AUD TH HO MODULE A SAMPLING

DR. JACKQUI MORENO

AUDIT SAMPLING RN:2020-09-15-3-04

AUDIT SAMPLING —The application of audit procedures to less than 100% of items within an account balance or
class of transactions such that all sampling units have a chance of selection. This will enable the auditor to obtain and
evaluate audit evidence about some characteristic of the items selected in order to form or assist in forming a
conclusion concerning the population from which the sample is drawn. Audit sampling can use either a statistical or a
non-statistical approach.1

Terminology related to Audit Sampling2

Anomalous error—An error that arises from an isolated event that has not recurred other than on specifically
identifiable occasions and is therefore not representative of errors in the population.

Confidence levels or Reliability Levels—The mathematical complements of sampling risk; chance that the auditor is
willing to take that the sample estimate may be wrong.

Precision – The accuracy the auditor wants the sample estimate to be; the estimated difference between the actual
amount & the reported amount based on the sample; When this exceeds the tolerable error, then the discovered errors
are concluded as material.

Expected error—The error that the auditor expects to be present in the population.

Expected deviation rate -- The number of deviations the auditor expects to find in the population before testing begins.
The auditor can develop this expectation based on the prior year's results or by examining few items in the population
(pilot sample).

Non-sampling risk—Arises from factors that cause the auditor to reach an erroneous conclusion for any reason not
related to the size of the sample. For example, most audit evidence is persuasive rather than conclusive, the auditor
might use inappropriate procedures, or the auditor might misinterpret evidence and fail to recognize an error.

Non-statistical sampling—Any sampling approach that does not have the characteristics of statistical sampling.

Population—The entire set of data from which a sample is selected and about which the auditor wishes to draw
conclusions. A population may be divided into strata, or sub-populations, with each stratum being examined separately.
The term population is used to include the term stratum.

Sampling risk—Arises from the possibility that the auditor’s conclusion, based on a sample may be different from the
conclusion reached if the entire population were subjected to the same audit procedure.

Sampling unit—The individual items constituting a population, for example checks listed on deposit slips, credit entries
on bank statements, sales invoices or debtors’ balances, or a monetary unit.

Statistical sampling—Any approach to sampling that has the following characteristics:

(a) Random selection of a sample; and
(b) Use of probability theory to evaluate sample results, including measurement of sampling risk.

Stratification—The process of dividing a population into subpopulations, each of which is a group of sampling units
which have similar characteristics (often monetary value).

Tolerable error—The maximum error in a population that the auditor is willing to accept.

1
Revised Glossary of Terms
2
Revised Glossary of Terms
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

Tolerable deviation rate-- The maximum rate of deviations the auditor is willing to accept, without modifying the
planned degree of reliance on the internal control.

Total error—Either the rate of deviation or total misstatement. When added to anomalous error results to the
best estimate of population error.

WHEN DO YOU SAMPLE?3

Note that sampling is resorted to ONLY when the auditor believes that the sample may be a good representative of the
whole, in that, the findings on the sample is to be concluded as the same if the total population was tested. Hence, we
can say that sampling is not required to be part of any audit procedure.

Not all testing procedures performed by auditors involve audit sampling. For example, the auditor may decide that it
would be more appropriate to examine the entire population of items that make up an account balance because the
population constitutes a small number of large value items. Likewise, the auditor may decide to apply audit procedures
only to those items which have particular significance (e.g. all items over a certain amount).

What are the Risks involved in Sampling?

GENERALLY, any auditor is exposed to the “likelihood of drawing incorrect conclusions about the account balance or
class of transactions because of (1) human errors such as application of inappropriate audit procedures, (2) failure to
recognize errors in the sample tested, or (3) misinterpretation of evidence obtained”. This so-called likelihood is known
as Non-sampling risk.

