Audit Sampling

Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 10

AUDITING THEORY

Red Sirug
AUDIT SAMPLING
(And Other Means of Testing)

Means for Selecting Items for Testing to Obtain Audit Evidence:


When designing tests of controls and tests of details, the auditor shall determine the means of
selecting items for testing that are effective in meeting the purpose of the audit procedure.
The means available to the auditor for selecting items for testing are (the auditor may use any
one or combination of these):
a. Selecting all items (100% examination);
b. Selecting specific items; and
c. Audit sampling
Selecting All Items (100% Examination):
Selecting all items involves examining the entire population of items that make up a class
of transactions or account balance
100% examination is:
More common for tests of details
Unlikely for tests of controls
Appropriate when:
The population constitutes a small number of large value items
There is significant risk (high RMM) and other means do not provide sufficient
appropriate audit evidence, or
Cost effective the repetitive nature of a calculation or other process performed
automatically by an information system makes a 100% examination cost
effective (for example, using computer-assisted audit techniques (CAATs)
Selecting Specific Items:
Selecting specific items judgmental selection of specific items from a population based on
the following factors:
Auditors understanding of the entity
Assessed risk of material misstatement
Characteristics of the population being tested
Specific items that may be selected by the auditor include:
High value or key items. The auditor may examine items of high value or items that
exhibit some other characteristic (for example, items that are suspicious, unusual,
particularly risk-prone or that have a history of error).
All items over a certain amount. The auditor may decide to examine items whose
values exceed a certain amount so as to verify a large proportion of the total amount of
a class of transactions or an account balance.
Items to obtain information. The auditor may examine items to obtain information about
matters such as the nature of the entity or the nature of transactions.
Items to test control activities. The auditor may use judgment to select and examine
specific items to determine whether or not a particular control activity is being
performed.

Selecting specific items for examination does not constitute audit sampling and
therefore, not subject to sampling risk.
The results of audit procedures applied to selected specific items cannot be projected to
the entire population; accordingly, selective examination of specific items does not
provide audit evidence concerning the remainder of the population.

Audit Sampling:
Audit sampling (sampling) the application of audit procedures to less than 100% of the
items within a population of audit relevance (account balance or class of transactions) such that all
sampling units have a chance of selection in order to provide the auditor with a reasonable basis
on which to draw conclusions about the entire population
Audit sampling is the means that enable the auditor to draw conclusions about the
population on the basis of testing a sample drawn from it.
Sampling is essential throughout audits as auditors attempt to gather sufficient
appropriate evidence in a cost efficient manner.
Audit sampling is not required part of any audit procedure because when designing
audit procedures, the auditor should determine appropriate means of selecting items
for testing.
Audit sampling is used for both tests of controls (attributes sampling) and for tests of
AT Audit Sampling

Red Sirug

Page 1

details of transactions and balances (variables sampling). In both attributes sampling


