Full Applied Statistics in Business and Economics 4Th Edition Doane Test Bank Online PDF All Chapter

Applied Statistics in Business and
Economics 4th Edition Doane Test

Bank
Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/applied-statistics-in-business-and-economics-4th-edition-doane-test-bank/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Applied Statistics in Business and Economics 4th

Edition Doane Solutions Manual
https://testbankdeal.com/product/applied-statistics-in-business-
and-economics-4th-edition-doane-solutions-manual/

Edition Doane Test Bank
and-economics-5th-edition-doane-test-bank/
Applied Statistics in Business and Economics Canadian

1st Edition Doane Test Bank
and-economics-canadian-1st-edition-doane-test-bank/

and-economics-5th-edition-doane-solutions-manual/
Essential Statistics in Business and Economics 3rd
Edition Doane Test Bank
https://testbankdeal.com/product/essential-statistics-in-
business-and-economics-3rd-edition-doane-test-bank/
Essential Statistics in Business and Economics 3rd

https://testbankdeal.com/product/essential-statistics-in-
business-and-economics-3rd-edition-doane-solutions-manual/
Practice of Statistics for Business and Economics 4th

Edition Moore Test Bank
https://testbankdeal.com/product/practice-of-statistics-for-
business-and-economics-4th-edition-moore-test-bank/
Basic Statistics for Business and Economics Canadian

4th Edition Lind Test Bank
https://testbankdeal.com/product/basic-statistics-for-business-
and-economics-canadian-4th-edition-lind-test-bank/
Practice of Statistics for Business and Economics 4th

Edition Moore Solutions Manual
https://testbankdeal.com/product/practice-of-statistics-for-
business-and-economics-4th-edition-moore-solutions-manual/
Chapter 10
Two-Sample Hypothesis Tests
True / False Questions
1. A pooled proportion is calculated by giving each sample proportion an equal

weight.
True False
2. The difference between two sample proportions p1 - p2 may be assumed

normally distributed if each sample has at least 10 "successes" and 10
"failures."
True False
3. When testing the difference between two population proportions, it is

necessary to use the same size sample from each population.
True False
4. When using independent samples to test the difference between two
population means, a pooled variance is used if the population variances are
unknown and assumed equal.
True False
5. In comparing the means of two independent samples, if the test statistic

indicates a significant difference at α =.05, it will also be significant at α =
.10.
True False
6. The degrees of freedom for the t-test used to compare two population means
(independent samples) with unknown variances (assumed equal) will be n1 +
n2 - 2.
True False

population means, it is desirable but not necessary for the sample sizes to be
the same.
True False
8. The Welch-Satterthwaite test is more conservative than the pooled variance
test to compare two population means with unknown variances in
independent samples.
True False
9. When sample data occur in pairs, an advantage of choosing a paired t-test is

that it tends to increase the power of a test, as compared to treating each
sample independently.
True False
10. A paired t-test with two columns of 10 observations in each column would
use d.f. = 18.
True False
11. In conducting a paired t-test of the difference between two population

means, the usual null hypothesis is that the mean of the population of paired
differences is zero.
True False
12. The t-test for two samples of paired data will use n differences, making it a
one-sample t-test.
True False
13. The F test is used to test for the equality of two population variances.
True False
14. The F distribution is never negative and is always skewed right.
True False
15. In an F test for the ratio of two population variances, the degrees of freedom
in both the numerator and the denominator must be equal.
True False
16. The critical value in an F test for equal variances is the ratio of the sample
variances.
True False
17. The test statistic in an F test for equal variances is the ratio of the sample
variances.
True False
18. We could use the same data set for two independent samples (i.e., two
columns of data) either to compare the means (t-test) or to compare the
variances (F test).
True False
19. The Welch-Satterthwaite t-test for two means has the same degrees of
freedom as the pooled samples t-test for two means.
True False
20. When the variances are known, a test comparing two independent sample
means would use the normal distribution.
True False
21. When the variances are unknown, a test comparing two independent sample
means would use the Student's t distribution.
True False
22. If the sample proportions are p1 = 15/60 and p2 = 20/90, normality may be
assumed in a test comparing the two population proportions.
True False
True False
24. A paired t-test with two columns of 8 observations in each column would use
d.f. = 7.
True False
25. If the population variances are exactly equal, the sample F test statistic will
be zero.
True False
Multiple Choice Questions
26. In a right-tailed test comparing two proportions, the test statistic was zcalc =
+1.81. The p-value is:
A. .9649
B. .0351
C. .4649
D. Must know n to answer.

27. In a left-tailed test comparing two means with unknown variances assumed
to be equal, the test statistic was t = -1.81 with sample sizes of n1 = 8 and n2
= 12. The p-value would be:
A. between .025 and .05.
B. between .01 and .025.
C. between .05 and .10.
D. Must know α to answer.
28. In a left-ailed test comparing two means with variances unknown but
assumed to be equal, the sample sizes were n1 = 8 and n2 = 12. At α = .05,
the critical value would be:
A. -1.960
B. -2.101
C. -1.734
D. -1.645
29. In a right-tailed test comparing two means with known variances, the sample
sizes were n1 = 8 and n2 = 12. At α = .05, the critical value would be:
A. 1.960
B. 1.645
C. 1.734
D. 1.282
30. In a test for equality of two proportions, the sample proportions were p1 =
12/50 and p2 = 18/50. The test statistic is approximately:
A. -1.44.
B. -1.31.
C. -1.67.
12/50 and p2 = 18/50. The pooled proportion is:
A. .20
B. .24
C. .36
D. .30
32. If the sample proportions were p1 = 12/50 and p2 = 18/50, the 95 percent
confidence interval for the difference of the population proportions is
approximately:
A. [-.144, +.244].
B. [-.120, +.120].
C. [-.298, +.058].
D. [-.011, .214].
33. John wants to compare two means. His sample statistics
were and . Assuming equal

variances, the 95 percent confidence interval for the difference of the
population means is approximately:
A. [2.44, 6.19].
B. [1.17, 5.08].
C. [0.08, 4.32].
D. [-0.09, 3.19].

variances, the pooled variance is:
A. 4.5
B. 4.9
C. 5.1
D. 3.8

variances, the test statistic is:
A. 2.37
B. 2.20
C. 1.96

variances, the degrees of freedom for his test will be:
A. 16.
B. 18.
C. 9.
D. 8.
37. In a random sample of patient records in Cutter Memorial Hospital, six-month
postoperative exams were given in 90 out of 200 prostatectomy patients,
while in Paymor Hospital such exams were given in 110 out of 200 cases. In
comparing these two proportions, normality of the difference may be
assumed because:
A. the populations are large enough to be assumed normal.
B. the probability of success can reasonably be assumed constant.
C. the samples are random, so the proportions are unbiased estimates.
D. nπ ≥ 10 and n(1 - π) ≥ 10 for each sample taken separately.

while in Paymor Hospital such exams were given in 110 out of 200 cases. In a
left-tailed test for equality of proportions, the test statistic is:
A. -1.96
B. -2.00
C. -4.00
D. -3.48
while in Paymor Hospital such exams were given in 110 out of 200 cases. In a
left-tailed test for equality of proportions, the p-value is:
A. .9772
B. .0228
C. .4772
D. .0517

postoperative exams were given in 90 out of 200 cases, while in Paymor
Hospital such exams were given in 110 out of 200 cases. The pooled
proportion is:
A. .50
B. .40
C. .30
D. .20
41. Management of Melodic Kortholt Company compared absenteeism rates in
two plants on the third Monday in November. Of Plant A's 800 employees, 120
were absent. Of Plant B's 1200 employees, 144 were absent. To compare the
two proportions, the pooled proportion is:
A. .130
B. .140
C. .132
D. .135
were absent. Of Plant B's 1200 employees, 144 were absent. MegaStat's
results for a two-tailed test are shown below.
The test statistic (shown as z = x.xx) is approximately:
A. 2.022
B. 1.960
C. 1.942
D. 1.645
were absent. Of Plant B's 1200 employees, 144 were absent. MegaStat's
results for a two-tailed test are shown below.
At α = .05, the two-tailed test for a difference in proportions is:
A. just barely significant.
B. not quite significant.
C. not feasible due to nonnormality.

44. A new policy of "flex hours" is proposed. Random sampling showed that 28 of
50 female workers favored the change, while 22 of 50 male workers favored
the change. Management wonders if there is a difference between the two
groups. For a test comparing the two proportions, the assumption of normality
for the difference of proportions is:
A. clearly justified.
B. clearly unjustified.
C. a borderline call.
groups. What is the test statistic to test for a zero difference in the population
proportions?
A. 1.321
B. 1.287
C. 1.200
D. -1.255
groups. What is the p-value for a two-tailed test?
A. .3849
B. .1151
C. .2301
D. .3453
47. At Huge University, a sample of 200 business school seniors showed that 26
planned to pursue an MBA degree, compared with 120 of 800 arts and
sciences seniors. We want to know if the proportion is higher in the arts and
sciences group. The pooled proportion for this test is:
A. .130
B. .140
C. .145
D. .146
sciences group. For this test, the assumption of normality for the difference of
proportions is:
A. clearly unjustified.
B. clearly justified.
sciences group. What is the z test statistic?
A. -1.322
B. -1.122
C. -0.716
D. We must first know α.

sciences group. The p-value for a left-tailed test is approximately:
A. .38
B. .48
C. .24
D. .51
51. Two well-known aviation training schools are being compared using random
samples of their graduates. It is found that 70 of 140 graduates of Fly-More
Academy passed their FAA exams on the first try, compared with 104 of 260
graduates of Blue Yonder Institute. To compare the pass rates, the pooled
proportion would be:
A. .500
B. .435
C. .400
D. .345
graduates of Blue Yonder Institute. To compare the two proportions, the
assumption of normality of the test statistic is:
A. justified, but it is a borderline case.
C. clearly not justified.
graduates of Blue Yonder Institute. The test statistic to test the pass rates for
equality is:
A. 2.141
B. 1.298
C. 1.227
D. 1.924
graduates of Blue Yonder Institute. To compare the pass rates, find the
critical value for a right-tailed test at α = .05.
A. 1.960
B. 1.645
C. 2.326
graduates of Blue Yonder Institute. To compare the pass rates, the p-value
for a right-tailed test is approximately:
A. .054
B. .027
C. .155
D. .013
graduates of Blue Yonder Institute. In a right-tailed test, the p-value is .0275,
so at α = .025 we should:
A. reject the hypothesis of equal proportions.
B. not reject the hypothesis of equal proportions.
C. change the α to .05 to get a rejection.
57. Of 200 youthful gamers (under 18) who tried the new Z-Box-Plus game, 160
rated it "excellent," compared with only 144 of 200 adult gamers (18 or over).
The pooled proportion for a test to compare the two proportions would be:
A. .76
B. .72
C. .77

The test statistic to test the two proportions for equality would be:
A. 1.645
B. 1.960
C. 1.873
D. 1.448
The p-value for a right-tailed test to compare the two proportions would be
approximately:
A. .042
B. .031
C. .054
D. .095
The 95 percent confidence interval for the difference of proportions would be
approximately:
A. [+.013, +.263].
B. [-.014, +.188].
C. [-.003, +.163].
D. [+.057, +.261].
61. Carver Memorial Hospital's surgeons have a new procedure that they think
will decrease the time to perform an appendectomy. A sample of 8
appendectomies using the old method had a mean of 38 minutes with a
variance of 36 minutes, while a sample of 10 appendectomies using the
experimental method had a mean of 29 minutes with a variance of 16
minutes. For a right-tail test for equal means (assume equal variances), the
critical value at α = .10 is:
A. 1.746
B. 1.337
C. 2.120
D. 2.754
minutes. For a right-tail test of equal means (assume equal variances), the
pooled variance is:
A. 14.76
B. 26.00
C. 24.75
D. 27.54
minutes. For a right-tail test of means (assume equal variances), the test
statistic is:
A. 3.814
B. 2.365
C. 3.000
D. 1.895
64. Based on a random sample of 13 tire changes, the mean time to change a tire
on a Boeing 777 has a mean of 59.5 minutes with a standard deviation of 8.4
minutes. For 10 tire changes on a Boeing 787 the mean time was 64.3
minutes with a standard deviation of 12.4 minutes. To test for equal variances
in a two-tailed test at α = .10, the critical values are:
A. 3.73 and 0.228
B. 2.51 and 3.67
C. 3.07 and 0.398
D. 3.07 and 0.357

65. A certain psychological theory predicts that men want bigger families than
women. Kate asked each student in her psychology class how many children
he or she considered ideal for a married couple, and obtained the Excel
results shown below at α = .05.
What conclusion can you draw in a two-tailed test at α = .05?
A. Men want larger families on average than women.
B. Women want larger families on average than men.
C. We cannot reject the hypothesis of equal population means.
D. The decision depends on whether or not the variances are equal.

