Audit Sampling Guidelines and Reference Materials for Internal Auditors

A

UDIT

 S

AMPLING

L

OUISIANA

 A

SSOCIATION

OF

 C

OLLEGE

AND

U

NIVERSITY

 A

UDITORS

 (LACUA)

O

CTOBER

 28, 2022

O

BJECTIVES

•

To review authoritative guidance for audit

sampling:

•

Potential for external auditor reliance on

internal auditors (AU-C §610).

•

To project results of our work with a greater

degree of certainty.

O

BJECTIVES

•

Guidance on sampling related topics:

•

Understand sampling

•

Apply concepts

•

To respond to questions related to audit

sampling.

A

VAILABLE

R

EFERENCE

 M

ATERIAL

•

AICPA 

Codification of Statements on

Auditing Standards

, AU-C §530, 

Audit

Sampling

•

AICPA 

Codification of Statements on

Auditing Standards

, AU-C §935,

Compliance Audits

, ¶.A21

A

VAILABLE

R

EFERENCE

 M

ATERIAL

•

AICPA 

Audit Guide

, 

Audit Sampling

•

AICPA Audit Guide, 

Government Auditing

Standards and Single Audits

A

VAILABLE

R

EFERENCE

 M

ATERIAL

•

The Institute of Internal Auditors (IIA)

International Professional Practices

Framework (IPPF)

•

IPPF Practice Guide, Assisting Small

Internal Audit Activities in Implementing

the Standards

IIA S

MALL

 I

NTERNAL

 A

UDIT

A

CTIVITIES

•

One to five auditors

•

Productive hours less than 7,500/year

•

Limited co-sourcing or out-sourcing

IPPF S

TANDARDS

 §2310

•

Auditors must identify sufficient, reliable,

relevant, and useful information to achieve

the engagement’s objectives.

IPPF I

MPLEMENTATION

 G

UIDANCE

•

If internal auditors choose to select a

sample, they are responsible for applying

methods to assure that the sample selected

represents the whole population and/or time

period to which the results will be

generalized.

IPPF P

RACTICE

 G

UIDE

•

External audit requirements for sampling

may not correspond to that of the internal

audit activity.

IPPF P

RACTICE

 G

UIDE

•

The Standards encourage collaboration with

external auditing and such collaboration

may include discussion of external audit

sampling and scoping parameters.

•

Such discussions in advance of audit plan

development and execution may allow for

greater reliance on the work of internal

auditing.

A

UDIT

 S

AMPLING

 D

EFINED

(AU-C §530.05)

•

The selection and evaluation of less than

100% of the population of audit relevance

such that the auditor expects the items

selected (the sample) to be representative of

the population and, thus, likely to provide a

reasonable basis for conclusions about the

population.

A

UDIT

 S

AMPLING

 D

EFINED

(AU-C §530.05)

•

Representative

•

Sample conclusions, subject to the limitations

of sampling risk, are similar to those that would

be drawn if the same procedures were applied

to the entire population.

S

AMPLING

 R

ISK

 D

EFINED

(AU-C §530.05)

•

The risk that the auditor’s conclusion based

on the sampling procedures may be

different than a conclusion based on

applying the same audit procedure to the

entire population.

S

AMPLING

 R

ISK

 D

EFINED

(AU-C §530.05)

•

Ineffective sample:

•

Concluding controls are 

more

 effective than

they actually are, or a material

misstatement/material noncompliance does not

exist when it does and is:

•

Likely to lead to an inappropriate conclusion.

•

Risk of overreliance

•

Potential audit failure

S

AMPLING

 R

ISK

 D

EFINED

(AU-C §530.05)

•

Inefficient sample:

•

Concluding controls are 

less

 effective than they

actually are or a material misstatement/material

noncompliance exists when it does not.

•

May result in additional work to determine

initial conclusions were incorrect.

•

Underreliance.

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

Auditor judgment is critical for

determining/assessing:

•

Risks (differs from the annual internal audit

risk assessments)

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

Auditor judgment is critical for

determining:

•

Materiality

•

Performance Materiality and Tolerable

Misstatement/Deviation

•

Immaterial Misstatements

•

Trivial Misstatements

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

However

•

“We judgmentally selected a sample of . . .”