and variables sampling, the plans may be either non-statistical or statistical.
Audit Sampling Terms:
Population the entire set of data from which a sample is selected and about which the
auditor wishes to draw conclusions
Sample the portion of the population that will be subjected to audit testing
Sampling unit the individual items constituting a population; also known as population
item, observation, or elementary unit
Representative sample a sample in which the characteristics of the sample are the
same as those of the population
Sampling risk The risk that the auditor's conclusion based on a sample may be different
from the conclusion if the entire population were subjected to the same audit procedure.
Anomaly a misstatement or deviation that is demonstrably not representative of
misstatements or deviations in a population
Confidence level the mathematical complements of sampling risks; also known as
reliability or confidence
Stratification the process of dividing a population into sub-populations, each of which is
a group of sampling units which have similar characteristics (often monetary value)
Tolerable misstatement (in substantive procedures) a monetary amount set by the
auditor in respect of which the auditor seeks to obtain an appropriate level of assurance
that the monetary amount set by the auditor is not exceeded by the actual misstatement in
the population; it is the maximum total error in a population that the auditor is willing to
accept
Tolerable rate of deviation (in tests of controls) a rate of deviation from prescribed
internal control procedures set by the auditor in respect of which the auditor seeks to
obtain an appropriate level of assurance that the rate of deviation set by the auditor is not
exceeded by the actual rate of deviation in the population; it is the maximum rate of
deviation from the prescribed control procedure the auditor is willing to accept without
changing control risk assessment or planned reliance on internal control
Error either control deviations, when performing tests of control, or misstatements, when
performing substantive procedures.
Total error either the rate of deviation (in case of tests of control) or total misstatement
(in case of substantive procedures)
Anomalous error means 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
Expected error
a. Expected error amount in substantive tests, it is the auditor's best estimate of the
amount of error the auditor expects to find in the population
b. Expected deviation rate in tests of control, it is the auditor's best estimate of the
rate of deviation from a prescribed control procedure in the population
Applicability of Audit Sampling:
Where an auditor has no special knowledge about likely misstatements contained in account
balances and transactions
When the auditor believes that the sample is to be a good representative of the population
Inapplicability of audit sampling: Situations where audit sampling generally do not
apply:
a. Risk assessment procedures performed to obtain an understanding of internal
control.
b. Tests of automated application controls when effective general controls are present.
(Generally, such controls would only be tested once or a few times.)
c. Analyses of security and access controls, or other controls that do not provide
documentary evidence of performance (e.g., controls related to segregation of
duties).
d. Some tests related to the operation of the control environment or the accounting
system (e.g., examination of the effectiveness of activities performed by those
charged with governance).
Lists procedures that do not involve sampling:
a. Inquiry and observation
b. Analytical procedures
c. Procedures applied to every item in a population
d. Tests of controls where application is not documented
e. Procedures from which the auditor does not intend to extend a conclusion to the
remaining item in the account (for example, tracing several transactions through
AT Audit Sampling

Red Sirug

Page 2

f.

accounting system to obtain understanding)


Untested balances

General Approaches to Audit Sampling:


1. Statistical sampling an 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

Statistical sampling applies the law of probability theory to aid the auditor in designing a
sampling plan and evaluating sample results.
In statistical sampling, auditors specify the sampling risk they are willing to accept and
then calculate the sample size that provides that degree of reliability. Results are
evaluated quantitatively.
Statistical sampling measures quantitatively the sampling risk (the risk from testing only
part of an audit population).
Advantages of statistical sampling: Conclusions may be drawn in more precise
ways when using statistical sampling because it enables the auditor to:
a. Measure the sufficiency of the audit evidence obtained.
b. Provide an objective basis for quantitatively evaluating sample results more
objective audit evidence
c. Design an efficient sample.
d. Quantify/measure sampling risk so as to limit it to an acceptable level.
e. Measure reliability (confidence level), precision, and sampling error (sampling risk).
Disadvantages of statistical sampling:
Danger of accepting statistical evidence at face value without sufficient skepticism
Its cost could exceed the benefits
Inappropriate it some cases (for example, test of controls that depend on segregation
fo duties or otherwise provide no audit trail of documentary evidence)
In statistical sampling, random sample selection methods should be used to give all
items in the population an equal chance to be included in the sample to be audited.

2. Non-statistical sampling a sampling approach that does not have both characteristics of
statistical sampling
Non-statistical sampling (or judgment sampling) is based solely on the auditors judgment.
The sample size is not determined mathematically. Auditors rely exclusively on subjective
judgment to determine sample size and to evaluate sample results.
A properly designed non-statistical sampling application can be as effective as statistical
sampling application.
One disadvantage is that it can misdirect an auditor to unreliable sampling units.
Additional notes on statistical and non-statistical sampling:
Statistical sampling is a mathematical approach to inference, whereas non-statistical
sampling is a more subjective approach.
Conclusions may be drawn in more precise ways when using statistical sampling
methods.
Both sampling approaches involve judgment in planning, executing the sampling plan,
and evaluating the results of the sample.
It is acceptable for auditors to use either or combination of statistical and non-statistical
sampling. The choice is based primarily on the auditors assessment of the relative
costs and benefits. Such choice is independent of the selection of audit procedures
because audit sampling is merely a means for accomplishing audit procedures.
Both sampling approaches can provide sufficient appropriate evidence.
Sampling methods are used by auditors in both control testing and substantive testing.
Auditors professional judgment:
Although statistical sampling aids the auditor in quantitative ways, it is not a substitute for
professional judgment. In other words, statistical sampling does not eliminate the need for the
auditors professional judgment.
The auditor must exercise professional judgment in both statistical and non-statistical sampling
to:
a. Define the population and the sampling unit;
b. Select the appropriate sampling method;
c. Evaluate the appropriateness of audit evidence;
d. Evaluate the nature of deviations or errors;
e. Consider sampling risk; and
f. Evaluate the results obtained from the sample and project those results to the population.
AT Audit Sampling