66. Nacirema Airlines is buying a fleet of new fuel-efficient planes. The HogJet
and the LitheJet both meet their price and performance needs, and both
planes meet EPA noise guidelines. However, the quieter plane is preferred.
Each plane is flown through a typical takeoff and landing sequence 10 times,
while remote sensors at ground level record the noise levels (in decibels). The
table below summarizes the sound level tests using Excel's default level of
significance (α = 0.05).
In a left-tailed test comparing the means at α = .05, we would:
A. not reject H0.
B. reject H0.
C. have insufficient information to make a decision.

After inspecting this table, we would most likely:
A. use the test assuming unequal variances.
B. use the test for equal variances.
C. perform another test to determine if the variances are equal before

proceeding.
D. realize that the decision isn't affected by our assumptions concerning the
variance.
If we switched from α = .05 to α = .005 in a two-tailed test of means, our

assumption about variances (assumed equal or assumed unequal) would:
A. affect the decision.
B. not affect the decision.
C. require a new analysis.

69. A psychology researcher has a theory that predicts women will tend to carry
more cash than men. A random sample of Ersatz University students revealed
that 16 females had a mean of $22.30 in their wallets with a standard
deviation of $3.20, while 16 males had a mean of $17.30 with a standard
deviation of $9.60. The researcher's hypothesis would lead us to perform a"
A. right-tailed test.
B. left-tailed test.
C. two-tailed test.
70. A psychology researcher has a theory that predicts women will tend to carry
more cash than men. A random sample of Ersatz University students revealed
that 16 females had a mean of $22.30 in their wallets with a standard
deviation of $3.20, while 16 males had a mean of $17.30 with a standard
deviation of $9.60. The test statistic for the researcher's hypothesis is:
A. impossible to determine without knowing α.
B. 1.250.
C. 1.504.
D. 1.976.
71. A random sample of Ersatz University students revealed that 16 females had
a mean of $22.30 in their wallets with a standard deviation of $3.20, while 16
males had a mean of $17.30 with a standard deviation of $9.60. In comparing
the population variances at α = .10 in a two-tailed test, we conclude that:
A. the variances are equal.
B. the variances are unequal.
C. the variances are incomparable (different sample sizes).

72. Randomly chosen MBA students were asked their opinions about the ideal
number of children for a married couple. The sample data were entered into
MegaStat, and the following results were produced.
To compare the means, would it be appropriate to use a test that assumes

equal variances?
A. Yes, because by pooling our variances we can simplify our calculations.
B. Yes, we should assume equal variances when comparing independent

samples.
C. No, because the sample statistics show that the variances are unequal at
α = .05.
D. No, because the variances will differ because the means differ
significantly.
73. Litter sizes (number of pups) for randomly chosen dogs from two breeds were
compared. The sample data were entered into Excel, and the following results
were produced.
What is the critical value for a left-tailed test comparing the means at α =
.05?
A. -1.645
B. -1.721
C. -1.703
D. -1.699
number of children for a married couple. The sample data were entered into
MegaStat, and the following results were produced.
What conclusion can you draw from this analysis at α = .05?
C. This is insufficient evidence to suggest a difference in means.
D. We could conclude that men want larger families if we used a two-tailed

test.
75. Litter sizes (number of pups) for randomly chosen dogs from two breeds were
compared. The sample data were entered into Excel, and the following results
were produced.
What is the p-value for a left-tailed test comparing the means at α = .05?
A. Less than .10
B. More than .10
C. Between .10 and .05
D. Between .05 and .01

76. During a test period, an experimental group of 10 vehicles using an 85
percent ethanol-gasoline mixture showed mean CO2 emissions of 667 pounds
per 1000 miles, with a standard deviation of 20 pounds. A control group of 14
vehicles using regular gasoline showed mean CO2 emissions of 679 pounds
per 1000 miles with a standard deviation of 15 pounds. At α = 0.05, in a left-
tailed test, the critical value to compare the means (assuming equal
variances) is:
A. -2.508
B. -2.074
C. -1.321
D. -1.717

per 1000 miles with a standard deviation of 15 pounds. Assuming equal
A. 296.59
B. 225.00
C. 400.00
D. 522.16
per 1000 miles with a standard deviation of 15 pounds. At α = 0.05, in a left-
tailed test (assuming equal variances) the test statistic is:
A. -1.310
B. -2.042
C. -1.645
D. -1.683

per 100 miles with a standard deviation of 15 pounds. A quick comparison of
the sample variances suggests that the population variances are:
A. probably equal.
B. probably unequal.
C. incomparable.
80. Mary did an analysis of acute care variances in samples of occupancy rates at
two community hospitals and obtained the following results (some
information is omitted).
Can Mary conclude that the variances are unequal at α = .05?
A. No, there isn't enough evidence to believe the variances are unequal.
B. Yes, the analysis shows that the variances are unequal.
C. You can't tell without knowing the sample sizes.

81. Mary analyzed occupancy rates at two community hospitals and obtained the
following Excel results.
Which conclusion is correct in a two-tailed test at α = .05?
A. There appears to be no difference in the mean occupancy rates.
B. HealthPro has a significantly higher mean occupancy rate.
C. There is a significant difference in the mean occupancy rates.

82. A medical researcher wondered if there is a significant difference between
the mean birth weight of boy and girl babies. Random samples of 5 babies'
weights (pounds) for each gender showed the following:
To test the researcher's hypothesis, we should use the:
A. paired (dependent) samples t-test.
B. independent samples t-test.
C. large-sample z-test.
D. t-test for correlation.

the mean birth weight of boy and girl babies. A random sample of babies'
weights (pounds) showed the following.
How many degrees of freedom would be used to test for a zero difference in
means?
A. 4
B. 8
C. 10
D. Must know α to say.

84. In a test of a new surgical procedure, the five most respected surgeons in
FlatBroke Township were invited to Carver Hospital. Each surgeon was
assigned two patients of the same age, gender, and overall health. One
patient was operated upon in the old way, and the other in the new way. Both
procedures are considered equally safe. The surgery times are shown below:
Which test should we use to test for zero difference in mean times?
A. Paired t-test
B. Independent samples t-test
C. Independent sampes z test
D. Cannot be sure without knowing α.

patient was operated upon in the old way, and the other in the new way. Both
procedures are considered equally safe. The surgery times are shown below:
The time (in minutes) to complete each procedure was carefully recorded. In
a right-tailed test for a difference of means, the test statistic is:
A. 3.162
B. 1.645
C. 1.860
D. 2.132
86. A corporate analyst is testing whether mean inventory turnover has increased.
Inventory turnover in six randomly chosen product distribution centers (PDCs)
is shown.
The degrees of freedom for the appropriate test would be:
A. 6.
B. 5.
C. 4.
D. 12.
is shown.
The right-tailed critical value at α = 0.005 is:
A. 1.645
B. 1.479
C. 4.032
D. 2.015
is shown.
The value of the test statistic is:
A. 3.798
B. 2.449
C. 1.225
D. 3.503
89. The table below shows the mean number of daily errors by seven air traffic
controller trainees during the first two weeks on the job. We want to perform
a paired t-test at α = .05 to see if the mean daily errors have decreased from
Week 1 to Week 2.
A. 1.895
B. 1.943
C. 2.447
D. 2.365
90. The table below shows the mean number of daily errors by air traffic
a paired t-test at α = .05 to see if the mean daily errors decreased
significantly.
The test statistic is:
A. 1.25
B. 1.75
C. 0.87
D. 0.79
a paired t-test at α = .05 to see if the mean daily errors decreased
significantly.
What would be the degrees of freedom for the appropriate test?
A. 14
B. 12
C. 7
D. 6
92. The F-test for equality of variances assumes:
A. normal populations.
B. equal means.
C. equal sample sizes.
D. equal means and sample sizes.

93. Which is not true of the two-tailed F-test for equality of variances?
A. It requires reversing the numerator and denominator d.f. to obtain the left-
tail critical value.
B. It can be avoided by "folding" the larger variance into the numerator and
adjusting α.
C. It is fairly robust to the presence of nonnormality in the populations being

sampled.
94. Which of the following is not a characteristic of the F distribution?
A. It is a continuous distribution.
B. It is always a positive number.
C. It is a family based on two sets of degrees of freedom.
D. It describes the ratio of two sample means.

will decrease the variance in the time it takes to perform an appendectomy. A
sample of 8 appendectomies using the old method had a variance of 36
minutes, while a sample of 10 appendectomies using the experimental
method had a variance of 16 minutes. At α = .10 in a two-tailed test for equal
variances, the critical values are:
A. 0.272 and 3.29
B. 0.299 and 3.07
C. 0.368 and 2.51
D. -1.645 and +1.645
96. The folded F-test for equality of variances:
A. is rarely used because of its complexity.
B. requires looking up the critical value for α/2.
C. puts the smaller variance in the numerator.
D. requires looking up two critical values of F instead of one.

97. An F-test for equality of variances gives a p-value of .003. At α = .05, what
conclusion can be made about the preferred test to compare the means for
the same sample?
A. We would prefer a pooled variance t-test for equality of means.
B. We would not wish to pool the variances in a t-test for equality of means.
C. We would prefer a paired t-test for equality of means.
D. The variances have nothing to do with the t-test for equality of means.
males had a mean of $17.30 with a standard deviation of $9.60. At α = .10, to
test for equal variances in a two-tailed test, the critical values are:
A. 0.441 and 3.24
B. 0.556 and 2.27
C. 0.345 and 4.62
D. 0.387 and 2.90

males had a mean of $17.30 with a standard deviation of $9.60. The value of
the test statistic for a folded F-test for equal variances is:
A. 0.333
B. 0.111
C. 9.00
D. 3.00
100.Assuming unequal variances in a t-test for a zero difference of two means,

we would:
A. sum the degrees of freedom for each sample.
B. use the larger degrees of freedom for simplicity.
C. use a complicated formula for the degrees of freedom.
D. use a z-test to be conservative in the calculation.
101.The z-test for zero difference in two means:
A. is generally the preferred test for means.
B. is rarely suitable for business data.
C. is the most powerful test for means.
D. is not available in Excel's Data Analysis.

102.A confidence interval for the difference of two population means:
A. must pool the sample variances.
B. may or may not pool the sample variances.
C. cannot be used to test for equal population means.
103.A medical researcher compared the variances in birth weights for five
randomly chosen babies of each gender, with the MegaStat results shown
below.
The population variances:
A. may be assumed equal at any customary α.
B. should be assumed unequal at any customary α.
C. are not relevant to this paired t-test.

104.A medical researcher wondered if there is a significant difference between
the mean birth weight of boy and girl babies. She weighed a random sample
of five babies of each gender. Their weights (pounds) are shown below,
along with some MegaStat results.
The population means:

105.The coach of an adult Master's Swim class selected eight swimmers within
each of the two age groups shown below. A 50-yard freestyle time is
recorded for each swimmer. The resulting times (seconds) are shown below.
Which statistical test would you choose to compare the two groups?
A. t-test for independent samples with known variances
B. t-test for independent samples with unknown variances
C. t-test for paired samples
D. z-test for two independent proportions

106.Jason wants to perform a two-tailed test for equality between two
independent sample proportions. Each sample has at least 10 "successes"
and 10 "failures." Jason's test statistic is -1.44. What is his p-value?
A. .1498
B. .0749
C. .9251
107.Does the Speedo Fastskin II Male Hi-Neck Bodyskin competition racing

swimsuit improve a swimmer's 200-yard individual medley performance
times? A test of 100 randomly chosen male varsity swimmers at several
different universities showed that 66 enjoyed improved times, compared with
only 54 of 100 female varsity swimmers. To test for equality in the
proportions of men versus women who experienced improvement, the test
statistic is approximately:
A. 1.73
B. 1.47
C. 2.31
D. Can't tell without knowing the tail of the test.

