AUDIT

SAMPLING
LEARNING OBJECTIVES
1. Explain the rationale for audit sampling.
2. Distinguish between sampling and non-sampling risk.
3. Differentiate statistical sampling from non-statistical sampling, attribute
sampling from variable sampling.
4. Discuss the basic steps in audit sampling.
5. Identify the commonly used sample selection methods and projection
techniques.
6. Explain how sample results are evaluated.

AUDPRIN 2
BASIS
PSA 500 (Redrafted) “Audit Evidence”
▪ Para. 10. Selecting Items for Testing to Obtain Audit Evidence . When designing
tests of controls and tests of details, the auditor shall determine means of
selecting items for testing that are effective in meeting the purpose of the audit
procedure.
▪ Para. A52-56. Means of selecting items for testing
➢ Selecting all items (100% examination)
➢ Selecting specific items
➢ Audit sampling

AUDPRIN 3
SELECTING ALL ITEMS
▪ Unlikely in tests of controls; more common for tests of details
▪ Appropriate when
✓ The population consists of small number of large value items
✓ There is significant risk and other means do not provide sufficient
appropriate audit evidence
✓ Repetitive nature of a calculation or process performed automatically by an
information system makes 100% examination cost effective

AUDPRIN 4
SELECTING SPECIFIC ITEMS
▪ Selective testing: auditor may decide to select specific items based on such
factors as his understanding of the entity, assessed risk of material
misstatement and characteristics of population being tested
▪ Examples
❑ Items with certain characteristics (high value, suspicious, unusual, risk-prone,
with error history)
❑ Items to obtain information about certain matters (nature of entity,
transactions, internal control)
❑ Items to test internal control activities

AUDPRIN 5
AUDIT SAMPLING

▪ Auditors usually draw conclusions about class

of transactions or account balance (the audit
population) by examining only a sample of
evidence on the premise that a sample can
be sufficiently representative of the audit
population and thus, an inference or
conclusion can be drawn about the
characteristics of the population based on
the sample results.

AUDPRIN 6
DEFINITION
▪ Audit Sampling – application of audit procedures to less than 100% of items
within a class of transactions or an account balance (the audit population) such
that all sampling units have a chance of selection in order to provide auditor
with a reasonable basis on which to draw conclusions about entire population

AUDPRIN 7
KEY TERMS
▪ Error – either control deviations, when performing tests of controls, or
misstatements, when performing tests of details
o Deviation – difference between what was expected, based on
documentation of controls, and what actually occurred; stated in terms of
percentages
o Misstatement – difference between amount computed by auditor and
amount actually reflected in accounting records; presented in terms of
monetary amounts
o Total error – rate of deviation or total misstatement
o Expected error – error that auditor expects to be present in population

AUDPRIN 8
KEY TERMS
▪ Anomaly – a misstatement or deviation that is demonstrably not representative
of misstatements or deviations in a population
▪ Population – entire set of data from which a sample is selected and about
which auditor wishes to draw conclusions; may be divided into strata, or sub-
populations, with each stratum being examined separately
▪ Sampling unit – individual items constituting a population e.g. credit entries on
bank statements

AUDPRIN 9
WHEN TO USE AUDIT SAMPLING
• YES
• Nature and materiality of balance or class of transaction does
not demand 100% audit.
• Decision must be made about the balance or class of
transaction.
• Time and cost to audit 100% of population would be too great.

• NO
• Tests performed on 100% of items within a population
• Inquiry and observation
• Analytical procedure
AUDPRIN 10
AUDIT SAMPLING VS. AUDIT PROCEDURES
▪ Risk assessment procedures
➢ Ordinarily do not involve use of audit sampling
▪ Tests of controls (TOC)
➢ Generally appropriate when application of control leaves audit evidence of
performance (e.g. initials of credit manager on sales invoice evidencing
credit approval)
▪ Substantive procedures (ST)
➢ Relate only to test of details; may be used to verify one or more assertions
about F/S amount (e.g. existence of accounts receivable)