However, when the auditor chooses to elect sampling, wherein 100% testing is not employed, he is further exposed to
the uncertainty of not detecting material errors sitting on items excluded from the sample chosen to be examined. Such
risk is referred to as Sampling risk. Events that give rise to this risk will have to include: (1) The improper selection of
sample units and (2) Inappropriate determination of sample size.

There are two types of sampling risk that could adversely affect the audit. The alpha risk and the beta risk.
 Alpha Risk is the risk the auditor will conclude, in the case of tests of control, that control risk is higher than it
actually is (i.e. Risk of Under reliance), or in the case of substantive test, that material error exist when in fact
it does not (i.e. Risk of Incorrect Rejection). Alpha risk results to auditor performing audit procedures more
than necessary. Hence, alpha risk affects the audit efficiency.
 Beta Risk is the risk the auditor will conclude, in-the case of tests of control that control risk is lower than it
actually is (i.e. Risk of Over reliance), or in the case of substantive test, that material error does not exist when
in fact it does(i.e. Risk of Incorrect Acceptance). Beta risk results in materially misstated financial statements
from not expanding audit tests to a necessary level. This type of risk is of greater concern to the auditor
because this risk affects the audit effectiveness.
How can these Risks be controlled?
Sampling Risks may be further reduced4 by:
It is important to realize that both
 Increasing the sample size; and
 Using an appropriate sample selection method.
statistical and non-statistical sampling
methods
Non-sampling risk cannot be eliminated but can be minimized by…  Are acceptable;
 Proper planning  Require the use of auditor's judgment
 Adequate direction, review, and supervision of the audit team. in designing and selecting the
sample, performing audit procedures
Does Sampling always involve Mathematical Formula? and evaluating the results; and
NO. Actually, sampling may be any of these two approaches:  Cannot assure that the sample will be
1. Statistical representative of the population.
…When sample is selected:

3
When designing audit procedures, the auditor should determine appropriate means for selecting items for
testing so as to gather audit evidence to meet the objectives of audit tests.(PSA 530, par 2)
4
Sampling risks is eliminated when 100% examination is elected.
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

1. Randomly
2. With the use of probability theory, you:
i. Design an efficient sample
ii. Measure sufficiency of sample
iii. Evaluate sample results
iv. Measures sampling risk quantitatively

2. Non-Statistical (or Judgmental)

…When sample is selected:
a) Not statistically 
b) Based on relative costs & benefits
c) The auditor being guided by:
i. Experience
ii. Prior knowledge
iii. Current information
d) Without the need of quantitative measures of sampling risk

Is JUDGMENT only USED when using NON-Statistical Approach?

NO. Both methods cannot do without the auditor’s professional judgment, well, at least in the evaluation and
conclusion parts of the testing.

When is it PREFERRED to employ STATISTICAL APPROACH TO SAMPLING?

 Items to be examined involve a very large population
 There is an audit software available for selection
 A single sample is holding significant (or ‘material’) part of the total evidence

In planning what Audit Procedures can we make use of Sampling?

Audit sampling may be used when performing tests of controls or substantive tests. When statistical sampling is used,
the auditor may use either attribute or variable sampling plan.

Attribute sampling is a sampling plan used to estimate the occurrence rate. This is generally used when performing tests
of controls to estimate the rate of deviations from prescribed internal control policies or procedures.

Variable sampling, on the other hand, is a sampling plan used to estimate a numerical measurement of a population
such as peso value. This is useful when performing substantive tests to estimate the amount of misstatements in the
financial statements

What are Basic Steps in Audit Sampling?

Audit procedures carried out by means of sampling techniques require consideration of at least the following basic
steps.
1. Define the objective of the test (whether to test for “deviation” or for “misstatements”)
2. Determine the appropriate population
3. Determine the audit procedure to be performed.
4. Determine the sample size.*
5. Select the sample.*
6. Audit the sample.
7. Evaluate the results.

