1Chapter 15

Audit Sampling for Tests of Controls and

Substantive Tests of Transactions
 Concept Checks

P. 529

1. A representative sample is one in which the characteristics of interest for the

sample are approximately the same as for the population (that is, the sample
accurately represents the total population). If the population contains
significant misstatements, but the sample is practically free of
misstatements, the sample is nonrepresentative, which is likely to result in
an improper audit decision. The auditor can never know for sure whether he
or she has a representative sample because the entire population is
ordinarily not tested, but certain things, such as the use of random
selection, can increase the likelihood of a representative sample.

2. Statistical sampling is the use of mathematical measurement techniques to

calculate formal statistical results. The auditor therefore quantifies
sampling risk when statistical sampling is used. In nonstatistical sampling, the
auditor does not quantify sampling risk. Instead, conclusions are reached
about populations on a more judgmental basis. Auditing standards permit
auditors to use either statistical or nonstatistical sampling methods.

P. 551

1. Analysis of exceptions is the investigation of individual exceptions to

determine the cause of the breakdown in internal control. Such
analysis is important because by discovering the nature and causes of
individual exceptions, the auditor can more effectively evaluate the
effectiveness of internal control. The analysis attempts to tell the “why” and
“how” of the exceptions after the auditor already knows how many and what
types of exceptions have occurred.

2. The steps of the audit sampling process that differ across statistical and
nonstatistical sampling are as follows.

 Specify acceptable risk of overreliance: the concepts are the same

across the two methods, but quantifying the risk is usually handled
differently. In nonstatistical sampling, auditors use categories such as
high, medium, or low to quantify acceptable risk, whereas auditors using
statistical sampling assign a specific amount (e.g., 10 percent) so that a
statistical sample size can be determined and results can be statistically
evaluated.

15-1
Copyright © 2017 Pearson Education, Ltd.
Concept check, P. 551 (continued)

 Determine initial sample size: In statistical sampling, auditors use

software or tables to determine sample size based on a statistical
formula. In both statistical and nonstatistical sampling approaches,
auditors need to consider population size, TER, ARO, and EPER in
determining sample sizes and should arrive at a similar sample size as a
result.
 Select the sample: The only difference in sample selection is the
requirement to use a probabilistic method when using statistical
sampling.
 Generalize from the sample to the population: For statistical sampling,
the auditor calculates the computed upper exception rate (CUER) at a
specified ARO using audit software or tables based on a statistical
formula. In nonstatistical sampling, the auditor compares the sample
exception rate (SER) to the estimated population exception rate (EPER)
in determining the acceptability of the population.

 Review Questions

15-1 Sampling risk is the risk that the auditor reaches the wrong conclusion
because a sample is not representative of the population. Sampling risk is an
inherent part of sampling because the auditor is not testing the entire
population. To reduce sampling risk, the auditor could increase sample size or
use a sampling method that increases the likelihood of having a representative
sample. Nonsampling risk is the risk that an auditor reaches the wrong
conclusion for any reason other than sampling risk. For example, the auditor
may perform procedures incorrectly or fail to correctly interpret the audit
evidence. To reduce nonsampling risk, the auditor can follow auditing standards
related to careful design of audit procedures, and proper supervision and
review of audit work performed.

15-2 For both statistical and nonstatistical methods, the three main phases
of audit sampling are:
1. Plan the sample
2. Select the sample and perform the tests
3. Evaluate the results

15-3 In probabilistic sample selection, the auditor randomly selects items from
a population such that each item has a known probability of being included in
the sample. Three examples of probabilistic sample methods include simple
random sample selection, systematic sample selection, and probability
proportional to size sample selection. In nonprobabilistic sample selection
methods, such as haphazard sample selection and block sample selection, the
auditor uses nonprobabilistic methods that approximate a random sampling
approach.

15-2
Copyright © 2017 Pearson Education, Ltd.
15-4 Haphazard sample selection is an example of nonprobabilistic sample
selection and refers to a selection of items without any conscious bias by the
auditor. The auditor selects population items without regard to their size, source,
or other distinguishing characteristics. The most serious shortcoming of this
sample selection is the difficulty of remaining completely unbiased in the
selection. Owing to the auditor’s training and unintentional bias, certain
population items are more likely than others to be included in the sample.

15-5 In replacement sampling, an element in the population can be included

in the sample more than once if the random number corresponding to that
element is selected more than once. In nonreplacement sampling, an element
can be included only once. If the random number corresponding to an element
is selected more than once, it is simply treated as a discard the second time.
Although both selection approaches are consistent with sound statistical theory,
auditors rarely use replacement sampling; it seems more intuitively satisfying
to auditors to include an item only once.

