei 79.074 81.972 77.418 74.520
 51  64.170  18  17.802   31  19.044 
2 2 2

    12.374
64.170 17.802 19.044

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  7  1  6 . From
Table IV, Appendix D,  .05  12.5916 . The rejection region is  2  12.5916 .
2

Since the observed value of the test statistic does not fall in the rejection region (  2  12.374  12.5916) , H0
is not rejected. There is insufficient evidence to indicate the number of overweight trucks per week is
distributed over the 7 days of the week is not in direct proportion to the volume of truck traffic at   .05 .

10.18 Some preliminary calculations are:

E1  np1,0  435(.28)  121.8 E2  np2,0  435(.04)  17.4


E3  np3,0  435(.02)  8.7 E4  np4,0  435(.66)  287.1

To determine if the House of Representatives is not statistically representative of the religious affiliations
of their constituents, we test:

H 0 : p1  .28, p2  .04, p3  .02, and p4  .66


H a : At lease one of the probabilities differs from the hypothesized value

The test statistic is


 ni  Ei 
2
(117  121.8) 2 (61  17.4) 2 (30  8.7) 2 (227  287.1) 2
 
2
     174.169
Ei 121.8 17.4 8.7 287.1

Since no value of  was given, we will use   .05 . The rejections region requires   .05 in the upper tail
of the  2 distribution with df  k  1  4  1  3 . From Table IV, Appendix D,  .05
2
 7.81473 . The
rejection region is  2  7.81473 .

Since the test statistic falls in the rejection region (  2  174.169  7.81473) , Ho is rejected. There is sufficient
evidence to indicate the House of Representatives is not statistically representative of the religious
affiliations of their constituents at   .05 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 573

10.19 a. df  ( r  1)(c  1)  (5  1)(5  1)  16 . From Table IV, Appendix D,  .05


2
 26.2962 . The rejection
region is  2  26.2962 .

b. df  ( r  1)(c  1)  (3  1)(6  1)  10 . From Table IV, Appendix D,  .10


2
 15.9871 . The rejection
region is  2  15.9871 .

c. df  ( r  1)(c  1)  (2  1)(3  1)  2 . From Table IV, Appendix D,  .01


2
 9.21034 . The rejection
region is  2  9.21034 .

10.20 a. H0: The row and column classifications are independent


Ha: The row and column classifications are dependent

[nij  Eˆij ]2
The test statistic is   
2
b. .
Eˆij

The rejection region requires   .01 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(3  1)  2 . From Table IV, Appendix D,  .01
2
 9.21034 . The rejection
region is  2  9.21034 .

c. The expected cell counts are:

R C 96(25) RC 96(64) R C 96(78)


Eˆ11  1 1   14.37 Eˆ12  1 2   36.79 Eˆ13  1 3   44.84
n 167 n 167 n 167

R C 71(25) RC 71(64) R C 71(78)


Eˆ 21  2 1   10.63 Eˆ 22  2 2   27.21 Eˆ 23  2 3   33.16
n 167 n 167 n 167

d. The test statistic is


[nij  Eˆ ij ]2 (9  14.37)2 (34  36.79) 2 (53  44.84) 2
2     
Eˆ ij
14.37 36.79 44.84
(16  10.63) 2 (30  27.21) 2 (25  33.16) 2
    8.71
10.63 27.21 33.16

Since the observed value of the test statistic does not fall in the rejection region
(  2  8.71  9.21034) , H0 is not rejected. There is insufficient evidence to indicate the row and
column classifications are dependent at   .01 .

Copyright © 2014 Pearson Education, Inc.


574 Chapter 10

10.21 a. To convert the frequencies to percentages, divide the numbers in each column by the column total
and multiply by 100. Also, divide the row totals by the overall total and multiply by 100. The
column totals are 25, 64, and 78, while the row totals are 96 and 71. The overall sample size is 165.
The table of percentages are:

Column
1 2 3
Row 1 9 34 53 96
 100  36%  100  53.1%  100  67.9%  100  57.5%
25 64 78 167
2 16 30 25 71
 100  64%  100  46.9%  100  32.1%  100  42.5%
25 64 78 167

b. Using MINITAB, the graph is:

70

60
57.5

50

40
Percent

30

20

10

0
1 2 3
Column

c. If the rows and columns are independent, the row percentages in each column would be close to the
row total percentages. This pattern is not evident in the plot, implying the rows and columns are not
independent. In Exercise 10.20, we did not have enough evidence to say the rows and columns were
not independent. If the sample sizes were bigger, we would have been able to reject H0.

10.22 Some preliminary calculations are:

R C 154(134) 154(163) 154(142)


Eˆ11  1 1   47.007 Eˆ12   57.180 Eˆ13   49.813
n 439 439 439

186(134) 186(163) 186(142)


Eˆ 21   56.774 Eˆ 22   69.062 Eˆ 23   60.164
439 439 439

99(134) 99(163) 99(142)


Eˆ 31   30.219 Eˆ 32   36.759 Eˆ 33   32.023
439 439 439

To determine if the row and column classifications are dependent, we test:

H0: The row and column classifications are independent


Ha: The row and column classifications are dependent

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 575

The test statistic is

[nij  Eˆ ij ]2 (40  47.007)2 (72  57.180)2 (42  49.813)2 (63  56.774)2


2      
Eˆ ij
47.007 57.180 49.813 56.774
(53  69.062) 2 (70  60.164) 2 (31  30.219) 2 (38  36.759)2 (30  32.023)2
      12.36
69.062 60.164 30.219 36.759 32.023

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(3  1)  4 . From Table IV, Appendix D,  .05
2
 9.48773 . The rejection region is
 2  9.48773 .

Since the observed value of the test statistic falls in the rejection region (  2  12.36  9.48773) , H0 is
rejected. There is sufficient evidence to indicate the row and column classification are dependent at
  .05 .

10.23 a-b. To convert the frequencies to percentages, divide the numbers in each column by the column total
and multiply by 100. Also, divide the row totals by the overall total and multiply by 100.

B
B1 B2 B3 Totals
A1 40 72 42 154
 100  29.9%  100  44.2% 100  29.6%  100  35.1%
134 163 142 439
A2 63 53 70 186
Row  100  47.0%  100  32.5% 100  49.3%  100  42.4%
134 163 142 439
A3 31 38 30 99
 100  23.1%  100  23.3% 100  21.1%  100  22.6%
134 163 142 439

c. Using MINITAB, the graph of A1 is:

50

40

35.1

30
Percent

20

10

0
1 2 3
B

The graph supports the conclusion that the rows and columns are not independent. If they were, then
the height of all the bars would be essentially the same.

Copyright © 2014 Pearson Education, Inc.


576 Chapter 10

d. Using MINITAB, the graph of A2 is:

50

42.4
40

30
Percent

20

10

0
1 2 3
B

The graph supports the conclusion that the rows and columns are not independent. If they were, then
the height of all the bars would be essentially the same.

e. Using MINITAB, the graph of A3 is:

25

22.6

20

15
Percent

10

0
1 2 3
B

The graph does not support the conclusion that the rows and columns are not independent. All the
bars would be essentially the same.

10.24 a. The two qualitative variables are model of Accord and injury (yes or no).
b. The contingency table is:

Injury No Injury Total


Conventional 5,364 44,768 50,132
Hybrid 137 1,368 1,505
Total 5,501 46,136 51,637

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 577

c. To determine if the injury rate for collision claims depends on Accord model, we test:

H 0 : Model and Injury rate are independent


H a : Model and Injury rate are dependent

d. The expected cell counts are:

R C 50,132(5,501) RC 50,132(46,136)
Eˆ11  1 1   5,340.67 Eˆ12  1 2   44, 791.33
n 51, 637 n 51, 637

R C 1, 505(5, 501) R C 1, 505(46,136)


Eˆ 21  2 1   160.33 Eˆ 22  2 2   1, 344.67
n 51, 637 n 51, 637

e. The test statistic is:

2
 nij  Eˆ ij 
   5,364  5,340.67    44, 768  44, 791.33
2 2

   
2

Eˆ ij 5,340.67 44, 791.33

137  160.33 1,368  1,344.67 


2 2

   3.91
160.33 1,344.67

This agrees with the test statistic found on the XLSTAT printout.

f. The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 . The rejection
region is  2  3.84146 . This is the same critical value found on the XLSTAT printout.

g. Since the observed value of the test statistic falls in the rejection region (  2  3.91  3.84146) , H0 is
rejected. There is sufficient evidence to indicate the injury rate for collision claims depends on
Accord model at   .05 .

Since the p-value is less than  ( p  .0479  .05) , H0 is rejected. There is sufficient evidence to
indicate the injury rate for collision claims depends on Accord model at   .05 .

h. Some preliminary calculations are:

x1 5,364 x2 137
pˆ1    .107 pˆ 2    .091
n1 50,132 n2 1,505

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ1qˆ1 pˆ 2 qˆ2 .107(.893) .091(.909)


( pˆ1  pˆ 2 )  z.025   (.107  .091)  1.96 
n1 n2 50,132 1,505
 .016  .015  (.001, .031)

Since the interval contains only positive numbers, the injury rate for hybrid Accords is less than the
injury rate for conventional Accords.

Copyright © 2014 Pearson Education, Inc.


578 Chapter 10

10.25 a. Yes, it appears that the male and female tourists differ in their responses to purchasing photographs,
postcards, and paintings. The values in the ‘Always’ and ‘Rarely or Never’ categories are quite
different. The percentages are insufficient to draw a conclusion because the sample sizes must be
taken into account.

b. The counts are found by changing the percentages to proportions and multiplying the proportions by
the sample sizes in each gender. The counts are:

Male Tourist Female Tourist Total


Always 240 476 716
Often 405 527 932
Occasionally 525 493 1018
Rarely or Never 330 204 534
Total 1500 1700 3200

c. To determine whether male and female tourists differ in their responses to purchasing photographs,
postcards, or paintings, we test:

H 0 : Gender and purchasing are independent


H a : Gender and purchasing are dependent

d. The test statistic is  2  112.433 and the p-value is p  .000 .

e. Since the p-value is less than  ( p  .000  .01) , H0 is rejected. There is sufficient evidence to
indicate male and female tourists differ in their responses to purchasing photographs, postcards, or
paintings at   .01 .

10.26 a. The sample proportion of negative tone news stories that are deceptive is 111 / 170  .653 .

b. The sample proportion of neutral tone news stories that are deceptive is 61 / 110  .555 .

c. The sample proportion of positive tone news stories that are deceptive is 11 / 31  .355 .

d. Yes, it appears that the proportion of news stories that are deceptive depends on the story tone. The
proportion that is deceptive for negative tone stories is .653, while the proportion that is deceptive for
positive tone stories is only .355. These proportions look much different.

e. To determine if the authenticity of a news story depends on tone, we test:

H 0 : Authenticity and tone are independent


H a : Authenticity and tone are dependent

f. The test statistic is  2  10.427 and the p-value is p  .005 .

Since the p-value is less than  ( p  .005  .05) , Ho is rejected. There is sufficient evidence to
indicate authenticity of a news story depends on tone at   .05 .

10.27 a. To compare the two proportions, we could use either a test of hypothesis or a confidence interval. I
will use a 95% confidence interval.

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 579

Some preliminary calculations are:

xM 1 29 xF 1 89
pˆ M 1    .282 pˆ F 1    .511
nM 103 nF 174

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ M 1qˆ M 1 pˆ F 1qˆ F 1 .282(.718) .511(.489)


 pˆ M 1  pˆ F1   z.025   .282  .511  1.96 
nM nF 103 174
 .229  .114  (.343,  .115)

We are 95% confident that the difference in the proportions of male and female professionals who
believe their salaries are too low is between .343 and .115. Since 0 is not in this interval, there is
evidence that the two proportions are different.

b. Some preliminary calculations are:

xM 2 58 xF 2 64
pˆ M 2    .563 pˆ F 2    .368
nM 103 nF 174

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ M 2 qˆ M 2 pˆ F 2 qˆ F 2 .563(.437) .368(.632)
 pˆ M 2  pˆ F 2   z.025   .563  .368   1.96 
nM nF 103 174
 .195  .120  (.075, .315)
We are 95% confident that the difference in the proportions of male and female professionals who
believe their salaries are equitable/fair is between .075 and .315. Since 0 is not in this interval, there
is evidence that the two proportions are different.

c. Some preliminary calculations are:

xM 3 16 xF 3 21
pˆ M 3    .155 pˆ F 3    .121
nM 103 nF 174

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ M 3 qˆ M 3 pˆ F 3 qˆ F 3 .155(.845) .121(.879)
 pˆ M 3  pˆ F 3   z.025   .155  .121  1.96 
nM nF 103 174
 .034  .085  (.051, .119)

We are 95% confident that the difference in the proportions of male and female professionals who
believe they are well paid is between .051 and .119. Since 0 is in this interval, there is no evidence
that the two proportions are different.