108.Does the Speedo Fastskin II Male Hi-Neck Bodyskin competition racing
different universities showed that 66 enjoyed improved times, compared with
only 54 of 100 female varsity swimmers. In comparing the proportions of
males versus females, is it safe to assume normality?
A. Yes, clearly.
B. Yes, but just barely.
C. No.
D. Can't tell without knowing α.
109.The table below shows two samples taken to compare the mean age of
individuals who purchased the iPhone 3G at two AT&T store locations.
What are the critical values for a two-tailed test for equal variances at α =
.05?
A. 0.275, 3.14
B. 0.244, 3.37
C. 0.210, 3.95
D. 0.181, 4.32
110.The table below shows two samples taken to compare the mean age of
At α = .05, can you conclude that the first sample has a larger variance than
the second sample?
A. Yes, clearly.
C. No.
111.Group 1 has a mean of 13.4 and group 2 has a mean of 15.2. Both
populations are known to have a variance of 9.0 and each sample consists of
18 items. What is the test statistic to test for equality of population means?
A. -1.755
B. -1.643
C. -1.800
D. -1.285
112.Which is not a type of comparison for which you would anticipate a two-
sample test?
A. Before versus After.
B. Old versus New.
C. Current versus Target.
D. Experimental versus Control.
Short Answer Questions

113.Random samples of students were compared to see whether or not there
was a difference in the proportion favoring the university's proposed switch
from MWF (three-day) classes to MW and TR (two-day) classes. The results
shown below are from MegaStat. Analyze these results thoroughly.
114.The table below compares two samples taken to compare the mean age of
individuals who purchased the iPhone 3G at two AT&T store locations. Very
briefly, what conclusions can you draw?
115.Random samples of Tuesday and Friday withdrawals from a college-campus
ATM were compared to see whether or not there was a difference in the
means. The results shown below are from MegaStat. Analyze these results
thoroughly.
Chapter 10 Two-Sample Hypothesis Tests Answer Key
True / False Questions
1. A pooled proportion is calculated by giving each sample proportion an

equal weight.
FALSE
In pc = (x1 + x2)/(n1 + n2) the sample successes are combined, as are the
sample sizes.
AACSB: Analytic
Blooms: Remember
Difficulty: 2 Medium
Learning Objective: 10-05 Perform a test to compare two proportions using z.
Topic: Comparing Two Proportions
2. The difference between two sample proportions p1 - p2 may be assumed

normally distributed if each sample has at least 10 "successes" and 10
"failures."
TRUE
This rule of thumb assures near-normality of the difference p1 - p2.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
Learning Objective: 10-06 Check whether normality may be assumed for two proportions.
3. When testing the difference between two population proportions, it is

necessary to use the same size sample from each population.
FALSE
The formula for the test statistic does not require equal sample sizes.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy

population means, a pooled variance is used if the population variances are
unknown and assumed equal.
TRUE
Sample variances are combined using weights equal to their respective

degrees of freedom.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
Learning Objective: 10-02 Recognize and perform a test for two means with unknown σ1 and σ2.
Topic: Comparing Two Means: Independent Samples
5. In comparing the means of two independent samples, if the test statistic
indicates a significant difference at α =.05, it will also be significant at α
= .10.
TRUE
Rejection in a two-tailed test implies rejection in a one-tailed test.
AACSB: Analytic
Blooms: Understand
Difficulty: 1 Easy
Learning Objective: 10-01 Recognize and perform a test for two means with known σ1 and σ2.
6. The degrees of freedom for the t-test used to compare two population
means (independent samples) with unknown variances (assumed equal)
will be n1 + n2 - 2.
TRUE
For the pooled variance case, we can add the degrees of freedom n1 - 1
and n2 - 1.
AACSB: Analytic
Blooms: Remember
population means, it is desirable but not necessary for the sample sizes to
be the same.
TRUE
Equal sample sizes may improve power but are not required.
AACSB: Analytic
Blooms: Remember
Learning Objective: 10-04 Explain the assumptions underlying the two-sample test of means.
8. The Welch-Satterthwaite test is more conservative than the pooled

variance test to compare two population means with unknown variances in
independent samples.
TRUE
When in doubt, assume unequal variances (but power may be lost if

variances were equal).
AACSB: Analytic
Blooms: Remember
9. When sample data occur in pairs, an advantage of choosing a paired t-test
is that it tends to increase the power of a test, as compared to treating
each sample independently.
TRUE
Assuming independent samples loses information, so power may be lost.
AACSB: Analytic
Blooms: Understand
Learning Objective: 10-03 Recognize paired data and be able to perform a paired t test.
Topic: Comparing Two Means: Paired Samples
use d.f. = 18.
FALSE
We have 9 differences so we would use d.f. = n - 1 = 10 - 1 = 9.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
11. In conducting a paired t-test of the difference between two population
means, the usual null hypothesis is that the mean of the population of
paired differences is zero.
TRUE
Testing for a zero difference is common.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
12. The t-test for two samples of paired data will use n differences, making it a
one-sample t-test.
TRUE
There are n data pairs so it is a one-sample test.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
13. The F test is used to test for the equality of two population variances.
TRUE
The ratio of variances should be near 1 if the population variances are the
same.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
Learning Objective: 10-08 Carry out a test of two variances using the F distribution.
Topic: Comparing Two Variances
14. The F distribution is never negative and is always skewed right.
TRUE
The ratio of variances has no upper limit but is bounded at zero on the low
end.
AACSB: Analytic
Blooms: Remember
15. In an F test for the ratio of two population variances, the degrees of
freedom in both the numerator and the denominator must be equal.
FALSE
Degrees of freedom are equal only if the sample sizes were the same.
AACSB: Analytic
Blooms: Understand
16. The critical value in an F test for equal variances is the ratio of the sample
variances.
FALSE
The test statistic is the ratio of variances, but the critical value is from a
table (or Excel).
AACSB: Analytic
Blooms: Remember
17. The test statistic in an F test for equal variances is the ratio of the sample
variances.
TRUE
The ratio of variances should be near 1 if the population variances are the
same.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
18. We could use the same data set for two independent samples (i.e., two
columns of data) either to compare the means (t-test) or to compare the
variances (F test).
TRUE
It depends on whether we are interested in means or variances, but both

tests use the same data.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
19. The Welch-Satterthwaite t-test for two means has the same degrees of
freedom as the pooled samples t-test for two means.
FALSE
The statement is true if the sample variances are identical, but it is not true
in general.
AACSB: Analytic
Blooms: Understand
20. When the variances are known, a test comparing two independent sample
means would use the normal distribution.
TRUE
In this (rare) case we would utilize the normal distribution.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
21. When the variances are unknown, a test comparing two independent
sample means would use the Student's t distribution.
TRUE
Student's t is needed when we don't know the population variances.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
TRUE
We have at least 10 successes (x1 = 15, x2 = 20) and 10 failures (n1 - x1 =

45, n2 - x2 = 70).
AACSB: Analytic
Blooms: Apply
FALSE
We do not have at least 10 successes (x1 = 6, x2 = 4) in each sample.
AACSB: Analytic
Blooms: Apply
use d.f. = 7.
TRUE
There are 8 differences so d.f. = n - 1 = 7.
AACSB: Analytic
Blooms: Understand
Difficulty: 1 Easy
25. If the population variances are exactly equal, the sample F test statistic will
be zero.
FALSE
The ratio of variances should be near 1 (not 0) if the population variances

are the same.
AACSB: Analytic
Blooms: Understand
Difficulty: 1 Easy
Multiple Choice Questions

26. In a right-tailed test comparing two proportions, the test statistic was zcalc =
+1.81. The p-value is:
A. .9649
B. .0351
C. .4649
D. Must know n to answer.
Appendix C gives P(Z ≥ 1.81) = .0351 or Excel =1-NORM.S.DIST(1.81,1).
AACSB: Analytic
Blooms: Apply
Learning Objective: 10-07 Use Excel to determine p-values for two-sample tests using z or t.
27. In a left-tailed test comparing two means with unknown variances assumed
to be equal, the test statistic was t = -1.81 with sample sizes of n1 = 8 and
n2 = 12. The p-value would be:
A. between .025 and .05.
B. between .01 and .025.
C. between .05 and .10.
For d.f. = 18, Appendix D gives t.05 = 1.734 and t.025 = 2.101, or for an exact
answer you can use the Excel function =T.DIST(-1.81,8+12-2,1) = .04351.
AACSB: Analytic
Blooms: Apply
28. In a left-ailed test comparing two means with variances unknown but
assumed to be equal, the sample sizes were n1 = 8 and n2 = 12. At α = .05,
the critical value would be:
A. -1.960
B. -2.101
C. -1.734
D. -1.645
For d.f. = 18, Appendix D gives t.05 = -1.734.
AACSB: Analytic
Blooms: Apply
29. In a right-tailed test comparing two means with known variances, the
sample sizes were n1 = 8 and n2 = 12. At α = .05, the critical value would
be:
A. 1.960
B. 1.645
C. 1.734
D. 1.282
For a right-tailed test with known variances, we would use z.05 = 1.645.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
12/50 and p2 = 18/50. The test statistic is approximately:
A. -1.44.
B. -1.31.
C. -1.67.
Use combined proportion pc = (x1 + x2)/(n1 + n2) = (12 + 18)/(50 + 50) =

.30 in zcalc.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
12/50 and p2 = 18/50. The pooled proportion is:
A. .20
B. .24
C. .36
D. .30
Use combined proportion pc = (x1 + x2)/(n1 + n2) = (12 + 18)/(50 + 50) =

.30 in the calculation.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
32. If the sample proportions were p1 = 12/50 and p2 = 18/50, the 95 percent
confidence interval for the difference of the population proportions is
approximately:
A. [-.144, +.244].
B. [-.120, +.120].
C. [-.298, +.058].
D. [-.011, .214].
Enter the proportions separately in the formula (we are not combining the
samples).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
Learning Objective: 10-09 Construct a confidence interval for μ1 - μ2 or π1 - π2.
Topic: Confidence Interval for the Difference of Two Proportions, π1 - π2

variances, the 95 percent confidence interval for the difference of the
population means is approximately:
A. [2.44, 6.19].
B. [1.17, 5.08].
C. [0.08, 4.32].
D. [-0.09, 3.19].
Pool the variances and add the degrees of freedom because equal
variances are assumed.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
Topic: Confidence Interval for the Difference of Two Means, μ1 - μ2

A. 4.5
B. 4.9
C. 5.1
D. 3.8
The pooled variance is [(n1 - 1)s12 + (n2 - 1)s22]/[(n1 - 1) + (n2 - 1)] = 4.5.
AACSB: Analytic
Blooms: Apply

variances, the test statistic is:
A. 2.37
B. 2.20
C. 1.96
The pooled variance is (n1 - 1)s12 + (n2 - 1)s22 = (9 - 1)(5.4) + (9 - 1)(3.6) =

4.5.
AACSB: Analytic
Blooms: Apply

variances, the degrees of freedom for his test will be:
A. 16.
B. 18.
C. 9.
D. 8.
With pooled samples, we add the d.f. = (n1 - 1) + (n2 - 1) = (9 - 1) + (9 - 1)

= 16.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
37. In a random sample of patient records in Cutter Memorial Hospital, six-
month postoperative exams were given in 90 out of 200 prostatectomy
patients, while in Paymor Hospital such exams were given in 110 out of 200
cases. In comparing these two proportions, normality of the difference may
be assumed because:
A. the populations are large enough to be assumed normal.
B. the probability of success can reasonably be assumed constant.
C. the samples are random, so the proportions are unbiased estimates.
D. nπ ≥ 10 and n(1 - π) ≥ 10 for each sample taken separately.

110, n2 - x2 = 90).
AACSB: Analytic
Blooms: Apply
cases. In a left-tailed test for equality of proportions, the test statistic is:
A. -1.96
B. -2.00
C. -4.00
D. -3.48
Combined pc = (90 + 110)/(200 + 200) = .50, so zcalc = (p1 - p2)/[pc(1 -

pc)/n1 + pc(1 - pc)/n2]1/2 = -2.000.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
cases. In a left-tailed test for equality of proportions, the p-value is:
A. .9772
B. .0228
C. .4772
D. .0517
Combined pc = (90 + 110)/(200 + 200) = .50, so zcalc = (p1 - p2)/[pc(1 -

pc)/n1 + pc(1 - pc)/n2]1/2 = -2.000 and using Appendix C we get P(Z ≤ -
2.000) = .0228.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
month postoperative exams were given in 90 out of 200 cases, while in
Paymor Hospital such exams were given in 110 out of 200 cases. The
pooled proportion is:
A. .50
B. .40
C. .30
D. .20
Use combined proportion pc = (x1 + x2)/(n1 + n2) = (90 + 110)/(200 + 200)

= .50 in the calculation.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
two plants on the third Monday in November. Of Plant A's 800 employees,
120 were absent. Of Plant B's 1200 employees, 144 were absent. To
compare the two proportions, the pooled proportion is:
A. .130
B. .140
C. .132
D. .135
Combined proportion is pc = (x1 + x2)/(n1 + n2) = (120 + 144)/(800 + 1200)

= .132.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
120 were absent. Of Plant B's 1200 employees, 144 were absent.
MegaStat's results for a two-tailed test are shown below.
The test statistic (shown as z = x.xx) is approximately:
A. 2.022
B. 1.960
C. 1.942
D. 1.645
Combined proportion is pc = (120 + 144)/(800 + 1200) = .132, so zcalc = (.15

- .12)/[.132(1 - .132)/800 + .132(1 - .132)/1200]1/2 = 1.942.
AACSB: Analytic
Blooms: Apply
120 were absent. Of Plant B's 1200 employees, 144 were absent.
MegaStat's results for a two-tailed test are shown below.
At α = .05, the two-tailed test for a difference in proportions is:
A. just barely significant.
B. not quite significant.
C. not feasible due to nonnormality.
Because the p-value is slightly greater than .05, we cannot reject H0.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
44. A new policy of "flex hours" is proposed. Random sampling showed that 28
of 50 female workers favored the change, while 22 of 50 male workers
favored the change. Management wonders if there is a difference between
the two groups. For a test comparing the two proportions, the assumption
of normality for the difference of proportions is:
A. clearly justified.
B. clearly unjustified.