15-6 The two common ways to obtain samples with larger recorded amounts
are:
 Sampling with probability proportional to size (PPS) by taking a
sample in which the probability of selecting any individual population
item is proportional to its recorded amount
 Stratified sampling by dividing the population into subpopulations,
usually by dollar size, and taking larger samples from the
subpopulations with larger sizes

15-7 In systematic sampling, the auditor calculates an interval and then

methodically selects the items for the sample based on the size of the interval.
The interval is set by dividing the population size by the number of sample
items desired.
To select 40 numbers from a population of 2,800, the auditor divides 40
into 2,800 and gets an interval of 70. He or she then selects a random number
between 0 and 69. Assume the auditor chooses 17. The first item is the number
17. The next is 87, then 157, 227, and so on.
The advantage of systematic sampling is its ease of use. In most
populations a systematic sample can be drawn quickly, the approach automatically
puts the numbers in sequential order, and documentation is easy.
A major problem with the use of systematic sampling is the possibility
of bias. Because of the way systematic samples are selected, once the first
item in the sample is selected, other items are chosen automatically. This
causes no problems if the characteristics of interest, such as control deviations,
are distributed randomly throughout the population; however, in many cases
they are not. If all items of a certain type are processed at certain times of the
month or with the use of certain document numbers, a systematically drawn
sample has a higher likelihood of failing to obtain a representative sample. This
shortcoming is sufficiently serious that some CPA firms prohibit the use of
systematic sampling.

15-3
Copyright © 2017 Pearson Education, Ltd.
15-8 Random (probabilistic) selection is a part of statistical sampling, but it is
not, by itself, statistical measurement. To have statistical measurement, it is
necessary to mathematically generalize from the sample to the population.
Probabilistic selection must be used if the sample is to be evaluated
statistically, although it is also acceptable to use probabilistic selection with
a nonstatistical evaluation. If nonprobabilistic selection is used, nonstatistical
evaluation must be used.

15-9 The auditor should examine the audit program and select those audit
procedures where audit sampling applies. Some of the audit procedures where
audit sampling is not applicable are:

 Performing analytical procedure (for example, reviewing sales

transactions for large and unusual amounts)
 Observation procedure (for example, observing whether the duties
of the accounts receivable clerk are separate from handling cash as
the test of control)
 Testing automated controls (with the help of computer assisted
auditing techniques)

15-10 An attribute is the definition of the characteristic being tested and the
exception conditions whenever audit sampling is used. The attributes of interest
are determined directly from the audit program. In a test of control, that attribute
is evidence of the operation of the control consistent with the design. In a
substantive test, the attribute is the absence of monetary misstatement.

15-11 An attribute is the characteristic being tested for in a population. An

exception occurs when the attribute being tested for is absent. The exception
for the audit procedure, the duplicate sales invoice has been initialed indicating
the performance of internal verification, is the lack of initials on duplicate sales
invoices.

15-4
Copyright © 2017 Pearson Education, Ltd.
15-12 The terms below are defined as follows:

TERM DEFINITION
a. Acceptable risk of The risk the auditor is willing to take of accepting a control
overreliance (ARO) as effective or a rate of monetary misstatements as
tolerable, when the true population exception rate is
greater than the tolerable exception rate.
b. Computed upper The highest estimated exception rate in the population
exception rate (CUER) at a given ARO.
c. Estimated population The exception rate the auditor expects to find in the
exception rate (EPER) population before testing begins. It is necessary to
plan the appropriate sample size.
d. Sample exception rate The actual rate of exception discovered in the sample.
(SER) It is calculated by dividing the actual number of
exceptions in the sample by the sample size.
e. Tolerable exception The exception rate the auditor will permit in the population
rate (TER) and still be willing to use the assessed control risk
and/or the amount of monetary misstatements in the
transactions established during planning.

15-13 The sampling unit is the population item from which the auditor selects
sample items. The major consideration in defining the sampling unit is making
it consistent with the objectives of the audit tests. Thus, the definition of the
population and the planned audit procedures usually dictate the appropriate
sampling unit.
The sampling unit for verifying the occurrence of recorded sales would
be the entries in the sales journal since this is the record the auditor wishes to
validate. The sampling unit for testing the possibility of omitted sales is the
shipping document from which sales are recorded, because the failure to bill a
shipment is the exception condition of interest to the auditor.
15-14 The tolerable exception rate (TER) represents the exception rate that
the auditor will permit in the population and still be willing to use the assessed
control risk and/or the amount of monetary misstatements in the transactions
established during planning. TER is determined by choice of the auditor on the
basis of his or her professional judgment.
The computed upper exception rate (CUER) is the highest estimated
exception rate in the population, at a given ARO. For nonstatistical sampling,
CUER is determined by adding an estimate of sampling risk to the SER
(sample exception rate). For statistical sampling, CUER is determined by using
a statistical sampling table after the auditor has completed the audit testing and
therefore knows the number of exceptions in the sample.

15-5
Copyright © 2017 Pearson Education, Ltd.
15-15 Tolerable exception rate (TER) is the result of an auditor’s judgment.
The suitable TER is a question of materiality and is therefore affected by both
the definition and the importance of the attribute in the audit plan. It represents
the exception rate that the auditor will permit in the population and still be
willing to conclude that the control is operating effectively and/or the amount of
monetary misstatements in the transactions established during planning is
acceptable.
The sample size for a TER of 7% would be smaller than that for a TER
of 4%, all other factors being equal.

15-16 If the auditor places extensive reliance on internal controls, the auditor
needs greater assurance that the internal controls are effective. Control risk is
assessed as low and therefore ARO will also be assessed as low, that is, the
auditor requires a low risk of overrelying on controls. ARO represents the
auditor’s risk of incorrectly accepting the control as effective. A larger sample
size is therefore required to support the lower risk of overreliance and this
corresponds to a reduction in substantive tests of details of balances.