Copyright © 2014 Pearson Education, Inc.


580 Chapter 10

d. Yes. Since there were differences between the proportions of males and females on 2 of the 3 levels,
there is evidence that the opinions of males and females are different.

e. Some preliminary calculations are:

R C 118(103) RC 118(174)
Eˆ11  1 1   43.877 Eˆ12  1 2   74.123
n 277 n 277

R C 122(103) RC 122(174)
Eˆ 21  2 1   45.365 Eˆ 22  2 2   76.635
n 277 n 277

R C 37(103) RC 37(174)
Eˆ 31  3 1   13.758 Eˆ 33  3 3   23.242
n 277 n 277

To determine if the opinion on the fairness of a travel professional’s salary differ for males and
females, we test:

H0: Opinion and Gender are independent


Ha: Opinion and Gender are dependent

The test statistic is


2
 nij  Eˆ ij 
   29  43.877    89  74.123   58  45.365 
2 2 2

   
2

Eˆ ij
43.877 74.123 45.365

 64  76.635  16  13.758  21  23.242 


2 2 2

    14.214
76.635 13.758 23.242

The rejection region requires   .10 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .10
2
 4.60517 . The rejection
region is  2  4.60517 .
Since the observed value of the test statistic falls in the rejection region (  2  14.214  4.60517) , H0
is rejected. There is sufficient evidence to indicate that the opinions on the fairness of a travel
professional’s salary differ for males and females at   .10 .

f. For confidence coefficient .90,   .10 and  / 2  .10 / 2  .05 . From Table II, Appendix D,
z.05  1.645 . The 90% confidence interval is:

pˆ M 1qˆM 1 pˆ F 1qˆ F 1 .282(.718) .511(.489)


 pˆ M 1  pˆ F 1   z.05   .282  .511  1.645 
nM nF 103 174
 .229  .096  (.325,  .133)

We are 90% confident that the difference in the proportions of male and female professionals who
believe their salaries are too low is between -.325 and -.133. Since 0 is not in this interval, there is
evidence that the two proportions are different.

10.28 a. Let p3  proportion of the 3-photos per page group who selected the target mugshot, p6  proportion
of the 6-photos per page group who selected the target mugshot, and p12  proportion of the 12-photos
per page group who selected the target mugshot.

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 581

19 19 15
pˆ 3   .594 , pˆ 6   .594 , pˆ12   .469
32 32 32

The 12-photos per page group had the lowest proportion.

b. The contingency table is:

Target Target
Mugshot Mugshot not Total
selected selected
3-photos per page 19 13 32
3-photos per page 19 13 32
3-photos per page 15 17 32
Total 53 43 96

c. Some preliminary calculations are:


R C 32(53) R C 32(53) R C 32(53)
Eˆ11  1 1   17.667 Eˆ 21  2 1   17.667 Eˆ 31  3 1   17.667
n 96 n 96 n 96

RC 32(43) RC 32(43) RC 32(43)


Eˆ12  1 2   14.333 Eˆ 22  2 2   14.333 Eˆ 32  3 2   14.333
n 96 n 96 n 96

To determine if there are differences in the proportions who selected the target mugshot among the
three photo groups, we test:

H 0 : Photo group and Mugshot selection are independent


H a : Photo group and Mugshot selection are dependent

The test statistic is:

2
 nij  Eˆ ij 
  19  17.667   13  14.333  19  17.667   13  14.333
2 2 2 2

   
2

Eˆ ij 17.667 14.333 17.667 14.333

15  17.667  17  14.333


2 2

   1.348
17.667 14.333

The rejection region requires   .10 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .10
2
 4.60517 . The rejection
region is  2  4.60517 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  1.348  4.60517) , H0 is not rejected. There is insufficient evidence to indicate that there are
differences in the proportions who selected the target mugshot among the three photo groups at
  .10 .

Copyright © 2014 Pearson Education, Inc.


582 Chapter 10

10.29 Using MINITAB, the contingency table analysis is:

Tabulated statistics: Position, Nationality

Using frequencies in Fr

Rows: Position Columns: Nationality

1 2 3 4 All

1 126 75 35 93 329
2 72 36 10 27 145
3 30 9 4 6 49
4 372 180 51 174 777
All 600 300 100 300 1300

Cell Contents: Count

Pearson Chi-Square = 21.242, DF = 9, P-Value = 0.012


Likelihood Ratio Chi-Square = 21.327, DF = 9, P-Value = 0.011

To determine if a firm’s position on off-shoring depends on the firm’s nationality, we test:

H 0 : Position and Nationality are independent


H a : Position and Nationality are dependent

From the printout, the test statistic is  2  21.242 and the p-value is p  .012 . Since the p-value is less than
 ( p  .012  .05) , H0 is rejected. There is sufficient evidence to indicate a firm’s position on off-shoring
depends on the firm’s nationality at   .05 .

10.30 Some preliminary calculations are:

R C 57(60) RC 58(60) R C 56(60)


Eˆ11  1 1   20 Eˆ 21  2 1   20.35 Eˆ 31  3 1   19.65
n 171 n 171 n 171

RC 57(111) RC 58(111) RC 56(111)


Eˆ12  1 2   37 Eˆ 22  2 2   37.65 Eˆ 32  3 2   36.35
n 171 n 171 n 171

To determine if the option choice depends on emotion state, we test:

H0: Option choice and emotion state are independent


Ha: Option choice and emotion state are dependent

The test statistic is


2
 nij  Eˆ ij 
  (45  20)  (12  37)  (8  20.35)  (50  37.65)
2 2 2 2
 2   
Eˆ ij 20 37 20.35 37.65
(75  19.65) 2 (49  36.35) 2
   72.234
19.65 36.35

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 583

The rejection region requires   .10 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .10
2
 4.60517 . The rejection region is
 2  4.60517 .

Since the observed value of the test statistic falls in the rejection region (  2  72.234  4.60517) , H0 is
rejected. There is sufficient evidence to indicate that the option choice depends on emotion state at   .10 .

10.31 Some preliminary calculations are:

R C 396(335) R C 311(335) R C 70(335)


Eˆ11  1 1   154.435 Eˆ 21  2 1   121.286 Eˆ 31  3 1   27.299
n 859 n 859 n 859

R C 39(335) R C 18(335) R C 25(335)


Eˆ 41  4 1   15.210 Eˆ 51  5 1   7.020 Eˆ 61  6 1   9.750
n 859 n 859 n 859

RC 396(524) RC 311(524) RC 70(524)


Eˆ12  1 2   241.565 Eˆ 22  2 2   189.714 Eˆ 32  3 2   42.701
n 859 n 859 n 859

R C 39(524) R C 18(524) RC 25(524)


Eˆ 42  4 2   23.790 Eˆ 52  5 2   10.980 Eˆ 62  6 2   15.250
n 859 n 859 n 859

To determine if the proportions of mobile device users in the six texting style categories depend on whether
a male or female are texting, we test:

H 0 : Texting style and sex are independent


H a : Texting style and sex are dependent

The test statistic is:

2
 nij  Eˆ ij 
  161  154.435    235  241.565     14  15.250   4.209
2 2 2

   
2

Eˆ ij
154.435 241.565 15.250

The rejection region requires   .10 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (6  1)(2  1)  5 . From Table IV, Appendix D,  .10
2
 9.23635 . The rejection region is
 2  9.23635 .

Since the observed value of the test statistic does not fall in the rejection region (  2  4.209  9.23635) , H0
is not rejected. There is insufficient evidence to indicate the proportions of mobile device users in the six
texting style categories depend on whether a male or female are texting at   .10 .

10.32 Some preliminary calculations are:

R C 234(40) RC 234(397)
Eˆ11  1 1   21.419 Eˆ12  1 2   212.581
n 437 n 437

R C 203(40) RC 203(397)
Eˆ 21  2 1   18.581 Eˆ 22  2 2   184.419
n 437 n 437

Copyright © 2014 Pearson Education, Inc.


584 Chapter 10

To determine if the response rate of air traffic controllers to mid-air collision alarms differs for true and
false alerts, we test:

H 0 : Responses and alerts are independent


H a : Responses and alerts are dependent

The test statistic is:

2
 nij  Eˆ ij 
   3  21.419    231  212.581   37  18.581  166  184.419   37.533
2 2 2 2

   
2

Eˆ ij
21.419 212.581 18.581 184.419

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 . The rejection region is
 2  3.84146 .

Since the observed value of the test statistic falls in the rejection region (  2  37.533  3.84146) , H0 is
rejected. There is sufficient evidence to indicate the response rate of air traffic controllers to mid-air
collision alarms differs for true and false alerts at   .05 .

10.33 Some preliminary calculations are:

R C 32(32) RC 32(32) RC 32(32)


Eˆ11  1 1   10.667 Eˆ 21  2 1   10.667 Eˆ 31  3 1   10.667
n 96 n 96 n 96

RC 32(64) RC 32(64) RC 32(64)


Eˆ12  1 2   21.333 Eˆ 22  2 2   21.333 Eˆ 32  3 2   21.333
n 96 n 96 n 96
To determine if the proportion of subjects who selected menus consistent with the theory depends on goal
condition, we test:

H0: Goal condition and Consistent with theory are independent


Ha: Goal condition and Consistent with theory are dependent

The test statistic is

2
 nij  Eˆ ij 
  15  10.667   17  21.333  14  10.667   18  21.333
2 2 2 2

   
2

Eˆ ij
10.667 21.333 10.667 21.333

 3  10.667   29  21.333
2 2

   12.469
10.667 21.333

The rejection region requires   .01 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .01
2
 9.21034 . The rejection region is
 2  9.21034 .

Since the observed value of the test statistic falls in the rejection region (  2  12.469  9.21034) , H0 is
rejected. There is sufficient evidence to indicate that the proportion of subjects who selected menus
consistent with the theory depends on goal condition at   .01 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 585

10.34 Using MINITAB, the results of the table comparing type of coupon user and gender are:

Tabulated statistics: USER, GENDER

Rows: USER Columns: GENDER

Female Male All

both 104 31 135


mail 178 84 262
net 36 7 43
All 318 122 440

Cell Contents: Count

Pearson Chi-Square = 6.797, DF = 2, P-Value = 0.033


Likelihood Ratio Chi-Square = 7.105, DF = 2, P-Value = 0.029

To determine if type of coupon user depends on gender, we test:

H 0 : Type of coupon user and gender are independent


H a : Type of coupon user and gender are dependent

The test statistic is  2  6.797 and the p-value is p  .033 . Since the p-value is not less than
 ( p  .033  .01) , H0 is not rejected. There is insufficient evidence to indicate type of coupon user
depends on gender at   .01 .

Using MINITAB, the results of the table comparing type of coupon user and coupon usage satisfaction
level are:

Tabulated statistics: USER, SATISF

Rows: USER Columns: SATISF

No Some Yes All

both 3 9 123 135


mail 28 62 172 262
net 4 9 30 43
All 35 80 325 440

Cell Contents: Count

Pearson Chi-Square = 30.418, DF = 4, P-Value = 0.000


Likelihood Ratio Chi-Square = 34.934, DF = 4, P-Value = 0.000

To determine if type of coupon user depends on coupon usage satisfaction level, we test:

H 0 : Type of coupon user and coupon usage satisfaction level are independent
H a : Type of coupon user and coupon usage satisfaction level are dependent

The test statistic is  2  30.418 and the p-value is p  .000 . Since the p-value is less than
 ( p  .000  .01) , H0 is rejected. There is sufficient evidence to indicate type of coupon user depends on
coupon usage satisfaction level at   .01 .

Copyright © 2014 Pearson Education, Inc.


586 Chapter 10

10.35 Using MINITAB, the results are:

Tabulated statistics: Instruction, Strategy

Rows: Instruction Columns: Strategy

Guess Other TTBC All

Cue 5 6 13 24
20.83 25.00 54.17 100.00
35.71 35.29 76.47 50.00

Pattern 9 11 4 24
37.50 45.83 16.67 100.00
64.29 64.71 23.53 50.00

All 14 17 17 48
29.17 35.42 35.42 100.00
100.00 100.00 100.00 100.00

Cell Contents: Count


% of Row
% of Column

Pearson Chi-Square = 7.378, DF = 2, P-Value = 0.025


Likelihood Ratio Chi-Square = 7.668, DF = 2, P-Value = 0.022

To determine if the choice of heuristic strategy depends on type of instruction, we test:

H0: Heuristic strategy and type of instruction are independent


Ha: Heuristic strategy and type of instruction are dependent

From the printout, the test statistic is  2  7.378 and the p-value is p  .025 .
Since the p-value is less than  ( p  .025  .05) , H0 is rejected. There is sufficient evidence to indicate the
choice of heuristic strategy depends on type of instruction at   .05 .