Red Sirug

Page 3

Types of Audit Sampling Plan:


Audit sampling is used for both tests of controls (attributes sampling) and for tests of details of
transactions and balances (usually, variables sampling). In both attributes sampling and variables
sampling, the plans may be either non-statistical or statistical.
1. Attribute sampling estimates the attribute or quality characteristic of a population
Applicable to tests of controls because attribute sampling deals with estimating
deviation rate (also called rate of occurrence) from prescribed control procedures that
the auditor plans to rely upon
2. Variables sampling estimates the numerical quantity of a population
Applicable to substantive testing because variables sampling deals with peso or
monetary amount of misstatement in account balances

Ordinarily, risk assessment procedures to obtaining understanding of the entity and its
environment, including internal control, do not involve the use of audit sampling.
Audit sampling for substantive procedures applies to tests of details only.

Sampling Risk:
Sampling risk the possibility that the auditors conclusion, based on a sample may be
different from the conclusion reached if the entire population were subjected to the same audit
procedure.
The risk that the sample is not representative of the population and that the auditor's
conclusion will be different from the conclusion had the auditor examined 100% of the
population.
The possibility that even though a sample is properly chosen, it may not be
representative of the population.
Sampling risk can be reduced by increasing the sample size.
Sampling risk is an inherent part of sampling that results from testing less than the
entire population.
Two Types of Sampling Risk:
1. Risk that affects audit effectiveness and may lead to an inappropriate audit
opinion (Beta risk or Type II error) the risk the auditor will conclude that:
a. Risk of assessing control risk too low in case of a test of control, that the assessed
control risk is lower than it actually is
b. Risk of incorrect acceptance in case of a substantive test, conclusion that a material
error does not exist when in fact it does
2. Risk that affects audit efficiency as it would usually lead to additional work to
establish that initial conclusions were incorrect (Alpha risk or Type I error)
the risk the auditor will conclude that:
a. Risk of assessing control risk too high in case of a test of control, that the assessed
control risk is higher than it actually is
b. Risk of incorrect rejection in case of a substantive test, conclusion that a material error
exists when in fact it does not
Aspects of Audit Risk(Aspect of detection risk):
Audit risk is a combination of the risk that a material misstatement will occur (inherent risk
and control risk) and the risk that it will not be detected by the auditor (detection risk).
1. Sampling risk the risk or the possibility that, when a test of controls or a substantive test is
restricted to a sample, the auditor's conclusions base on a sample may be different from the
conclusions which would have been reached had the tests been applied to all items in the
population
Sampling risk is the aspect of audit risk and of detection risk that is due to sampling.
Aspects of sampling risk:
a. Substantive testing sampling risks:
1) Risk of incorrect acceptance the risk that the auditor will conclude that a material
error in an account balance (based on the sample) does not exist when in fact it does
(i.e., sample results fail to identify an existing material misstatement).
2) Risk of incorrect rejection the risk that the auditor will conclude that a material
error in an account balance exists when in fact it does not (i.e., sample results
mistakenly indicate a material misstatement).
b. Tests of controls sampling risks:
a. Risk of assessing control risk too high or the risk of under reliance (Alpha risk
or Type I error) the risk that the assessed level of control risk (based on the sample)
is greater than the true level of control risk (i.e., sample results indicate a greater
deviation rate than actually exists in the population).

