AUD1

Audit Sampling

Audit Sampling – application of audit procedures to less than 100% of the items
Sufficiency of audit evidence
Audit sampling methods:
1. Statistical sampling – characterized by:
 Random selection of sample items, and
 The use of probability theory to evaluate sample results, including measurement of sampling risk
2. Non-statistical sampling
 Does not have the characteristics of a statistical sampling
 Both sampling methods are allowed under Philippine Standards on Auditing; when properly applied both
methods result in sample size that provides sufficient audit evidence
 Both sampling methods require auditor’s professional judgment
Attribute vs Variable Sampling:
Attribute sampling – primarily used for testing internal controls
Variables sampling – typically used in substantive testing of account balances
Situations where sampling may not apply:
1. Risk assessment procedures performed to obtain an understanding of internal control
2. Tests performed on 100% of the items within a population
3. Procedures that depend on segregation of duties or that otherwise provide no documentary evidence
4. Analysis of significant ratios and trends
Audit risk – includes both:
1. Sampling risk – risk that auditor’s conclusion based on a sample may be different from the conclusion if the
entire population were subjected to the same audit procedure; the risk that sample may not be representative of
the population
2. Non-sampling risk – risk that the auditor reaches an erroneous conclusion for any reason not related to
sampling risk; examples:
 Inappropriate audit procedures
 Inappropriate audit evidence
 Failure to recognize misstatements in documents examined or improper evaluation of results

Sampling Risk + confidence level = 100%

Sampling Risk in Test of Controls

a. Risk of Assessing Control Risk Too Low b. Risk of Assessing Control Risk Too High
 Risk that assessed level of control risk based  Risk that assessed level of control risk based
on the sample is less than the true risk based on the sample is greater than the true risk
on the actual operating effectiveness of the based on the actual operating effectiveness of
control the control
 Overreliance  Underreliance
 Ineffective  Inefficient
 Beta risk  Alpha risk

Sampling Risk in Substantive Testing

a. Risk of Incorrect Acceptance b. Risk of Incorrect Rejection
 Risk that the sample supports the conclusion  Risk that the sample supports the conclusion
that the recorded account balance is not that the recorded account balance is
materially misstated when in fact it is materially misstated when in fact it is not
materially misstated materially misstated
 Ineffective  Inefficient
 Beta risk  Alpha risk
SAMPLING IN TEST OF CONTROLS – ATTRIBUTE SAMPLING
 To test the operating effectiveness of controls
 Estimates the rate of occurrence (deviation) of a specific characteristic (attribute)

Attribute – the characteristic of control

Deviation – the absence of an attribute

Tolerable deviation rate – the maximum rate of deviation that the auditor is willing to accept without modifying the planned reliance on
internal control. (Note that such deviations do not necessarily result in misstatements but do increase the risk of misstatement)

Expected deviation rate – the auditor’s best estimate of the deviation from a prescribed control procedure he expects to find in the entire
population

Relationship between tolerable deviation rate and expected deviation rate

Factors affecting sample sizes for test of controls:

Factor Effect on Sample Size

1. An increase in the extent to which the auditor’s risk assessment takes into
account relevant controls
2. Increase in expected deviation rate of the population to be tested
3. Increase in the auditor’s desired level of assurance that the tolerable rate of
deviation is not exceeded by the actual rate of deviation in the population
4. Increase in the tolerable deviation rate
5. Number of sampling units in the population

Sample Selection
1. Random number selection – each item has an equal opportunity to be included in the sample. Usually accomplished by
generating use of random number tables or computer programs generating random numbers
2. Systematic selection – (i.e. every nth item); disadvantageous if errors occur in a systematic pattern and the sampling interval
corresponds with that pattern; to reduce his risk, the auditor may use multiple random start
3. Block (cluster) sampling – a block of contagious items are selected; groups of adjacent items are selected
4. Haphazard selection – selects a sample without a structured technique
5. Stratification – divides the population into sub-populations, each of which is a group of sampling units which have similar
characteristics

Evaluation of Sample Results

Sample deviation rate = Number of deviations observed
Sample size
Upper deviation rate = Sample deviation rate + Allowance for sampling risk
Tolerable deviation rate ≥ Upper deviation rate → Rely or Do not rely?
Tolerable deviation rate < Upper deviation rate → Rely or Do not rely?

Example
1. An auditor wants to test whether sales orders are properly approved by the credit department. The auditor identified the
population to be the client’s sales orders and identified each sales order as one sampling unit. Deviations are defined as missing
credit approvals and missing sales orders.
Population = 200 items
Acceptable risk of assessing control risk too low = 5%
Tolerable deviation rate = 15%
Expected deviation rate = 1%
Using the table 1 below, sample size will be _____ items

2. Assuming based on the auditor’s testing, 1 of the sample selected did not have proper credit approval.
Sample deviation rate is ______.
Upper deviation rate is _____.
Allowance for sampling risk is _____.