Since the p-value is not less than  ( p  .025  .01) , H0 is not rejected. There is insufficient evidence to
indicate the choice of heuristic strategy depends on type of instruction at   .01 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 587

10.36 a. Using MINITAB, the results for the First Trial are:

Chi-Square Test: Switch Boxes, No Switch


Expected counts are printed below observed counts
Chi-Square contributions are printed below expected counts

Switch
Boxes No Switch Total
1 10 17 27
6.50 20.50
1.885 0.598

2 3 24 27
6.50 20.50
1.885 0.598

3 5 22 27
6.50 20.50
0.346 0.110

4 8 19 27
6.50 20.50
0.346 0.110

Total 26 82 108
Chi-Sq = 5.876, DF = 3, P-Value = 0.118

To determine if the likelihood of switching boxes depends on condition for the first trial, we test:
H0: Likelihood of switching boxes and condition are independent
Ha: Likelihood of switching boxes and condition are dependent

From the printout above, the test statistic is  2  5.876 and the p-value is p  0.118 . Since the p-value
is not small, Ho is not rejected. There is insufficient evidence to indicate that the likelihood of
switching boxes depends on condition for the first trial for any value of   .118 .

Using MINITAB, the results for the Last Trial are:

Chi-Square Test: Switch Boxes, No Switch

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts
Switch
Boxes No Switch Total
1 23 4 27
18.75 8.25
0.963 2.189

2 12 15 27
18.75 8.25
2.430 5.523

3 21 6 27
18.75 8.25
0.270 0.614

4 19 8 27
18.75 8.25
0.003 0.008

Total 75 33 108
Chi-Sq = 12.000, DF = 3, P-Value = 0.007

Copyright © 2014 Pearson Education, Inc.


588 Chapter 10

To determine if the likelihood of switching boxes depends on condition for the last trial, we test:

H0: Likelihood of switching boxes and condition are independent


Ha: Likelihood of switching boxes and condition are dependent

From the printout above, the test statistic is  2  12.00 and the p-value is p  0.007 . Since the p-value
is small, H0 is rejected. There is sufficient evidence to indicate that the likelihood of switching boxes
depends on condition for the last trial for any value of   .007 .

b. Using MINITAB, the results from the Empty condition are:

Chi-Square Test: Switch Boxes, No Switch

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts

Switch
Boxes No Switch Total
1 10 17 27
16.50 10.50
2.561 4.024

2 23 4 27
16.50 10.50
2.561 4.024

Total 33 21 54
Chi-Sq = 13.169, DF = 1, P-Value = 0.000

To determine if the likelihood of switching boxes depends on trial number for the Empty condition, we
test:

H0: Likelihood of switching boxes and trial number are independent


Ha: Likelihood of switching boxes and trial number are dependent

From the printout above, the test statistic is  2  13.169 and the p-value is p  0.000 . Since the p-
value is so small, H0 is rejected. There is sufficient evidence to indicate that the likelihood of
switching boxes depends on trial number for the Empty condition for any value of   .000 .

Using MINITAB, the results from the Vanish condition are:

Chi-Square Test: Switch Boxes, No Switch

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts

Switch
Boxes No Switch Total
1 3 24 27
7.50 19.50
2.700 1.038

2 12 15 27
7.50 19.50
2.700 1.038

Total 15 39 54
Chi-Sq = 7.477, DF = 1, P-Value = 0.006

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 589

To determine if the likelihood of switching boxes depends on trial number for the Vanish condition,
we test:

H0: Likelihood of switching boxes and trial number are independent


Ha: Likelihood of switching boxes and trial number are dependent

From the printout above, the test statistic is  2  7.477 and the p-value is p  0.006 . Since the p-value
is so small, Ho is rejected. There is sufficient evidence to indicate that the likelihood of switching
boxes depends on trial number for the Vanish condition for any value of   .006 .

Using MINITAB, the results from the Steroids condition are:

Chi-Square Test: Switch Boxes, No Switch

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts
Switch
Boxes No Switch Total
1 5 22 27
13.00 14.00
4.923 4.571

2 21 6 27
13.00 14.00
4.923 4.571

Total 26 28 54
Chi-Sq = 18.989, DF = 1, P-Value = 0.000

To determine if the likelihood of switching boxes depends on trial number for the Steroids condition,
we test:

H0: Likelihood of switching boxes and trial number are independent


Ha: Likelihood of switching boxes and trial number are dependent

From the printout above, the test statistic is  2  18.989 and the p-value is p  .000 . Since the p-value
is so small, H0 is rejected. There is sufficient evidence to indicate that the likelihood of switching
boxes depends on trial number for the Steroids condition for any value of   .000 .

Using MINITAB, the results from the Steroids2 condition are:

Chi-Square Test: Switch Boxes, No Switch

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts

Switch
Boxes No Switch Total
1 8 19 27
13.50 13.50
2.241 2.241

2 19 8 27
13.50 13.50
2.241 2.241

Total 27 27 54
Chi-Sq = 8.963, DF = 1, P-Value = 0.003

Copyright © 2014 Pearson Education, Inc.


590 Chapter 10

To determine if the likelihood of switching boxes depends on trial number for the Steroids2 condition,
we test:
H0: Likelihood of switching boxes and trial number are independent
Ha: Likelihood of switching boxes and trial number are dependent

From the printout above, the test statistic is  2  8.963 and the p-value is p  .003 . Since the p-value is
so small, H0 is rejected. There is sufficient evidence to indicate that the likelihood of switching boxes
depends on trial number for the Steroids2 condition for any value of   .003 .

c. Of all the tests performed, only one was not significant. There was no evidence that the likelihood of
switching boxes depended on condition for the first trial. All other tests indicated that the variables
were dependent. Thus, both condition and trial number influence a subject to switch.

10.37 a. To determine if the vaccine is effective in treating the MN strain of HIV, we test:
H0: Vaccine status and MN strain are independent
Ha: Vaccine status and MN strain are dependent
From the printout the test statistic is  2  4.411 and the p-value is p  0.036 . Since the p-value is
less than  ( p  .036  .05) , H0 is rejected. There is sufficient evidence to indicate that the vaccine is
effective in treating the MN strain of HIV at   .05 .

b. We must assume that we have a random sample from the population of interest. We cannot really
check this assumption. The second assumption is that all expected cell counts will be 5 or more. In
this case, since there are only 7 observations in the second row, there is no way that the expected cell
counts in that row will both be 5 or more (the sum of the expected cell counts in the row must sum to
the observed row total).

 7   31  7! 31! 7  6 1 31  30 1


  
c.  2   22   2!(7  2)! 22!(31  22)!  2  5  4  3  2 1 22  21 1  9  8 1  .04378
 38  38! 38  37 1
  24!(38  24)! 24  23 1 14 13 1
 24 

d. If vaccine status and MN are independent, then the proportion of those in each group that are positive
should be very similar. In these two additional tables, the proportion of positive results for the
unvaccinated group is increasing and the proportion of positive results for the vaccinated group is
decreasing.

Table 1:

 7  31  7! 31! 7  6 1 31  30 1


  
1  23   1!(7  1)! 23!(31  23)!  1  6  5  4  3  2 1 23  22 1  8  7 1  .00571
 38  38! 38  37 1
  24!(38  24)! 24  23 1 14 13 1
 24 

Table 2:

 7   31  7! 31! 7  6 1 31  30 1


  
     0!(7  0)! 24!(31  24)!  1  7  6  5  4  3  2 1 24  23 1  7  6 1  .00027
0 24
 38  38! 38  37 1
  24!(38  24)! 24  23 1 14 13 1
 24 

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 591

e. The p-value is 04378  .00571  .00027  .04976 . Since the p-value is less than  ( p  .04976  .05) ,
H0 is rejected. There is sufficient evidence to indicate that the vaccine is effective in treating the MN
strain of HIV at   .05 .

10.38 a. Some preliminary calculations are:

50(50) 50(90) 50(110)


Eˆ11   10 Eˆ12   18 Eˆ13   22
250 250 250
100(50) 100(90) 100(110)
Eˆ 21   20 Eˆ 22   36 Eˆ 23   44
250 250 250

100(50) 100(90) 100(110)


Eˆ 31   20 Eˆ 32   36 Eˆ 33   44
250 250 250

To determine if the rows and columns are dependent, we test:

H0: Rows and columns are independent


H a: Rows and columns are dependent

2
 nij  Eˆ ij 
  (20  10)    (30  44)  54.14
2 2
The test statistic is    
2

Eˆ ij 10 44

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(3  1)  4 . From Table IV, Appendix D,  .05
2
 9.48773 . The rejection
region is  2  9.48773 .

Since the observed value of the test statistic falls in the rejection region (  2  54.15  9.48773) , H0
is rejected. There is sufficient evidence to indicate a dependence between rows and columns at
  .05 .

b. No, the analysis remains identical.

c. Yes, the assumptions differ. If the row and column totals are not fixed, then we assume that we take
a random sample form a multinomial distribution. If the row totals are fixed, then we assume that we
are taking k random samples from k multinomial populations.

d. The percentages are in the table below.

Column
1 2 3 Totals
20 20 10 50
1  100%  40%  100%  22.2% 100%  9.1%  100%  20%
50 90 110 250
Row 10 20 70 100
2  100%  20%  100%  22.2% 100%  63.6%  100%  40%
50 90 110 250
20 50 30 100
3  100%  40%  100%  55.6% 100%  27.3%  100%  40%
50 90 110 250

Copyright © 2014 Pearson Education, Inc.


592 Chapter 10

e. Using MINITAB, the bar graph is:

40

30

Percent
20 20

10

0
1 2 3
Column

The graph supports the decision in part a. In part a, we rejected the null hypothesis and concluded
that the rows and columns were dependent. If they were independent, then we would expect the three
bars to be the same height. In this graph, they are not the same height.

10.39 a. If all the categories are equally likely, then p1,0  p2,0  p3,0  p4,0  p5,0  .2 .

E1  E2  E3  E4  E5  npi ,0  150(.20)  30

To determine if the categories are not equally likely, we test:

H 0 : p1  p2  p3  p4  p5  .2
H a : At lease one of the probabilities differs from .2

[ni  Ei ]2 (28  30) 2 (35  30) 2 (33  30) 2 (25  30) 2


The test statistic is  2        2.133
Ei 30 30 30 30

The rejection region requires   .10 in the upper tail of the  2 distribution with df  k  1  5  1  4 .
From Table IV, Appendix D,  .10
2
 7.77944 . The rejection region is  2  7.77944 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  2.133  7.77944) , H0 is not rejected. There is insufficient evidence to indicate the categories
are not equally likely at   .10 .

35
b. pˆ 2   .233
150

For confidence coefficient .90,   .10 and  / 2  .10 / 2  .05 . From Table II, Appendix D,
z.05  1.645 . The confidence interval is:
pˆ 2 qˆ2 .233(.767)
pˆ 2  z.05  .233  1.645  .233  .057  (.176, .290)
n2 150

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 593

10.40 a. The qualitative variable of interest is the location of professional sports stadiums and ballparks.
There are 3 levels or categories of this variable – downtown, central city, and suburban.

b. Let p1  proportion of major sports facilities located in downtown areas, p2  proportion of major
sports facilities located in central city areas, and p3  proportion of major sports facilities located in
suburban areas in 1997.

To determine if the proportions of major sports facilities in downtown, central city, and
suburban areas in 1997 are the different than in 1985, we test:

H 0 : p1  .40, p2  .30, p3  .30


H a : At lease one of the probabilities differs from its hypothesized value

c. E1  np1,0  113(.40)  45.2 ; E2  np2,0  113(.30)  33.9 ; E3  np3,0  113(.30)  33.9

[ni  Ei ]2 (58  45.2) 2 (26  33.9) 2 (29  33.9) 2


d. The test statistic is  2       6.174
Ei 45.2 33.9 33.9

e. The degrees of freedom for the test statistic is df  k  1  3  1  2 . The p-value is p  P (  2  6.174) .