SAMPLING FOR TESTS OF CONTROLS

Audit sampling for tests of control is generally appropriate when application of the control leaves evidence of
performance. For those controls that leave no documentary evidence of performance, non-sampling procedures such as
inquiries and observation would be more appropriate.
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

How large should the Attribute sample size be?

There are three major factors affecting the determination of sample size for tests of controls. These are the

 Acceptable sampling risk;

 Tolerable deviation rate; and
 Expected deviation rate

Rules of Thumb:

Relationship Application of relationship

(Direct/Inverse)
Tolerable rate of The lower the tolerable rate, the LARGER the
deviation sample
Allowable risk of The higher the allowable risk of over-reliance, the
Over-reliance SMALLER the sample
Expected population The higher the expected population deviation rate
deviation rate the LARGER the sample

How will the Attribute Sample units be selected?

PSA 530 has identified three principal methods of selecting samples namely, random number selection, systematic
selection, and haphazard selection.

Random number selection

Under this method, the auditor selects the sample by matching random numbers, generated by random number table or
computer software generator, with the population numbering system such as document number. An advantage of this
selection technique is that it gives each item in the population an equal opportunity to be selected.

Systematic selection
Another method of selecting sample that is easier to use compared to random number selection is the
systematic selection. This method involves determining a constant sampling interval and then selects the
sample based on the size of that interval. 5

Systematic selection is useful when identification numbers lacking. It is appropriate for both statistical & non-statistical
sampling.

Haphazard selection
When using this method, the sample is selected without following an organized or structured technique. Haphazard
selection is useful for non-statistical sampling, but it is not used for statistical sampling because the auditor can not
measure the probability of an item being selected when using this method.

Block sampling
This type of selection provides the least assurance for an auditor. It is composed of items so arranged to each other
within a large group of sampling unit. It yields usually the smallest number of sample size and hence exposes the
auditor to high sampling risk. Block selection involves selecting a block(s) of contiguous items from within the
population.
Important notes about block sampling are: In selecting the sample and
□ Block sampling is best used when a group of items arranged applying the appropriate audit
contiguously within a larger grouping procedures, the auditor may
□ Results from block sampling are inefficient & not generalizable encounter the following
□ Block sampling should not be used for statistical or non-statistical situations.
sampling without care in controlling sampling risk.

5
The auditor should determine that the population is not structured in such manner that the sampling interval
corresponds with a particular pattern in the population.
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

 Voided documents. The auditor may occasionally select a voided or cancelled document in a sample. If the
document has been properly voided, such document should be replaced by another sample item.
 Missing documents. If the auditor encounters missing document and he is unable to determine whether the
control has been properly performed, such item should be treated as a deviation for the purpose of evaluating
sample results.
How to evaluate results from Attribute Sampling?
When evaluating sample results, both the qualitative and the quantitative factors of deviations should be considered.
Here are some general guidelines that may be used when evaluating sample results for tests of controls.

1. Determine the Sample Deviation Rate (SDR).

The sample deviation rate is computed by dividing the number of deviations found in the sample by the sample
size. For example, if the auditor found 4 deviations out of the 200 sample items, the sample deviation rate is 2% (4
/ 200). This sample deviation rate represents the auditor's best estimate of the deviation rate in the population.
2. Compare the Sample Deviation Rate (SDR) with the Tolerable Deviation Rate (TDR) and draw an overall
conclusion about the population.