15-17 The relationship between sample size and the four factors determining
sample size are as follows:
a. As the ARO increases, the required sample size decreases.
b. As the population size increases, the required sample size is
normally unchanged, or may increase slightly.
c. As the TER increases, the sample size decreases.
d. As the EPER increases, the required sample size increases.

15-18 When the CUER exceeds the TER, the auditor may do one or more of
the following:

1. Revise the TER or the ARO. This alternative should be followed

only when the auditor has concluded that the original specifications
were too conservative, and when he or she is willing to accept the
risk associated with the higher specifications.
2. Expand the sample size. This alternative should be followed when
the auditor expects the additional benefits to exceed the additional
costs. That is, the auditor believes that the sample tested was
not representative of the population and that exceptions are not
expected in the expanded sample.
3. Revise assessed control risk upward. This is likely to increase
substantive procedures. Revising assessed control risk may be
done if 1 or 2 is not practical and additional substantive procedures
are possible.

15-6
Copyright © 2017 Pearson Education, Ltd.
15-18 (continued)

4. Write a letter to management. This action should be done in

conjunction with each of the three alternatives above. Management
should always be informed when its internal controls are not
operating effectively. If a deficiency in internal control is considered
to be a significant deficiency in the design or operation of internal
control, professional standards require the auditor to communicate
the significant deficiency to the audit committee or its equivalent in
writing. If the client is an accelerated filer publicly traded company,
the auditor must evaluate the deficiency to determine the impact
on the auditor’s report on internal control over financial reporting.
If the deficiency is deemed to be a material weakness, the auditor’s
report on internal control would contain an adverse opinion.

15-19 Attributes sampling is a statistical, probabilistic sampling method that

results in an estimate of the proportion of items in a population containing a
characteristic or attribute of interest. Attributes sampling is commonly used for
tests of controls and substantive tests of transactions.

15-20 In this situation, the sample exception rate (SER) is 4%, the sample
size is 100 and the ARO is 10%. From the 10% ARO table (Table 15-9) then,
the CUER is 7.9%. This means that the auditor can state with a 10% risk of
being wrong that the true population exception rate does not exceed 7.9%.

15-21 The decisions the auditor must make in using attributes sampling are:
 What are the objectives of the audit test?
 Does audit sampling apply?
 What attributes are to be tested and what exception conditions are
identified?
 What is the population?
 What is the sampling unit?
 What should the TER be?
 What should the ARO be?
 What is the EPER?
 What generalizations can be made from the sample to the population?
 What are the causes of the individual exceptions?
 Is the population acceptable?

In making the above decisions, the following should be considered:

 The individual situation.
 Time and budget constraints.
 The availability of additional substantive procedures.
 The professional judgment of the auditor.

15-7
Copyright © 2017 Pearson Education, Ltd.
 Multiple Choice Questions From CPA Examinations

15-22 a. (3) b. (1) c. (1)

15-23 a. (3) b. (4) c. (4)

15-24 a. (4) b. (3) c. (2)

 Discussion Questions and Problems

15-25 Note: The PCAOB has reorganized PCAOB auditing standards effective
December 31, 2016. The responses below reflect the location of the appropriate
guidance in the reorganized standards.

a. AS 2315, Audit Sampling, lists benefits and costs of statistical

sampling in section .46. Statistical sampling allows the auditor to
design an efficient sample, measure the sufficiency of evidence,
and evaluate the sample results. Statistical sampling allows the
auditor to quantify sampling risk. However, the costs of using a
statistical sampling approach include the costs of training the
auditors in appropriate methods, designing samples that satisfy
the statistical requirements, and selecting items to be examined.
Auditors trade off these benefits and costs in determining whether
to use a statistical or nonstatistical approach.

b. According to AS 2315.09-2315.11, audit risk includes both

uncertainties from sampling (sampling risk) as well as
uncertainties unrelated to sampling (nonsampling risk).

c. According to AS 2315.32, sampling is not always appropriate for

tests of controls. For example, procedures performed during the
planning process to obtain an understanding of internal controls
do not involve sampling. Sampling is not appropriate for tests of
controls related to separation of duties or when there is no
documented evidence the control has been performed. Sampling
also may not apply to procedures performed to obtain an
understanding of the control environment (e.g., an inquiry or
observation) or the accounting system.

15-8
Copyright © 2017 Pearson Education, Ltd.
15-25 (continued)

d. These risks are defined in AS 2315.12. The risk of assessing

control risk too low represents the risk that the level of control risk
assessed by the auditor based on the sample is lower than the
true operating effectiveness of the control warrants. This has the
potential to impact the effectiveness of the audit if the auditor
inappropriately relies on internal controls to reduce substantive
testing. The risk of assessing control risk too high represents the
risk that the level of control risk assessed by the auditor based on
the sample is higher than the true operating effectiveness of the
control warrants. This risk relates to the efficiency of the audit in
that it could result in the auditor increasing the level of substantive
testing unnecessarily.

15-26 a. The sampling unit is the sales invoice.

b. (1) All invoices numbered 0001 to 9300.