Using MINITAB,

Cumulative Distribution Function

Chi-Square with 2 DF

x P( X <= x )
6.174 0.954361

The p-value is p  1  .954361  .045639 .

Since the p-value is smaller than  ( p  .0456  .05) , H0 is rejected. There is sufficient evidence to
indicate the proportions of major sports facilities in downtown, central city, and suburban areas in
1997 are the different than in 1985.

10.41 a. The qualitative variable in this exercise is what “Made in the USA” means. There are 4 levels or
categories for this variable: 100% of labor and materials are produced in the US, 75-99% of labor
and materials are produced in the US, 50-74% of labor and materials are produced in the US, and less
than 50% of labor and materials are produced in the US.

b. The consumer advocate group hypothesized that p1  1 / 2  .5 , p2  1 / 4  .25 , p3  1 / 5  .20 , and


p4  .05 .

c. To determine if the consumer advocate group’s claim is correct, we test:

H 0 : p1  .5, p2  .25, p3  .20 and p4  .05


H a : At lease one of the probabilities differs from its hypothesized value

Copyright © 2014 Pearson Education, Inc.


594 Chapter 10

d. Some preliminary calculations are:

n  64  20  18  4  106 .

E1  np1,0  106(.50)  53 ; E2  np2,0  106(.25)  26.5 ;

E3  np3,0  106(.20)  21.2 ; E4  np4,0  106(.05)  5.3

[ni  Ei ]2 (64  53) 2 (20  26.5) 2 (18  21.2) 2 (4  5.3) 2


2        4.68
Ei 53 26.5 21.2 5.3

e. The rejection region requires   .10 in the upper tail of the  2 distribution with df  k  1  4  1  3 .
From Table IV, Appendix D,  .10
2
 6.25139 . The rejection region is  2  6.25139 .

f. Since the observed value of the test statistic does not fall in the rejection region
(  2  4.68  6.25139) , H0 is not rejected. There is insufficient evidence to indicate the consumer
advocate group’s claim is incorrect at   .10 .

n1 64
g. pˆ1    .604
n 106

For confidence coefficient .90,   .10 and  / 2  .10 / 2  .05 . From Table II, Appendix D,
z.05  1.645 . The 90% confidence interval is:

pˆ1 (1  pˆ1 ) .604(.396)


pˆ1  z.05  .604  1.645  .604  .078  (.526, .682)
n 106

We are 90% confident that the proportion of all consumers who believe “Made in the USA” means
“100%” of labor and material are produced in the US” is between .526 and .682.

10.42 a. The contingency table would be:

Tax- Itemize Deductions


motivation Yes No Total
Yes 691 381 1,072
No 794 899 1,693
Total 1,482 1,280 2,765

R C 1, 072(1, 485) RC 1, 072(1, 280)


b. Eˆ11  1 1   575.7 Eˆ12  1 2   496.3
n 2, 765 n 2, 765

R C 1, 693(1, 485) RC 1, 693(1, 280)


Eˆ 21  2 1   909.3 Eˆ 22  2 2   783.7
n 2, 765 n 2, 765

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 595

c. The test statistic is:

[nij  Eˆij ]2 [691  575.7]2 [381  496.3]2 [794  909.3]2 [899  783.7]2
 2        81.46
Eˆij
575.7 496.3 909.3 783.7

d. To determine if tax-motivation and itemize-deduction are related for charitable givers, we test:

H0: Tax-motivation and itemize-deduction are independent


Ha: Tax-motivation and itemize-deduction are dependent

The test statistic is  2  81.46 .

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 . The rejection
region is  2  3.84146 .

Since the observed value of the test statistic falls in the rejection region (  2  81.46  3.84146) , H0 is
rejected. There is sufficient evidence to indicate that tax-motivation and itemize-deduction are
related for charitable givers at   .05 .

e. To compute the bar graph, we first convert frequencies to percentages by dividing the numbers in
each column by the column total and multiplying by 100%. Also, divide the row totals by the overall
total and multiply by 100%.

Tax- Itemize Deductions


motivation Yes No Total
Yes 691 381 1072
100%  46.5% 100%  29.8% 100%  38.8%
1485 1280 2765
No 794 899 1693
100%  53.5% 100%  70.2% 100%  61.2%
1485 1280 2765
Total 1,485 1,280 2,765

Using MINITAB, the bar graph is:

50

40
38.8 %

30
Percent

20

10

0
Yes No
Itemize

Copyright © 2014 Pearson Education, Inc.


596 Chapter 10

10.43 a. Some preliminary calculations are:

xC1 175 xC 2 236 xC 3 319


pˆ C1    .028 pˆ C 2    .050 pˆ C 3    .045
n1 6, 222 n2 4, 692 n3 7,140

xC 4 231 xC 5 480 xC 6 187


pˆ C 4    .038 pˆ C 5    .046 pˆ C 6    .039
n4 6,120 n5 10,353 n6 4794

The proportions range from .028 to .050. Since .050 is about twice as big as .028, there may be
evidence to conclude some of the proportions are different.

b. Some preliminary calculations are:

RC 6, 222(37, 693) RC 6, 222(1628)


Eˆ11  1 1   5,964.39 Eˆ12  1 2   257.61
n 39,321 n 39,321

RC 4, 692(37, 693) RC 4, 692(1, 628)


Eˆ 21  2 1   4497.74 Eˆ 22  2 2   194.26
n 39,321 n 39,321

RC 7,140(37, 693) RC 7,140(1, 628)


Eˆ 31  3 1   6,844.38 Eˆ 32  3 2   295.62
n 39,321 n 39,321

RC 6,120(37, 693) RC 6,120(1, 628)


Eˆ 41  4 1   5,866.61 Eˆ 42  4 2   253.39
n 39,321 n 39,321

R C 10,353(37, 693) RC 10,353(1, 628)


Eˆ 51  5 1   9,924.36 Eˆ 52  5 2   428.64
n 39,321 n 39,321

RC 4, 794(37, 693) RC 4, 794(1, 628)


Eˆ 61  6 1   4,595.51 Eˆ 62  6 2   198.49
n 39,321 n 39,321

To determine if the proportions of censored measurements differ for the six tractor lines, we test:

H0: Tractor lines and Censored measurements are independent


Ha: Tractor lines and Censored measurements are dependent

The test statistic is


2
 nij  Eˆ ij 
   6047  5964.39   175  257.61   4456  4497.74 
2 2 2

   
2

Eˆ ij
5964.39 257.61 4497.74

187  198.49 
2

    48.0978
198.49

The rejection region requires   .01 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (6  1)(2  1)  5 . From Table IV, Appendix D,  .01
2
 15.0863 . The rejection
region is  2  15.0863 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 597

Since the observed value of the test statistic falls in the rejection region (  2  48.0978  15.0863) ,
H0 is rejected. There is sufficient evidence to indicate that the proportions of censored measurements
differ for the six tractor lines at   .01 .

c. Even though there are differences in the proportions of censured data among the 6 tractor lines, these
proportions range from .028 to .050. In practice, there is very little difference between .028 and .050.

n1 146
10.44 a. Let p1 = proportion of abstainers with congestive heart failure. Then pˆ1    .163 .
n 896

n2 106
b. Let p2 = proportion of moderate drinkers with congestive heart failure. Then pˆ 2    .152 .
n 696

n3 29
c. Let p3 = proportion of heavy drinkers with congestive heart failure. Then pˆ 3    .090 .
n 321

d. The three sample proportions found in parts a, b, and c appear to be different. It appears that the
proportion of AMI patients with congestive heart failure depends on alcohol consumption.

e. To determine if the proportion of AMI patients with congestive heart failure depends on alcohol
consumption, we test:

H0: The proportion of AMI patients with congestive heart failure is independent of alcohol
consumption

f. H a: The proportion of AMI patients with congestive heart failure depends on alcohol consumption

Using MINITAB, the results are:


Chi-Square Test: Abstain, Less 7, 7 or more

Expected counts are printed below observed counts


Chi-Square contributions are printed below expected counts

Abstain Less 7 7 or more Total


1 146 106 29 281
131.61 102.24 47.15
1.573 0.139 6.988

2 750 590 292 1632


764.39 593.76 273.85
0.271 0.024 1.203

Total 896 696 321 1913

Chi-Sq = 10.197, DF = 2, P-Value = 0.006

The test statistics is  2  10.197 , and the p-value p  0.006 .

Since the p-value is less than  ( p  .006  .05) , H0 is rejected. There is sufficient evidence to
indicate that the proportion of AMI patients with congestive heart failure depends on alcohol
consumption at   .05 .

Copyright © 2014 Pearson Education, Inc.


598 Chapter 10

10.45 a. Some preliminary calculations are:

E1  np1,0  400 .30   120 E2  np2,0  400 .20   80 E3  np3,0  400 .20   80

E4  np4,0  400 .10   40 E5  np5,0  400 .10   40 E6  np6,0  400 .10   40

b. The test statistic is

 ni  Ei 
2
(100  120) 2 (75  80) 2 (85  80) 2 (50  40) 2
2      
Ei 120 80 80 40
(40  40) 2 (50  40) 2
   8.958
40 40

c. To determine if the true percentages of the colors produced differ from the manufacturer’s stated
percentages, we test:

H 0 : p1  .30, p2  .20, p3  .20, p4  .10, p5  .10, and p6  .10


H a : At least one of the probabilities differs from the hypothesized value

The test statistic is  2  8.958 .

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  6  1  5 .
From Table IV, Appendix D,  .05
2
 11.0705 . The rejection region is  2  11.0705 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  8.958  11.0705) , H0 is not rejected. There is insufficient evidence to indicate the true
percentages of the colors produced differ from the manufacturer’s stated percentages at   .05 .

10.46 a. Some preliminary calculations are:

E1  np1,0  1000(.50)  500 E2  np2,0  1000(.22)  220

E3  np3,0  1000(.11)  110 E4  np4,0  1000(.17)  170

To determine if the percentages disagree with the percentages reported by Nielson/NetRatings,


we test:

H 0 : p1  .50, p2  .22, p3  .11 and p4  .17


H a : At lease one of the probabilities differs from its hypothesized value

The test statistic is

 ni  Ei   487  500   245  220  121  110  147  170 


2 2 2 2 2

2        7.391
Ei 500 220 110 170

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 599

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  k  1  4  1  3 . From Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is
 2  7.81473 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  7.391  7.81473) , H0 is not rejected. There is insufficient evidence to indicate the percentages
disagree with the percentages reported by Nielson/NetRatings at   .05 .

b. Some preliminary calculations are:

x1 487
pˆ1    .487
n 1000

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ1qˆ1 .487(.513)
pˆ1  z.025  .487  1.96  .487  .031  (.456, .518)
n 1000

We are 95% confident that the percentage of all Internet searches that use the Google Search Engine
is between 45.6% and 51.8%.

10.47 a. Some preliminary calculations are:

R C 53(35) RC 53(35)
Eˆ11  1 1   26.5 Eˆ12  1 2   26.5
n 70 n 70

R C 17(35) RC 17(35)
Eˆ 21  2 1   8.5 Eˆ 22  2 2   8.5
n 70 n 70

To determine if the severity of the ethical issue influenced whether the issue was identified or not by
the auditors, we test:

H0: Severity of ethical issue and identification are independent


Ha: Severity of ethical issue and identification are dependent

The test statistic is


2
 nij  Eˆ ij  2 2 2 2
   
2   (27  26.5)  (26  26.5)  (8  8.5)  (9  8.5)  .078
Eˆ ij 26.5 26.5 8.5 8.5

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 . The rejection
region is  2  3.84146 .

Since the observed value of the test statistic does not fall in the rejection region
(  2  .078  3.84146) , H0 is not rejected. There is insufficient evidence to indicate that the severity
of the ethical issue influenced whether the issue was identified or not by the auditors at   .05 .

Copyright © 2014 Pearson Education, Inc.


600 Chapter 10

b. No. If there were 0 in the bottom cell of the column, then the expected count for that cell will be less
than 5. One of the assumptions necessary for the test statistic to have a  2 distribution will not hold.

c. Suppose we change the numbers in the table to be as follows:

Severity of Ethical Issue


Moderate Severe
Ethical Issue Identified 32 21
Ethical Issue Not Identified 3 14

Since the row and column totals are the same, the expected cell counts are the same as above.

The test statistic is

2
 nij  Eˆ ij  2 2 2 2
 2      (32  26.5)  (21  26.5)  (3  8.5)  (14  8.5)  9.401
Eˆ ij 26.5 26.5 8.5 8.5

Now the test statistic would fall in the rejection region.