This comparison may result to the following situations:

 The sample deviation rate exceeds the tolerable deviation rate.(DR>TDR)
Conclusion/s:
The sample results do not support the auditor's planned degree of reliance on internal control.
Final Action 1: CR is set at maximum
Final Action 2: ST are increased
 The sample deviation rate is less than the tolerable deviation rate. (DR<TDR)
Action: Consider the Maximum Population Deviation (MPD) rate or Upper Precision Limit6
The auditor should determine how close the sample deviation rate is to the tolerable deviation rate. As the
sample deviation rate approaches the tolerable deviation rate the allowance for sampling risk decreases, or the
precision narrows.
Further assume two cases for illustration:
Case 1:
SDR= 2% ; Acceptable Risk for Over reliance (or Sampling Risk) = 5%
Thus, MPD = 7.00%
Case 2:
SDR= 2%; Acceptable Risk for Over reliance (or Sampling
Risk) = 10% GENERALLY, as the “Acceptable”
Thus, MPD = 12% Sampling risk increases, the
confidence in the procedures
Conclusion/s : decreases, thus requiring a smaller
HOWEVER, If the TDR is equal to 10%, sample size. This is true since it
…AND in Case 1, where 7% UPL< 10% TDR produces a higher “maximum
Final Action 1: CR is set is retained population deviation”.
Final Action 2: ST are reasonably performed as
planned (i.e. effective)

…AND in Case 2, where 12% UPL= 10% TDR

Final Action 1: CR is set at maximum
Final Action 2: ST are increased, sample size is increased

6
(Sample deviation rate + Allowable risk of Over-reliance) ; sum of SDR and Acceptable Sampling Risk
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

WHAT APPROACHES can you choose from to perform ATTRIBUTE SAMPLING?

Sequential Sampling (A.K.A. Stop or Go)
Best Employed when:
1. auditor expects very FEW DEVIATIONS within the population
Basic Description:
1. does not use fixed sample size
2. separates the sampling process into several stages
3. determine efficient sample size after each test
Basic Mechanics:
1. After a step, the auditor determines if it is warranted to accept or increase the preliminary level of control
risk that is if he believes he had acquired sufficient data.

Discovery Sampling
Best Employed when:
1. NO DEVIATIONS are expected in the population
2. Normally used when the auditor suspects that an irregularity might have been committed.
Basic Description:
1. assures the auditor will note at least one exception
Basic Mechanics:
1. Any 1 noted error results in rejection of the control as to be relied upon.

SAMPLING FOR SUBSTANTIVE TESTS

Substantive tests are concerned with the amounts reported in the financial statements. Audit sampling is appropriate
when performing tests of details to estimate the amount of misstatements in the financial statements.

How large does the Variable Sample size have to be?

AT – determining
When Audit Sampling (AT-0510.Lec)
sample size for substantive tests, the following factors must be considered:
 Acceptable sampling risk 7

 Tolerable misstatement
 Expected misstatement
 Variation in the population8

Rules of Thumb:
Factors Relationship Sample Size
Tolerable misstatement Inverse The lower the tolerable misstatement, the LARGER the
sample
Allowable risk of incorrect The higher the allowable risk or incorrect acceptance,
acceptance Inverse the LOWER the expected reliability, and the SMALLER
the sample
Expected amount of misstatement Direct The higher the expected amount of misstatement, THE
LARGER the sample
Assessed level of control risk Direct The high the assessed level of control risk, the LARGER
the sample
Population size Direct The larger the population, the LARGER the sample
An increase in the auditor’s Direct The higher the inherent risk, the LARGER the sample
assessment of inherent risk*

7
For practical purposes, the auditor uses the acceptable level of detection risk as the acceptable sampling after
giving adequate consideration to the risk that analytical procedures may fail to detect material misstatement in
account balance.
8
When using statistical sampling, this variability is measured by the standard deviation.
When a population consists of highly variable recorded amounts, it is difficult to select a representative
sample. Consequently, a larger sample size is required as the degree of variability within the population
increases. The auditor can estimate the variation based on the prior year's tests results or a pilot sample.
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

Factors Relationship Sample Size

An increase in the use of other The more alternative procedures there are, the SMALLER
substantive procedures directed at Inverse the sample is necessary for each test
the same financial statement
assertion*
Stratification of the population when Inverse
appropriate *

How will the Variable Sample units be selected?9

Stratified sampling
Best Employed when the auditor wants to:
1. Improve audit efficiency, by reducing sample size without increasing sampling risk and
2. Reduce the variability of items within each stratum

Basic Description and Mechanics:

Stratification is the process of dividing a population into subpopulations, each of which is a group of sampling
units which have similar characteristics (often monetary value) or which have an identifying characteristic.