22, n2 - x2 = 28).
AACSB: Analytic
Blooms: Apply
the two groups. What is the test statistic to test for a zero difference in the
population proportions?
A. 1.321
B. 1.287
C. 1.200
D. -1.255
Combined proportion is pc = (28 + 22)/(50 + 50) = .50, so zcalc = (.56 -

.44)/[.50(1 - .50)/50 + .50(1 - .50)/50]1/2 = -1.255.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
the two groups. What is the p-value for a two-tailed test?
A. .3849
B. .1151
C. .2301
D. .3453

.44)/[.50(1 - .50)/50 + 50(1 - .50)/50]1/2 = -1.20 and 2 × P(Z < -1.20) = 2
× .1151 = .2302 (or .2301 using Excel).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
47. At Huge University, a sample of 200 business school seniors showed that
26 planned to pursue an MBA degree, compared with 120 of 800 arts and
sciences seniors. We want to know if the proportion is higher in the arts
and sciences group. The pooled proportion for this test is:
A. .130
B. .140
C. .145
D. .146
Combined proportion is pc = (26 + 120)/(200 + 800) = .146.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
and sciences group. For this test, the assumption of normality for the
difference of proportions is:
A. clearly unjustified.

174, n2 - x2 = 680).
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
and sciences group. What is the z test statistic?
A. -1.322
B. -1.122
C. -0.716
D. We must first know α.

.15)/[.146(1 - .146)/200 + .145(1 - .146)/800]1/2 = -0.716.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
and sciences group. The p-value for a left-tailed test is approximately:
A. .38
B. .48
C. .24
D. .51

.15)/[.146(1 - .146)/200 + .145(1 - .146)/800]1/2 = -0.716, so from Appendix
C we get P(Z ≤ -0.72) = .2358.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
51. Two well-known aviation training schools are being compared using
random samples of their graduates. It is found that 70 of 140 graduates of
Fly-More Academy passed their FAA exams on the first try, compared with
104 of 260 graduates of Blue Yonder Institute. To compare the pass rates,
the pooled proportion would be:
A. .500
B. .435
C. .400
D. .345
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
104 of 260 graduates of Blue Yonder Institute. To compare the two
proportions, the assumption of normality of the test statistic is:
A. justified, but it is a borderline case.
C. clearly not justified.

70, n2 - x2 = 156).
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
104 of 260 graduates of Blue Yonder Institute. The test statistic to test the
pass rates for equality is:
A. 2.141
B. 1.298
C. 1.227
D. 1.924

.40)/[.435(1 - .435)/140 +.435(1 - .435)/260]1/2 = 1.924.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
find the critical value for a right-tailed test at α = .05.
A. 1.960
B. 1.645
C. 2.326
z.05 = 1.645.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
the p-value for a right-tailed test is approximately:
A. .054
B. .027
C. .155
D. .013

.40)/[.435(1 - .435)/140 +.435(1 - .435)/260]1/2 = 1.924, so from Appendix
C we get P(Z ≥ 1.92) = .0274.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
104 of 260 graduates of Blue Yonder Institute. In a right-tailed test, the p-
value is .0275, so at α = .025 we should:
A. reject the hypothesis of equal proportions.
B. not reject the hypothesis of equal proportions.
C. change the α to .05 to get a rejection.
The p-value is less than the chosen level of significance.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
57. Of 200 youthful gamers (under 18) who tried the new Z-Box-Plus game,
160 rated it "excellent," compared with only 144 of 200 adult gamers (18 or
over). The pooled proportion for a test to compare the two proportions
would be:
A. .76
B. .72
C. .77
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
over). The test statistic to test the two proportions for equality would be:
A. 1.645
B. 1.960
C. 1.873
D. 1.448

.72)/[.76(1 - .76)/200 + .76(1 - .76)/200]1/2 = 1.873.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
over). The p-value for a right-tailed test to compare the two proportions
would be approximately:
A. .042
B. .031
C. .054
D. .095

.72)/[.76(1 - .76)/200 +.76(1 - .76)/200]1/2 = 1.873, so from Appendix C we
get P(Z ≥ 1.87) = .0307.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
over). The 95 percent confidence interval for the difference of proportions
would be approximately:
A. [+.013, +.263].
B. [-.014, +.188].
C. [-.003, +.163].
D. [+.057, +.261].
Do not pool the proportions when you calculate the standard error of p1 -
p2.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
Topic: Confidence Interval for the Difference of Two Proportions, π1 - π2
minutes. For a right-tail test for equal means (assume equal variances), the
critical value at α = .10 is:
A. 1.746
B. 1.337
C. 2.120
D. 2.754
For d.f. = (n1 - 1) + (n2 - 1) = 7 + 9 = 16, we get t.10 = 1.337.
AACSB: Analytic
Blooms: Apply
minutes. For a right-tail test of equal means (assume equal variances), the
pooled variance is:
A. 14.76
B. 26.00
C. 24.75
D. 27.54
The pooled variance is [(8 - 1)36 + (10 - 1)16]/[(8 - 1) + (10 - 1)] = 24.75.
AACSB: Analytic
Blooms: Apply
minutes. For a right-tail test of means (assume equal variances), the test
statistic is:
A. 3.814
B. 2.365
C. 3.000
D. 1.895
The pooled variance is [(n1 - 1)s12 + (n2 - 1)s22]/[(n1 - 1) + (n2 - 1)] = [(8 -
1)36 + (10 - 1)16]/[(8 - 1) + (10 - 1)] = 24.75, so tcalc = (38 - 29)/(24.75/8
+ 24.75/10)1/2 = 3.814.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
64. Based on a random sample of 13 tire changes, the mean time to change a
tire on a Boeing 777 has a mean of 59.5 minutes with a standard deviation
of 8.4 minutes. For 10 tire changes on a Boeing 787 the mean time was
64.3 minutes with a standard deviation of 12.4 minutes. To test for equal
variances in a two-tailed test at α = .10, the critical values are:
A. 3.73 and 0.228
B. 2.51 and 3.67
C. 3.07 and 0.398
D. 3.07 and 0.357
Using α/2 in Appendix F we get F12,9 = 3.07 and 1/F9,12 = 1/(2.80) = 0.357
(or else use Excel's functions =F.INV.RT(.05,12,9) and =F.INV(.05,12,9).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
65. A certain psychological theory predicts that men want bigger families than
women. Kate asked each student in her psychology class how many
children he or she considered ideal for a married couple, and obtained the
Excel results shown below at α = .05.
What conclusion can you draw in a two-tailed test at α = .05?
C. We cannot reject the hypothesis of equal population means.
D. The decision depends on whether or not the variances are equal.
Neither assumption about variances leads to a p-value smaller than .05.
AACSB: Analytic
Blooms: Apply
Each plane is flown through a typical takeoff and landing sequence 10
times, while remote sensors at ground level record the noise levels (in
decibels). The table below summarizes the sound level tests using Excel's
default level of significance (α = 0.05).
In a left-tailed test comparing the means at α = .05, we would:
A. not reject H0.
B. reject H0.
C. have insufficient information to make a decision.
Either assumption about variances leads to a p-value smaller than .05.
AACSB: Analytic
Blooms: Apply
After inspecting this table, we would most likely:
A. use the test assuming unequal variances.
B. use the test for equal variances.
C. perform another test to determine if the variances are equal before

proceeding.
D. realize that the decision isn't affected by our assumptions concerning

the variance.
Fcalc = 4.7385/0.7178 = 6.60 > F.05 = 5.35 for d.f. = (9, 9), so we would want
to assume unequal variances. However, we notice that our decision is the
same under either assumption about variances (one-tailed p-values are
less than .05 in either test).
AACSB: Analytic
Blooms: Apply
If we switched from α = .05 to α = .005 in a two-tailed test of means, our

assumption about variances (assumed equal or assumed unequal) would:
A. affect the decision.
B. not affect the decision.
C. require a new analysis.
For equal variances the p-value is .0049 (reject equal means) while for
unequal variances the p-value = .0069 (don't reject equal means), so the
choice of tests does matter.
AACSB: Analytic
Blooms: Apply
69. A psychology researcher has a theory that predicts women will tend to
carry more cash than men. A random sample of Ersatz University students
revealed that 16 females had a mean of $22.30 in their wallets with a
standard deviation of $3.20, while 16 males had a mean of $17.30 with a
standard deviation of $9.60. The researcher's hypothesis would lead us to
perform a"
A. right-tailed test.
B. left-tailed test.
C. two-tailed test.
The a priori theory requires a right-tailed test.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
70. A psychology researcher has a theory that predicts women will tend to
carry more cash than men. A random sample of Ersatz University students
revealed that 16 females had a mean of $22.30 in their wallets with a
standard deviation of $3.20, while 16 males had a mean of $17.30 with a
standard deviation of $9.60. The test statistic for the researcher's
hypothesis is:
A. impossible to determine without knowing α.
B. 1.250.
C. 1.504.
D. 1.976.
Using a folded F test at α = .05 we get Fcalc = 9.602/3.202 = 9.00 > F.025 =
2.86 for d.f. = (15, 15), so we would want to assume unequal variances. For
unequal variances tcalc = (22.30 - 17.30)/[(3.2)2/16 + (9.6)2/16]1/2 =
(5)/(6.4)1/2 = 1.976.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
71. A random sample of Ersatz University students revealed that 16 females
had a mean of $22.30 in their wallets with a standard deviation of $3.20,
while 16 males had a mean of $17.30 with a standard deviation of $9.60. In
comparing the population variances at α = .10 in a two-tailed test, we
conclude that:
A. the variances are equal.
B. the variances are unequal.
C. the variances are incomparable (different sample sizes).
Folded Fcalc = (9.60)2/(3.20)2 = 9.00 > F.01 = 3.52 for d.f. = (15, 15), so
unequal variances.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
number of children for a married couple. The sample data were entered
into MegaStat, and the following results were produced.
To compare the means, would it be appropriate to use a test that assumes

equal variances?
A. Yes, because by pooling our variances we can simplify our calculations.
B. Yes, we should assume equal variances when comparing independent

samples.
C. No, because the sample statistics show that the variances are unequal
at α = .05.
D. No, because the variances will differ because the means differ
significantly.
Folded Fcalc = (1.2505)2/(0.5547)2 = 5.08 > F.025 = 3.37 for d.f. = (10, 12), so
unequal variances.
AACSB: Analytic
Blooms: Apply
73. Litter sizes (number of pups) for randomly chosen dogs from two breeds
were compared. The sample data were entered into Excel, and the
following results were produced.
What is the critical value for a left-tailed test comparing the means at α =
.05?
A. -1.645
B. -1.721
C. -1.703
D. -1.699
t.05 = - 1.721 for Welch's test with d.f. = 21.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
number of children for a married couple. The sample data were entered
into MegaStat, and the following results were produced.
What conclusion can you draw from this analysis at α = .05?
C. This is insufficient evidence to suggest a difference in means.
D. We could conclude that men want larger families if we used a two-tailed

test.
Cannot reject equal means because the p-value (.065) is not less than α
(.05).
AACSB: Analytic
Blooms: Apply
75. Litter sizes (number of pups) for randomly chosen dogs from two breeds
were compared. The sample data were entered into Excel, and the
following results were produced.
What is the p-value for a left-tailed test comparing the means at α = .05?
A. Less than .10
B. More than .10
C. Between .10 and .05
The test statistic (-1.261) is not less than t.10 (-1.372) for d.f. = 21. For a
more precise answer, we can use the Excel function =T.DIST(-
1.261183,21,1) = .11054.
AACSB: Analytic
Blooms: Apply
percent ethanol-gasoline mixture showed mean CO2 emissions of 667
pounds per 1000 miles, with a standard deviation of 20 pounds. A control
group of 14 vehicles using regular gasoline showed mean CO2 emissions of
679 pounds per 1000 miles with a standard deviation of 15 pounds. At α =
0.05, in a left-tailed test, the critical value to compare the means (assuming
equal variances) is:
A. -2.508
B. -2.074
C. -1.321
D. -1.717
Assuming equal variances, d.f. = (10 - 1) + (14 - 1) = 22, so t.05 = -1.717.
AACSB: Analytic
Blooms: Apply
679 pounds per 1000 miles with a standard deviation of 15 pounds.
Assuming equal variances, the pooled variance is:
A. 296.59
B. 225.00
C. 400.00
D. 522.16
The pooled variance is [(10 - 1)202 + (14 - 1)152]/[(10 - 1) + (14 - 1)] =

296.59.
AACSB: Analytic
Blooms: Apply
679 pounds per 1000 miles with a standard deviation of 15 pounds. At α =
0.05, in a left-tailed test (assuming equal variances) the test statistic is:
A. -1.310
B. -2.042
C. -1.645
D. -1.683
The pooled variance is [(10 - 1)202 + (14 - 1)152]/[(10 - 1) + (14 - 1)] =