10.48 Some preliminary calculations are:

RC 95(118) RC 69(118) RC 42(118)


Eˆ11  1 1   42.79 Eˆ 21  2 1   31.08 Eˆ 31  3 1   18.92
n 262 n 262 n 262

RC 56(118) RC 95(144) RC 69(144)


Eˆ 41  4 1   25.22 Eˆ12  1 2   52.21 Eˆ 22  2 2   37.92
n 262 n 262 n 262

RC 42(144) RC 56(144)
Eˆ 32  3 2   23.08 Eˆ 42  4 2   30.78
n 262 n 262

To determine whether a pig farmer’s education level has an impact on the size of the pig farm, we test:

H0: Pig farmer’s education level and size of pig farm are independent
Ha: Pig farmer’s education level and size of pig farm are dependent

The test statistic is

[nij  Eˆ ij ]2 (42  42.79) 2 (53  52.21) 2 (27  31.08) 2 (42  37.92) 2 (22  18.92) 2
 2       
Eˆ ij 42.79 52.21 31.08 37.92 18.92
(20  23.08) 2 (27  25.22) 2 (29  30.78) 2
    2.14
23.08 25.22 30.78

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (4  1)(2  1)  3 . From Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is
 2  7.81473 .

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 601

Since the observed value of the test statistic does not fall in the rejection region (  2  2.14  7.81473) , H0
is not rejected. There is insufficient evidence to indicate that a pig farmer’s education level has an impact
on the size of the pig farm at   .05 .

To compute the bar graph, we first convert frequencies to percentages by dividing the numbers in each row
by the row total and multiplying by 100%. Also, divide the column totals by the overall total and multiply
by 100%.

Farm Size Education Level


No college College Total
<1,000 pigs 42 53 95
 100%  44.2%  100%  55.8%
95 95
1,000-2,000 27 42 69
pigs  100%  39.1%  100%  60.9%
69 69
2,000-5,000 22 20 42
pigs 100%  52.4% 100%  47.6%
42 42
> 5,000 27 29 56
pigs  100%  48.2%  100%  51.8%
56 56
Total 118 144 262
 100%  45.0%  100%  55.0%
262 262

Using MINITAB, the bar graph is:

50

45

40
Percent

30

20

10

0
<1,000 1,000-2,000 2,000-5,000 >5,000
Farm Size

Since the bars are all similar in height, it supports the conclusion to the test above.

10.49 Some preliminary calculations are: E1  E2  E3  E4  np1,0  83(.25)  20.75

To determine if there are differences in the percentages of incidents in the four cause categories, we test:

H 0 : p1 = p2  p3  p4  .25
H a : At lease one of the probabilities differs from its hypothesized value

Copyright © 2014 Pearson Education, Inc.


602 Chapter 10

The test statistic is


 n  Ei    27  20.75 2   24  20.75 2   22  20.75 2  10  20.752  8.036
2

2   i
Ei 20.75 20.75 20.75 20.75

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  4  1  3 . From
Table IV, Appendix D,  .05
2
 7.81473 . The rejection region is  2  7.81473 .

Since the observed value of the test statistic falls in the rejection region (  2  8.036  7.81473) , H0 is
rejected. There is sufficient evidence to indicate there are differences in the percentages of incidents in the
four cause categories at   .05 .

10.50 a. The two qualitative variables are years (1990, 1991, . . . , 2000) and acquisition status (yes or no).

b. To determine if year and acquisition status are dependent, we test:

H0: Year and acquisition status are independent


Ha: Year and acquisition status are dependent

c. From the printout, the test statistic is  2  297.048 and the p-value is p  0.000 . Since the p-value is
less than  ( p  0.000  .05) , H0 is rejected. There is sufficient evidence to indicate that year and
acquisition status are dependent at   .05 .

10.51 a. The contingency table is:

Committee
Acceptable Rejected Totals
Acceptable 101 23 124
Inspector
Rejected 10 19 29
Totals 111 42 153

b. Yes. To plot the percentages, first convert frequencies to percentages by dividing the numbers in
each column by the column total and multiplying by 100. Also, divide the row totals by the overall
total and multiply by 100.

Acceptable Rejected Totals


Acceptable 101 23 124
100  90.99% 100  54.76% 100  81.05%
111 42 153
Inspector
Rejected 10 19 29
100  9.01% 100  45.23% 100  18.95%
111 42 153

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 603

Using MINITAB, the graph of the data is:

90

80 81.1

70

Percent 60

50

40

30

20

10

0
Acceptable Rejected
Committee

Since the heights of the bars are not similar, it appears there is a relationship.

c. Some preliminary calculations are:

R C 124(111) R C 124(42)
Eˆ11  1 1   89.691 Eˆ 21  1 2   34.039
n1 153 n1 153

R C 29(111) RC 29(42)
Eˆ 21  2 1   21.039 Eˆ 22  2 2   7.961
n1 153 n1 153

To determine if the inspector's classifications and the committee's classifications are related, we test:

H0: The inspector's and committee's classification are independent


Ha: The inspector's and committee's classifications are dependent

The test statistic is

2
 nij  Eˆ ij 
  (101  89.961)  (23  34.039)  (10  21.039)  (19  7.961)  26.034
2 2 2 2
   
2

Eˆ ij 89.961 34.039 21.039 7.961

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 . The rejection
region is  2  3.84146 .

Since the observed value of the test statistic falls in the rejection region (  2  26.034  3.84146) , H0
is rejected. There is sufficient evidence to indicate the inspector's and committee's classifications are
related at   .05 . This indicates that the inspector and committee tend to make the same decisions.

10.52 a. Some preliminary calculations are:

E1  np1,0  85(.26)  22.1 E2  np2,0  85(.30)  25.5 E3  np3,0  85(.11)  9.35

E4  np4,0  85(.14)  11.9 E5  np2,0  85(.19)  16.15

Copyright © 2014 Pearson Education, Inc.


604 Chapter 10

To determine if probabilities differ from the hypothesized values, we test:

H 0 : p1  .26, p2  .30, p3  .11, p4  .14 and p5  .19


H a : At lease one of the probabilities differs from its hypothesized value

The test statistic is

2
 ni  E i  (32  22.1) 2 (26  25.5) 2 (15  9.35) 2 (6  11.9) 2 (6  16.15) 2
 2
      17.16
Ei 2 22.1 25.5 9.35 11.9 16.15

The rejection region requires   .05 in the upper tail of the  2 distribution with df  k  1  5  1  4 .
From Table IV, Appendix D,  .05
2
 9.48773 . The rejection region is  2  9.48773 .

Since the observed value of the test statistic falls in the rejection region (  2  17.16  9.48773) ,
reject H0. There is sufficient evidence to indicate the probabilities differ from their hypothesized
values at   .05 .

n1 32
b. pˆ1    .376
n 85

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ1 (1  pˆ1 ) .376(1  .376)


pˆ1  z.025  .376  1.96  .376  .103  .273, .479 
n 85

c. The interval tells us that between 27.3% and 47.9% of the Avonex MS patients are exacerbation-free
during a two-year period. Since this interval is completely above the percentage of placebo patients
(26%), it seems that the Avonex patients are more likely to have no exacerbations than placebo
patients.

10.53 a. The contingency table is:

Flight Response
Altitude Low High Totals
< 300 85 105 190
300-600 77 121 198
 600 17 59 76
Totals 179 285 464

b. Some preliminary calculations are:

R C 190(179) RC 190(285)
Eˆ11  1 1   73.297 Eˆ12  1 2   116.703
n 464 n 464

R C 198(179) RC 198(285)
Eˆ 21  2 1   76.384 Eˆ 22  2 2   121.616
n 464 n 464

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 605

RC 76(179) RC 76(285)
Eˆ 31  3 1   29.319 Eˆ 32  3 2   46.681
n 464 n 464

To determine if flight response of the geese depends on the altitude of the helicopter, we test:

H0: Flight response and Altitude of helicopter are independent


Ha: Flight response and Altitude of helicopter are dependent

The test statistic is


2
 nij  Eˆ ij  85  73.297  105  116.703  77  76.384 
2 2 2

  
2     
Eˆ ij 73.297 116.703 76.384

121  121.616  17  29.319   59  46.681


2 2 2

    11.477
121.616 29.319 46.681

The rejection region requires   .01 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .01
2
 9.21034 . The rejection
region is  2  9.21034 .

Since the observed value of the test statistic falls in the rejection region (  2  11.477  9.21034) , H0
is rejected. There is sufficient evidence to indicate that the flight response of the geese depends on
the altitude of the helicopter at   .01 .

c. The contingency table is:

Flight Response
Lateral
Distance Low High Totals
< 1000 37 243 280
1000-2000 68 37 105
2000-3000 44 4 48
 3000 30 1 31
Totals 179 285 464

d. Some preliminary calculations are:

RC 280(179) RC 280(285)
Eˆ11  1 1   108.017 Eˆ12  1 2   171.983
n 464 n 464

R C 105(179) RC 105(285)
Eˆ 21  2 1   40.506 Eˆ 22  2 2   64.494
n 464 n 464

RC 48(179) RC 48(285)
Eˆ 31  3 1   18.517 Eˆ 32  3 2   29.483
n 464 n 464

R C 31(179) RC 31(285)
Eˆ 41  4 1   11.959 Eˆ 42  4 2   19.041
n 464 n 464

Copyright © 2014 Pearson Education, Inc.


606 Chapter 10

To determine if flight response of the geese depends on the lateral distance of the helicopter, we test:
H0: Flight response and Lateral distance of the helicopter are independent
Ha: Flight response and Lateral distance of the helicopter are dependent

The test statistic is


2
 nij  Eˆ ij   37  108.017   243  171.983  68  40.506   37  64.494 
2 2 2 2

  
2      

ij
108.017 171.983 40.506 64.494

 44  18.517   4  29.494   30  11.959  1  19.041


2 2 2 2

     207.814
18.517 29.494 11.959 19.041

The rejection region requires   .01 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (4  1)(2  1)  3 . From Table IV, Appendix D,  .01
2
 11.3449 . The rejection
region is  2  11.3449 .

Since the observed value of the test statistic falls in the rejection region (  2  207.814  11.3449) , H0
is rejected. There is sufficient evidence to indicate that the flight response of the geese depends on
the lateral distance of the helicopter at   .01 .

e. Using SAS, the contingency table for altitude by response with the column percents is:

Table of ALTGRP by RESPONSE

ALTGRP RESPONSE

Frequency|
Percent |
Row Pct |
Col Pct |LOW |HIGH | Total
---------+--------+--------+
<300 | 85 | 105 | 190
| 18.32 | 22.63 | 40.95
| 44.74 | 55.26 |
| 47.49 | 36.84 |
---------+--------+--------+
300-600 | 77 | 121 | 198
| 16.59 | 26.08 | 42.67
| 38.89 | 61.11 |
| 43.02 | 42.46 |
---------+--------+--------+
600+ | 17 | 59 | 76
| 3.66 | 12.72 | 16.38
| 22.37 | 77.63 |
| 9.50 | 20.70 |
---------+--------+--------+
Total 179 285 464
38.58 61.42 100.00

Statistics for Table of ALTGRP by RESPONSE

Statistic DF Value Prob


------------------------------------------------------
Chi-Square 2 11.4770 0.0032
Likelihood Ratio Chi-Square 2 12.1040 0.0024
Mantel-Haenszel Chi-Square 1 10.2104 0.0014
Phi Coefficient 0.1573
Contingency Coefficient 0.1554
Cramer's V 0.1573
Sample Size = 464

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 607

From the row percents, it appears that the lower the plane, the lower the response. For altitude
<300m, 55.26% of the geese had a high response. For altitude 300-600m, 61.11% of the geese had a
high response. For altitude 600+m, 77.63% of the geese had a high response. Thus, instead of
setting a minimum altitude for the planes, we need to set a maximum altitude. For this data, the
lowest response is at an altitude of < 300 meters.

