Question 1
Question
We should determine our objectives
of a study before we collect the data.
Question 2
Question
The numbers on a basketball jersey
are an example of qualitative data.
Question 3
Question
A cumulative relative frequency
totals to
Question 4
Question
A data set that is skewed left
(negative) has most of the data
values on the left with a few data
values trailing off to the right.
Question 5
Question
A descriptive measure of a sample is
a parameter.
Question 6
Question
A stem and leaf plot is useful
because it
Answer

shows the distribution of the data

contains all the original data values

shows the distribution of the data AND contains all the original data values

it is a cumulative distribution
Question 7
Question
The first quartile of a distribution can
never be less than zero.
Question 8
Question
A boxplot is a good way to show the
mean of a data set.
Question 9
Question
Generally speaking, a stem and leaf
plot cannot be constructed from a
boxplot, but a boxplot can be
constructed from a stem and leaf
plot.
Question 10
Question
A population consists of the numbers 3,
8, 4. The mean and median, respectively,
are:
Question 11
Question
The median and mode are not
affected by outlier values.
Question 12
Question
Every data set has a mode.
Question 13
Question
One section of SCM 200 has 20 students,
another has 25 students, and a third
section has 55 students. A weighted
average of the number of students in
these sections is:
Question 14
Question
The following measures of variation
measure distance from the mean:
Question 15
Question
Median is a measure of variability of
a data set.
Question 16
Question
Range is sensitive to all data values.
Question 17
Question
If we want the average of all the
deviations from the mean of a data
set, we can simply add the deviations
and divide by n.
Question 18
Question
The mean absolute deviation of the four
numbers 6, 14, 14, 14 is equal to
Question 19
Question
The units of a population mean absolute
deviation for a problem in which the units
are feet are
Question 20
Question
A population has three values: 3, 7, 5.
Find the population variance.
Question 21
Question
A population consisted of the values 3, 7,
11, 11, 13. The population standard
deviation is equal to:
Question 22
Question
A sample consisting of the numbers 3, 6,
9 has a sample variance equal to 12.
Question 23
Question
The number of hours a child spent
watching TV over the past three days
was 15, 3, 9. Compute the sample
standard deviation.
Question 24
Question
A standard deviation can sometimes be
larger in numerical value than a variance.
Question 25
Question
The coefficient of variation computed
from the three sample values 25, 30, 35
would be equal to 20.
Question 26
Question
95% of data values that conform to a bellshaped
distribution lie within how many
standard deviations of the mean?
Question 27
Question
Student score = 70
Class mean = 80
Standard deviation = 5
The standard score is equal to 2.
Question 28
Question
Drawing an ace of spades and a three
of hearts are complementary events.
Question 29
Question
The items in a sample space must be
Question 30
Question
When a probability is described in terms of the proportion of times that and event can be theoretically expected to occur, it is an example of relative frequency probability.
Question 31
Question
A couple plans to have 3 children. Assuming malefemale births are equally likely, what is the probability that couple will have at least one girl among the three children?
Question 32
Question
When one throws a die 1,000 times
and determines that the probability of
obtaining a six on a die is onesixth,
that person has used the theoretical
approach to probability.
Question 33
Question
A student has three pairs of socks in her
closet: one black pair, one red pair, and
one white pair. If she randomly selects
two socks from her closet, what is the
probability that they will be the same
color? (i.e. They will be a matching pair.)
Question 34
Question
The distribution of people’s heights
is an example of a discrete
probability distribution.
Question 35
Question
The sum of the probabilities in a
discrete probability distribution could
total 1.2.
Question 36
Question
Probabilities associated with random
variables must all be equal.
Question 37
Question
A quiz with 10 questions on it was given
to a class of 10 students. Each question
was worth one point and the results are
summarized below:
X = number of correct answers = 10, 9, 8,
7, 6, 5 or less, with frequencies f(X) = 3, 2,
1, 2, 2, 0 respectively.
The quiz score mean (expected value) is:
Question 38
Question
Find the variance of X for the following
probability distribution:
X P(X)
1 .4
3 .2
5 .4
Question 39
Question
Find the standard deviation of X for the
following probability distribution:
X P(X)
20 .50
30 .10
60 .40
Question 40
Question
Nonrandom samples involve
unequal probabilities.
Question 41
Question
For a random variable X to have a
binomial distribution, it is necessary that:
Answer