(2) =RANDBETWEEN(1,9300)
(3) Students are expected to list out 60 samples generated in
an Excel spreadsheet using the function in (2) above..

c. (1) Interval = 9,300/60 = 155

(2) =RANDBETWEEN(1,155)
(3) The next four samples are Invoice #207 (52+155), Invoice
#362 (207+155), Invoice #517 (362+155), Invoice #672
(517+155).
(4) The advantage of systematic selection is its ease of use. In
most populations, a systematic sample can be drawn
quickly and the approach automatically puts the numbers in
sequence, making it easy to develop the appropriate
documentation.
A concern with systematic selection is the possibility
of bias. Once the first item in the sample is selected, all
other items are automatically chosen. This causes no
problem if the characteristics of interest is distributed
randomly throughout the population, but this may not
always be the case (for example, with a control deviation
occurred at a certain particular quarter). Auditors must
therefore consider possible patterns in the population data
that can cause sample bias.

d. In a stratified sampling method, an auditor will take larger samples

from the subpopulations with larger sizes. The company generates
the largest dollar value of sales in the 2nd quarter, therefore, the
auditor selects the largest samples from the 2nd quarter.

15-9
Copyright © 2017 Pearson Education, Ltd.
15-27 a. Control (1) is an automated control and can be tested using the
computer assisted auditing technique, which does not require
sampling. Control (2) can be tested by sampling. To test whether
recorded sales are for shipments actually made, sales invoices are
traced to supporting documents including shipping documents and
customer orders to test for occurrence.

b. The sampling unit in the above audit procedure is the sales invoice.

c. At a 5% ARO, with a 0% EPER and a 6% TER, the initial sample

size is determined to be 49.

d. The selection method of the audit trainee is inappropriate:

 Bias has been introduced to the selection to avoid small
amounts and insignificant overseas sales. Every possible
combination of population items does not have an equal
chance of being included in the sample. This is a common
shortcoming in haphazard sample selection as it is difficult
to keeping the auditor completely unbiased.
 The trainee is incorrect in defining the population for testing
completeness of sales. To test completeness of sales (i.e.
whether existing sales transactions are recorded), the
population should be the shipping documents. He cannot
use the same set of samples for testing completeness of
sales.

If systematic sample selection method is used, the first step is to

calculate the interval. The population size of the sales invoices is
20802 (74243-53419+1). The interval is 425 (20802/49). Use the
Excel RANDBETWEEN function to choose the random starting
point for the selection: =RANDBETWEEN(53419,53844).
Assuming the first starting point generated by the above Excel
function is Invoice #53500, the next two samples will be Invoice
#53925 (53500+425) and Invoice #54350 (53925+425).

15-10
Copyright © 2017 Pearson Education, Ltd.
15-28 a. It would be appropriate to use attributes sampling for all audit
procedures except audit procedures 1 and 7. Procedure 1
involves comparing current year’s sales figure against that of
prior years which is an analytical procedure. Procedure 7 is
an observation procedure for which no documentation exists
to perform audit sampling.

b. The sample sizes for each attribute are as follows:

Audit procedure 2: Test of control – ARO = 10%; TER = 7%; EPER
= 1.25%; Initial sample size = 55
Audit procedure 3: Test of control – ARO = 10%; TER = 7%; EPER
= 1.25%; Initial sample size = 55
Audit procedure 4: Substantive test of transaction – ARO = 10%;
TER = 6%; EPER = 1.0%; Initial sample size = 64
Audit procedure 5: Substantive test of transaction – ARO = 10%;
TER = 6%; EPER = 1.0%; Initial sample size = 64
Audit procedure 6: Substantive test of transaction – ARO = 10%;
TER = 6%; EPER = 1.0%; Initial sample size = 64.

c. Using an ARO of 10%, the CUER is determined to be 9.4%. In

other words, if the auditor concludes that the true exception rate
does not exceed 9.4%, there is a 90% probability that the
conclusion is right and a 10% that it is wrong

15-11
Copyright © 2017 Pearson Education, Ltd.
15-29 a.
(1) Common objectives of audit tests in the sales and collection
cycle are:
 test the effectiveness of internal controls over sales and
cash receipts; and
 determine whether sales and cash receipts transactions
contain monetary misstatements.
(2) Sampling applies to the audit procedure which is considered
as test of control.
(3) Attribute – Credit is approved; Exception condition – Lack of
initials indicating the credit approval.
(4) Population – Sales invoices issued during the year ended
31 December 20X6; Sampling unit – Sales invoice
(5) The above steps are the same for nonstatistical and
attributes sampling.

b.
(1) There may a variation in the answers for this part. Students
may discuss about the following points. ARO represents the
auditor’s measure of sampling risk. It depends on the
assessment of control risk and the importance of the
attribute. For a private company, it depends on whether the
auditor plans to reply on internal control. For audits where
there is extensive reliance on internal control, control risk
will be assessed as low and therefore ARO will also be
assessed as low. Conversely, if the auditor plans to rely on
internal controls only to a limited extent, control risk will be
assessed as high and so will ARO. Further, auditor can
establish different ARO levels for different attributes of an
audit test depending on the importance of the attribute and
related control. It is common for auditors to use higher ARO
levels for tests of credit approval (as in this case) than for
tests of the occurrence of sales invoices and bills of lading
because the exceptions for the latter are likely to have a
more direct impact on the correctness of the financial
statements than the former.
(2) Some of the common factors for consideration in
determining the initial sample size under nonstatistical
sampling include:
 Population (not a significant factor);
 TER
 ARO
 EPER
 Professional judgment and other factors including
whether the sources of evidence are persuasive,
whether there are concerns about the operations of the
control, whether the controls have been changed, and
whether deficiencies have been experienced in the past.
15-12
Copyright © 2017 Pearson Education, Ltd.
15-29 (continued)

c.