Using SAS, the contingency table for lateral distance by response with the column percents is:
The FREQ Procedure

Table of LATGRP by RESPONSE

LATGRP RESPONSE

Frequency |
Percent |
Row Pct |
Col Pct |LOW |HIGH | Total
----------+--------+--------+
<1000 | 37 | 243 | 280
| 7.97 | 52.37 | 60.34
| 13.21 | 86.79 |
| 20.67 | 85.26 |
----------+--------+--------+
1000-2000 | 68 | 37 | 105
| 14.66 | 7.97 | 22.63
| 64.76 | 35.24 |
| 37.99 | 12.98 |
----------+--------+--------+
2000-3000 | 44 | 4 | 48
| 9.48 | 0.86 | 10.34
| 91.67 | 8.33 |
| 24.58 | 1.40 |
----------+--------+--------+
3000+ | 30 | 1 | 31
| 6.47 | 0.22 | 6.68
| 96.77 | 3.23 |
| 16.76 | 0.35 |
----------+--------+--------+
Total 179 284 464
38.58 61.42 100.00

Statistics for Table of LATGRP by RESPONSE

Statistic DF Value Prob


------------------------------------------------------
Chi-Square 3 207.0812 <.0001
Likelihood Ratio Chi-Square 3 227.5212 <.0001
Mantel-Haenszel Chi-Square 1 189.2843 <.0001
Phi Coefficient 0.6692
Contingency Coefficient 0.5562
Cramer's V 0.6692

Sample Size = 464

From the row percents, it appears that the greater the lateral distance, the lower the response. For a
lateral distance of 3000+m only 3.23% of the geese had a high response. Thus, the further away the
plane is laterally, the lower the response. For this data, the lowest response is when the plane is
further than 3000 meters.

Thus, the recommendation would be a maximum height of 300 m and a minimum lateral distance of
3000 m.

Copyright © 2014 Pearson Education, Inc.


608 Chapter 10

10.54 a. Some preliminary calculations are:

The contingency table is:

Defectives Non-Defectives Total


Shift 1 25 175 200
2 35 165 200
3 80 120 200
Total 140 460 600

RC 200(140) 200(140)
Eˆ11  1 1 =  46.667 Eˆ 21  Eˆ 31   46.667
n 600 600

200(460)
Eˆ12  Eˆ 22  Eˆ 32   153.333
600

To determine if quality of the filters are related to shift, we test:

H0: Quality of filters and shift are independent


H a: Quality of filters and shift are dependent

The test statistic is

[nij  Eˆij ]2  25  46.667   35  46.667   80  46.667  175  153.333


2 2 2 2

 2      
Eˆij
46.667 46.667 46.667 153.333

165  153.333 120  153.333


2 2

   47.98
153.333 153.333

The rejection region requires   .05 in the upper tail of the  2 distribution with
df  ( r  1)(c  1)  (3  1)(2  1)  2 . From Table IV, Appendix D,  .05
2
 5.99147 . The rejection
region is  2  5.99147 .

Since the observed value of the test statistic falls in the rejection region (  2  47.98  5.99147) , H0
is rejected. There is sufficient evidence to indicate quality of filters and shift are related at   .05 .

25
b. pˆ1   .125
200

For confidence coefficient .95,   .05 and  / 2  .05 / 2  .025 . From Table II, Appendix D,
z.025  1.96 . The 95% confidence interval is:

pˆ1 (1  pˆ1 ) .125(.875)


pˆ1  z.025  .125  1.96  .125  .046  .079, .171
n 200

Copyright © 2014 Pearson Education, Inc.


Categorical Data Analysis 609

[ni  Ei ]2 (26  23) 2 (146  136) 2 (361  341) 2 (143  136) 2 (13  23) 2
10.55 a. 2         9.647
Ei 23 136 341 136 23
b. From Table IV, Appendix D, with df  5 ,  .05 2
 11.0705
c. No. Since the observed value of the test statistics does not fall in the rejection region
(  2  9.647  11.0705) , H0 is not rejected. There is insufficient evidence to indicate the salary
distribution is non-normal for   .05 .

d. The p-value is p  P (  2  9.647) . Using MINITAB,

Cumulative Distribution Function

Chi-Square with 5 DF

x P( X <= x )
9.647 0.914122

The p-value is p  P (  2  9.647)  1  .914122  .085878 .

10.56 Using MINITAB, the results are:

Tabulated statistics: Defect, PredEVG

Using frequencies in Fr

Rows: Defect Columns: PredEVG

1 2 All

1 441 8 449
2 47 2 49
All 488 10 498

Cell Contents: Count

Pearson Chi-Square = 1.188, DF = 1


Likelihood Ratio Chi-Square = 0.948, DF = 1

To determine if Defect and Pred_EVG are dependent, we test:

H 0 : Defect and Pred_EVG are independent


H a : Defect and Pred_EVG are dependent

The test statistic is  2  1.188 .

Since no  level was given, we will use   .05 . The rejection region requires   .05 in the upper tail of
the  2 distribution with df  ( r  1)(c  1)  (2  1)(2  1)  1 . From Table IV, Appendix D,  .05
2
 3.84146 .
The rejection region is  2  3.84146 .

Since the observed value of the test statistic does not fall in the rejection region (  2  1.188  3.84146) , H0
is not rejected. There is insufficient evidence to indicate that Defect and Pred_EVG are dependent at
  .05 . If Defect and Pred_EVG are independent, then the Pred_EVG is no better predicting defects than
just guessing. I would not recommend the essential complexity algorithm be used as a predictor of
defective software modules.

Copyright © 2014 Pearson Education, Inc.


610 Chapter 10

10.57 Using SAS, the output is:

The FREQ Procedure

Table of CANDIDATE by TIME

CANDIDATE TIME

Frequency|
Col Pct | 1| 2| 3| 4| 5| 6| Total
---------+--------+--------+--------+--------+--------+--------+
SMITH | 208 | 208 | 451 | 392 | 351 | 410 | 2020
| 52.53 | 55.32 | 55.34 | 55.92 | 56.16 | 55.33 |
---------+--------+--------+--------+--------+--------+--------+
COPPIN | 55 | 51 | 109 | 98 | 88 | 104 | 505
| 13.89 | 13.56 | 13.37 | 13.98 | 14.08 | 14.04 |
---------+--------+--------+--------+--------+--------+--------+
MONTES | 133 | 117 | 255 | 211 | 186 | 227 | 1129
| 33.59 | 31.12 | 31.29 | 30.10 | 29.76 | 30.63 |
---------+--------+--------+--------+--------+--------+--------+
Total 396 376 815 701 625 741 3654

Statistics for Table of CANDIDATE by TIME

Statistic DF Value Prob


------------------------------------------------------
Chi-Square 10 2.2839 0.9937
Likelihood Ratio Chi-Square 10 2.2722 0.9938
Mantel-Haenszel Chi-Square 1 0.9851 0.3209
Phi Coefficient 0.0250
Contingency Coefficient 0.0250
Cramer's V 0.0177

Sample Size = 3654

To determine if candidates received votes independent of time period, we test:

H0: Voting and Time period are independent


Ha: Voting and Time period are dependent

The test statistic is  2  2.2839 .

Since no value of  was given, we will use   .05 . The rejection region requires   .05 in the upper tail
of the  2 distribution with df  ( r  1)(c  1)  (3  1)(6  1)  10 . From Table IV, Appendix D,
 .05
2
 18.3070 . The rejection region is  2  18.3070 .

Since the observed value of the test statistic does not fall in the rejection region (  2  2.2839  18.3070) ,
H0 is not rejected. There is insufficient evidence to indicate Voting and Time period are dependent at
  .05 . Thus, we can conclude that voting and time period are independent. This means that regardless
of time period, the percentage of votes received by each candidate is the same. In the table created by SAS,
the bottom number in each cell is the column percent. This is the percent of votes received by the
candidate in each time period. An inspection of these percents indicates that candidate Smith received
approximately 55.3% of the votes each time period, candidate Coppin received approximately 13.8% of the
vote, and candidate Montes received approximately 30.9% of the vote. All of this indicates that the
election was rigged

Copyright © 2014 Pearson Education, Inc.