X represents the number of successes

the outcome of each trial is a success

the n trials are statistically dependent on eachother

X represents the number of successes AND the outcome of each trial is a success
Question 42
Question
A success in a binomial distribution
indicates that something positive has
occurred.
Question 43
Question
A weighted die is thrown. Success is
defined as getting an even number.
The probability of success is .55.
The probability of getting 2 even
numbers when throwing the die four
times is approximately .3675.
Question 44
Question
Samantha calls on 10 houses per day
selling Girl Scout cookies. Historically,
3 out of 4 customers buy cookies.
Assuming the binomial distribution
applies here, the variance of this
distribution is:
Question 45
Question
The standard deviation of the
distribution in the previous problem
is 1.37.
Question 46
Question
The standard normal distribution has
a mean of 1 and a standard deviation
of 0.
Question 47
Question
If X is a normal random variable with
a mean of 10 and a standard
deviation of 1/2, then X = 14 is 8
standard deviations away from the
mean.
Question 48
Question
Use the standard normal curve to
determine the probability that the random
variable z will fall between –0.04 and 0.44.
That is, find the following probability:
P(–0.04 < z < 0.44):
Question 49
Question
In the standard normal zdistribution,
the probability between z = –1 and z =
+1 is the same as the probability
between z = –0.5 and z = +1.5.
Question 50
Question
The lifetime of tires is normally distributed
with a mean of 50,000 miles and a standard
deviation of 3,000 miles. The warranty is
for 46,000 miles. What proportion of the
tires will fail after the warranty but before
52,000 miles? Indicate the interval below
that contains this probability.
Answer

.0000 to .2000

.2001 to .4000

.6001 to .8000

.8001 to 1.000

.4001 to .6000
Question 51
Question
When 6.3% of the data values fall
below a normally distributed random
variable, the correct zvalue is 1.53.
Question 52
Question
Weights of cereal boxes are normally
distributed with a mean of 15 oz. and a
standard deviation of .5 oz. What is the
minimum weight a box could be and
remain in the top 14.46% of all boxes
filled?
Question 53
Question
An unbiased estimator is:
Answer

no more likely to be above the
population parameter than below.

always better than a biased
estimator.

is a specific observed value of a
statistic.

always better than a biased
estimator AND is a specific
observed value of a statistic.
Question 54
Question
The sample range is generally an
unbiased estimator of the population
range.
Question 55
Question
A disadvantage of a point estimate is
that we don’t know how accurate that
estimate is.
Question 56
Question
A sampling distribution is a
distribution of all possible values of a
statistic for a given sample size.
Question 57
Question
All standard deviations are standard
errors, but not all standard errors are
standard deviations.
Question 58
Question
If a random sample of size 16 is taken
from a skewed population whose mean is
equal to 360 and standard deviation is 36,
the standard error of the mean would
equal:
Question 59
Question
Given a population standard
deviation, as sample size increases,
standard error also increases.
Question 60
Question
For a test, the pop. mean score is 1100 and
the pop. st. dev. is 100. If the test is given
to 36 randomly selected individuals, what
is the probability that the sample mean will
lie between 1090 and 1122? After finding
the appropriate probability, indicate the
interval that includes this probability:
Answer

.0000 to .3000

.8501 to 1.000

.6001 to .7000

.3001 to .6000

.7001 to .8500
Question 61
Question
The Central Limit Theorem assures
us that the sampling distribution for
the sample mean approaches a
normal distribution as the sample
size increases, regardless of the
shape of the population distribution.
Question 62
Question
Records have shown that 15% of patients
are not satisfied with their care. A poll of
100 patients was conducted. What is the
probability that more than 18 patients will
not be satisfied with their care? After
finding the probability, indicate the
interval that includes this probability.
Answer

.0000 to .3000

.8501 to 1.000

.6001 to .7000

.3001 to .6000

.7001 to .8500
Question 63
Question
Confidence intervals specify
parameter values in advance.
Question 64
Question
In using the standard normal distribution
to establish a confidence interval for the
average number of hours that a light bulb
will last, what is the appropriate zvalue to
use for a 34% level of confidence:
Question 65
Question
Other things being equal, a 90%
confidence interval is wider than a
95% confidence interval.
Question 66
Question
A bank wants to determine mean waiting
time. It samples 100 customers and the
mean time is 6.9 minutes. Population
standard deviation is assumed to be 4
minutes. Find a 91.98% confidence
interval of mean waiting time:
Answer