296.591, so tcalc = (667 - 679)/(296.591/10 + 296.591/14)1/2 = -1.683.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
252 pounds per 100 miles with a standard deviation of 15 pounds. A quick
comparison of the sample variances suggests that the population variances
are:
A. probably equal.
B. probably unequal.
C. incomparable.
Folded Fcalc = (20)2/(15)2 = 1.78 < F.025 = 3.31 for d.f. = (9, 13), so equal
variances.
AACSB: Analytic
Blooms: Apply
80. Mary did an analysis of acute care variances in samples of occupancy rates
at two community hospitals and obtained the following results (some
information is omitted).
Can Mary conclude that the variances are unequal at α = .05?
A. No, there isn't enough evidence to believe the variances are unequal.
B. Yes, the analysis shows that the variances are unequal.
C. You can't tell without knowing the sample sizes.
The p-value (.905) is far greater than .05, so we cannot reject equal
variances.
AACSB: Analytic
Blooms: Apply
81. Mary analyzed occupancy rates at two community hospitals and obtained
the following Excel results.
Which conclusion is correct in a two-tailed test at α = .05?
A. There appears to be no difference in the mean occupancy rates.
B. HealthPro has a significantly higher mean occupancy rate.
C. There is a significant difference in the mean occupancy rates.
The p-value (.1791) is greater than .05, so we cannot reject equal means.
AACSB: Analytic
Blooms: Apply
the mean birth weight of boy and girl babies. Random samples of 5 babies'
weights (pounds) for each gender showed the following:
To test the researcher's hypothesis, we should use the:
A. paired (dependent) samples t-test.
B. independent samples t-test.
C. large-sample z-test.
D. t-test for correlation.
Although arranged side by side, these are unrelated data (independent

samples).
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
the mean birth weight of boy and girl babies. A random sample of babies'
weights (pounds) showed the following.
How many degrees of freedom would be used to test for a zero difference
in means?
A. 4
B. 8
C. 10
D. Must know α to say.
Although arranged side by side, these are unrelated data (independent

samples).
AACSB: Analytic
Blooms: Apply
patient was operated upon in the old way, and the other in the new way.
Both procedures are considered equally safe. The surgery times are shown
below:
Which test should we use to test for zero difference in mean times?
A. Paired t-test
B. Independent samples t-test
C. Independent sampes z test
D. Cannot be sure without knowing α.
There are five pairs of data, so we would calculate the differences.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
patient was operated upon in the old way, and the other in the new way.
Both procedures are considered equally safe. The surgery times are shown
below:
The time (in minutes) to complete each procedure was carefully recorded.
In a right-tailed test for a difference of means, the test statistic is:
A. 3.162
B. 1.645
C. 1.860
D. 2.132
The test statistic is tcalc = (5 - 0)/[(3.5355)/51/2] = 3.162.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
86. A corporate analyst is testing whether mean inventory turnover has
increased. Inventory turnover in six randomly chosen product distribution
centers (PDCs) is shown.
The degrees of freedom for the appropriate test would be:
A. 6.
B. 5.
C. 4.
D. 12.
These are paired samples, so d.f. = 6 - 1 = 5.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
A. 1.645
B. 1.479
C. 4.032
D. 2.015
Paired data with d.f. = 6 - 1 = 5 gives us t.005 = 4.032.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
The value of the test statistic is:
A. 3.798
B. 2.449
C. 1.225
D. 3.503
Paired data test statistic is tcalc = (1.5 - 0)/[(1.0488)/61/2] = 3.503.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
89. The table below shows the mean number of daily errors by seven air traffic
controller trainees during the first two weeks on the job. We want to
perform a paired t-test at α = .05 to see if the mean daily errors have
decreased from Week 1 to Week 2.
A. 1.895
B. 1.943
C. 2.447
D. 2.365
Paired data with d.f. = 7 - 1 = 6 gives us t.05 = 1.943.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
perform a paired t-test at α = .05 to see if the mean daily errors decreased
significantly.
The test statistic is:
A. 1.25
B. 1.75
C. 0.87
D. 0.79
Paired data test statistic is tcalc = (0.8286 - 0)/[(1.7547)/71/2] = 1.249.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
perform a paired t-test at α = .05 to see if the mean daily errors decreased
significantly.
What would be the degrees of freedom for the appropriate test?
A. 14
B. 12
C. 7
D. 6
Paired data with d.f. = 7 - 1 = 6.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
92. The F-test for equality of variances assumes:
A. normal populations.
B. equal means.
C. equal sample sizes.
D. equal means and sample sizes.
Severely skewed populations could pose a problem for the F-test.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
93. Which is not true of the two-tailed F-test for equality of variances?
A. It requires reversing the numerator and denominator d.f. to obtain the

left-tail critical value.
B. It can be avoided by "folding" the larger variance into the numerator and
adjusting α.
C. It is fairly robust to the presence of nonnormality in the populations

being sampled.
Simulation studies suggest that the F-test is robust to modest departures

from normality.
AACSB: Analytic
Blooms: Remember
94. Which of the following is not a characteristic of the F distribution?
A. It is a continuous distribution.
B. It is always a positive number.
C. It is a family based on two sets of degrees of freedom.
D. It describes the ratio of two sample means.
The F-test is a ratio of two variances.
AACSB: Analytic
Blooms: Remember
Difficulty: 1 Easy
will decrease the variance in the time it takes to perform an appendectomy.
A sample of 8 appendectomies using the old method had a variance of 36
minutes, while a sample of 10 appendectomies using the experimental
method had a variance of 16 minutes. At α = .10 in a two-tailed test for
equal variances, the critical values are:
A. 0.272 and 3.29
B. 0.299 and 3.07
C. 0.368 and 2.51
D. -1.645 and +1.645
Using α/2 = .05 in each tail, from Appendix F, F7,9 = 3.29 and 1/F9,7 =
1/(3.68) = 0.272, or else use Excel's functions =F.INV.RT(.05,7,9) and
=F.INV(.05,7,9).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
96. The folded F-test for equality of variances:
A. is rarely used because of its complexity.
B. requires looking up the critical value for α/2.
C. puts the smaller variance in the numerator.
D. requires looking up two critical values of F instead of one.
Putting the larger variance in the numerator makes it into a right-tailed test
using α/2.
AACSB: Analytic
Blooms: Remember
97. An F-test for equality of variances gives a p-value of .003. At α = .05, what
conclusion can be made about the preferred test to compare the means for
the same sample?
A. We would prefer a pooled variance t-test for equality of means.
B. We would not wish to pool the variances in a t-test for equality of

means.
C. We would prefer a paired t-test for equality of means.
D. The variances have nothing to do with the t-test for equality of means.
The results of the F-test suggest that we should assume unequal variances
in the t-test.
AACSB: Analytic
Blooms: Apply
while 6 males had a mean of $17.30 with a standard deviation of $9.60. At
α = .10, to test for equal variances in a two-tailed test, the critical values
are:
A. 0.441 and 3.24
B. 0.556 and 2.27
C. 0.345 and 4.62
D. 0.387 and 2.90
=F.INV(.05,15,5).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
while 6 males had a mean of $17.30 with a standard deviation of $9.60. The
value of the test statistic for a folded F-test for equal variances is:
A. 0.333
B. 0.111
C. 9.00
D. 3.00
Fcalc = (9.60)2/(3.20)2 = 9.00 because a folded F-test puts the larger

variance in the numerator.
AACSB: Analytic
Blooms: Apply
100. Assuming unequal variances in a t-test for a zero difference of two means,
we would:
A. sum the degrees of freedom for each sample.
B. use the larger degrees of freedom for simplicity.
C. use a complicated formula for the degrees of freedom.
D. use a z-test to be conservative in the calculation.
The formula for Welch's adjusted degrees of freedom is not easy without a
computer.
AACSB: Analytic
Blooms: Remember
101. The z-test for zero difference in two means:
A. is generally the preferred test for means.
B. is rarely suitable for business data.
C. is the most powerful test for means.
D. is not available in Excel's Data Analysis.
We rarely know the population variances, so we would generally use a t-

test.
AACSB: Analytic
Blooms: Remember
102. A confidence interval for the difference of two population means:
A. must pool the sample variances.
B. may or may not pool the sample variances.
C. cannot be used to test for equal population means.
Whether to pool or not is a choice the researcher can make.
AACSB: Analytic
Blooms: Apply
Topic: Confidence Interval for the Difference of Two Means, μ1 - μ2
103. A medical researcher compared the variances in birth weights for five
randomly chosen babies of each gender, with the MegaStat results shown
below.
The population variances:
The test statistic Fcalc = (3.537)2/(3.288)2 does not differ significantly from 1
at any typical α.
AACSB: Analytic
Blooms: Apply
the mean birth weight of boy and girl babies. She weighed a random
sample of five babies of each gender. Their weights (pounds) are shown
below, along with some MegaStat results.
The population means:
The p-value (.4407) does not lead to rejection of equal means at any
typical α.
AACSB: Analytic
Blooms: Apply
105. The coach of an adult Master's Swim class selected eight swimmers within
each of the two age groups shown below. A 50-yard freestyle time is
recorded for each swimmer. The resulting times (seconds) are shown
below. Which statistical test would you choose to compare the two groups?
A. t-test for independent samples with known variances
B. t-test for independent samples with unknown variances
C. t-test for paired samples
D. z-test for two independent proportions
Despite being arranged side-by-side, there is no link between the columns.

The similar standard deviations suggest that it would be reasonable to
"pool" the variances (pun intended) although this question was not posed.
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
106. Jason wants to perform a two-tailed test for equality between two
independent sample proportions. Each sample has at least 10 "successes"
and 10 "failures." Jason's test statistic is -1.44. What is his p-value?
A. .1498
B. .0749
C. .9251
From Appendix C we find P(Z ≤ -1.44) = .0749, which we double for a two-
tailed test.
AACSB: Analytic
Blooms: Apply
107. Does the Speedo Fastskin II Male Hi-Neck Bodyskin competition racing
different universities showed that 66 enjoyed improved times, compared
with only 54 of 100 female varsity swimmers. To test for equality in the
proportions of men versus women who experienced improvement, the test
statistic is approximately:
A. 1.73
B. 1.47
C. 2.31
D. Can't tell without knowing the tail of the test.

.54)/[.60(1 - .60)/100 + .60(1 - .60)/100]1/2 = 1.73.
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
108. Does the Speedo Fastskin II Male Hi-Neck Bodyskin competition racing
different universities showed that 66 enjoyed improved times, compared
with only 54 of 100 female varsity swimmers. In comparing the proportions
of males versus females, is it safe to assume normality?
A. Yes, clearly.
C. No.
D. Can't tell without knowing α.

34, n2 - x2 = 46).
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
109. The table below shows two samples taken to compare the mean age of
What are the critical values for a two-tailed test for equal variances at α =
.05?
A. 0.275, 3.14
B. 0.244, 3.37
C. 0.210, 3.95
D. 0.181, 4.32
=F.INV(.025,6,9).
AACSB: Analytic
Blooms: Apply
Difficulty: 3 Hard
110. The table below shows two samples taken to compare the mean age of
At α = .05, can you conclude that the first sample has a larger variance
than the second sample?
A. Yes, clearly.
C. No.
The test statistic Fcalc = (4.389)2/(1.874)2 = 5.59 exceeds F.05 = 3.37 with d.f.
= (6, 9).
AACSB: Analytic
Blooms: Apply
Difficulty: 1 Easy
111. Group 1 has a mean of 13.4 and group 2 has a mean of 15.2. Both
populations are known to have a variance of 9.0 and each sample consists
of 18 items. What is the test statistic to test for equality of population
means?
A. -1.755
B. -1.643
C. -1.800
D. -1.285
With known variances, zcalc = (13.4 - 15.2)/[9.0/18 + 9.0/18]1/2 = -1.800.
AACSB: Analytic
Blooms: Apply
112. Which is not a type of comparison for which you would anticipate a two-
sample test?
A. Before versus After.
B. Old versus New.
C. Current versus Target.
D. Experimental versus Control.
The point of comparison is between two samples, not a benchmark or

target.
AACSB: Analytic
Blooms: Understand
Topic: Two-Sample Tests
Short Answer Questions

113. Random samples of students were compared to see whether or not there
was a difference in the proportion favoring the university's proposed switch
from MWF (three-day) classes to MW and TR (two-day) classes. The
results shown below are from MegaStat. Analyze these results thoroughly.
The large test statistic (z = -4.000) and small p-value (.001) indicate a
clear difference in the proportions in a two-tailed test. Samples are large
enough to assume normality of the difference of proportions since nπ ≥ 10
and n(1 - π) ≥ 10 in each sample.
Feedback: The large test statistic (z = - 4.000) and small p-value (.001)
indicate a clear-cut difference in the proportions in a two-tailed test. The
samples are large enough to assume normality because nπ ≥ 10 and n(1 -
π) ≥ 10 in each sample.
AACSB: Reflective Thinking

Blooms: Evaluate
Difficulty: 1 Easy
114. The table below compares two samples taken to compare the mean age of
individuals who purchased the iPhone 3G at two AT&T store locations. Very
briefly, what conclusions can you draw?
In either test, the large test statistic and small p-value indicate a clear-cut
difference in the means at α = .01 in a one-tailed test. The conclusions
agree, so it doesn't matter whether or not equal variances are assumed.
While we suspect unequal variances (by casual inspection of the sample
standard deviations), the test statistic (F = 3.27) is not significant in a two-
tailed test for equal variances at α = .05.
Feedback: In either test, the large test statistic and small p-value indicate a
clear-cut difference in the means even at α = .01 in a one-tailed test.
Since the conclusions agree, it doesn't matter whether or not equal
variances are assumed. There is a suspicion of unequal variances, by
casual inspection of the sample standard deviations, but the test statistic
(F = 3.27) is not significant in a two-tailed test for equal variances at α =
.05.