3. Assuming based on the auditor’s testing, another 1 of the selected sales order was missing and could not be presented by the
client to the auditor.
Sample deviation rate is ______.
Upper deviation rate is _____.
Allowance for sampling risk is _____.
Conclusion: __________

Other attribute sampling models

1. Discovery sampling – used for detecting fraud/critical items
2. Stop-or-go sampling (sequential sampling) – designed to avoid oversampling for attributes by allowing the auditor to stop an
audit test before completing all steps. It is used when few errors are expected in the population.
 Table 1 – Sample Sizes for Test of Controls At 5% Risk of
Assessing Control Risk Too Low
Expected Deviation Rate Tolerable Deviation Rate
2% 3% 4% 5% 6% 7% 8% 9% 10% 15% 20%
0.00% 149 99 74 59 49 42 36 32 29 19 14
0.25 236 157 117 93 78 66 58 51 46 30 22
0.50 157 117 93 78 66 58 51 46 30 22
0.75 208 117 93 78 66 58 51 46 30 22
1.00 156 93 78 66 58 51 46 30 22
1.25 156 124 78 66 58 51 46 30 22
1.50 192 124 103 66 58 51 46 30 22

Table 2 – Statistical Sample Results Evaluation Table for Test of Controls

Upper Deviation Rate At 5% Risk of Assessing Control Risk Too Low
Actual Number of Deviations Found
Sample Size 0 1 2 3 4 5 6 7 8 9 10
25 11.3 17.6
30 9.5 14.9 19.6
35 8.3 12.9 17.0
40 7.3 11.4 15.0 18.3
45 6.5 10.2 13.4 16.4 19.2
50 5.9 9.2 12.1 14.8 17.4 19.9
55 5.4 8.4 11.1 13.5 15.9 18.2

SAMPLING IN SUBSTANTIVE TESTS – VARIABLES SAMPLING

 To obtain evidence about the reasonableness of monetary amounts
 Estimates the numerical measurement of a population (eg. Php value of receivables).
 Estimates the true value of the population by computing a point estimate of the population and computing a
precision interval around this point estimate.

Tolerable misstatement – maximum monetary misstatement in the related account balance or class of transactions that
the auditor is willing to accept
Relationship between tolerable misstatement and performance materiality
Factors affecting sample sizes for test of details:
Factor Effect on Sample Size
1. An increase in the auditor’s assessment of the risk of material
misstatement
2. An increase in the use of other substantive procedures directed at
the same assertion
3. An increase in the auditor’s desired level of assurance that tolerable
misstatement is not exceeded by actual misstatement in the
population
4. An increase in tolerable misstatement
5. An increase in the amount of misstatement that the auditor expects
to find in the population
6. Stratification of the population (when appropriate; commonly used
when population has highly variable recorded amounts)
7. Increase in the number of sampling units in the population

Evaluation of Sample Results

Projected misstatements
Point estimate
Allowance for sampling risk (precision interval)
Determine whether client’s recorded value falls within the acceptable range (point estimate +/- allowance for
sampling risk)
Classical Variables Sampling Plans
1. Mean per unit estimation
 uses the average value of the items in the sample to estimate the true population value
  does not require the book value of the population to estimate the true population value
2. Ratio estimation
 uses the ratio of the audited values of items to their book values to project the true population value
 a highly efficient technique when the calculated audit amounts are approximately proportional to the
client’s book amounts
3. Difference estimation
 uses the average difference between the audited values of items and their book values to project the
actual population value
 used instead of ratio estimation when the differences are not nearly proportional to book values

Probability-Proportional to Size Sampling (PPS)

 Uses individual peso amount as sampling unit
 Once a peso amount is selected, the entire account containing that peso is selected
 Automatically emphasizes larger items by stratifying the sample. The chance of an item being selected is
proportional to its peso amount.
 Useful only for tests of overstatements
 Zero balances, negative balances and understated balances generally require special design considerations
 Sample size determination:
Sampling interval = Tolerable misstatement
Reliability Factor

Sample size = Recorded amount of the population

Sampling Interval

Reliability factor is determined in relation to risk of incorrect acceptance. (The higher the risk of
incorrect acceptance, the lower the reliability factor.)
 Sample selection
A random number between 1 and the sampling interval is selected as the random start, which will also
determine the first item selected. Systematic selection is then used to select the remainder of the sample
 Evaluation of sample results
If no errors are found in the sample, the error projection is zero
If errors are found in an account, the errors need to be projected to the interval.
If the account selected has a balance greater than the interval, the actual peso amount of the error
should be used.
As with other variables sampling plans, an allowance for sampling risk would be calculated and added
to the projected error, and the result would be compared to the tolerable misstatement.