6.9 +/ 0.7

6.9 +/ 0.92

6.9 +/ 0.5

6.9 +/ 1.75
Question 67
Question
A tdistribution with 5 degrees of
freedom has less area in the tails
than a standard normal distribution.
Question 68
Question
The following is true about the tdistribution:
Answer

like the standard normal distribution, there is only one tdistribution

the mean is 0

is determined by the parameter mu

approaches the standard normal as degrees of freedom become smaller
Question 69
Question
The important distinction between
the zstatistic and tstatistic is that z
is used for large sample sizes and t is
used for small sample sizes.
Question 70
Question
A confidence interval for a true population
mean is to be constructed from sample
data with size n = 23. The tvalue to use
for setting a 90% level of confidence is:
Answer

1.321

1.319

1.717

1.714

1.645
Question 71
Question
A random sample of 4 glass rods is tested
and reveals the following breaking
strength in pounds: 8, 4, 2, 6. Construct
an 80% confidence interval for the true
mean breaking strength:
Answer

(1.73, 8.27)

(3.74, 6.26)

(.46, 10.46)

(2.89, 7.11)
Question 72
Question
Increasing the sample size, but using
the same level of confidence,
produces a confidence interval that
has a greater likelihood of containing
the parameter being estimated.
Question 73
Question
The mean age of viewers of TV shows
interests advertisers. A major network
believes that the mean age of viewers of a
show is more than 30. Many affiliate
stations claim the mean age is not greater
than 30. What is H0 for testing the major
network’s claim. Population mean mu is:
Answer

less than 30

greater than 30

not equal to 30

at least 30

at most 30
Question 74
Question
In testing the hypothesis below, a
statistician found that z = 0.44. What is
the pvalue?
H0: mu = 10
Ha: mu does not equal 10
Question 75
Question
In order to compute the pvalue from
sample data, we need to know both
the alternative hypothesis and the
level of significance.
Question 76
Question
To test that the mean lifetime of light
bulbs is at least 900 hours (pop. normally
distributed and pop. st. dev. is 20), a
random sample of 25 bulbs is tested,
yielding a sample mean of 894 hours.
Find the pvalue for the test. Indicate
which interval below contains the pvalue.
Answer

.0301 to .1000

.5001 to 1.000

.2001 to .5000

.0000 to .0300

.1001 to .2000
Question 77
Question
If the pvalue for a given hypothesis
testing problem is .055 and the level
of significance is .05, the null
hypothesis should be rejected.
Question 78
Question
In the following hypothesis test,
H0: mu = 4
Ha: mu > 4
the tvalue was computed to be –2
and degrees of freedom are 11. The
correct pvalue is .025 < pvalue < .05.
Question 79
Question
Given an uppertailed ttest for one
mean with 30 degrees of freedom and
the value of the test statistic computed
from the sample data t = 2.75 would
yield a pvalue equal to .01.
Question 80
Question
Four runners were randomly sampled and
it was found they ran 23, 19, 23, and 23
miles per week. If we wish to test the
claim that the population mean running
distance is less than 25 miles per week,
what conclusion should be reached at the
1% level of significance?
Answer

Reject the claim by rejecting Ho

Accept the claim by rejecting Ho

Reject the claim by accepting Ho

Accept the claim by accepting Ho
Question 81
Question
In statistical process control, if the pvalue
is less than or equal to alpha, we
should conclude the process is under
control.
Question 82
Question
If mu = 40 pounds, sigma = 4 pounds, and
sample size is 36, the LCL and UCL on a
control chart for x would be:
Question 83
Question
A poll in a senatorial contest revealed that
40 out of 400 randomly selected people
indicated a preference for candidate A.
Find a 95.44% confidence interval for the
proportion of voters who favor candidate
A.
Answer

.07 to .13

.06 to .14

.05 to .15

.09 to .11

.38 to .42
Question 84
Question
A promoter is deciding whether to book a
new band. The promoter decides to do a
survey to try to estimate the true
proportion of individuals in the area who
will attend the concert. What should
sample size be to estimate the proportion
to within 5% with a 68.26% confidence
level? (Assume maximum error possible.)
Question 85
Question
In selecting a sample size to estimate
a population proportion, the sample
size will depend, among other things,
upon the level of confidence that is to
be used for the estimate.
Question 86
Question
A company is planning to test whether the
market share of a new product during its
first year on the market is more than 20
percent. The appropriate null hypothesis
would be that the market share percentage
is
Question 87
Question
A manufacturer claims that no more than
20% of all units will experience a failure
within the first 5 years. 100 purchasers
randomly selected were asked to report
failures. If 28 failures were reported, what
conclusion should be reached about the
manufacturer’s claim? alpha = 5%.
Answer

Accept the claim by rejecting Ho

Reject the claim by rejecting Ho

Accept the claim by accepting Ho

Reject the claim by accepting Ho
Question 88
Question
Assume for a given hypothesis testing
problem, the test statistic was computed
and led to a rejection of the null at alpha of
.01. Based on this information, we can
further conclude that since H0 was rejected
at the 1% level, then H0
Answer