•

“We judgmentally determined to use a sample

size of . . .”

•

May not be appropriate.

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

Auditor understanding of population:

•

Allows judgment when determining whether to

stratify the population before sampling, or

whether to use sampling at all.

•

Determining if population consists primarily of

significant transactions.

•

Using substantive analytical procedures to

provide sufficient appropriate audit evidence.

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

D

ATA

 A

NALYTICS

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

Acceptable sampling methods:

•

Random (statistical or nonstatistical)

•

Systematic (starting point, then every 

nth

 item

in the population)

•

Haphazard

A

UDITOR

 J

UDGMENT

IN

 S

AMPLING

•

Acceptable sampling methods include:

•

Monetary Unit Sampling (statistical) – used to

assess the amount of monetary misstatement

that may exist in a population.

R

ELEVANT

 A

SSERTIONS

(M

ANAGEMENT

’

S

 A

SSERTIONS

)

A

ND

 C

OMPLIANCE

(

E

.

G

., 

LAWS

AND

REGULATIONS

,

FEDERAL

PROGRAM

REQUIREMENTS

)

P

CARDS

/T

RAVEL

 C

ARDS

?

S

TUDENT

 F

INANCIAL

 A

ID

?

I

S

 S

AMPLING

 A

PPROPRIATE

•

Determine if sampling is most appropriate/

effective:

•

Homogeneous population

•

Control tests/substantive tests of details vs

substantive analytical procedures?

•

Data Analytics

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

Homogeneous Population:

•

College/university’s tuition and fee revenues totaled

$400 million for the fiscal year.

•

Enrollment = 30,000

•

Full-time COA = $7,500 per semester

•

Sample of 100 students = $750,000 (maximum)

Will a substantive analytical procedure provide

better, sufficient appropriate substantive evidence?

•

Homogeneous Population:

•

Testing controls (sampling) - effective or efficient?

•

Test of controls with a limited number of transactions

through/from the software application?

•

Fewer students?

•

More students?

•

Substantive analytical procedure?

•

Data analytics?

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

Homogeneous

 Population:

•

Substantive analytical procedure:

•

Published tuition and fee costs per credit hour (A)

•

Validate schedule - i.e., approved and agrees with the fee

tables in the software application

•

Student enrollment (B)

•

Break down enrollment by credit hours

•

Tolerable limit (C)

•

Acceptable difference from reported amount

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

Homogeneous

 Population:

•

A x B 

≈

 Tuition and fees on income statement.

•

Result within C (+ or -)?

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

Homogeneous Population:

•

Evaluate effectiveness:

•

Does the calculation provide evidence that the fee

schedule is applied to each student appropriately

(provides evidence that controls are effective)?

•

Is the calculated total within the expectation/

tolerable limit (substantive result)?

•

Does the calculation provide assurance that the reported

balance is materially correct?

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

Homogeneous Population:

•

If a fee is not correct in the IT system, is sampling

individual students necessary?

•

Conclude based on a substantive procedure

(calculation)?

•

A $10 fee charged per 

student

 is posted in the system as

$9.50 maximum error  = $15,000 (0.00375%).

•

A $10 fee charged per 

credit hour

 (360,000 hours) is

posted as $9.50.  Maximum error = $180,000 (0.045%).

•

Is either material?

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

•

PCard/Travel Expenses:

•

Not homogeneous

•

Airfare, hotel, car rental, food, incidentals, etc.

•

Sample?

•

Stratify population?

•

Combination of the two?

D

ETERMINING

IF

 A

UDIT

 S

AMPLING

I

S

 A

PPROPRIATE

FOR

THE

 T

ESTS

D

OCUMENTING

THE

 S

AMPLE

D

EFINING

THE

 O

BJECTIVE

OF

THE

S

AMPLE

•

The control or amount (substantive) of the

account balance, class of transactions, or

disclosure component being tested.