(1) The concepts of specifying ARO are the same for both
attributes and nonstatistical sampling, however the method
of quantification is usually different. For nonstatistical
sampling in part b., it will be expressed as low, medium and
high for ARO, whereas a specific amount, such as 10
percent or 5 percent, will be assigned for ARO if attributes
sampling is used.
(2) Referring to Table 15-8, the initial sample size for Cases 1, 2
and 3 are 93, 156, and 96 respectively.
(3) There is a key difference in sample selection for attributes
and nonstatistical sampling. Either simple random or
systematic sampling (both being probabilistic methods) will
be used for attributes sampling. For nonstatistical sampling,
auditors may use any of the probabilistic or nonprobabilistic
methods (such as simple random, systematic, haphazard
sampling, or block sampling method).

d. It can be concluded that the results of the audit tests are not
acceptable. There are typically four courses of actions that an
auditor can choose from:
 Revise TER or ARO when the auditor has concluded
that the original specifications were too conservative;
 Expand the sample size when the initial sample was not
representative;
 Revise assessed control risk; and
 Communicate with the audit committee or management.

15-13
Copyright © 2017 Pearson Education, Ltd.
15-30 a.

ALLOWANCE FOR SAMPLING

SER TER-SER RISK SUFFICIENT?

1. 2% 3% Probably*

2. 2% 3% No (due to smaller sample size)*

3. 2% 3% Yes

4. 2% 3% Probably*

5. 2% 3% No (due to small sample size)

6. 10% NA No (SER exceeds TER)

7. 0% 5% Yes

8. 0% 5% No due to small sample size

* Students’ answers as to whether the calculated allowance for

sampling risk is sufficient will vary, depending on their judgment.
However, they should recognize the effect that lower sample sizes
have on the allowance for sampling risk in situations 2, 5, and 8.

b. Using the attributes sampling table in Table 15-9, the CUERs for
columns 1-8 are:
1. 4.6%
2. 6.2%
3. 4.0%
4. 4.6%
5. 9.2%
6. 16.4%
7. 3.0%
8. 11.3%

15-14
Copyright © 2017 Pearson Education, Ltd.
15-30 (continued)

c.

CHANGE IN EFFECT ON ILLUSTRATIONS IN PART

FACTORS CUER a. or b.

1 Decrease in ARO Increase Compare columns 3 and 4

2 Decrease in No effect or Compare columns 1 and 4
population size minor decrease
3 Decrease in Increase Compare columns 4 and 5
sample size (both sample exception
rates are 2%)
4 Decrease in the Decrease Compare columns 6 and 7
number of
exceptions in the
sample

d. The factor that appears to have the greatest effect on CUER is the
number of exceptions found in the sample compared to sample
size. For example, in columns 2 and 6, the increase from 2% to
10% SER dramatically increased the CUER. Population size
appears to have the least effect. For example, in columns 1 and
4, the CUER was the same using the attributes sampling table
even though the population in column 1 was 10 times larger.

e. The CUER represents the results of the actual sample whereas

the TER represents what the auditor will allow. They must be
compared to determine whether or not the population is acceptable.

15-15
Copyright © 2017 Pearson Education, Ltd.
15-31 a. The four populations have different EPER, TER and ARO. It is
unlikely to reach the same conclusion of having 100 sample size.
 Comparing population 1 and population 2, they both have the
same EPER (1%) and TER (5%). Yet the higher ARO in
population 1 (10% vs 5% in population 2) should give a
smaller sample size than that in population 2.
 Comparing population 2 and population 3, they both have the
same EPER (1%) and ARO (5%). Yet the higher TER in
population 3 (20% vs 5% in population 2) should give a
smaller sample size than that in population 2.
 Comparing population 2 and population 4, they both have the
same TER (5%) and ARO (5%). Yet the lower EPER in
population 4 (0% vs 1% in population 2) should give a smaller
sample size than that in population 2.

b. The initial sample sizes determined by Table 15-8 are shown

below. The effect of selecting a smaller sample size than the initial
sample size required from the table (as in Populations 1, 2, and 4)
is the increased likelihood of having the CUER exceed the TER. If
a larger sample size is selected (as in Population 3), the result
may be a sample size larger than needed to satisfy TER. That
results in excess audit cost. Ultimately, however, the comparison
of CUER to TER determines whether the sample size was too
large or too small.
1 2 3 4
EPER (in percent) 1 1 1 0
TER (in percent) 5 5 20 5
ARO (in percent) 10 5 5 5
Initial sample size (from Table 15-8) 77 93 22 59
Actual sample size 50 50 50 50
Conclusion (Actual sample size
Too Too Too Too
compared to sample size determined
small small large small
by Table 15-8)

c. It is correct to ignore the size of the population in determining the

initial sample size. Statistical theory shows that in populations
where attributes sampling applies, population size is a minor
consideration in determining sample sizes. As the populations
here are considerably large, the population size effect may be
ignored.