AT Audit Sampling

Red Sirug

Page 4

This risk means that the auditor wrongly concludes that the control risk is higher
than it actually is.
This risk relates to audit efficiency as it would lead o additional work. If the
auditor assesses control risk too high, substantive tests will consequently be
expanded beyond the necessary level, leading to audit inefficiency.
b. Risk of assessing control risk too low or the risk of over reliance (Beta risk or
Type II error) the risk that the assessed level of control risk (based on the sample) is
less than the actual/true level of control risk (i.e., sample results indicate a lower
deviation rate than actually exists in the population).
This risk means that the auditor wrongly concludes that the control risk is lower than
it actually is.
This risk relates to audit effectiveness. If the auditor assesses control risk too low,
substantive tests will not be expanded to the necessary level to ensure an effective
audit. This would more likely lead to an inappropriate audit opinion.
Analysis of Sampling Risks:
Aspects of
sampling risks
Risk of incorrect
acceptance

Risk of incorrect
rejection

Risk
of
assessing
control risk too
low (risk of over
reliance)
Risk
of
assessing
control risk too
high
(risk
of
under reliance)

Auditors
wrong
conclusion
Not materially
misstated
when in fact
materially
misstated

Effect on audit work


because of wrong
conclusion
Performance of less
extensive substantive
tests

Materially
misstated
when in fact
not materially
misstated
CR than actual
CR internal
control is
reliable

Additional work
(performance of
unnecessary more
extensive substantive
tests)
Performance of tests
of controls and less
extensive substantive
tests

CR than
actual CR
internal control
is not reliable

Additional work
(because nonperformance of tests
of controls would lead
to the performance of
unnecessary more
extensive substantive
tests)

Sacrificed
Effectiveness of the
audit because it may
lead to inappropriate
opinion
due
to
inappropriate
less
extensive
substantive
tests
Efficiency of the audit
because of unnecessary
additional work
Effectiveness of the
audit because it may
lead to inappropriate
opinion
due
to
inappropriate
less
extensive
substantive
tests
Efficiency of the audit
because of unnecessary
additional work

2. Non-sampling risk the risk that the auditor reaches an erroneous conclusion for any
reason not related to sampling risk
Examples of non-sampling risk:
The auditor might use/select inappropriate audit procedures (audit procedures that are
not appropriate to achieve a specific objective)
The auditor might misinterpret evidence or the results of audit tests
The auditor may fail to recognize an error (for example, failure by the auditor to
recognize a misstatement or deviation in documents examined)

Non-sampling risk pertains to all aspects of audit risk that are not due to sampling. It
refers to the possibility that auditors will arrive at an erroneous conclusion not
because of the chosen sample but due to other factors.
Non-sampling risk is always present and cannot be measured.
Non-sampling risk can be controlled by adequate planning and supervision of audit
work and proper adherence to quality control standards.

Sampling risk and non-sampling risk can affect the components of audit risk.
For example, when performing tests of control, the auditor may find no errors in a
sample and conclude that control risk is low, when the rate of error in the population is,
in fact, unacceptably high (sampling risk). Or there may be errors in the sample which
AT Audit Sampling

Red Sirug

Page 5

the auditor fails to recognize (non-sampling risk).