D

EFINING

THE

 O

BJECTIVE

OF

THE

S

AMPLE

•

Control

•

To determine whether key controls relating to

PCard/travel expenses are in place and

operating effectively

•

To support the preliminary risk assessment

(identified for each key control), and

•

To determine if the institution has complied with

applicable, material purchasing laws/regulations

D

EFINING

THE

 O

BJECTIVE

OF

THE

S

AMPLE

•

Substantive

•

To determine if PCard/travel expenses are

materially correct.

D

EFINING

THE

 P

OPULATION

FROM

W

HICH

THE

 S

AMPLE

 I

S

 T

AKEN

•

Control/Substantive or Both

•

The population is all PCard/travel expenses

recorded in the general ledger.

D

EFINING

THE

 P

OPULATION

FROM

W

HICH

THE

 S

AMPLE

 I

S

 T

AKEN

•

Control/Substantive or Both

•

Transactions to be included in the population:

•

All expenses coded to travel accounts (e.g., GL

accounts 7451-7460).

•

PCard transactions processed through the issuing

bank.

D

EFINING

THE

 P

ERIOD

 C

OVERED

FOR

THE

 P

OPULATION

•

Control/Substantive or Both

•

All expenses as defined above included in the

period from July 1, 2021, through June 30,

2022.

C

ONSIDERATION

OF

THE

C

OMPLETENESS

OF

THE

 P

OPULATION

•

Control/Substantive or Both

•

Reconcile detail expense transactions processed

through AP to the GL accounts.

•

Reconcile PCard transactions to bank

statements.

D

EFINING

THE

 S

AMPLING

 U

NIT

•

Control/Substantive or Both

•

Any individual transaction from the population

from which the sample will be taken.

I

DENTIFICATION

OF

 I

NDIVIDUALLY

S

IGNIFICANT

 I

TEMS

 (S

TRATIFYING

)

•

Substantive (using the same population for

control testing)

I

DENTIFICATION

OF

 I

NDIVIDUALLY

S

IGNIFICANT

 I

TEMS

 (S

TRATIFYING

)

•

Substantive

•

Justify the portion of the population not tested:

•

All transactions <= $100 totaled $3,750, 0.375% of

dollars and 25% of transactions

•

Transactions in total are less than the level of

immateriality and represent little to no risk of

material misstatement for the population as a whole.

D

EFINING

THE

 E

RROR

•

Deviation condition (control) or error

(substantive - i.e., misstatement/exception):

•

An answer of “X” to any of the control/

substantive attributes listed at the test.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Control test:

•

Determine confidence level and risk of

overreliance – a 95% confidence level results in

a 5% risk of overreliance (¶¶3.38-3.61).

•

Determine the tolerable deviation rate

(normally expressed as a percentage).

•

Determine the expected deviation rate – the

higher the rate, the larger the sample.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Control test:

•

If practicable, use the 

Audit Guide

 Table A-1.

•

Adjust the base sample size if the population is

small or if operating controls are infrequent.

•

See the 

Audit Guide

, Tables 3-4 and 3-5 for

sample sizes for small populations.  Note that

for populations of 4 - 52, Table 3-5 suggests

sample sizes (requires sample to be

documented).

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine the objective:

•

To determine if PCard/travel expenses are materially

correct.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine the RMM.

•

Determine analytical procedures risk:

•

Will substantive analytical procedures provide

sufficient appropriate audit evidence?

•

Will substantive analytical procedures reduce

sample sizes?

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Substantive analytical procedures as a primary

test generally aren’t appropriate for non-

homogeneous populations.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details - Payroll:

•

Determine analytical procedures risk:

•

Payroll analytical procedures:

•

Divide employer contributions to pension plans by

the contribution rate to determine covered payroll,

and then compare the results to total payroll.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details - Payroll:

•

Determine analytical procedures risk:

•

Analytical procedures may not be appropriate for:

•

Testing time and attendance; or

•

Census data testing for pensions and OPEB.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine tolerable misstatement as a

percentage of the population subject to testing.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Materiality:

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine tolerable misstatement as a

percentage of the population subject to testing

(7.5% above)

•

Note in the previous table that tolerable

misstatement for the account tested is

$7,500,000 and 

not

 $75,000,000

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

In the above, if the population tested is the $100

million, and if financial statement tolerable

misstatement is used, then the auditor is saying

that if a $74,999,999 error is identified, the

population is materially correct because the

error does not exceed the tolerable

misstatement.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine the expected misstatement as a

percentage of the population subject to

sampling:

•

The greater this percentage, the larger the sample.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Determine the base sample size (

Audit

Sampling

, Table 4-5 for one example).