15-16
Copyright © 2017 Pearson Education, Ltd.
15-31 (continued)

d. The SER and CUER from Table 15-9 for the four populations are
shown below. Before a population can be considered acceptable,
the CUER determined on the basis of the actual sample results
must be less than or equal to TER when both are based on the
same ARO. Such requirements are not met for populations 1, 2,
and 4. Only population 3 can be considered as acceptable. It is
necessary to take specific actions for populations 1, 2, and 4.
Taking population 4 as an example, an increase in sample size
may be worthwhile in because the CUER exceeds the TER by
only a small amount.

1 2 3 4
Actual sample size 50 50 50 50
Actual number of exceptions in the
2 3 1 0
sample
SER (in percent) 4 6 2 0
CUER (in percent) (from Table 15-9) 10.3 14.8 9.2 5.9
TER (in percent) 5 5 20 5
Conclusion: Acceptable? No No Yes No

e. The conclusion is incorrect. The true exception rate should be

unknown. The CUER should be interpreted as this: if the auditor
concludes that the true exception rate does not exceed 9.2
percent for this population, there is a 95 percent probability that
the conclusion is right and a 5 percent chance that it is wrong

15-17
Copyright © 2017 Pearson Education, Ltd.
15-32 a. Based on the given ARO of 5% (confidence level of 95%), a
CUER of 7.92% indicates the auditor can conclude that the
exception rate in the population is no greater than 7.92% with a
5% risk of the exception rate exceeding 7.92%. Stated differently,
the auditor is 95% confident that the population exception rate
does not exceed 7.92%.

b. Given that Annie established a tolerable exception rate (TER) of

5%, she cannot conclude that the control is effective because the
CUER of 7.92% (based on one identified deviation) exceeds the
TER of 5%.

c. Annie has the following options given that CUER exceeds TER:
1. Revise the TER or the ARO. She would choose this
alternative only if she concludes that the original
specifications were too conservative, or if she is willing to
accept the risk associated with the higher specifications.
2. Expand the sample size. She would choose this alternative
if she believes that the sample tested was not
representative of the population and that exceptions are
not expected in the expanded sample.
3. Revise assessed control risk upward. This is likely to
increase substantive procedures. Revising assessed
control risk may be done if 1 or 2 is not practical and
additional substantive procedures are possible.
4. Disclose the information to management. This action
should be done in conjunction with each of the three
alternatives above to inform management that internal
controls are not operating effectively.

d. The controller’s explanation that the deviation occurred while a

temporary employee was covering for the regular clerk could
impact which of the alternatives listed in part c. would be most
appropriate. If the exception is isolated to a brief time period, then
Annie may choose to expand the sample. She would expect zero
deviations for any observations drawn outside the time period
covered by the temporary employee. In expanding the sample, she
could test additional transactions occurring outside the time period,
and assuming zero exceptions are identified in the expanded
sample and the CUER is less than TER, she could conclude
controls are effective. Deciding between the options discussed in
part c. would still be a matter of weighing the costs and benefits of
each.

15-18
Copyright © 2017 Pearson Education, Ltd.
15-33 a. The following shows which are exceptions and why:

INVOICE
NUMBER EXCEPTION? TYPE OF EXCEPTION
5028 No Error was detected and corrected by
client.
6791 No Sales invoice was voided.
6810 Yes Proof of shipment not presented.
7364 No Credit collection problem; should be
noted for review of allowance for
doubtful accounts.
7625 Yes Duplicate sales invoice not properly
filed.
8431 Yes Invoices not recorded by proper date;
represents potential cutoff problem.
8528 Yes Customer orders not included in
invoice package to verify compliance
with the order.
8566 Yes Error in pricing. No internal
verification.
8780 Yes Duplicate sales invoice not properly
filed.
9169 Yes Credit not authorized.
9974 Yes Internal verification of price extensions
and postings of sales invoices was
not included.

b. It is inappropriate to set a single acceptable tolerable exception

rate and estimated population exception rate for the combined
exceptions because each attribute has a different significance to
the auditor and should be considered separately in analyzing the
results of the test.

c. The SER assuming a 5% ARO for each attribute and a sample

size of 150 is as shown in the table on the following page:

15-19
Copyright © 2017 Pearson Education, Ltd.
15-33 (continued)

INVOICE NUMBER OF
NUMBER DESCRIPTION OF ATTRIBUTE EXCEPTIONS SER

6810 Shipping document not located 1 .67%

7625 Duplicate sales invoice/ 3 2.00%
8528 customer order not located
8780
8431 Invoice date improper 1 .67%
8566 Price extensions and postings 1 .67%
incorrect
8566 Price extensions and postings 2 1.33%
9974 not internally verified
9169 Credit not authorized 1 .67%

d.

ALLOWANCE FOR
INVOICE SAMPLING RISK
NUMBER TER-SER SUFFICIENT?