Value weighted selection

Best Employed when:
1. It will often be efficient in substantive testing, particularly when testing for overstatements, to identify the
sampling unit as the individual monetary units (e.g., pesos) that make up an account balance or class of
transactions.
2. This approach is ordinarily used in conjunction with the systematic method of sample selection and is most
efficient when selecting from a computerized database

Basic Description:
This method sets the high-value items as priority to be included in the sample that can result to
smaller sample sizes.

Basic Mechanics:
1. Each peso is treated as one sampling unit Disadvantage of PPS utility:
2. This method of sample selection is similar to stratified (a) Understated items have a lower
sampling in that large monetary values are given greater
probability of being selected
representation in the sample
(b) Items with zero or negative
Probability-proportional to Size (PPS) balance are not generally included in
Best Employed when: the sample
1. Used especially for testing overstatement especially for asset (c) A high frequency of
and revenue accounts misstatements, results in an increase
2. Few or no errors are expected in sample size

Basic Description:
1. Automatically stratifies the population to the larger peso amounts and lower peso items
Basic Mechanics:
1. Items with larger peso amounts have a greater probability of being selected
2. An item that is individually material will automatically be selected
3. Sample size may be reduced as the same item may be selected more than once
4. The sample distribution does not have to be close to the distribution in the population for the sample to be
valid
5. Sampling can be initiated prior to year end more easily

9
The other methods discussed in Attribute sampling can also be employed for Variable sampling.
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

Variable Sampling- Mean Per Unit Estimation

It enables the auditors to estimate the average peso value of items in the sample that will be extended to the population.

How to Evaluate the Variable Sampling results?

Evaluating sample results for substantive tests will involve the following steps:
1. Project the misstatements to the population using:
 Ratio estimation; or
 Difference estimation
The only difference between the two methods is that ratio estimation uses the book values of the
population size and sample size to project the misstatement, while difference estimation uses the number of
customers to project the misstatements to the population. Hence, the use of ratio estimation is appropriate
when the amount of misstatements found is approximately proportional to the client's book amount.

2. Compare the projected misstatements together with the tolerable misstatements and draw an overall
conclusion.
If the projected misstatement is greater than the tolerable misstatement, the auditor will conclude that the
account balance is materially misstated. In this case, the auditor may:
 Examine additional units,
 Perform suitable alternative procedures, or
 Request the client to adjust the account balance.

However, if the projected misstatement is less than the tolerable misstatement, the auditor should consider the
allowance for sampling risk presented in pages 5 to 6.

Summary of Concepts

SAMPLING RISKS : ATTRIBUTE SAMPLING

When by When in fact if the
The auditor is exposed examining the POPULATION …Thus,
to the following risks SAMPLE …. was considered… Making the Affecting
such as… The Auditor What is TRUE Auditor… the auditor’s
BELIEVES the about the
INTERNAL INTERNAL
CONTROL is… CONTROL is that
it is…
Alpha Risk
Risk of Under
reliance Weak or lacking Functioning Perform Efficiency
(i.e. when Error rate is satisfactorily MORE
lower in population Substantive
than in sample tested) tests than
necessary…
Risk of assessing
control risk too high
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

SAMPLING RISKS :ATTRIBUTE SAMPLING

When by When in fact if the
The auditor is examining the POPULATION …Thus,
exposed to the SAMPLE …. was considered… Making the Affecting
following risks such The Auditor What is TRUE Auditor… the auditor’s
as… BELIEVES the about the
INTERNAL INTERNAL
CONTROL CONTROL is that
is… it is…
Beta Risk
Risk of Over
reliance Strong Perform LESS
(i.e. when Error and Not to be relied effective Effectiveness
rate is higher in Operating upon procedures
population than in
sample tested)
Risk of assessing
control risk too
low