AUDPRIN 11
RISKS IN SAMPLING
▪ Auditor faces uncertainty of not detecting material errors in an account balance or class of
transactions due to
➢ Sampling risk
o Arises from possibility that auditor’s conclusion, based on a sample, may be different from
conclusion reached if entire population were subjected to same audit procedure
➢ Non-sampling risk
o Uncertainties arising from factors unrelated to sampling
o Risk that auditor may draw incorrect conclusions about account balance or transaction class
because of human errors
o application of inappropriate audit procedures
o failure to recognize errors in sample tested
o misinterpretation of evidence obtained

AUDPRIN 12
SAMPLING RISK: Types
▪ Alpha Risk – risk that auditor will conclude
❖ TOC: that internal control is not reliable when in fact it is effective and can
be relied upon (risk of assessing control risk too high, risk of under reliance)
❖ ST: that material misstatement exists in an account balance or transaction
class when in fact such misstatement does not exist (risk of incorrect
rejection)
This results when the auditor performs audit procedure more than what is
necessary, thus affecting audit efficiency -> overauditing

AUDPRIN 13
SAMPLING RISK: Types
▪ Beta Risk – risk that auditor will conclude
❖ TOC: that internal control is reliable when in fact it is not effective and
cannot be relied upon (risk of assessing control risk too low, risk of over
reliance)
❖ ST: that material misstatement does not exist in an account balance or
transaction class when in fact such misstatement exists (risk of incorrect
acceptance)
This results when the auditor performs audit procedure less than what is
necessary, thus affecting audit effectiveness -> underauditing

AUDPRIN 14
SAMPLING RISK: TOC
Auditor’s
Assessment of Client’s Control Risk is
Control Risk is
< Maximum Maximum
Type II error
< Maximum level Correct decision (Beta Risk – over
reliance)
Type I error (Alpha Risk -
Maximum level Correct decision
under reliance)
AUDPRIN 15
SAMPLING RISK for TOC: Example
Assume that the auditor performed audit sampling and sample results
were estimated at 5% deviation rate. The auditor was willing to
tolerate a deviation rate of only 3%. Assume further that the true
population rate of deviation was 2%.
What would be the likely conclusion of the auditor? Discuss the
applicable sampling risk.

AUDPRIN 16
SAMPLING RISK: ST
Auditor’s Decision Based
True State of F/S Account
on Sample Results

Not materially misstated Materially misstated

Type II error
Fairly stated Correct decision
(Beta Risk – incorrect acceptance)

Type I error
Not fairly stated (Alpha Risk –incorrect Correct decision
rejection)

AUDPRIN 17
AUDIT SAMPLING AND AUDIT RISK
AR = IR x CR x DR

Sampling Risk
Risk of material
misstatement

Non-sampling
Risk

AUDPRIN 18
CONTROLLING THE RISKS
▪ Sampling risk may be decreased by
➢ Increasing sample size
➢ Using appropriate sample selection method
▪ Non-sampling risk may be minimized by
➢ Proper planning
➢ Adequate direction, review and supervision of audit team

AUDPRIN 19
AUDIT SAMPLING: BASIC STEPS
1. Define the audit objective.

2. Determine the audit procedure. Define the population and its characteristics.

3. Determine the sample size.

4. Select the sample.

5. Apply the procedures.

6. Project errors/Evaluate the results.

AUDPRIN 20
GENERAL APPROACHES TO AUDIT
SAMPLING
• Statistical Sampling • Non-statistical Sampling
• Random selection of sample items; and • Purely uses auditor’s judgment in
• Use of probability theory to evaluate estimating sampling risks, determining
sample results, including measurement sample size, and evaluating sample
of sampling risk results

AUDPRIN 21
STATISTICAL SAMPLING: Advantages &
Disadvantages
• Advantages
• Aids in determining sample size required to meet given objectives
• Provides more objective audit evidence
• Allows auditor to measure precision, reliability and sampling error
• Disadvantages
• Danger of accepting statistical evidence at face value, without sufficient
skepticism
• Cost of statistical sampling could exceed benefits
• May be less appropriate than non-statistical sampling in some cases

AUDPRIN 22
STATISTICAL SAMPLING VS. NON-STATISTICAL
SAMPLING
• Similarities
• Both acceptable
• Both require use of auditor’s judgment in designing and selecting sample, performing audit procedures, and evaluating
results
• Both cannot assure that sample will be representative of population
• Difference
• Statistical sampling allows auditor to measure sampling risk by using mathematical formula.