6810 5.33% Yes

7625
8780 4.0% Probably*
8528
8431 5.33% Yes
8566 5.33% Yes
8566
4.67% Probably*
9974
9169 5.33% Yes

* Students’ answers will most likely vary for this attribute.

e. For each exception, the auditor should check with the controller to
determine an explanation for the cause. In addition, the appropriate
analysis for each type of exception is as follows:

15-20
Copyright © 2017 Pearson Education, Ltd.
15-33 (continued)

INVOICE
NUMBER EXCEPTION ANALYSIS
6810 Confirm the account balances to the customers; examine the
reduction in the perpetual inventory records.
7625 Trace the amount to the sales journal and accounts receivable
master file; examine the shipping document and recompute
the sale amount.
8431 Determine who recorded the invoice and check several others
prepared by him or her to determine if the problem consistently
occurs.
8528 Examine the accounts receivable master file for subsequent
cash receipt; examine sales invoices for other invoices to the
same customer to determine if customer orders were attached.
8566 Check the price on other invoices to the same customer. Check
the price on other invoices that have the same product.
8780 See 7625
9169 Check credit history of customer and evaluate collectibility of the
customer’s account.
9974 Recheck actual price, extensions, and postings; determine who
the clerk was and check several other invoices for proper
indication of performance.

15-34 a.

ATTRIBUTE MISSING ELEMENT

1 CUER is 9.2%
2 Initial sample size is 77
3 Tolerable exception rate is 7%
4 The actual number of exceptions in the sample was 1
5 ARO is 5%
6 Sample size is 70

b. The sample results are unacceptable for Attributes 1 and 6 because

CUER exceeds TER for those attributes.

c. Attribute 1 has a lower ARO, which normally would lead to a large

sample size. TER-EPER is 6% for both attribute 1 and attribute 3.
The sample size for attribute 1 is smaller than the sample size for
attribute 3 because the estimated population exception rate is
0%

15-21
Copyright © 2017 Pearson Education, Ltd.
15-34 (continued)

for attribute 1 but 1% for attribute 3. Because there is no

expectation of errors in the population, the sample size is smaller
for attribute 1.

d. The CUER is smaller for attribute 2 relative to attribute 5 because

ARO is higher in attribute 2 relative to attribute 5, which means
the auditor has a greater willingness to accept the risk that the
assessment of control risk is too low for attribute 2.

■ Integrated Case Application

15-36 (see text Web site for Excel formatted sampling data sheet for part a.-
Filename P1537.xls)

a.
and
d. PINNACLE MANUFACTURING―PART VI

Client: Pinnacle Manufacturing

Audit Area: Tests of Controls and Substantive Test of Transactions

― Acquisitions.

Define the Objective(s): Examine vendors’ invoices, receiving

reports, purchase orders, and other related documents to
determine whether the system has functioned as intended and as
described in the audit program.

15-22
Copyright © 2017 Pearson Education, Ltd.
 15-36 (continued)

Define the population precisely (including stratification, if any):

Vouchers from 1/1/2016 to 10/31/2016. First voucher number –
6734. Last voucher number – 33722.

Define the sampling unit, organization of population items,

and random selection procedures: Voucher number, recorded
sequentially in the acquisitions journal; random number function in
electronic spreadsheet.

Planned Audit Actual Results

Calculated
Initial Number Sampling
Sample Sample of Error
Description of Attributes EPER TER ARO Size Size Exceptions SER (TER-SER)

1. Evidence of internal
verification of voucher
package including propriety
-.7% (note
of purchase, dates, unit
SER
costs, prices, extensions, 0 6% 10% 30 30 2 6.7%
exceeds
footings, account classification,
TER)
recording in journal, and
posting and summarization.
(6a, b)

2. Prices on vendors’ invoices

conform to approved price
0 5% 10% 40 40 0 0 5%
limits established by
management. (6c)

3. Price times quantity and

other calculations on the
1% 5% 10% 50 50 0 0 5%
vendor’s invoice are correct.
(6d)

4. Evidence of proper account

classification on vendors’ 2% 5% 10% 70 70 0 0 5%
invoices. (6e)

5. Dates on entries in purchases

journal agree with dates on 1% 5% 10% 50 50 2 4% 1%
receiving reports. (6f)

6. Evidence of internal
verification of each purchase 0 6% 10% 30 30 0 0% 6%
voucher. (6g)

Results: Based on the results of the tests, all controls appear effective except for evidence of internal
verification. Since there were also two errors on timing and an error in comparing the vendor’s
invoice amount to the acquisitions journal that is not included as an attribute, a larger than
normal sample in year-end testing of accounts payable is appropriate.

Notes: 1. The planned audit variables are judgmental. The results of the prior year from part III were used
to decide EPER. Initial sample size and CUER are judgments.
2. There was an error discovered where there was no attribute. This happens in practice, too. The
auditor should not ignore the exception even though it is an unplanned discovery.

15-23
Copyright © 2017 Pearson Education, Ltd.
 15-36 (continued)

b. Client: Pinnacle Manufacturing

Audit Area: Tests of Controls and Substantive Test of Transactions

― Cash Disbursements

Define the Objective(s): Examine cancelled checks and other

related documents to determine whether the system has functioned
as intended and as described in the audit program.

Define the population precisely (including stratification, if any):

Cancelled checks from 1/1/2016 to 10/31/2016. First check number
– 12376. Last check number – 37318.

Define the sampling unit, organization of population items,

and random selection procedures: Check number, recorded
sequentially in the cash disbursements journal; random number
function in electronic spreadsheet.