Types of Statistical Sampling Plans:
1. Attribute sampling the method used to estimate the rate (%) of occurrence (or exception)
of a specific characteristic or attribute
Attribute sampling used in tests of controls.
In attribute sampling, samples taken to test the operating effectiveness of controls are
intended to provide a basis for the auditor to conclude whether the controls are being
applied as prescribed.
Attribute sampling generally deals with yes/no questions. For example, "Are time cards
properly authorized (i.e., to assure recorded hours were worked)?", or "Are invoices properly
voided (e.g., stamped "paid") to prevent duplicate payments?"
Commonly used attribute sampling techniques/models:
a. Attribute estimation sampling a statistical sampling plan that uses a fixed
sampling plan (for example, testing a single sample)
It is used when the auditor wishes to estimate a true but unknown population
deviation rate.
b. Discovery sampling a special type of attribute sampling appropriate when the
auditor believes the expected population deviation rate is zero or near zero and
when the auditors objective is to find at least one deviation in the sample if actual
population deviation rate exceeds or equal a predetermined critical rate (tolerable
deviation rate)
Discovery sampling should be used to estimate whether a population contains
critical deviations.
It is used when the auditor is looking for a very critical characteristic or
deviations (e.g., fraud). The auditor predetermines the desired reliability
(confidence) level (e.g., 95%) and the maximum acceptable tolerable rate (e.g.,
1%), and a table is then used to determine sample size. If no deviations are
found in the sample, the auditor can be 95% certain that the rate of deviation in
the population does not exceed 1%. If deviations are found, a regular attribute
sampling table may be used to estimate the deviation rate in the population,
and audit procedures may need to be expanded.
c. Stop-or-go sampling (sequential sampling) is designed to avoid
oversampling for attributes by allowing the auditor to stop an audit test before
completing all steps
It is used when the auditor expects zero or very few deviations in the
population.
It separates the sampling process into several states or steps. After a step, the
auditor decides whether to stop or to go on to the next step. In each step, the
auditor determines if it is warranted to accept or increase the preliminary level
of control risk.
2. Variables sampling method used in reaching a conclusion in peso amounts
Variables sampling is used in substantive tests.
a. Probability-proportional-to-size (PPS) sampling sampling technique where the
sampling unit is defined as an individual peso in a population
PS is a sampling plan that automatically stratifies the population.
PPS uses a peso as a sampling unit. Once a peso is selected, the entire account
(containing that peso) is audited.
PPS is only useful for tests of overstatements (for example, of assets). Thus, it is not
appropriate for testing liabilities because understatement is the primary audit
consideration.
b. Classical variables sampling a statistical sampling method used to estimate the
numerical measurement of a population, such as a peso value (e.g., accounts receivable
balance)
The objective of variables sampling is to obtain evidence about the reasonableness of
monetary amounts. The auditor estimates the true value of the population by
computing a point estimate of the population and computing a precision interval around
this point estimate.
Classical variables sampling measures sampling risk by using the variation of the
underlying characteristic of interest.
Three commonly used classical variables sampling:
1. Mean-per-unit estimation a sampling plan that uses the average value of the items
in the sample to estimate the true population value by multiplying average sample
AT Audit Sampling

Red Sirug

Page 6

value by the number of items in population. MPU does not require the book value of the
population to estimate true population value.
2. Ratio estimation a sampling plan that uses the ratio of the audited (correct)
values/amount to their book values to project the true population value and an
allowance for sampling risk
Ratio estimation is a highly efficient technique when the calculated audit amounts
are approximately proportional to the client's book amounts.
3. Difference estimation a sampling plan that uses the average difference between
the audited (correct) values of items and their book values to project the actual
population value.
Difference estimation is used instead of ratio estimation when the differences are
not nearly proportional to book values.
Comparison of PPS sampling to classical variables sampling
Advantages of PPS sampling
1. Generally easier to use
2. Size of sample not based on variation of
audited amounts
3. Automatically results in a stratified
sample
4. Individually
significant
items
are
automatically identified
5. Usually results in a smaller sample size if
no misstatements are expected
6. Can be easily designed and sample
selection can begin before the complete
population is available

1.
2.
3.
4.

Advantages of classical variables


sampling
May result in a smaller sample size if
there are many differences between
audited and book values
Easier to expand sample size if that
becomes necessary
Selection of zero balances does not
require
special
sample
design
considerations
Inclusion of negative balances does not
require
special
sample
design
considerations

Steps in Sampling for Substantive Testing (Variables sampling)


1. Determine the objective of the test
The objective of the test must be to satisfy an audit objective pertaining to a particular
account balance, that is, to test the reasonableness of a recorded account balance. For
example, the auditor wishes to estimate the value of the client's accounts receivable
balance to satisfy valuation audit objective.
2. Define the population and the sampling unit
This is to provide assurance that the audit sample to be selected and examined will satisfy
the objective of the test. For example, the population might consist of 5,000 accounts with
a recorded book value of P4,500,000. The auditor would examine 100% of accounts for
which potential errors could equal or exceed the tolerable error and would exclude those
accounts from the population to be sampled.
The auditor should consider the completeness of the population in defining the sampling
unit. For example, each of the 5,000 accounts is a sampling unit.
3. Select an appropriate audit sampling technique
Select either non-statistical or statistical sampling. If statistical sampling is used, either
probability-proportional-to-size sampling (PPS) or classical variables techniques may be
selected.
4. Determine the sample size
Factors to consider in determining sample size for substantive tests of details:
a. Acceptable risk of incorrect acceptance
b. Acceptable risk of incorrect rejection
c. Tolerable error (tolerable misstatement) maximum amount of errors (or monetary
misstatement) that may exist without causing the account balance or class of
transactions to be materially misstated (or maximum amount of error that the auditor is
willing to accept)
d. Expected error/misstatement (size, frequency, etc.)
e. Variation within the population (e.g., an estimate of the standard deviation, or
variability, of the population)
Summary of relationships between the above factors and the sample size:
Increases in
Effect on
Explanation
sample size
Risk of incorrect
Decrease
This is a sampling risk and sampling risk is
acceptance
reduced by increasing the sample size.
AT Audit Sampling