Another random document with
no related content on Scribd:
estate of course. Now it’s up to us to get in on the next great clean-
up.... It’s almost here.... Buy Forty....”
The man with the diamond stud raised one eyebrow and shook
his head. “For one night on Beauty’s lap, O put gross care away ... or
something of the sort.... Waiter why in holy hell are you so long with
the champagne?” He got to his feet, coughed in his hand and began
to sing in his croaking voice:
O would the Atlantic were all champagne
Bright billows of champagne.
Everybody clapped. The old waiter had just divided a baked
Alaska and, his face like a beet, was prying out a stiff
champagnecork. When the cork popped the lady in the tiara let out a
yell. They toasted the man in the diamond stud.
For he’s a jolly good fellow ...
“Now what kind of a dish d’ye call this?” the man with the
bottlenose leaned over and asked the girl next to him. Her black hair
parted in the middle; she wore a palegreen dress with puffy sleeves.
He winked slowly and then stared hard into her black eyes.
“This here’s the fanciest cookin I ever put in my mouth.... D’ye
know young leddy, I dont come to this town often....” He gulped down
the rest of his glass. “An when I do I usually go away kinder
disgusted....” His look bright and feverish from the champagne
explored the contours of her neck and shoulders and roamed down a
bare arm. “But this time I kinder think....”
“It must be a great life prospecting,” she interrupted flushing.
“It was a great life in the old days, a rough life but a man’s life....
I’m glad I made my pile in the old days.... Wouldnt have the same
luck now.”
She looked up at him. “How modest you are to call it luck.”
Emile was standing outside the door of the private room. There
was nothing more to serve. The redhaired girl from the cloakroom
walked by with a big flounced cape on her arm. He smiled, tried to
catch her eye. She sniffed and tossed her nose in the air. Wont look
at me because I’m a waiter. When I make some money I’ll show ’em.
“Dis; tella Charlie two more bottle Moet and Chandon, Gout
Americain,” came the old waiter’s hissing voice in his ear.
The moonfaced man was on his feet. “Ladies and Gentlemen....”
“Silence in the pigsty ...” piped up a voice.
“The big sow wants to talk,” said Olga under her breath.
“Ladies and gentlemen owing to the unfortunate absence of our
star of Bethlehem and fulltime act....”
“Gilly dont blaspheme,” said the lady with the tiara.
“Ladies and gentlemen, unaccustomed as I am....”
“Gilly you’re drunk.”
“... Whether the tide ... I mean whether the waters be with us or
against us...”
Somebody yanked at his coat-tails and the moonfaced man sat
down suddenly in his chair.
“It’s terrible,” said the lady in the tiara addressing herself to a man
with a long face the color of tobacco who sat at the end of the table
... “It’s terrible, Colonel, the way Gilly gets blasphemous when he’s
been drinking...”
The Colonel was meticulously rolling the tinfoil off a cigar. “Dear
me, you dont say?” he drawled. Above the bristly gray mustache his
face was expressionless. “There’s a most dreadful story about poor
old Atkins, Elliott Atkins who used to be with Mansfield...”
“Indeed?” said the Colonel icily as he slit the end of the cigar with
a small pearlhandled penknife.
“Say Chester did you hear that Mabie Evans was making a hit?”
“Honestly Olga I dont see how she does it. She has no figure...”
“Well he made a speech, drunk as a lord you understand, one
night when they were barnstorming in Kansas...”
“She cant sing...”
“The poor fellow never did go very strong in the bright lights...”
“She hasnt the slightest particle of figure...”
“And made a sort of Bob Ingersoll speech...”
“The dear old feller.... Ah I knew him well out in Chicago in the old
days...”
“You dont say.” The Colonel held a lighted match carefully to the
end of his cigar...
“And there was a terrible flash of lightning and a ball of fire came
in one window and went out the other.”
“Was he ... er ... killed?” The Colonel sent a blue puff of smoke
towards the ceiling.
“What, did you say Bob Ingersoll had been struck by lightning?”
cried Olga shrilly. “Serve him right the horrid atheist.”
“No not exactly, but it scared him into a realization of the important
things of life and now he’s joined the Methodist church.”
“Funny how many actors get to be ministers.”
“Cant get an audience any other way,” creaked the man with the
diamond stud.
The two waiters hovered outside the door listening to the racket
inside. “Tas de sacrés cochons ... sporca madonna!” hissed the old
waiter. Emile shrugged his shoulders. “That brunette girl make eyes
at you all night...” He brought his face near Emile’s and winked.
“Sure, maybe you pick up somethin good.”
“I dont want any of them or their dirty diseases either.”
The old waiter slapped his thigh. “No young men nowadays....
When I was young man I take heap o chances.”
“They dont even look at you ...” said Emile through clenched
teeth. “An animated dress suit that’s all.”
“Wait a minute, you learn by and by.”
The door opened. They bowed respectfully towards the diamond
stud. Somebody had drawn a pair of woman’s legs on his shirtfront.
There was a bright flush on each of his cheeks. The lower lid of one
eye sagged, giving his weasle face a quizzical lobsided look.
“Wazzahell, Marco wazzahell?” he was muttering. “We aint got a
thing to drink.... Bring the Atlantic Ozz-shen and two quarts.”
“De suite monsieur....” The old waiter bowed. “Emile tell Auguste,
immediatement et bien frappé.”
As Emile went down the corridor he could hear singing.
O would the Atlantic were all champagne
Bright bi-i-i....
The moonface and the bottlenose were coming back from the
lavatory reeling arm in arm among the palms in the hall.
“These damn fools make me sick.”
“Yessir these aint the champagne suppers we used to have in
Frisco in the ole days.”
“Ah those were great days those.”
“By the way,” the moonfaced man steadied himself against the
wall, “Holyoke ole fella, did you shee that very nobby little article on
the rubber trade I got into the morning papers.... That’ll make the
investors nibble ... like lil mishe.”
“Whash you know about rubber?... The stuff aint no good.”
“You wait an shee, Holyoke ole fella or you looshing opportunity of
your life.... Drunk or sober I can smell money ... on the wind.”
“Why aint you got any then?” The bottlenosed man’s beefred face
went purple; he doubled up letting out great hoots of laughter.
“Because I always let my friends in on my tips,” said the other
man soberly. “Hay boy where’s zis here private dinin room?”
“Par ici monsieur.”
A red accordionpleated dress swirled past them, a little oval face
framed by brown flat curls, pearly teeth in an open-mouthed laugh.
“Fifi Waters,” everyone shouted. “Why my darlin lil Fifi, come to
my arms.”
She was lifted onto a chair where she stood jiggling from one foot
to the other, champagne dripping out of a tipped glass.
“Merry Christmas.”
“Happy New Year.”
“Many returns of the day....”
A fair young man who had followed her in was reeling intricately
round the table singing:
O we went to the animals’ fair
And the birds and the beasts were there
And the big baboon
By the light of the moon
Was combing his auburn hair.
“Hoopla,” cried Fifi Waters and mussed the gray hair of the man
with the diamond stud. “Hoopla.” She jumped down with a kick,
pranced round the room, kicking high with her skirts fluffed up round
her knees.
“Oh la la ze French high kicker!”
“Look out for the Pony Ballet.”
Her slender legs, shiny black silk stockings tapering to red
rosetted slippers flashed in the men’s faces.
“She’s a mad thing,” cried the lady in the tiara.
Hoopla. Holyoke was swaying in the doorway with his top hat
tilted over the glowing bulb of his nose. She let out a whoop and
kicked it off.
“It’s a goal,” everyone cried.
“For crissake you kicked me in the eye.”
She stared at him a second with round eyes and then burst into
tears on the broad shirtfront of the diamond stud. “I wont be insulted
like that,” she sobbed.
“Rub the other eye.”
“Get a bandage someone.”
“Goddam it she may have put his eye out.”
“Call a cab there waiter.”
“Where’s a doctor?”
“That’s hell to pay ole fella.”
A handkerchief full of tears and blood pressed to his eye the
bottlenosed man stumbled out. The men and women crowded
through the door after him; last went the blond young man, reeling
and singing:
An the big baboon by the light of the moon
Was combing his auburn hair.
Fifi Waters was sobbing with her head on the table.
“Dont cry Fifi,” said the Colonel who was still sitting where he had
sat all the evening. “Here’s something I rather fancy might do you
good.” He pushed a glass of champagne towards her down the
table.
She sniffled and began drinking it in little sips. “Hullo Roger, how’s
the boy?”
“The boy’s quite well thank you.... Rather bored, dont you know?
An evening with such infernal bounders....”
“I’m hungry.”
“There doesnt seem to be anything left to eat.”
“I didnt know you’d be here or I’d have come earlier, honest.”
“Would you indeed?... Now that’s very nice.”
The long ash dropped from the Colonel’s cigar; he got to his feet.
“Now Fifi, I’ll call a cab and we’ll go for a ride in the Park....”
She drank down her champagne and nodded brightly. “Dear me
it’s four o’clock....” “You have the proper wraps haven’t you?”
She nodded again.
“Splendid Fifi ... I say you are in form.” The Colonel’s cigarcolored
face was unraveling in smiles. “Well, come along.”
She looked about her in a dazed way. “Didnt I come with
somebody?”
“Quite unnecessary!”
In the hall they came upon the fair young man quietly vomiting
into a firebucket under an artificial palm.
“Oh let’s leave him,” she said wrinkling up her nose.
“Quite unnecessary,” said the Colonel.
Emile brought their wraps. The redhaired girl had gone home.
“Look here, boy.” The Colonel waved his cane. “Call me a cab
please.... Be sure the horse is decent and the driver is sober.”
“De suite monsieur.”
The sky beyond roofs and chimneys was the blue of a sapphire.
The Colonel took three or four deep sniffs of the dawnsmelling air
and threw his cigar into the gutter. “Suppose we have a bit of
breakfast at Cleremont. I haven’t had anything fit to eat all night.
That beastly sweet champagne, ugh!”
Fifi giggled. After the Colonel had examined the horse’s fetlocks
and patted his head, they climbed into the cab. The Colonel fitted in
Fifi carefully under his arm and they drove off. Emile stood a second
in the door of the restaurant uncrumpling a five dollar bill. He was
tired and his insteps ached.
When Emile came out of the back door of the restaurant he found
Congo waiting for him sitting on the doorstep. Congo’s skin had a
green chilly look under the frayed turned up coatcollar.
“This is my friend,” Emile said to Marco. “Came over on the same
boat.”
“You havent a bottle of fine under your coat have you? Sapristi
I’ve seen some chickens not half bad come out of this place.”
“But what’s the matter?”
“Lost my job that’s all.... I wont have to take any more off that guy.
Come over and drink a coffee.”
They ordered coffee and doughnuts in a lunchwagon on a vacant
lot.
“Eh bien you like it this sacred pig of a country?” asked Marco.
“Why not? I like it anywhere. It’s all the same, in France you are
paid badly and live well; here you are paid well and live badly.”
“Questo paese e completamente soto sopra.”
“I think I’ll go to sea again....”
“Say why de hell doan yous guys loin English?” said the man with
a cauliflower face who slapped the three mugs of coffee down on the
counter.
“If we talk Engleesh,” snapped Marco “maybe you no lika what we
say.”
“Why did they fire you?”
“Merde. I dont know. I had an argument with the old camel who
runs the place.... He lived next door to the stables; as well as
washing the carriages he made me scrub the floors in his house....
His wife, she had a face like this.” Congo sucked in his lips and tried
to look crosseyed.
Marco laughed. “Santissima Maria putana!”
“How did you talk to them?”
“They pointed to things; then I nodded my head and said Awright.
I went there at eight and worked till six and they gave me every day
more filthy things to do.... Last night they tell me to clean out the
toilet in the bathroom. I shook my head.... That’s woman’s work....
She got very angry and started screeching. Then I began to learn
Angleesh.... Go awright to ’ell, I says to her.... Then the old man
comes and chases me out into a street with a carriage whip and
says he wont pay me my week.... While we were arguing he got a
policeman, and when I try to explain to the policeman that the old
man owed me ten dollars for the week, he says Beat it you lousy
wop, and cracks me on the coco with his nightstick.... Merde alors...”
Marco was red in the face. “He call you lousy wop?”
Congo nodded his mouth full of doughnut.
“Notten but shanty Irish himself,” muttered Marco in English. “I’m
fed up with this rotten town....
“It’s the same all over the world, the police beating us up, rich
people cheating us out of their starvation wages, and who’s fault?...
Dio cane! Your fault, my fault, Emile’s fault....”
“We didn’t make the world.... They did or maybe God did.”
“God’s on their side, like a policeman.... When the day comes
we’ll kill God.... I am an anarchist.”
Congo hummed “les bourgeois à la lanterne nom de dieu.”
“Are you one of us?”
Congo shrugged his shoulders. “I’m not a catholic or a protestant;
I haven’t any money and I haven’t any work. Look at that.” Congo
pointed with a dirty finger to a long rip on his trouserknee. “That’s
anarchist.... Hell I’m going out to Senegal and get to be a nigger.”
“You look like one already,” laughed Emile.
“That’s why they call me Congo.”
“But that’s all silly,” went on Emile. “People are all the same. It’s
only that some people get ahead and others dont.... That’s why I
came to New York.”
“Dio cane I think that too twentyfive years ago.... When you’re old
like me you know better. Doesnt the shame of it get you sometimes?
Here” ... he tapped with his knuckles on his stiff shirtfront.... “I feel it
hot and like choking me here.... Then I say to myself Courage our
day is coming, our day of blood.”
“I say to myself,” said Emile “When you have some money old
kid.”
“Listen, before I leave Torino when I go last time to see the mama
I go to a meetin of comrades.... A fellow from Capua got up to speak
... a very handsome man, tall and very thin.... He said that there
would be no more force when after the revolution nobody lived off
another man’s work.... Police, governments, armies, presidents,
kings ... all that is force. Force is not real; it is illusion. The working
man makes all that himself because he believes it. The day that we
stop believing in money and property it will be like a dream when we
wake up. We will not need bombs or barricades.... Religion, politics,
democracy all that is to keep us asleep.... Everybody must go round
telling people: Wake up!”
“When you go down into the street I’ll be with you,” said Congo.
“You know that man I tell about?... That man Errico Malatesta, in
Italy greatest man after Garibaldi.... He give his whole life in jail and
exile, in Egypt, in England, in South America, everywhere.... If I
could be a man like that, I dont care what they do; they can string me
up, shoot me ... I dont care ... I am very happy.”
“But he must be crazy a feller like that,” said Emile slowly. “He
must be crazy.”
Marco gulped down the last of his coffee. “Wait a minute. You are
too young. You will understand.... One by one they make us
understand.... And remember what I say.... Maybe I’m too old,
maybe I’m dead, but it will come when the working people awake
from slavery.... You will walk out in the street and the police will run
away, you will go into a bank and there will be money poured out on
the floor and you wont stoop to pick it up, no more good.... All over
the world we are preparing. There are comrades even in China....
Your Commune in France was the beginning ... socialism failed. It’s
for the anarchists to strike the next blow.... If we fail there will be
others....”
Congo yawned, “I am sleepy as a dog.”
Outside the lemoncolored dawn was drenching the empty streets,
dripping from cornices, from the rails of fire escapes, from the rims of
ashcans, shattering the blocks of shadow between buildings. The
streetlights were out. At a corner they looked up Broadway that was
narrow and scorched as if a fire had gutted it.
“I never see the dawn,” said Marco, his voice rattling in his throat,
“that I dont say to myself perhaps ... perhaps today.” He cleared his
throat and spat against the base of a lamppost; then he moved away
from them with his waddling step, taking hard short sniffs of the cool
air.
“Is that true, Congo, about shipping again?”
“Why not? Got to see the world a bit...”
“I’ll miss you.... I’ll have to find another room.”
“You’ll find another friend to bunk with.”
“But if you do that you’ll stay a sailor all your life.”
“What does it matter? When you are rich and married I’ll come
and visit you.”
They were walking down Sixth Avenue. An L train roared above
their heads leaving a humming rattle to fade among the girders after
it had passed.
“Why dont you get another job and stay on a while?”
Congo produced two bent cigarettes out of the breast pocket of
his coat, handed one to Emile, struck a match on the seat of his
trousers, and let the smoke out slowly through his nose. “I’m fed up
with it here I tell you....” He brought his flat hand up across his
Adam’s apple, “up to here.... Maybe I’ll go home an visit the little girls
of Bordeaux.... At least they are not all made of whalebone.... I’ll
engage myself as a volunteer in the navy and wear a red pompom....
The girls like that. That’s the only life.... Get drunk and raise cain
payday and see the extreme orient.”
“And die of the syph in a hospital at thirty....”
“What’s it matter?... Your body renews itself every seven years.”
The steps of their rooming house smelled of cabbage and stale
beer. They stumbled up yawning.
“Waiting’s a rotton tiring job.... Makes the soles of your feet
ache.... Look it’s going to be a fine day; I can see the sun on the
watertank opposite.”
Congo pulled off his shoes and socks and trousers and curled up
in bed like a cat.
“Those dirty shades let in all the light,” muttered Emile as he
stretched himself on the outer edge of the bed. He lay tossing
uneasily on the rumpled sheets. Congo’s breathing beside him was
low and regular. If I was only like that, thought Emile, never worrying
about a thing.... But it’s not that way you get along in the world. My
God it’s stupid.... Marco’s gaga the old fool.
And he lay on his back looking up at the rusty stains on the
ceiling, shuddering every time an elevated train shook the room.
Sacred name of God I must save up my money. When he turned
over the knob on the bedstead rattled and he remembered Marco’s
hissing husky voice: I never see the dawn that I dont say to myself
perhaps.