Blooms: Apply
115. Random samples of Tuesday and Friday withdrawals from a college-
campus ATM were compared to see whether or not there was a difference
in the means. The results shown below are from MegaStat. Analyze these
results thoroughly.
The test statistic (t = 2.430) and p-value (.0204) indicate a significant

difference in means at α = .025, but not at α = .01. The assumption of
unequal variances appears justified, since the ratio of variances Fcalc =
(75.35/35.53)2 = 4.498 far exceeds the right-tail critical value of F.05 with
d.f. = (24, 19) (between 2.11 and 2.16).
Feedback: The test statistic (t = 2.430) and p-value (.0204) indicate a

significant difference in means at α = .025, but not at α = .01. The
assumption of unequal variances appears justified since the ratio of
variances Fcalc = (75.35/35.53)2 = 4.498 far exceeds the right-tail critical
value of F.05 with d.f. = (24, 19) (between 2.11 and 2.16).
Blooms: Evaluate
Another random document with
no related content on Scribd:
“You told me that you had been a mechanic in the works before
your rheumatism got bad. Have you worked at any of those
duplicators like what was packed in the crate?”
“I worked at all kinds of erecting works—duplicators an’ files an’
indexes an’ addressing machines an’ all the rest o’ them. I knows
them all.”
“Good! Now I want you to come round to the store and show me
the different parts of a duplicator.”
Gurney led the way from the boiler-house.
“Don’t switch on the light,” French directed. “I don’t want the
windows to show lit up. I have a torch.”
They passed through the packing-shed and into the completed-
machine store adjoining. Here French called a halt.
“Just let’s look at one of these duplicators again,” he said.
“Suppose you wanted to take one of them to pieces, let me see how
you would set about it. Should I be correct in saying that if five or six
of the larger pieces were got rid of, all the rest could be carried in a
handbag?”
“That’s right, sir.”
“Now show me the bins where these larger parts are stocked.”
They passed on to the part store and across it to a line of bins
labelled “Duplicators.” In the first bin were rows of leg castings.
French ran his eye along them.
“There must be fifty or sixty here,” he said, slowly. “Let’s see if
that is a good guess.”
On every bin was a stock card in a metal holder. French lifted
down that in question. It was divided into three sets of columns, one
set showing incomes, the second outgoes, and the third the existing
stock. The date of each transaction was given, and for each entry
the stock was adjusted.
“Not such a bad guess,” French remarked, slowly, as he
scrutinised the entries. “There are just fifty-four.”
The card was large and was nearly full. French noticed that it
went back for some weeks before the tragedy. He stood gazing at it
in the light of his torch while a feeling of bitter disappointment grew in
his mind. Then suddenly he thought he saw what he was looking for,
and whipping out a lens, he examined one of the entries more
closely. “Got it, by Jove! I’ve actually got it!” he thought, delightedly.
His luck had held.
One of the entries had been altered. A loop had been skilfully
added to a six to make it an eight. The card showed that two
castings had been taken out which either had never been taken out
at all or, more probably, which had been taken out and afterwards
replaced.
Convinced that he had solved the last of his four test problems,
French examined the cards of the other bins. In all of those referring
to large parts he noticed the same peculiarity; the entries had been
tampered with to show that one more duplicator had been sent out
than really was the case. The cards for the small parts were
unaltered and French could understand the reason. It was easier to
get rid of the parts themselves than to falsify their records. The fraud
was necessary only in the case of objects too big and heavy to carry
away.
French was highly pleased. His discovery was not only valuable
in itself, but he had reached it in the way which most appealed to his
vanity—from his own imagination. He had imagined that the fraud
might have been worked in this way. He had tested it and found that
it had been. Pure brains! Such things were soothing to his self-
respect.
He stood considering the matter. The evidence was valuable, but
it was far from permanent. A hint that suspicion was aroused, and it
would be gone. The criminal, if he were still about, would see to it
that innocuous copies of the cards were substituted for these
dangerous ones. French felt he dare not run such a risk. Nor could
he let Gurney suspect his discovery, lest unwittingly the old man
might put the criminal on his guard. He therefore went on:
“Now all I want is to make a sketch of each of these parts. The
duplicator which went out in the crate may have been taken to
pieces and I want to be able to recognise them if they’re found. I
suppose I could get a sheet or two of paper in the storeman’s desk?”
In one corner a small box with glass sides constituted an office
for the storeman. French led the way thither. The door was closed
but not locked. The desk, which he next tried, was fastened. But
above it in a rack he saw what he was looking for, a pile of blank bin-
cards. He turned back.
“It doesn’t matter about the paper, after all,” he explained. “I see
the desk is locked. I can make my sketches in my notebook, though
it’s not so convenient. But many a sketch I’ve made in it before.”
Chatting pleasantly, he returned to the bins and began slowly to
sketch the leg casting. He was purposely extremely slow and
detailed in the work, measuring every possible dimension and noting
it on his sketch. Gurney, as he had hoped, began to get fidgety.
French continued talking and sketching. Suddenly he looked up.
“By the way,” he said, as if a new idea had suddenly entered his
mind, “there is no earthly need for me to keep you here while I am
working. It will take me an hour or two to finish these sketches. If you
want to do your rounds and to get your supper, go ahead. I’ll find you
in the boiler-house when I have done.”
Gurney seemed relieved. He explained that it really was time to
make his rounds and that if French didn’t mind he would go and do
so. French reassured him heartily, and he slowly disappeared.
No sooner had his shuffling footsteps died away than French
became an extremely active man. Quickly slipping the four faked
cards from their metal holders, he carried them to the office. Then
taking four fresh cards from the rack, he began slowly and carefully
to copy the others. He was not a skilful forger, but at the end of half
an hour’s work he had produced four passable imitations. Two
minutes later he breathed more freely. The copies were in the
holders and the genuine cards in his pocket. Hurriedly he resumed
his sketching.
French’s work amounted to genius in the infinite pains he took
with detail. In twenty minutes his sketches were complete and he
effectually banished any suspicion which his actions might have
aroused in Gurney’s mind by showing them to him when he rejoined
him in the boiler-house. Like an artist he proceeded to establish the
deception.
“Copies of these sketches sent to the men who are searching for
the duplicator will help them to recognise parts of it if it has been
taken to pieces,” he explained. “You see the idea?”
Gurney appreciated the point, and French, after again warning
him to be circumspect, left the works.
The problem of what he should do next was solved for French by
the receipt of a letter by the early post. It was written on a half sheet
of cheap notepaper in an uneducated hand and read:
Ashburton.
12th October.
Dear Sir,
If you would come round some time that suits you I have
something I could tell you that would maybe interest you. It’s
better not wrote about.
Lizzie Johnston.
French had received too many communications of the kind to be