•

In dual-purpose testing for control and

substantive tests, if the auditor has calculated a

sample size of 60 for control and 80 for

substantive testing, the sample size should be

80 (

Audit Sampling

, ¶2.13).

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size due to variation and

stratification:

•

Variation 

- 

effectiveness of the sample.

•

Not removing low dollar items, or not testing high

dollar 

- efficiency and effectiveness, respectively

.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size due to variation and

stratification:

•

For example, pulling a sample for a population of $1

million results in a sample whose total dollars were

$21,000.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size and/or approach due to

variation and stratification.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size due to variation and

stratification:

•

The 15 items (individually significant) tested are not

part of the sample

•

If calculated sample size is 80, then 95 transactions

are tested.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size due to variation and

stratification:

•

However, stratification of the population can reduce

sample size below 80 if the stratification reduces

audit risk.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

Adjust the sample size due to the number of

items in the population:

•

See 

Audit Sampling

, Tables 3-4 and 3-5 for sample

sizes for smaller populations.

D

ETERMINING

 S

AMPLE

 S

IZE

•

Substantive test of details:

•

State the final sample size.

P

ERFORMING

THE

 T

EST

(

S

)

•

Substantive test of details:

•

Describe how the sampling procedure was

performed.

E

VALUATING

THE

 S

AMPLE

 R

ESULTS

•

Number of deviations in the sample

(control) or the misstatement identified in

the sample (substantive).

•

Deviations or misstatement as a percentage

of the sample.

•

Project deviations or misstatement to the

population as a whole.

E

VALUATING

THE

 S

AMPLE

 R

ESULTS

•

Consider the qualitative aspects of sample

results:

•

Nature and cause (fraud or error).

•

Whether results lead to a conclusion that the

sample is not representative of the population.

E

VALUATING

THE

 S

AMPLE

 R

ESULTS

•

Consider the qualitative aspects of sample

results:

•

Is there a broader impact of the deviations/

misstatements from:

•

Management override of controls?

•

Identified weaknesses in an IT application that

processes other transactions?

E

VALUATING

THE

 S

AMPLE

 R

ESULTS

•

Consider sampling risk:

•

If sampling risk is high, the projected

deviations or misstatement may exceed the

expected error.

•

Determine if the sample should be expanded

(

Audit Sampling

, ¶¶3.82-.84 and 4.95-.100).

•

Determine if evidence supports writing a

finding.

E

VALUATING

THE

 S

AMPLE

 R

ESULTS

•

Consider sampling risk:

•

Are deviations and/or misstatements

determined to be isolated or anomalies, or are

they pervasive throughout the population?

•

Do compensating/mitigating controls exist, are

they operating as designed, and do they reduce

risk to an acceptable level?

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

AICPA 

Audit Guide

, 

Government Auditing

Standards and Single Audits

, Chapter 11

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Procedures not involving audit sampling:

•

Inquiry and observation

•

Analytical procedures

•

Testing 100% of a population

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Procedures not involving audit sampling:

•

Individually important items

•

Obtaining an understanding of and testing

operating effectiveness of controls over

compliance

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Sampling steps/procedures are similar to

those for control tests and substantive tests

of details.

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Additional steps include:

•

Considering multiple major programs:

•

Where controls for a type of compliance

requirement are the same for more than one major

federal program, the transactions for the programs

may be combined into one population for testing

(Audit Guide ¶11.42).

•

e.g., CARES Act/HEERF consist of numerous

programs/CFDAs that can be tested together.

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Additional steps include:

•

Considering multiple organizational units (e.g.,

colleges/universities with multiple campuses):

•

Are separate internal controls used – i.e., do controls

vary significantly?