• Statistical or non-statistical?
• Depends on auditor’s assessment of relative costs and benefits
• Use non-statistical sampling if cost of selecting statistical random sample is deemed too high
• Use statistical sampling if the cost is justifiable e.g. related controls were critical, thus measure of sampling risk was deemed
essential

AUDPRIN 23
SAMPLE SELECTION METHODS
METHOD DESCRIPTION
Random number Utilizes random-number tables or computer-generated numbers to
sampling select sampling units

Systematic sampling Selects items based on sampling interval

Divides population into groups (strata) according to some

Stratified sampling
characteristic, then applies random sampling to each stratum.

Selects group of items arranged contiguously within larger grouping of

Block selection
sample units

Haphazard sampling Selects sampling units without special reason, without conscious bias

AUDPRIN 24
RANDOM NUMBER SAMPLING

54463 41417 22662 98326 65905 87719

15389 28357 85205 94070 18850 20652
85941 17783 40756 00015 82414 10806
61149 40950 69440 84820 11268 29881
05219 82995 81619 64157 10651 66164

AUDPRIN 25
SYSTEMATIC SAMPLING: Example
Population: Sales invoices 652 -3151
Desired sample size is 125.
What is the sampling interval?

Answer:
(3151-651)/125 = 20
Then select a random start between 1 and 19 (say 10). First item in the sample is
inv 662 (652+10), remaining 124 items = 682 (662+20), 702 (682+20), 722
(702+20), etc.
AUDPRIN 26
AUDIT SAMPLING CONSIDERATIONS
• Characteristics to be tested
• Consider objectives of audit procedure and attributes of population from which sample will be drawn
• Characteristic of interest depends on type of test that will be performed on the sample selected (TOC: deviation or occurrence rate; ST: monetary
amount of misstatement)
• Population
• Should be complete and appropriate to audit objective (consider direction of testing)
• Sample Size
• Sufficient to reduce sampling risk to an acceptably low level
• Items for Testing
• Should be selected in such a way that each sampling unit in the population has a chance of selection
• Audit Procedures
• Should be appropriate for the purpose.
• If audit procedure is not applicable to selected item, perform procedure on a replacement item.
• If auditor is unable to apply designed or suitable alternative audit procedure to a selected item, treat the item as deviation (TOC) or
misstatement (TOD).

AUDPRIN 27
AUDIT SAMPLING PLANS
• Attribute sampling
• Estimates frequency of occurrence of certain characteristic in population
(occurrence rate)
• Used in TOC to estimate rate of deviations from prescribed internal control
policies or procedures
• Variable sampling
• Estimates numerical measurement of population e.g. peso value
• Used in ST to estimate amount of misstatements in F/S

AUDPRIN 28
SAMPLING FOR TESTS OF CONTROL
• Appropriate when application of control leaves evidence of performance, else, use non-sampling
procedures e.g. observation, inquiry
• Factors to consider in determining sample size
• Acceptable sampling risk: inversely related to sample size ->the smaller the sampling risk
the auditor is willing to accept, the larger the sample size to be
• Tolerable deviation rate: maximum rate of deviations auditor is willing to accept without
modifying planned degree of reliance on internal control; inversely related to sample size
• Expected deviation rate: rate of deviations auditor expects to find in population before
testing begins, usually based on prior year’s results or by doing pilot sample; directly
related to sample size; should not exceed tolerable deviation rate

AUDPRIN 29
SAMPLING FOR TESTS OF CONTROL:
Sample Selection Methods
• Random number selection
• Systematic selection
• Haphazard selection
• Other considerations
• Properly voided documents-> replace by another item
• Missing document-> treat as deviation for purpose of evaluating sample results
• Other sampling applications
• Sequential (stop-or-go) sampling
• Discovery sampling