hopeful that this one would result in anything valuable. However, he
thought he ought to see the ex-parlourmaid and once again he made
his way to her cottage.
“It’s my Alf,” she explained. “Alf Beer, they call him. We’re being
married as soon as he gets another job.”
“He’s out of a job, then?”
“Yes, he was in the sales department in the works; a packer, he
was. He left there six months ago.”
“How was that?” French asked, sympathetically.
“He wasn’t well and he stayed home a few mornings, and Mr.
Berlyn had him up in his office and spoke to him something wicked.
Well, Alf wouldn’t take that, not from no man living, so he said what
relieved his feelings and Mr. Berlyn told him he could go.”
“And has he been doing nothing since?”
“Not steady, he hasn’t. Just jobbing, as you might say.”
“Hard lines, that is. You say he had something to tell me?”
The girl nodded. “That’s right,” was her original reply.
“What is it, do you know?”
“He wouldn’t say. I told him you was in asking questions and he
seemed sort of interested. ‘Wants to know about Berlyn and Pyke
and Mrs. Berlyn’s goings-on with Pyke, does he?’ he sez. ‘I thought
some one would be wanting to know about that before long. Well, I
can tell him something,’ he sez.”
“But he didn’t mention what it was?”
“No. I asked him and he sez ‘Value for cash,’ he sez. ‘He puts
down the beans and I cough up the stuff. That’s fair, ain’t it?’ he sez.
‘Don’t be a silly guff, Alf,’ I sez. ‘He’s police and if he asks you
questions, why, you don’t half have to answer them.’ ‘The devil I
have,’ he sez. ‘I ain’t done no crime and he hasn’t nothing on me.
You tell him,’ he sez, ‘tell him I know something that would be worth
a quid or two to him.’ And so I wrote you that note.”
“Tell me why you thought I was police,” French invited.
Miss Johnston laughed scornfully.
“Well, ain’t you?” she parried.
“That’s hardly an answer to my question.”
“Well, everybody knows what you’re after. They say you think
Pyke was murdered on the moor and that Berlyn murdered him.
Leastways, that’s what I’ve heard said.”
This was something more than a blow to French, and his self-
esteem reeled under it. For the nth time he marvelled at the amazing
knowledge of other people’s business to be found in country districts.
The small country town, he thought, was the absolute limit! There he
was, moving continually among the townspeople, none of whom
gave the least sign of interest in his calling, yet evidently they had
discussed him and his affairs to some purpose. The garrulous
landlady, Mrs. Billing, was no doubt responsible for the murder of
Pyke becoming known, but the belief that he, French, suspected
Berlyn of murdering him was really rather wonderful.
“It seems to me,” he said with a rather sickly smile, “that your
townspeople are better detectives than ever came out of Scotland
Yard. So your young man thinks I’m police and wants to turn an
honest penny, does he? Where am I to find him?”
“He’ll be at home. He’s living with his father at the head of East
Street—a single red house on the left-hand side just beyond the
town.”
In the leisurely, holiday-like way he had adopted, French crossed
the town and half an hour later had introduced himself to Mr. Alfred
Beer. Lizzie’s Alf was a stalwart young man with a heavy face and a
sullen, discontented expression. French, sizing him up rapidly,
decided that the suave method would scarcely meet the case.
“You are Alfred Beer, engaged to Lizzie Johnston, the former
servant at Mr. Berlyn’s?” he began.
“That’s right, mister.”
“I am a police officer investigating the deaths of Mr. Berlyn and
Mr. Pyke. You have some information for me?”
“I don’t altogether know that,” Beer answered, slowly. “Just wot
did you want to know?”
“What you have to tell me,” French said, sharply. “You told Miss
Johnston you had some information and I’ve come up to hear it.”
The man looked at him calculatingly.
“Wot do you think it might be worth to you?” he queried.
“Not a brass farthing. You should know that witnesses are not
paid for their evidence. Don’t you misunderstand the situation, Beer,
or you’ll find things mighty unpleasant. Come along now. Out with it.”
“How can I tell you if you won’t say wot you want?”
“I wouldn’t talk to you any more, Beer, only, I think you don’t
understand where you are,” French answered, quietly. “This is a
murder case. Mr. Pyke has been murdered. If you know anything
that might help the police to discover the murderer and you don’t tell
it, you become an accessory after the fact. Do you realise that you’d
get a good spell of years for that?”
Beer gave an uncouth shrug and turned back to his digging.
“I don’t know nothing about no murder,” he declared,
contemptuously. “I was just pulling Lizzie’s leg.”
“You’ve done it now,” French said, producing his card. “There’s
my authority as a police officer. You’ve wasted my time and kept me
back from my work. That’s obstruction and you’ll get six months for
it. Come along to the station. And unless you want a couple of years
you’ll come quietly.”
This was not what the man expected.
“Wot’s that?” he stammered. “You ain’t going to arrest me? I ain’t
done nothin’ against the law, I ain’t.”
“You’ll soon find out about that. Look sharp now. I can’t spend the
day here waiting for you.”
“Aw!” The man shifted nervously. “See, mister, I ain’t done no
harm, I ain’t. I don’t know nothing about no murder. I don’t, honest.”
“I don’t want to be hard on you,” French answered. “If you tell
your story without any more humbugging I’ll let the rest go. But I
warn you, you needn’t start inventing any yarn. What you say will be
gone into, and Heaven help you if it’s not true.”
“I’ll take my davy it’s true, mister, but it ain’t about no murder.”
“Well, get along sharp and let’s hear it.”
“It was one night about six months ago,” said Beer, now speaking
almost eagerly. “Me and Lizzie were walking out at that time. Well,
that night we’d fixed up for to go for a walk, and then at the last
minute she couldn’t get away. Mrs. Berlyn was goin’ out or
somethin’, and she couldn’t get off. We’d ’ad it fixed up that when
that ’appened Lizzie would come down to the shrubbery after the
rest ’ad gone to bed. Well, I wanted to see ’er that night for to fix up
some little business between ourselves, so I went up to the ’ouse
and gave the sign—three taps with a tree branch at ’er window. You
understand?”
French nodded.
“Well, I went back into the shrubbery for to wait for her. It was
dark, but a quiet night. An’ then I ’eard voices an’ steps comin’ along
the path. So I got behind a bush so as they’d not see me. There was
a man and a woman, an’ when they came close I knew them by their
voices. It was Pyke and Mrs. Berlyn. I stayed still an’ they passed me
close.”
“Go ahead. Did you hear what they said, or what are you getting
at?”
“I ’eard wot they said when they were passing. ‘I tell you ’e
knows,’ she said. ‘I’m frightened,’ she said. ‘You don’t know him. If ’e
once thinks you’ve played ’im false ’e’ll make a ’ell of a trouble.’ An’
then Pyke says: ‘Nonsense!’ ’e said. ‘ ’E’s not that sort. Besides,’ ’e
said, ‘ ’e don’t know anything. ’E knows we’re friends, but that’s all.’
‘No,’ she said, ‘I’m sure ’e knows or ’e guesses, anyway. We’ll ’ave to
separate,’ she said. ’E said they ’ad been careful enough, and then
they went past an’ I didn’t ’ear no more.”
“That all?”
“That’s all,” said Beer, disgustedly. “Ain’t it enough?”
“Nothing to boast about,” French replied, absently. He
remembered that the man had been dismissed by Berlyn and he
wondered if this statement was merely the result of spite. He
therefore questioned him closely. But he was unable to shake him
and he formed the opinion that the story was true.
If so, it certainly had a pretty direct bearing on the theory he was
trying to evolve, for there could be little doubt as to who “ ’e” was. As
he considered the matter he was surprised to find how complete that
theory was and how much of it had been definitely established.
There were gaps, of course, but there was no doubt as to its general
correctness.
As French now saw it, the affair stood as follows:
Stanley Pyke and Phyllis Berlyn, friends during childhood, find
that they love each other when they renew their acquaintance in later
years. But it is then too late for the course of true love to run smooth
and a clandestine attachment follows. Berlyn learns of this some four
months before the tragedy and as a result of his interference the two
decide to discontinue their meetings—in public, at all events. The
flirtation with Colonel Domlio is possibly deliberately undertaken by
Mrs. Berlyn to prove to her husband that her interest in Pyke is over.
But the two find that they cannot give each other up and the
intrigue is continued secretly. Berlyn, however, is not hoodwinked.
He sees his friend betraying him and he determines on vengeance.
His first move is to get an accomplice to assist in the details.
Here French admitted to himself that he was out of his depth. He
could not imagine who the accomplice was or why he should have
been required. But if Berlyn were guilty, the murder was clearly a
two-man job. Simultaneous activities in different places proved it.
The arrangements about the crate are next made. French was
aware that these had not yet been properly followed up; other
matters had been more urgent. But they represented a second string
to his bow which he would develop if necessary.
Then comes the night of the crime. While Berlyn and Pyke are at
Tavistock, the accomplice drugs the watchman’s food. He then waits
for the car. Pyke is sandbagged and his body carried into the works.
One of the men then unpacks the crate, and taking the duplicator to
pieces, returns the larger parts to stock. He has already doctored the
cards, as well, necessarily, as the corresponding books. He then
strips the recognisable clothes off the body, puts the latter in the
crate, smashes in the face, closes the crate, and leaves all as
before. Finally he escapes with Berlyn’s outer clothes and the
smaller parts of the duplicator. He has only to get rid of these and his
part in the ghastly business is complete.
In the meantime his confederate has driven the car out to a lonely
part of the moor, changed the magneto, and made the tracks leading
from the road.
The facts which pointed to Berlyn’s guilt were sixfold:
1. Berlyn in all probability was consumed by jealousy, one of the
strongest of human motives for crime.
2. Berlyn had an unparalleled opportunity for the deed, which
only he could have arranged.
3. It was not easy to see how anyone but Berlyn could have
handled the magneto affair.
4. Berlyn had the necessary position in the Veda Works to carry
out the watchman and stock-card episodes.
5. Berlyn answered the description of the man who had called for
the crate.
6. Berlyn had disappeared, an incomprehensible action if he were
innocent.
As French thought again over the accomplice, he recognised that
here was the snag in his theory. Motives of personal jealousy and
private wrong leave no room for an accomplice. Moreover, it was
incredible that a man who had shown such ingenuity could not have
devised a scheme to carry out the crime single-handed.
But though French recognised that there were points in the case
as yet unexplained, he saw that his own procedure was clear. He
must start the search for Berlyn and he must learn the identity of the
accomplice.
The first of these was easy. He had compiled a pretty accurate
description of the junior partner and Daw had got hold of his
photograph. A note in the Police Gazette would start every police
officer in the country on the search.
The second problem he found more difficult. Rack his brains as
he would, he could think of no one who might have helped Berlyn.
He thought his next plan would be an enquiry into the
whereabouts at ten o’clock, on the night of the crime, of everyone
whom it was possible to suspect. That, coupled with an investigation
as to who was in London when the various letters were posted,
should yield results.
The fact that a number of possible suspects had been at Mrs.
Berlyn’s party from eight to eleven on the fatal evening seemed to
rule them out. But French thought he should get some more definite
information on the point. Accordingly, he went up to the works and
asked for Mr. Fogden, one of those whom Lizzie Johnston had
mentioned as being present.
“I heard a peculiar story about Mrs. Berlyn,” he said a propos of
nothing special when they had talked for some time. “I was told she
had a premonition of Mr. Berlyn’s death and was miserable and
upset all that evening of the crime. A peculiar thing, if true, isn’t it?”
“Who told you that?” Mr. Fogden asked, sceptically.
“A chance remark in the bar of the Silver Tiger; I don’t know the
speaker’s name nor, of course, do I know if his story was true.”
“Well, you may take it from me that it wasn’t. I was at Mrs.
Berlyn’s that evening and there was nothing wrong with her that I
saw.”
This gave French his lead. When he left the office he had
obtained all the details of the party that he wanted. On the day
before the crime Mr. Fogden had had a telephone call from Mrs.
Berlyn saying that Berlyn was to be out on the following evening and
that she would be alone, and asking if he and one or two of the
others would come and keep her company. Eight people had turned
up, including himself, Cowls and Leacock from the works, a Dr.
Lancaster and his wife, and two Miss Pyms and a Miss Nesbitt from
the town. All these people were very intimate and the party was quite
informal. Some of them had played billiards, and the others bridge.
This information seemed to French to eliminate Fogden, Cowls,
and Leacock, as well, of course, as Mrs. Berlyn herself. He spent the
remainder of the day in racking his brains for other possible
accomplices and in thinking out ways to learn their movements on
the night in question.
Next morning he took up the matter of the whereabouts of all
suspects when the incriminating letters were posted in London.
Fortunately, the enquiry presented but little difficulty. A further
application to Mr. Fogden revealed the fact there was an attendance
book at the works which all the officers signed, from Mr. Fogden
himself down. This book showed that everyone concerned was in
Ashburton on the dates of posting. Even Stanley Pyke, who was
absent five days out of six on his rounds, had been there. Further,
Mr. Fogden’s diary showed that he had had interviews with Colonel
Domlio on the critical days. From Lizzie Johnston, French learned
that Mrs. Berlyn had also been at home during the period.
French was more puzzled than ever. It looked as if someone
must have been mixed up in the affair of whose existence he was
still in ignorance.
Just as he was about to step into bed that night an idea struck
him which gave him sharply to think. As he considered it he began to
wonder if his whole view of the crime were not mistaken. He
suddenly saw that the facts could bear a quite different interpretation
from that which he had placed upon them, an interpretation,
moreover, which would go far towards solving the problem of the
accomplice.
Once again he swung from depression to optimism as, chuckling
gently to himself, he decided that next morning he would embark on
a line of enquiry which up to the present he had been stupid enough
entirely to overlook.
Chapter Thirteen: The Accomplice?
French’s new idea had been subconsciously in his mind from the
very first, but probably owing to his theory of the guilt of one of the
two men supposed to be lost, he had never given it the consideration
he now saw that it deserved.
Suppose that on the night of the tragedy the lines of footprints
had not been faked. Suppose that after leaving the car the two men
had walked across the moor and reached Domlio’s. Suppose that
Domlio was the moving spirit in the affair and Berlyn merely the
accomplice.
This idea, French thought, would account not only for the facts
which his previous theory had covered, but also for nearly all of
those which the latter had failed to meet.
As before, the affair hinged on the fatal attractiveness of Phyllis
Berlyn, but in this case Domlio was the victim. Suppose Domlio had
fallen desperately in love with Phyllis and that she had encouraged
him. So far from this being unlikely, the facts bore it out. Different
witnesses had testified to the flirtation and Mrs. Berlyn herself had
not denied it.
Domlio then would see that there was a double barrier to the
realisation of his desires. There was of course Berlyn, but if Berlyn
were out of the way there was still Pyke. How far Mrs. Berlyn loved
Pyke, Domlio might not know, but their “affair” was common
knowledge and he would want to be on the safe side. If murder were
the way out in one case, why not in both? The risk was probably no
greater, and once both his rivals were out of the way, his own
happiness was secured.
His plan decided on, he would approach his friend Berlyn with
insidious suggestions as to the part Pyke was playing with his wife.
Gradually he would let it be known that he also had occasion to hate
Pyke—obviously for some quite different reason. He would feed the
other’s jealousy until at last Berlyn would be as ready for the crime
as he was himself. Then he would put forward his proposals.
Pyke was a cause of misery in both their lives; they would
combine to remove his evil influence.
Between them they would obtain and damage the spare
magneto, then arrange the visit to Tavistock and the ordering of the
crate and crane lorry. Berlyn would require Pyke to accompany him
to Tavistock. All would be done without raising suspicion.
On the fatal night Domlio would go to the works and drug
Gurney’s supper. Later on, during the run back from Tavistock,
Berlyn would stop the car and pretend to Pyke that it had broken
down. He would suggest looking up Domlio, who would certainly run
them into Ashburton in his own car. A light in the colonel’s study
would lead them direct to his French window and Domlio would
admit them without letting his servants know of their call.
Domlio would immediately get out his car and they would start for
the town. A sandbag would be in the car and on the way Pyke would
be done to death. The two men would then leave the car in some
deserted place, and carrying the body to the works, would pack it in
the crate. When the ghastly work was done they would return to the
car, taking with them Pyke’s suit and the small parts of the duplicator.
These they would get rid of later. Lastly they would change the
magneto on Berlyn’s car.
So far French was well pleased with his new theory, but he
realised that it contained a couple of nasty snags.
In the first place, it did not account satisfactorily for the
disappearance of Berlyn. Presumably Domlio had manœuvred his
colleague into such a position that he could give him away to the
police with safety to himself. Berlyn would therefore have to do the
other’s bidding, which would be to disappear and to get rid of the
crate. This was possible, but there was not a shred of proof that it
had happened.
Secondly, the theory did not explain how the letters were posted
in London. However, though French was not entirely satisfied, he
grew more and more convinced that he was on sure ground in
suspecting Domlio. At all events, his next job must be to test the
point.
First he decided to find out what Sergeant Daw could tell him
about the colonel and early next morning saw him at the police
station. The sergeant greeted him with a peculiar smile.
“I suppose, sir, you’ve heard the rumor that’s going round?” he
asked at once.
“What’s that, Sergeant?”
“They say you’ve found out that Mr. Berlyn murdered Mr. Pyke
out on the moor that night. Mrs. Billing, Pyke’s landlady, is supposed
to have recognised the underclothes.”
French smiled.
“Well it’s quite true,” he admitted. “I didn’t mean to keep it from
you, Sergeant, but I went off to London as soon as I discovered it. I
warned Mrs. Billing not to talk, but I hardly believed she could help
herself.”
The sergeant was evidently upset.
“I’m sorry about the whole thing, Mr. French. I should have
thought Mr. Berlyn was the last man who would do such a thing.”
“You may be right. Indeed, it’s a matter arising out of that very
point that I want to see you about. I have a notion there was a
second person in it—some one who might even have taken the lead.
Tell me”—French’s voice became very confidential—“what sort of a
man is Colonel Domlio?”
The sergeant looked shocked.
“Colonel Domlio?” he repeated. “Surely, sir, you don’t mean to
suggest that the colonel was mixed up in a murder?”
“You don’t think it likely?”
“I don’t, sir, and that’s a fact. The colonel’s a very quiet man and
peculiar in some ways, but he’s well respected in the district.”
“So was many a murderer.”
The sergeant was clearly sceptical, though anxious to be polite.
He said he was sure Mr. French would not speak without good
reason, but his own view was evident.
“Well, tell me all you know about him, anyway.”
Domlio, it appeared, was a man of about forty-five, short,
thickset, and dark. (Not the man who called for the crate, thought
French.) He was very well off, and since his wife had died some six
years earlier, had lived alone with his servants in his house on the
moor. He held sufficient Veda stock to give him a controlling interest
in the firm, acted as consulting engineer, and was usually referred to
as the senior partner. Entomology was his pet hobby and it was
believed that he was writing a book on the insect life of the moor.
He had four servants. Inside was John Burt, valet, butler, and
general factotum, and his wife, Sarah Burt, who combined the offices
of cook and general servant. Outside was an ex-service man named
Coombe, who acted as chauffeur and general handy man, and an
old gardener called Mee. Mee lived with his wife and daughter in the
gate lodge and Coombe boarded with them. All, so far as the
sergeant knew, were reliable people of good character.
“I’ll go out and see the colonel after lunch,” French announced.
“Could you lend me a push bicycle? I don’t want all my movements
reported on by the driver of a car.”
“I can borrow one for you, but it’ll not be much use on these hilly
roads.”
“It’ll do all I want.”
A couple of hours later French set out. When near Colonel
Domlio’s gate he hid the bicycle in the brushwood and approached
the house on foot. It was a smallish, creeper-covered building, L-
shaped, with thick walls and heavy overhanging eaves. At least a
hundred years old, French thought. It stood some two hundred yards
back from the road and was approached by a drive which wound
between clumps of stunted trees and shrubs. In front was a small
lawn of mown grass, while between the trees to the right French
glimpsed the roofs of outbuildings. The place had a cared-for
appearance. The woodwork of the house had been freshly painted,
the flower beds were tidy, and the grass edges had recently been
cut.
The door was opened by an elderly man in butler’s dress, honest
and kindly-looking, but rather stupid. John Burt, evidently. He asked
French to step inside while he took his card to his master.
The hall was of fair size, with a large, old-fashioned fireplace and
lead-lighted windows. French had not much time to observe it, for
Burt called him almost immediately into a room on the left of the hall
door.
It was long, low, and delightfully furnished as a study. Bookcases
lined the walls and a couple of deep saddle-bag armchairs stood on
the soft Chinese carpet in front of the fireplace. A collector’s
entomological cabinet was in one corner, with close by a table
bearing books and a fine microscope. The room was evidently in the
corner of the house, for there were French windows in adjacent
walls. In one of these was a leather-topped desk and at the desk
was seated a shortish man with a strong, clean-shaven face, iron
grey hair, and a not too amiable expression. He rose as French
entered.
“Inspector French of Scotland Yard, is it not? I have heard that
you were in the town.”
“That’s correct, sir,” French answered, taking the chair to which
the other pointed. “You’ve probably heard enough, then, to guess my
business?”
Colonel Domlio squared his shoulders.
“I heard you were investigating the deaths of Mr. Berlyn and Mr.
Pyke. I don’t know the object of this call.”
“I’ve come, Colonel Domlio, in connection with my investigation. I
want to ask for your help in it.”
“What do you wish me to do?”
“Two things, sir. In the first place, I want any information you can
give me about either of the two gentlemen you mentioned or
anything which might throw light on the tragedy. Secondly, I would be
obliged if you would answer the purely formal question that we
inspectors have to ask all who were in any way connected with the
victim of such a tragedy. Where were you yourself at the time of the
occurrence?”
The colonel raised his eyebrows.
“Do you suspect me of murdering Mr. Pyke?” he asked, drily.
“I think, sir, you needn’t take up that line.” French’s tone was also
a trifle dry. “I have explained that my question is a formal one,
invariably put. You are not bound to answer it unless you wish.”
“If I don’t you will suspect me in reality, so I don’t see that I have
much option. I was here, in this room.”
“Between what hours?”
“During the whole evening. I finished dinner about eight or a
quarter past. Then I came in here and stayed here until I went to bed
between one and two.”
“And no one came in during that time?”
“No one came in. I take nothing after dinner except a little whisky
going to bed, and I have everything I want in the cupboard there. I’m
writing a book at present and I don’t like to be disturbed in the
evenings.”
“Then in the face of what you’ve said I presume I needn’t ask you
if you heard any sound at the door or windows?”
“You need not.”
“And—excuse this question—you didn’t sleep at all during the
evening? No one could have knocked while you were asleep?”
The Colonel smiled slightly.
“Ingenious,” he remarked, “but unproductive. No, I didn’t sleep at
all.”
“Thank you,” said French, “that disposes of one question. Now
the other. Can you tell me anything likely to be helpful to me about
either of the two gentlemen?”
The colonel regretted that in this case also he could do nothing to
oblige. He would answer Mr. French’s questions so far as he could,
but he had nothing to volunteer. And French found that after half an
hour’s interrogation he had learnt just nothing whatever.
“There is one other matter to which I must refer,” he said. “I regret
the necessity, as it’s somewhat delicate. Common report says that
Mrs. Berlyn was on very intimate terms first with Mr. Pyke and then
with yourself. Would you tell me how far that is true?”
The colonel squared his shoulders again and French presently
saw that it was an unconscious nervous trick.
“Is it really necessary that Mrs. Berlyn’s name should be dragged
in?” he asked, stiffly.
“I’m afraid so. You will recognise that I am trying to find motives.”
“I don’t think you will find one there.”
“On the contrary, Colonel Domlio, I have evidence that Mr. Berlyn
was acutely jealous.”
But the colonel was not to be drawn.
“That is news to me,” he declared.
“Well,” said French, doggedly, “I should like to have your definite
statement as to whether such jealousy would or would not have
been justified, in so far, at all events, as you yourself were
concerned.”
The colonel smiled sardonically.
“I should say that it would not have been justified.”
“Very good, Colonel. I have now only one other request to make.
I should like to interrogate your servants. Some of them may have
seen or heard something which might be useful to me. Would you
oblige me by calling them in and instructing them to reply to me?”
For the first time an uneasy look appeared in the colonel’s eyes.
“Surely that is unnecessary?” he demurred. “What could they
possibly tell you?”
“Nothing, I very greatly fear,” French admitted. “But it is a routine
enquiry and as such I dare not omit it.”
With an evident ill grace Colonel Domlio rang the bell. French,
sensing his opposition, had become keenly alert. It seemed to him
that he might be on the brink of learning something important. But
instantly he decided that he would postpone serious examination of
the staff until he had them to himself.
The butler, Burt, answered the bell.
“This gentleman is Mr. French, Burt,” said the colonel. “He wants
to ask you some questions. You might answer him so far as you
can.”
“It was only to know whether you heard or saw anything unusual
on the night of the deaths of Mr. Berlyn and Mr. Pyke,” French
explained.
The man denied with what French thought was over-earnestness.
Moreover, he looked acutely uneasy, even scared. French felt a
sudden thrill, but he merely nodded and said:
“You didn’t see any traces on the moor the next day?”
“Nothing whatever, sir,” said the man with evident relief.
“Thank you. That’s all I want. Now, Colonel, if I could see the
others to put the same questions I should be finished.”
Mrs. Burt and the two outside men were produced in turn, and
each denied having heard or seen anything unusual. Coombe and
Mee, the chauffeur and gardener, were interested, but evidently
nothing more. But Mrs. Burt reproduced all the signs of uneasiness
which her husband had exhibited, only in an intensified degree. She
was obviously terrified when French questioned her, and her relief
when her ordeal was over was unmistakable.
But French apparently saw nothing amiss and when the quartet
had gone he thanked Colonel Domlio for his assistance and
apologised for the trouble he had given. And in the colonel’s manner
he noticed the same repressed evidences of relief. That something
had taken place that night of which the master of the house and the
two domestics were aware, French was positive.
He left the house and regained the clump of brushwood in which
he had hidden the bicycle. But he did not withdraw the machine.
Instead, after a quick glance round he crept in beside it, pulling the
bushes over him to make sure that he was invisible from the road.
From his hiding-place he could see the entrance to “Torview,” as the
colonel had named his house.
He was waiting on a pure chance, but after an hour he found that
his luck was in. He heard the sounds of an engine being started up,
and presently saw a small green car turn out of the drive and
disappear in the direction of Ashburton. In the car was Colonel
Domlio.
French allowed another twenty minutes to pass, then crawling out
of the brushwood, he returned to the house. Burt again opened the
door.
“I’m sorry to trouble you again, Mr. Burt,” he apologised, with his
pleasant smile, “but I forgot to ask Colonel Domlio a question. Could
I see him again just for a moment?”
“Colonel Domlio went out about half an hour ago, sir.”
“Ah, that’s very unfortunate.” French paused and looked
disappointed, then brightened up. “Perhaps you could give me the
information, if you would be so kind? I don’t want to have to come
back another day.”
Burt was obviously disconcerted. But he tried to hide his feelings
and reluctantly invited the caller into the study.
“Yes, sir?” he said.
French instantly became official and very stern. He swung round,
frowning at the other and staring him full in the face. Then he said,
harshly: “It is you I want to see, Burt. You lied to me this afternoon. I
have come back to hear the truth.”
The man started and fell back a pace, while dismay and
something like terror showed on his features.
“I don’t understand,” he stammered. “What do you mean?”
“It’s no use, Burt. You’ve given yourself away. You saw or heard
something that night. What was it?”
“You’re mistaken, sir,” he declared, with a look of relief. “I neither
saw nor heard anything. I swear it.” And then gaining confidence: “I
don’t know what right you have to come here and tell me I was lying.
I’m sure——”
“Cut it out,” French said, sharply. “Look here, Burt, do you want to
be arrested on a charge of conspiracy to murder?”
Burt’s jaw dropped, but French did not give him time to reply.
“Because if you don’t you’ll tell what you know. Mr. Pyke was
murdered that night, and perhaps Mr. Berlyn as well. They were not
lost on the moor and it is believed they came here. If you keep back
any information that might lead to the arrest of the murderer, it’s
conspiracy—accessory after the fact. Ten years penal for that, Burt!
Come along, now. Make up your mind whether you’re going to tell or
face the judge.”
Burt’s face had grown pale, but he stuck to it that he had neither
seen nor heard anything. French cut his protestations short.
“Fetch your wife,” he ordered.
The man’s manner as he heard these words, coupled with Mrs.
Burt’s evident fear when originally questioned, assured French that
this time he was on the right track. With evident unwillingness the
woman appeared.
“Now, Mrs. Burt, I want to know what you heard or saw on the
night of the tragedy. There is no use in telling me there was nothing.
Now out with it!” And in terse language he explained what accessory
after the fact meant, and its penalty.
Mrs. Burt was of less stern stuff than her husband. Under
French’s examination she was soon in tears and presently, disjointed
and in fragments, her story came out.
It appeared that on the night of the tragedy she slept badly, owing
to some small indisposition. Shortly after one she woke in