SAMPLING RISKS :VARIABLE SAMPLING

When by When in fact if the
The auditor is examining the POPULATION was …Thus,
exposed to the SAMPLE …. considered… Making the Affecting
following risks The Auditor What is TRUE about Auditor… the auditor’s
such as… BELIEVES the the ACCOUNT is
ACCOUNT that it is…
is…
Alpha Risk
Risk of Incorrect
Rejection Fairly presented Materially misstated Perform Efficiency
(i.e. when Error LESS
rate is lower in Substantive
population than in tests than
sample tested) necessary…
SAMPLING RISKS :VARIABLE SAMPLING (continuation)
When by When in fact if the
The auditor is examining the POPULATION was …Thus,
exposed to the SAMPLE …. considered… Making the Affecting
following risks The Auditor What is TRUE about Auditor… the auditor’s
such as… BELIEVES the the ACCOUNT is
ACCOUNT that it is…
is…
Beta Risk
Risk of Incorrect
Acceptance Materially Fairly presented Perform
(i.e. when Error misstated MORE Effectiveness
rate is higher in Substantive
population than in tests than
sample tested) necessary…
AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

Sampling Used in Measures… From sample whose size is affected by the following
Plan Performing… factors:
Attribute Test of Control Deviation from (1) TOLERABLE DEVIATION RATE (which is determined
sampling significant based on the assessed level of control risk),
internal control (2) desired RISK OF ASSESSING CONTROL RISK TOO
procedures the LOW, and
auditor is (3) EXPECTED POPULATION DEVIATION RATE (which is
intending to rely the basis of assessed level of control risk).
on (4) SAMPLE DEVIATION RATE (i.e. the number of errors
divided over the number of items in sample)
(5) UPPER PRECISION LIMIT (i.e. the maximum
population error rate ; sample deviation rate +
acceptable risk of over-reliance or “sampling
risk”)10

Samplin Used in From sample whose size is affected by the

g Plan Performing… Measures… following factors:
Variable Substantive Misstatement or (1) TOLERABLE MISSTATEMENT (i.e. maximum
sampling Testing Discrepancy of the difference between the should be and
adjusted and reported by management balance that does
presented account not render the account materially misstated ),
(2) ALLOWABLE RISK OF INCORRECT ACCEPTANCE
Risk incorrect
acceptance = AR÷(CR x DR)
where DR may mean risk that analytical
procedures may fail to detect misstatements.
(3) EXPECTED AMOUNT OF MISSTATEMENT
(4) Assessed level of CONTROL RISK *
(5) POPULATION SIZE
(6) An increase in the auditor’s assessment of
INHERENT RISK*
(7) An increase in the use of other substantive
procedures directed at the same financial
statement assertion*
(8) STRATIFICATION of the population when
appropriate*

GENERAL CONSIDERATIONS ON SAMPLE SIZES

Conditions leading to
Smaller Sample Larger Sample Size
Size
Tolerable error Larger Smaller
(indirect effect)
Allowances for risks Larger Smaller
(indirect effect)
Expected error Lower Higher

10
if upper precision limit is lesser than or equal to tolerable error, control risk preliminarily assessed is maintained for
planning substantive tests (TEL>PL,CR is sustained, ST should be effective) ; if not, increased control risk is adopted
(TEL<PL,CR is increased, ST should be more effective)
 AUD TH HO MODULE A SAMPLING
DR. JACKQUI MORENO

(direct effect)
Degree of assurance Lower Higher
required
(same as expected error)
Number of items in Little effect on sample size unless there are fewer than 250 items in
population the population, in which case the sample size is likely to be smaller,
all other factors being equal

AT – Audit Sampling (AT-0510.Lec)