AUDPRIN 30
SAMPLING FOR SUBSTANTIVE TESTS
• Variables sampling plan
• Used to estimate total peso amounts (or possibly units) of a population or the
peso amount of an error in a population
• Substantive tests of transactions (STOT) - tests control effectiveness and
monetary correctness of transactions
• substantive tests of balances (STOB) - tests whether monetary amount of
an account balance is materially misstated

AUDPRIN 31
VARIABLES SAMPLING PLAN:
Steps
1. Determine objective/s of the test.
2. Determine appropriate audit procedures to be performed to satisfy objective/s. Define
population and its characteristics.
3. Choose audit sampling approach/technique.
4. Determine sample size.
5. Select the sample using chosen sample selection methods (e.g. random number, systematic,
haphazard).
6. Perform the audit procedures.
7. Evaluate sample results.
8. Document sampling plan, procedures performed, and conclusions reached.

AUDPRIN 32
SAMPLING TECHNIQUES
• Probability-proportional-to-size (PPS) sampling
• Uses attribute sampling theory to evaluate results when large number of
transactions are captured within a single account
• Most appropriate when no (or few) errors are expected or when testing for
overstatement (normally for assets and income)
• Auditor randomly selects individual pesos from population and then audits
balances, transactions or documents (called logical units) that include pesos
selected.
Each peso in population has equal chance of being selected, but likelihood of
selecting any one logical unit for testing is directly proportional to its size.

AUDPRIN 33
SAMPLING TECHNIQUES
• Classical Variables Sampling (CVS)
• Relies on normal distribution theory to evaluate audit samples
• May be appropriate when audit objective is to estimate true but unknown
monetary balance
• Projection Techniques
• Ratio estimation
• Difference estimation
• Mean-per-unit estimation

AUDPRIN 34
PROJECTION TECHNIQUES
• Ratio estimation
• Uses ratio of audited amounts to recorded amounts in sample to estimate
total peso amount of population and an allowance for sampling risk
• Difference estimation
• Uses average difference between audited amounts and individual recorded
amounts in sample to estimate total audited amount of population and
allowance for sampling risk
• Mean-per-unit estimation
• Projects sample average (mean) to total population by multiplying sample
average by number of items in population
AUDPRIN 35
PROJECTION TECHNIQUES:
Example
The auditor sent out positive confirmation requests to 2,000 customers with a population
size of 4,800 accounts and total recorded value of P380,000. The confirmation results are
summarized below:
Book value of samples selected P159,960
Audited value of samples selected 151,360
Difference P 8,600

Required:
Using CVS techniques, calculate the estimated audited value of the population and the
projected misstatement.
AUDPRIN 36
PROJECTION TECHNIQUES:
Example
Ratio Estimation
a. EPAV = SAV/SBV x PBV
= P151,360/P159,960 x P380,000
= P359,570
b. Projected misstatement = EPAV – PBV
= P359,570 – P380,000
= P20,430 overstated

AUDPRIN 37
PROJECTION TECHNIQUES
Example
Difference Estimation
a. Ave. diff= Ave. audited value – Ave. book value
= (P151,360/2,000) – (P159,960/2,000)
= P76 – P80
= P4 overstated
b. Projected misstatement = Ave. diff * number of items in population
= P4 x 4,800 accounts
= P19,200 overstated
c. EPAV = PBV +- projected misstatement
= P380,000 – P19,200
= P360,000

AUDPRIN 38
PROJECTION TECHNIQUES
EXAMPLE
Mean-per-unit Estimation
a. EPAV = Ave audited value x number of items in population
= P76 x 4,800 accounts
= P364,800
b. Projected misstatement = EPAV – PBV
= P364,800 – P380,000
= P15,200 overstated

AUDPRIN 39
 PROJECTION TECHNIQUES

Type Formula When appropriate?