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Additional steps include:

•

Considering clusters of programs (Student

Financial Assistance Cluster of Programs):

•

Sufficient appropriate audit evidence is to be

gathered for the compliance requirements for cluster

programs as a whole (i.e., treat the cluster as one

program).

•

i.e., is the sample representative of the program

cluster as whole?

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Dual purpose testing opportunities may

exist:

•

Allowable costs and cost principles.

•

Evidence that the recorded amount, account,

and period are correct for financial reporting.

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

See 

Audit Guide

 Table 11-1 for control

testing sample sizes based on the RMN.

•

Sample sizes in the table range from 25 to 60.

•

Inherent risks that may be applicable are at

¶11.67.

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

See 

Audit Guide

 Table 11-2 for compliance

testing sample sizes based on the desired

level of assurance for the remaining RMN,

with high, moderate, low risks and no

expected deviations.

•

Sample sizes in the table range from 25 to 60.

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Audit Guide

 small sample sizes:

•

Quarterly

   (2)

•

Monthly

(2-4)

•

Semimonthly

(3-8)

•

Weekly

(5-9)

S

AMPLING

FOR

 U

NIFORM

 G

UIDANCE

C

OMPLIANCE

 A

UDITS

•

Documenting the sample is critical.

•

Identifying specifically what controls are

tested is critical.

•

Auditor flexibility is limited when

determining if a finding is to be written:

•

$25,000 in 

known or 

likely

 questioned costs

•

Exception rate over the tolerable deviation

P

OINTS

OF

 E

MPHASIS

V

OIDED

 D

OCUMENTS

•

If a transaction has been properly voided

and does not represent a deviation from the

prescribed control:

•

Replace the voided transaction.

•

If simple random sampling is used, should

match a replacement random number with the

appropriate transaction (¶3.67).

U

NUSED

OR

 I

NAPPLICABLE

 D

OCUMENTS

•

Unused document number included in the

sample

•

Obtain evidence the number actually represents

an unused document and is not a deviation.

•

Replace the document number with an

additional document number.  (¶3.68)

U

NUSED

OR

 I

NAPPLICABLE

 D

OCUMENTS

•

Deviation defined as a transaction not

supported by a receiving report:

•

A utility expense selected not expected to be

supported by a receiving report.

•

Replace the item with another transaction.

(¶3.68)

V

ARIATION

IN

THE

 P

OPULATION

•

Stating there is little to no variation is

generally not appropriate in a substantive

test of details.

•

The auditor is required to understand the

population and the audit objectives before

sampling begins.

V

ARIATION

IN

THE

 P

OPULATION

(S

UBSTANTIVE

 T

ESTS

OF

 D

ETAILS

)

•

Variation example:

•

The transaction amounts range from $10 to $1,000,000.

To address the variation in the population, we have

removed the transactions below $100 (50% of the

transactions total $XX, which is below immateriality)

•

We will remove transactions above significance (or

above a certain dollar amount) and will test 100% of

the transactions.  These transactions total 50% of the

population dollars.

V

ARIATION

IN

THE

 P

OPULATION

•

Using monetary unit sampling (MUS) may

be more efficient than stratifying or

increasing the sample size.

V

ERIFYING

THE

 C

OMPLETENESS

OF

THE

P

OPULATION

•

Completeness of a population when

agreeing detail transactions to the general

ledger (GL) or financial statements is easier

than circumstances where other sources

must be used.

V

ERIFYING

THE

 C

OMPLETENESS

OF

THE

P

OPULATION

•

Number of employees

•

Payroll compared to organization charts in

smaller organizations.

•

HR employee lists compared to the payroll

transaction population to ensure employees are

active.

•

Conversely, determine if transactions represent

active employees reported by HR.

V

ERIFYING

THE

 C

OMPLETENESS

OF

THE

P

OPULATION

•

Number of tickets:

•

When prenumbered, obtain used and partially

used ticket books to determine a population

size.

•

For an event:

•

Number of seats in an arena/auditorium

compared to tickets sold and attendance

records.