Planned Audit Actual Results

Calculated
Initial Number Sampling
Description Sample Sample of Error
of Attributes EPER TER ARO Size Size Exceptions SER (TER-SER)

1. Payee, name,
amount, and date
on cancelled
check agrees with
0 5% 10% 40
related purchases
journal and cash
disbursements
entry. (9a)

2. Evidence of
signature, proper
endorsement and 0 5% 10% 40
cancellation of
each check. (9b)

3. Date on cancelled
check agrees with
2 5% 10% 70
bank cancellation
date. (9c)

4. Cash discounts
0 5% 10% 40
are correct. (9d)

15-24
Copyright © 2017 Pearson Education, Ltd.
15-36 (continued)

c. Population = voucher numbers 6734 to 33722

Sample size = 50

Random Selection:
If random selection is performed using Excel, the command to
select numbers randomly from the population is:
=RANDBETWEEN(6734, 33722)

The command for selecting the random number can be entered

directly onto the spreadsheet, or can be selected from the function
menu (math & trig) functions under the “insert” menu. It may be
necessary to add the analysis tool pack to access the
RANDBETWEEN function. Once the formula is entered, it can be
copied down to select additional random numbers. The random
numbers will vary for each student, so they are not listed in this
solution.

 15-37 ACL Problem

a. There are 4,082 records included in the Invoices file as noted on

the bottom of the screen.

b. The sample size is 39 and the sampling interval is 104 (rounded

down from 104.66).

c. If the confidence level is increased to 95% and the other sampling

parameters are not changed, the sample size increases to 50. If
the upper limit is increased to 8% and the confidence level is 90%,
the sample size decreases to 29.

(Note: answers to part b. and c. are similar to answers obtained

using Table 15-8 on p. 548. Differences in sample size of one or
two items likely reflect rounding.)

15-25
Copyright © 2017 Pearson Education, Ltd.
15-37 (continued)

d. The sample is included below (not all columns shown):

Customer Invoice
Invoice Number Sales Order Number Number Invoice Date Amount
173640036997 000008176082 0252432 2/10/2014 29144.36
173640037166 000008176251 0258424 3/9/2014 11167.69
173640037196 000008176281 0261014 3/12/2014 3723.44
173640037226 000008176311 0260835 3/22/2014 21582.56
173640037472 000008176558 0263402 4/15/2014 5009.58
173640037506 000008176592 0255268 4/16/2014 11707.48
173640037545 000008176631 0255998 4/28/2014 19450.53
173640037782 000008176869 0257490 5/11/2014 23638.49
173640037800 000008176887 0257884 5/11/2014 4639.11
173640037808 000008176895 0235520 5/17/2014 1595.31
173640037891 000008176978 0260600 5/20/2014 945.32
173640037931 000008177018 0245068 5/21/2014 27755.91
173640037940 000008177027 0262176 5/26/2014 19244.99
173640038114 000008177201 0257847 6/1/2014 7401.07
173640038204 000008177291 0240643 6/16/2014 15132.12
173640038292 000008177379 0249233 6/16/2014 15728.09
173640038449 000008177536 0244063 6/29/2014 22261.59
173640038458 000008177545 0241913 6/24/2014 6188.54
173640038611 000008177698 0239939 6/29/2014 4170.89
173640038717 000008177804 0256692 7/13/2014 15043.24
173640038905 000008177992 0240304 7/8/2014 9413.12
173640039091 000008178178 0250396 7/22/2014 3210.42
173640039277 000008178364 0239614 7/30/2014 2745.13
173640039343 000008178430 0244488 8/10/2014 20998.44
173640039430 000008178517 0234834 8/20/2014 15128.39
173640039457 000008178544 0251543 8/20/2014 3238.49
173640039487 000008178574 0252534 8/10/2014 768.13
173640039502 000008178589 0244430 8/16/2014 7268.08
173640039589 000008178677 0253965 8/27/2014 1276.47
173640039617 000008178705 0260685 8/27/2014 17376.00
173640039969 000008179057 0244080 9/23/2014 2010.65
173640040032 000008179121 0242752 9/20/2014 13466.11
173640040150 000008179239 0260166 9/30/2014 15636.04
173640040214 000008179303 0248413 9/30/2014 1137.00
173640040565 000008179655 0256038 10/21/2014 1538.06
173640040583 000008179673 0247128 10/29/2014 16293.16
173640040585 000008179675 0263951 10/29/2014 19807.72
173640040697 000008179787 0237042 11/16/2014 2967.13
173640040739 000008179829 0250143 11/25/2014 5182.95

15-26
Copyright © 2017 Pearson Education, Ltd.
15-37 (continued)

Students should generate the same solution if they use the assigned
seed number. Without the seed number, student’s solutions will
be unique. The formatting of students’ solutions may also appear
differently if printed in ACL. The solution above was exported to
Excel, and not all columns of the dataset are included above.

e. The largest invoice amount included in the sample above is

$29,144.36 (invoice number 173640036997). Since record
sampling was used, the invoice amount is not considered.

f. Because the sample size was based on zero planned exceptions,

the presence of one exception should make the results unacceptable.
Using a confidence of 90%, sample size of 39 and 1 error, ACL
calculates an upper error limit of 9.97%, which exceeds the tolerable
limit of 6%.

15-27
Copyright © 2017 Pearson Education, Ltd.