SAV/SBV * PBV = EPAV +(-) Misstatement in account is directly
Ratio estimation
sampling risk proportional to its book value

Difference (SAV-SBV)/SS * P = Misstatement in account is not affected

estimation projected error by book value of item being examined

Mean-per-unit SAV/SS * P = EPAV +(-) Individual population items do not have

estimation sampling risk recorded values

AUDPRIN 40
 PROJECTION
TECHNIQUES
Legend:
SAV = sample audited value
SBV = sample recorded book value
PBV =population book value
EPAV = estimated population audited value
SS = sample size
P = number of items in population

AUDPRIN 41
 ADVANTAGES

• Automatically results in stratified sample

because items are selected in proportion to their
peso amounts
• Results in smaller sample size than CVS when no
errors are expected
• Can be designed more easily and sample

PPS selection can begin before complete population

is available

SAMPLING DISADVANTAGES

• Sample evaluation will require special sample

design considerations if sample includes
understatement errors
• Evaluation may overstate allowance for sampling
risk when errors are found
• Generally includes assumption that audited
amount of sampling unit should not be <0 or
>recorded amount

AUDPRIN 42
 ADVANTAGES

• May result in smaller sample size if there are many

individual differences between recorded and
audited amounts in population
• Selection of zero or negative balances within
sample does not require special sample design
considerations
• If necessary, it is easier to expand samples than PPS

DISADVANTAGES
CVS SAMPLING • More complex than PPS
• To determine sample size, auditor must have
estimate of population standard deviation
• Normal distribution theory may not be appropriate
when sample size is not large and there are either
very large items or very large differences between
recorded and audited amounts in population

AUDPRIN 43
FACTORS INFLUENCING SAMPLE SIZE FOR
SUBSTANTIVE TESTS
EFFECT ON
FACTOR
SAMPLE SIZE
Increase in auditor’s assessment of inherent risk Increase
Increase in auditor’s assessment of control risk Increase
Increase in use of other ST directed at same F/S assertion Decrease
Increase in auditor’s required confidence level Increase
Increase in total error that auditor is willing to accept (tolerable error)
Decrease

Increase in amount of error that auditor expects to find in population

Increase
(expected error)
Number of sampling units in population Negligible effect
AUDPRIN 44
FURTHER READINGS
1. https://www.dallasiia.org/PDF/010809
_CAST.pdf

2. https://www.iia.nl/SiteFiles/PA_2320-
3%20(1).pdf

3. https://aasc.org.ph/downloads/PSA/pu
blications/psa-530-redrafted.php

AUDPRIN 45
SELF-TEST QUESTIONS
1. Can you eliminate sampling risk? Why or why not?
2. What are the two general approaches to audit sampling? Explain each briefly.
3. Explain briefly the following sample selection methods:
a. Random number sampling
b. Systematic sampling
c. Haphazard sampling
4. Give two determinants of the sample size for tests of controls.
5. Compare and contrast the following projection techniques:
a. Ratio estimation
b. Difference estimation
c. Mean-per-unit estimation

AUDPRIN 46
LEARNING ACTIVITY
IDENTIFICATION. Write the word or group of words that best describes or identifies each of the
statements below. Incomplete answers will not qualify for points.
1. entire set of data from which a sample is selected and about which the auditor wishes to draw
conclusions
2. misstatement or deviation that is demonstrably not representative of misstatements or
deviations in a population
3. process of dividing a population into sub-populations, each of which is a group of sampling
units which have similar characteristics, often monetary value
4. 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
5. either control deviations, when performing tests of controls, or misstatements, when
performing tests of details
AUDPRIN 47
LEARNING ACTIVITY
ABA Corporation has 2,000 accounts receivable with a total book value of
P2,000,000. Ric Ramos, CPA, has selected and audited a sample of 100 of the
accounts with a combined book value of P101,000; the total of their audited values
was P110,000.

Required:
Using the 3 common projection techniques, calculate the (a) accounts receivable
estimated audited value, (b) projected misstatement.

AUDPRIN 48
AUDPRIN 49