Red Sirug

Page 7

Risk of incorrect
rejection
Tolerable
misstatement
(error)
Expected
misstatement
(error)

Decrease

Decrease

Increase

Variation in the
population
(standard
deviation)
Increase
in
auditors
assessment
of
control
risk
or
inherent risk
Reliance on other
substantive
procedures

Increase

Increase

The higher the auditors assessment of inherent


risk and control risk, the larger the sample size
needs to be.

Decrease

Negligible
effect

The more the auditor intends to rely on other


substantive procedures to reduce to an
acceptable level the detection risk, the less
assurance the auditor will require from sampling
and, therefore, the smaller the sample size can
be.
The number of items in the population virtually
has no effect on sample size unless the
population is very small.
In other words,
population size is not an issue provided the
population is large.

Number of items in
the population

This is a sampling risk and sampling risk is


reduced by increasing the sample size.
The lower the total error that the auditor is
willing to accept, the larger the sample size
needs to be.
The greater the expected amount of error in the
population, the larger the sample size needs to
be in order to make a reasonable estimate of the
actual amount of error in the population.
Increases in variation (standard deviation in
classical sampling) result in increases in sample
size.

5. Determine the sample selection method


Generally random number or systematic sampling.
6. Perform the sampling plan
Sample items should be selected in such a way that the sample can be expected to be
representative of the population (e.g., random sampling). In the same example, an
appropriate sample would consist of individual account balances. Confirmations could then
be used to determine the audited values for sample items.
7. Evaluate the sample results: This includes the following procedure:
a. Projecting the sample error. The auditor projects the misstatements found in the sample to
the population using one of several methods (e.g., MPU, ratio, difference, etc.). The
projected misstatement is applied to the recorded balance to obtain a "point estimate" of
the true balance.
b. Considering sampling risk. The auditor must then add an allowance for sampling risk
(sometimes called a "precision interval") to this estimate.
c. Considering qualitative information
d. Reaching an overall conclusion
In deciding whether to accept the client's book value, the auditor determines whether
the recorded book value falls within the acceptable range (i.e., the point estimate +/the allowance for sampling risk). If so, the book value is fairly stated.
The auditor's treatment of items selected for sampling that cannot be located (e.g., are
"lost") will depend on their effect on the auditor's evaluation of the sample.
If the sample is representative of the population, the auditor generally will make a
correct decision regarding whether the account balance is fairly stated.
If the sample is not representative of the population, the auditor will make an incorrect
decision, either accepting a materially misstated balance, or rejecting a fairly stated
balance.
8. Document the sampling procedure
The auditor must document each step in audit sampling as well as the basis for overall
conclusions.
Steps in Sampling for Tests of Controls (Attribute sampling)
1. Define the objectives of the test

AT Audit Sampling

Red Sirug

Page 8

Tests of controls are designed to test the operating effectiveness of controls. For example,
the auditor might test controls for billing systems.