Full Applied Statistics in Business and Economics 4Th Edition Doane Test Bank Online PDF All Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Full Applied Statistics in Business and Economics 4Th Edition Doane Test Bank Online PDF All Chapter

Uploaded by

Copyright:

Available Formats

Applied Statistics in Business and

Economics 4th Edition Doane Test

Applied Statistics in Business and Economics 4th

Applied Statistics in Business and Economics 5th

Applied Statistics in Business and Economics Canadian

Applied Statistics in Business and Economics 5th

Essential Statistics in Business and Economics 3rd

Practice of Statistics for Business and Economics 4th

Basic Statistics for Business and Economics Canadian

Practice of Statistics for Business and Economics 4th

True / False Questions

1. A pooled proportion is calculated by giving each sample proportion an equal

2. The difference between two sample proportions p1 - p2 may be assumed

3. When testing the difference between two population proportions, it is

5. In comparing the means of two independent samples, if the test statistic

7. When using independent samples to test the difference between two

9. When sample data occur in pairs, an advantage of choosing a paired t-test is

11. In conducting a paired t-test of the difference between two population

14. The F distribution is never negative and is always skewed right.

Multiple Choice Questions

D. Must know n to answer.

A. between .025 and .05.

B. between .01 and .025.

C. between .05 and .10.

D. Must know α to answer.

D. Must know α to answer.

were and . Assuming equal

34. John wants to compare two means. His sample statistics

were and . Assuming equal

were and . Assuming equal

D. Must know α to answer.

36. John wants to compare two means. His sample statistics

were and . Assuming equal

A. the populations are large enough to be assumed normal.

B. the probability of success can reasonably be assumed constant.

C. the samples are random, so the proportions are unbiased estimates.

D. nπ ≥ 10 and n(1 - π) ≥ 10 for each sample taken separately.

38. In a random sample of patient records in Cutter Memorial Hospital, six-month

40. In a random sample of patient records in Cutter Memorial Hospital, six-month

The test statistic (shown as z = x.xx) is approximately:

At α = .05, the two-tailed test for a difference in proportions is:

A. just barely significant.

B. not quite significant.

C. not feasible due to nonnormality.

D. We must first know α.

A. justified, but it is a borderline case.

C. clearly not justified.

A. reject the hypothesis of equal proportions.

B. not reject the hypothesis of equal proportions.

C. change the α to .05 to get a rejection.

D. Must know α to answer.

A. 3.73 and 0.228

B. 2.51 and 3.67

C. 3.07 and 0.398

D. 3.07 and 0.357

What conclusion can you draw in a two-tailed test at α = .05?

A. Men want larger families on average than women.

B. Women want larger families on average than men.

C. We cannot reject the hypothesis of equal population means.

D. The decision depends on whether or not the variances are equal.

In a left-tailed test comparing the means at α = .05, we would:

A. not reject H0.

C. have insufficient information to make a decision.