Must be rejected at alpha of 10%

must be accepted at alpha of 10%

must be accepted at alpha of 5%

must be accepted at alpha of 5% and 10%
Question 89
Question
The ttest for n1 = 15 and n2 = 15 using
paired testing has
Answer

14 degrees of freedom

13 degrees of freedom

29 degrees of freedom

28 degrees of freedom

30 degrees of freedom
Question 90
Question
A paired difference experiment produced
the following data: n = 30, xbar1 = 72, xbar2 = 75.5,
Dbar = 3.5, and SD squared = 17. SE(Dbar) is equal to:
Answer

0.64

0.64

0.75

0.75

3.10
Question 91
Question
Assume a matched pairs test for a
mean difference with a twotailed
alternative hypothesis and the
number of paired differences n = 4. If
the computed test statistic t = 2.353,
then the pvalue would be equal to
.05.
Question 92
Question
An analyst is testing a new system to see if
it uses a different processing time than the
old system. The time for each in seconds
was recorded for 28 samples.
Old System: mean = 21.2, s = 3.2, n = 28
New System: mean = 24.3, s = 2.1, n = 28
Difference (Old – New): mean = – 3.1,
s = 1.4, n = 28
What is the appropriate statistical test?
Answer

Paired ttest for mean differences

Hypothesis test for one proportion

regression analysis

Independent ttest for mean differences
Question 93
Question
An analyst is testing a new system to see if
it uses a different processing time than the
old system. The time for each in seconds
was recorded for 28 samples.
Old System: mean = 21.2, s = 3.2, n = 28
New System: mean = 24.3, s = 2.1, n = 28
Difference (Old – New): mean = – 3.1,
s = 1.4, n = 28
What is the alternative hypothesis for the
question above?
Answer

Ha: muD = 0

Ha: MuD < 0

Ha: MuD > 0

Ha: MuD does not equal 0
Question 94
Question
In a simple linear regression analysis,
the pvalue associated with a test of
the slope coefficient was equal to .026,
which would lead us to conclude that a
linear relationship exists between the
two variables at the 5% level of
significance.
Question 95
Question
An analyst is testing a new system to see if
it uses a different processing time than the
old system. The time for each in seconds
was recorded for 28 samples.
Old System: mean = 21.2, s = 3.2, n = 28
New System: mean = 24.3, s = 2.1, n = 28
Difference (Old – New): mean = – 3.1,
s = 1.4, n = 28
The correct tvalue for the previous
question is:
Question 96
Question
An analyst is testing a new system to see if
it uses a different processing time than the
old system. The time for each in seconds
was recorded for 28 samples.
Old System: mean = 21.2, s = 3.2, n = 28
New System: mean = 24.3, s = 2.1, n = 28
Difference (Old – New): mean = – 3.1,
s = 1.4, n = 28
If the pvalue for this test is less than .002
and alpha is .05, analysts would conclude
that
Answer

the old system uses more processing time

the old system uses less processing time

the systems use the same processing time

the system processing times are unequal
Question 97
Question
A doctor claims the average person is
more than 11 pounds overweight. To test
the claim, the difference between actual
and ideal weight of 36 randomly selected
people was calculated. The sample mean
and sample standard deviation were 14
and 5 pounds respectively. At alpha of 1%,
can we conclude the claim is true?
Answer

Yes, the claim is true by rejecting Ho

Yes, the claim is true by accepting Ho

No, the claim is not true by rejecting Ho

No, the claim is not true by accepting Ho
Question 98
Question
The difference between the
independent sample and paired
sample approach is that with the
independent sample approach, a
background variable's effect is
controlled by pairing.
Question 99
Question
In general, the paired samples
method is preferred over the
independent samples method.
Question 100
Question
The ttest for n1 = 15 and n2 = 7 using the
independent samples approach has
(assuming equal population variances)
Answer