2. Define the population


For tests of controls, the population is the class of transactions being tested. Conclusions
based on sample results can be projected only to the population from which the sample
was selected.
3. Define the attribute and deviation conditions
An attribute (or characteristic) would indicate operation of the internal control procedures.
A deviation is a departure from the prescribed internal control policy or procedure. For
example, if the prescribed procedure to be tested requires the cancellation of each paid
voucher, a paid but uncanceled voucher would constitute a deviation.
4. Determine the sample size
The sample size is determined by considering the following factors:
a. Risk of assessing control risk too low inverse relationship with the sample size
b. Tolerable deviation rate inverse relationship with the sample size
c. Expected population deviation rate direct relationship with the sample size
Summary of relationships between the above factors and the sample size:
Increases in
Effect on
Explanation
sample
size
Risk of assessing
Decrease
The more assurance the auditor intends to
control risk too low
obtain from internal controls, the lower the
auditors assessment of control risk will be, and
the larger the sample size will need to be.
This is a sampling risk and sampling risk is
reduced by increasing the sample size.
Tolerable deviation
Decrease
The lower the rate of deviation that the auditor
rate
is willing to accept, the larger the sample size
needs to be.
Expected
Increase
The higher the rate of deviation that the auditor
population
expects, the larger the sample size needs to be
deviation rate
so as to be in a position to make a reasonable
estimate of the actual rate of deviation.
Number of items in
Negligible
The number of items in the population virtually
the population
effect
has no effect on sample size unless the
population is very small.
In other words,
population size is not an issue provided the
population is large.
5. Determine the sample selection method
Principal sample selection methods: Appropriate sample selection methods could
reduce sampling risk.
a. Random-number sampling this method uses of a computerized random number
generator or random number tables
Each item in the population has an equal chance and nonzero probability of
selection.
It is appropriate both for statistical and non-statistical sampling.
b. Systematic selection the number of sampling units in the population is divided
by the sample size to give a sampling interval regardless of the amount involved
(for example 50, and having determined a starting point within the first 50, each
50th sampling unit thereafter is selected)
It involves selecting every nth item from a population of sequentially ordered
items.
The auditor need to determine that the sampling units within the population are
not structured in such a way that the sampling interval corresponds with a
particular pattern in the population.
It is useful for non-statistical sampling, although it can also be useful for
statistical plan if the starting point is selected at random.
c. Haphazard selection the auditor selects the sample without following a
structured technique, but the method is intended to avoid or predictability (for
example avoiding difficult to locate items, or always choosing or avoiding the first or
last entries on a page) and thus attempt to ensure that all items in the population
have a chance of selection
Haphazard selection, although may be useful for non-statistical sampling, is not
AT Audit Sampling

Red Sirug

Page 9

appropriate when using statistical sampling.


d. Block selection involves selecting a block(s) of contiguous items from within the
population
Block selection cannot ordinarily be used in audit sampling because most
populations are structured such that items in a sequence can be expected to
have similar characteristics to each other, but different characteristics from
items elsewhere in the population.
It often results to excessively high sampling risk.
e. Stratification grouping of items of similar size and each group is treated as a
separate population; it involves subdiving a population into subpopulations or strata
For example, assume 1,000 items are stratified into two groups: the 100 largest
items will all be examined individually, but sampling techniques will be applied to
the remaining 900 items. In this case, the population size for the sampling
application would be 900, not 1,000.
The primary objective of using stratification as a sampling method is to decrease
the effect of variance or variability of items in the total population.
Stratification is used when there is a wide range (variability) in the monetary
size of items in the population.
This method will enable the auditor to direct his efforts towards the items he
considers he would potentially contain the greater monetary error.
f. Value-weighted selection sets the high-value items as priority to be included in
the sample
6. Perform the sampling plan
The sampling units selected should be examined for the attributes or quality characteristics
of interest and deviations should be documented in the working papers.
7. Evaluate and document results: These include the following:
a. Determine the sample deviation rate
= Number of deviations observed
Sample size
b. Determine the maximum population deviation rate (achieved upper deviation rate)
and the allowance for sampling risk (achieved precision)
The maximum deviation rate is based on the sample size and the number of
deviations discovered. The auditor uses standard tables that yield maximum population
deviation rates at specified risk of assessing control risk too low.
Allowance for sampling risk = Maximum deviation rate Sample deviation rate
When the deviation in the sample is at the expected deviation rate or less, the
auditor can continue using his planned assessment of control risk. If it happens to
be greater than expected, reassessment of risk is necessary. Usually, an increase
in such should be made.
The stronger the internal control, the lower the control risk, the lower the tolerable
deviation rate.
c. Consider qualitative consideration (such as the nature of each deviation, its importance,
and probable cause)
d. Reach an overall conclusion
The overall conclusion relates to assessing control risk after considering all available
quantitative and qualitative information.

AT Audit Sampling

Red Sirug

Page 10

You might also like