U

NIFORM

 G

UIDANCE

S

IGNIFICANCE

OF

THE

 C

ONTROL

(

S

)

•

Significance of the control is to be

determined for 

each

 control and 

not

overall

for the compliance 

requirement

.

U

NIFORM

 G

UIDANCE

I

NHERENT

 R

ISK

•

Inherent risk is to be assessed for 

each

compliance requirement tested and 

not

overall

 for the program.

R

ESULTS

, C

ONCLUSIONS

, 

AND

A

DDITIONAL

 P

ROCEDURES

•

Determinations of the sample size for tests

of controls and the sample size for

substantive tests of details and compliance

tests are separate.

R

ESULTS

, C

ONCLUSIONS

, 

AND

A

DDITIONAL

 P

ROCEDURES

•

The results of the tests of controls and substantive

tests of details/compliance tests can impact each

other:

•

Deviations in tests of controls may impact substantive/

compliance tests.

•

If the auditor does not expand tests of controls, thereby

increasing the assessed level of control risk (writing a

finding), the original sample sizes for substantive or

compliance tests may no longer be appropriate.

D

ATA

 A

NALYTICS

D

ATA

 A

NALYTICS

•

Can be used to test 100% of a population.

•

Advantages:

•

NO SAMPLING!

•

Gives the auditor a much deeper understanding

of populations and data.

•

Can save significant time.

•

Significantly more reliable results.

D

ATA

 A

NALYTICS

•

Advantages:

•

Easier to adjust criteria where the output is not

what the auditor is looking for.

D

ATA

 A

NALYTICS

•

Comparing different data sets:

Vendor Bank

Accounts

Employee Bank

Accounts

D

ATA

 A

NALYTICS

•

Coverage for areas that may not have been

given attention in previous work.

•

May apply to federal program compliance:

•

Student Financial Assistance:

•

All students – aid posted to student accounts within

the required timeframe.

•

Maximum award by academic years and aggregated

for lifetime limitation.

D

ATA

 A

NALYTICS

•

Highlighting anomalies within a population.

•

Finding errors in data sets.

•

Stratifying data.

D

ATA

 A

NALYTICS

 S

OFTWARE

•

Multiple options for visualizing data:

•

Spreadsheets and schedules

•

Line graphs

•

Charts

•

Reports

•

Data extraction based on predetermined

criteria.

D

ATA

 A

NALYTICS

 S

OFTWARE

•

Inputs from a variety of sources:

•

Excel

•

Power BI datasets

•

SQL

•

Text/CSV

•

Etc.

D

ATA

 A

NALYTICS

•

May not be practical for all tests:

•

Federal Program compliance tests – Student

Financial Assistance (e.g., eligibility).

•

Providing a substantive assurance in a non-

homogeneous population.

•

PCard tests

D

ATA

 A

NALYTICS

 S

OFTWARE

•

Microsoft Excel:

•

Pivot Tables

•

Filtering

•

V-Lookups

•

Sampling

D

ATA

 A

NALYTICS

 S

OFTWARE

•

Microsoft Power BI ($9.99/mo/user):

•

Parish libraries may have instructional videos

free to members

•

Data from numerous sources

•

Identify relationships

•

Provides measures

•

What ifs

D

ATA

 A

NALYTICS

 S

OFTWARE

•

TeamMate Analytics

•

Provides statistics

D

ATA

 A

NALYTICS

 S

OFTWARE

•

TeamMate Analytics

•

Visualizes data

D

ATA

 A

NALYTICS

 S

OFTWARE

•

TeamMate Analytics

•

Join

•

Append

•

Split

•

Data extraction

•

Regression analysis

•

Sampling

D

ATA

 A

NALYTICS

 S

OFTWARE

•

Other products based on a Google search:

•

Tableau - $70/mo/user

•

Qlik ($30/mo/user

•

Looker (Contact vendor)

•

Klipfolio (Free)

•

Zoho (Free)

•

Domo (Contact vendor)

https://www.forbes.com/advisor/business/software/best-data-analytics-tools/

C

ONCLUSION