“If you’ll excuse me just a moment Mr. Olafson,” said the


houseagent. “While you and the madam are deciding about the
apartment...” They stood side by side in the empty room, looking out
the window at the slatecolored Hudson and the warships at anchor
and a schooner tacking upstream.
Suddenly she turned to him with glistening eyes; “O Billy, just
think of it.”
He took hold of her shoulders and drew her to him slowly. “You
can smell the sea, almost.”
“Just think Billy that we are going to live here, on Riverside Drive.
I’ll have to have a day at home ... Mrs. William C. Olafson, 218
Riverside Drive.... I wonder if it is all right to put the address on our
visiting cards.” She took his hand and led him through the empty
cleanswept rooms that no one had ever lived in. He was a big
shambling man with eyes of a washed out blue deepset in a white
infantile head.
“It’s a lot of money Bertha.”
“We can afford it now, of course we can. We must live up to our
income.... Your position demands it.... And think how happy we’ll be.”
The house agent came back down the hall rubbing his hands.
“Well, well, well ... Ah I see that we’ve come to a favorable
decision.... You are very wise too, not a finer location in the city of
New York and in a few months you wont be able to get anything out
this way for love or money....”
“Yes we’ll take it from the first of the month.”
“Very good.... You wont regret your decision, Mr. Olafson.”
“I’ll send you a check for the amount in the morning.”
“At your own convenience.... And what is your present address
please....” The houseagent took out a notebook and moistened a
stub of pencil with his tongue.
“You had better put Hotel Astor.” She stepped in front of her
husband.
“Our things are stored just at the moment.”
Mr. Olafson turned red.
“And ... er ... we’d like the names of two references please in the
city of New York.”
“I’m with Keating and Bradley, Sanitary Engineers, 43 Park
Avenue...”
“He’s just been made assistant general manager,” added Mrs.
Olafson.
When they got out on the Drive walking downtown against a
tussling wind she cried out: “Darling I’m so happy.... It’s really going
to be worth living now.”
“But why did you tell him we lived at the Astor?”
“I couldnt tell him we lived in the Bronx could I? He’d have thought
we were Jews and wouldnt have rented us the apartment.”
“But you know I dont like that sort of thing.”
“Well we’ll just move down to the Astor for the rest of the week, if
you’re feeling so truthful.... I’ve never in my life stopped in a big
downtown hotel.”
“Oh Bertha it’s the principle of the thing.... I don’t like you to be
like that.”
She turned and looked at him with twitching nostrils. “You’re so
nambypamby, Billy.... I wish to heavens I’d married a man for a
husband.”
He took her by the arm. “Let’s go up here,” he said gruffly with his
face turned away.
They walked up a cross street between buildinglots. At a corner
the rickety half of a weatherboarded farmhouse was still standing.
There was half a room with blueflowered paper eaten by brown
stains on the walls, a smoked fireplace, a shattered builtin cupboard,
and an iron bedstead bent double.

Plates slip endlessly through Bud’s greasy fingers. Smell of swill


and hot soapsuds. Twice round with the little mop, dip, rinse and pile
in the rack for the longnosed Jewish boy to wipe. Knees wet from
spillings, grease creeping up his forearms, elbows cramped.
“Hell this aint no job for a white man.”
“I dont care so long as I eat,” said the Jewish boy above the rattle
of dishes and the clatter and seething of the range where three
sweating cooks fried eggs and ham and hamburger steak and
browned potatoes and cornedbeef hash.
“Sure I et all right,” said Bud and ran his tongue round his teeth
dislodging a sliver of salt meat that he mashed against his palate
with his tongue. Twice round with the little mop, dip, rinse and pile in
the rack for the longnosed Jewish boy to wipe. There was a lull. The
Jewish boy handed Bud a cigarette. They stood leaning against the
sink.
“Aint no way to make money dishwashing.” The cigarette wabbled
on the Jewish boy’s heavy lip as he spoke.
“Aint no job for a white man nohow,” said Bud. “Waitin’s better,
they’s the tips.”
A man in a brown derby came in through the swinging door from
the lunchroom. He was a bigjawed man with pigeyes and a long
cigar sticking straight out of the middle of his mouth. Bud caught his
eye and felt the cold glint twisting his bowels.
“Whosat?” he whispered.
“Dunno.... Customer I guess.”
“Dont he look to you like one o them detectives?”
“How de hell should I know? I aint never been in jail.” The Jewish
boy turned red and stuck out his jaw.
The busboy set down a new pile of dirty dishes. Twice round with
the little mop, dip, rinse and pile in the rack. When the man in the
brown derby passed back through the kitchen, Bud kept his eyes on
his red greasy hands. What the hell even if he is a detective.... When
Bud had finished the batch, he strolled to the door wiping his hands,
took his coat and hat from the hook and slipped out the side door
past the garbage cans out into the street. Fool to jump two hours
pay. In an optician’s window the clock was at twentyfive past two. He
walked down Broadway, past Lincoln Square, across Columbus
Circle, further downtown towards the center of things where it’d be
more crowded.
She lay with her knees doubled up to her chin, the nightgown
pulled tight under her toes.
“Now straighten out and go to sleep dear.... Promise mother you’ll
go to sleep.”
“Wont daddy come and kiss me good night?”
“He will when he comes in; he’s gone back down to the office and
mother’s going to Mrs. Spingarn’s to play euchre.”
“When’ll daddy be home?”
“Ellie I said go to sleep.... I’ll leave the light.”
“Dont mummy, it makes shadows.... When’ll daddy be home?”
“When he gets good and ready.” She was turning down the
gaslight. Shadows out of the corners joined wings and rushed
together. “Good night Ellen.” The streak of light of the door narrowed
behind mummy, slowly narrowed to a thread up and along the top.
The knob clicked; the steps went away down the hall; the front door
slammed. A clock ticked somewhere in the silent room; outside the
apartment, outside the house, wheels and gallumping of hoofs,
trailing voices; the roar grew. It was black except for the two strings
of light that made an upside down L in the corner of the door.
Ellie wanted to stretch out her feet but she was afraid to. She
didnt dare take her eyes from the upside down L in the corner of the
door. If she closed her eyes the light would go out. Behind the bed,
out of the windowcurtains, out of the closet, from under the table
shadows nudged creakily towards her. She held on tight to her
ankles, pressed her chin in between her knees. The pillow bulged
with shadow, rummaging shadows were slipping into the bed. If she
closed her eyes the light would go out.
Black spiraling roar outside was melting through the walls making
the cuddled shadows throb. Her tongue clicked against her teeth like
the ticking of the clock. Her arms and legs were stiff; her neck was
stiff; she was going to yell. Yell above the roaring and the rattat
outside, yell to make daddy hear, daddy come home. She drew in
her breath and shrieked again. Make daddy come home. The roaring
shadows staggered and danced, the shadows lurched round and
round. Then she was crying, her eyes were full of safe warm tears,
they were running over her cheeks and into her ears. She turned
over and lay crying with her face in the pillow.

The gaslamps tremble a while down the purplecold streets and


then go out under the lurid dawn. Gus McNiel, the sleep still
gumming his eyes, walks beside his wagon swinging a wire basket
of milkbottles, stopping at doors, collecting the empties, climbing
chilly stairs, remembering grades A and B and pints of cream and
buttermilk, while the sky behind cornices, tanks, roofpeaks,
chimneys becomes rosy and yellow. Hoarfrost glistens on doorsteps
and curbs. The horse with dangling head lurches jerkily from door to
door. There begin to be dark footprints on the frosty pavement. A
heavy brewers’ dray rumbles down the street.
“Howdy Moike, a little chilled are ye?” shouts Gus McNiel at a cop
threshing his arms on the corner of Eighth Avenue.
“Howdy Gus. Cows still milkin’?”
It’s broad daylight when he finally slaps the reins down on the
gelding’s threadbare rump and starts back to the dairy, empties
bouncing and jiggling in the cart behind him. At Ninth Avenue a train
shoots overhead clattering downtown behind a little green engine
that emits blobs of smoke white and dense as cottonwool to melt in
the raw air between the stiff blackwindowed houses. The first rays of
the sun pick out the gilt lettering of DANIEL McGILLYCUDDY’S
WINES AND LIQUORS at the corner of Tenth Avenue. Gus McNiel’s
tongue is dry and the dawn has a salty taste in his mouth. A can o
beer’d be the makin of a guy a cold mornin like this. He takes a turn
with the reins round the whip and jumps over the wheel. His numb
feet sting when they hit the pavement. Stamping to get the blood
back into his toes he shoves through the swinging doors.
“Well I’ll be damned if it aint the milkman bringin us a pint o cream
for our coffee.” Gus spits into the newly polished cuspidor beside the
bar.
“Boy, I got a thoist on me....”
“Been drinkin too much milk again, Gus, I’ll warrant,” roars the
barkeep out of a square steak face.
The saloon smells of brasspolish and fresh sawdust. Through an
open window a streak of ruddy sunlight caresses the rump of a
naked lady who reclines calm as a hardboiled egg on a bed of
spinach in a giltframed picture behind the bar.
“Well Gus what’s yer pleasure a foine cold mornin loike this?”
“I guess beer’ll do, Mac.”
The foam rises in the glass, trembles up, slops over. The barkeep
cuts across the top with a wooden scoop, lets the foam settle a
second, then puts the glass under the faintly wheezing spigot again.
Gus is settling his heel comfortably against the brass rail.
“Well how’s the job?”
Gus gulps the glass of beer and makes a mark on his neck with
his flat hand before wiping his mouth with it. “Full up to the neck wid
it.... I tell yer what I’m goin to do, I’m goin to go out West, take up
free land in North Dakota or somewhere an raise wheat.... I’m pretty
handy round a farm.... This here livin in the city’s no good.”
“How’ll Nellie take that?”
“She wont cotton to it much at foist, loikes her comforts of home
an all that she’s been used to, but I think she’ll loike it foine onct
she’s out there an all. This aint no loife for her nor me neyther.”
“You’re right there. This town’s goin to hell.... Me and the misses’ll
sell out here some day soon I guess. If we could buy a noice genteel
restaurant uptown or a roadhouse, that’s what’d suit us. Got me eye
on a little property out Bronxville way, within easy drivin distance.”
He lifts a malletshaped fist meditatively to his chin. “I’m sick o
bouncin these goddam drunks every night. Whade hell did I get
outen the ring for xep to stop fightin? Jus last night two guys starts
asluggin an I has to mix it up with both of em to clear the place out....
I’m sick o fighten every drunk on Tenth Avenoo.... Have somethin on
the house?”
“Jez I’m afraid Nellie’ll smell it on me.”
“Oh, niver moind that. Nellie ought to be used to a bit o drinkin.
Her ole man loikes it well enough.”
“But honest Mac I aint been slopped once since me weddinday.”
“I dont blame ye. She’s a real sweet girl Nellie is. Those little
spitcurls o hers’d near drive a feller crazy.”
The second beer sends a foamy acrid flush to Gus’s fingertips.
Laughing he slaps his thigh.
“She’s a pippin, that’s what she is Gus, so ladylike an all.”
“Well I reckon I’ll be gettin back to her.”
“You lucky young divil to be goin home to bed wid your wife when
we’re all startin to go to work.”
Gus’s red face gets redder. His ears tingle. “Sometimes she’s
abed yet.... So long Mac.” He stamps out into the street again.
The morning has grown bleak. Leaden clouds have settled down
over the city. “Git up old skin an bones,” shouts Gus jerking at the
gelding’s head. Eleventh Avenue is full of icy dust, of grinding rattle
of wheels and scrape of hoofs on the cobblestones. Down the
railroad tracks comes the clang of a locomotive bell and the clatter of
shunting freightcars. Gus is in bed with his wife talking gently to her:
Look here Nellie, you wouldn’t moind movin West would yez? I’ve
filed application for free farmin land in the state o North Dakota,
black soil land where we can make a pile o money in wheat; some
fellers git rich in foive good crops.... Healthier for the kids anyway....
“Hello Moike!” There’s poor old Moike still on his beat. Cold work
bein a cop. Better be a wheatfarmer an have a big farmhouse an
barns an pigs an horses an cows an chickens.... Pretty curlyheaded
Nellie feedin the chickens at the kitchen door....
“Hay dere for crissake....” a man is yelling at Gus from the curb.
“Look out for de cars!”
A yelling mouth gaping under a visored cap, a green flag waving.
“Godamighty I’m on the tracks.” He yanks the horse’s head round. A
crash rips the wagon behind him. Cars, the gelding, a green flag, red
houses whirl and crumble into blackness.

You might also like