14 degrees of freedom

6 degrees of freedom

21 degrees of freedom

20 degrees of freedom

22 degrees of freedom
Question 101
Question
Given the following information about a
hypothesis test of the difference between
two means based on independent random
samples. (Assume normal distributions
with equal variances.) The correct pvalue
is between .005 and .01.
H0: μ1 – μ2 = 0
Ha: μ1 – μ2 ≠ 0
x1 = 16.32, x2 = 17.44,
s1squared = 4.3,
s2squared = 2.2,
n1 = 30, n2 = 32
Question 102
Question
The best statistic for pi1  pi2 is p1 – p2.
Question 103
Question
For n1 = 70 and n2 = 100 from populations
1 and 2, the number of successes are 35
and 35 respectively. The estimate of the
difference between pi1  pi2 is
Question 104
Question
For n1 = 200 and n2 = 100 from
populations 1 and 2, the numbers of
successes are 35 and 25 respectively.
The value of the pooled sample
proportion phat is 0.20.
Question 105
Question
In an uppertailed test of the difference
of two proportions, the zvalue was
calculated to be 2.69. The pvalue for
this test would then be .0036.
Question 106
Question
The purpose of a scatterplot is to
visually determine if a relationship
exists between two variables.
Question 107
Question
In using the regression model for
forecasting the next value (or an
individual value) of y, the prediction
interval will be
Answer

the same as the estimating interval

narrower than the estimating interval

wider than the estimating interval

wider or narrower, depending on the data
Question 108
Question
The coefficient of determination (R2) is
equal to 0.64, and the linear regression
equation which indicates an inverse
relationship between x and y is equal to
ŷ = 2  .74x, then the correlation coefficient
must necessarily be equal to:
Answer

.80

+.64

.74

either .80 or +.80
Question 109
Question
In a simple linear regression and
correlation problem, a correlation
coefficient of .80 means that 64% of
the variation in y can be explained by
x.
Question 110
Question
When R2 = 1, then Se = Sy.
Question 111
Question
A correlation coefficient of –1.0 would
imply that the standard error of the
estimate Se would necessarily be equal
to 0.
Question 112
Question
Standard error is measured in the units
of the x variable.
Question 113
Question
The range of a regression coefficient is
– 1 to + 1.
Question 114
Question
Standard deviation of height = 2.45
inches.
Standard deviation of weight = 23
pounds.
Correlation = .56
If we are predicting weight from height,
the regression coefficient is 5.26.
Question 115
Question
If ŷ = 15 + 10x, then the estimated value of
y when x = 5 is:
Question 116
Question
When H0 is accepted in a regression
model, we conclude:
Answer

A linear relationship exists
between x and y.

The x variable is statistically
significant.

The x variable is not statistically
significant.

A linear relationship exists
between x and y AND the x
variable is statistically significant.
Question 117
Question
Seasonality can be incorporated into
regression models with dummy
variables.
Question 118
Question
For the linear equation ŷ = 50 – 2x where y
is the number of items sold and x is the
price of the product in $, which is true?
Answer

The equation represents an inverse
relationship between y and x.

The equation represents a positive
relationship between y and x
because the constant term 50 is
greater than 0.

The slope of the line is – 2 dollars
per each item sold.

The slope of the line is +50 dollars
per each item sold.
Question 119
Question
If a correlation exists between y and x,
then necessarily either y causes x or x
causes y.
Question 120
Question
The advantage of multiple regression
over simple regression is that we can
change more than one variable at a
time.
Question 121
Question
If ŷ = 31 + 1.2x1 – 3.4x2 + 5x3, then the
estimated value of y when x1 = – 6, x2 = 3,
and x3 = 2 is:
Question 122
Question
The alternative hypothesis in a multiple
regression problem is that no linear
relationship exists between a given
independent variable and y.
Question 123
Question
The correlation coefficient between college
GPA and H.S. GPA is .70. The correlation
coefficient between college GPA and H.S.
rank is .60. Which of the following is true?
Answer

70% of the variation in college GPA can
be explained by H.S. GPA.

60% of the variation in college GPA can
be explained by H.S. GPA.

H.S. GPA is a better predictor of college
GPA than H.S. rank.

There is an inverse relationship between
H.S. GPA and College GPA.
Question 124
Question
Adding another variable to a regression
equation will make R2
Answer

decrease

increase

stay the same or increase

stay the same or decrease

increase, decrease, or stay the same
Question 125
Question
The purpose of R2 adjusted is to
discern the effect of adding a variable
to a model.
Question 126
Question
Adding another variable to a
regression equation will necessarily
make R2 adjusted increase.
Question 127
Question
It is possible for a variable to be
significant in multiple regression when
that same variable is not significant in
simple regression.
Question 128
Question
Stable environments are critical for
effective time series analysis.
Question 129
Question
Which of the following is not a component
of time series?
Answer

Seasonality

Cycle

Randomness

Trend

Contingency
Question 130
Question
The regression model based on annual
data for the last 20 years is
ŷ = 20.35 + 2.4x. The actual value of y
when x = 12 is 52. The value of the
forecasting error is:
Answer

68.35

16.35

2.85

49.15

40