AUD 5 - Audit Sampling

Question 1

Audit Sampling

Answer 1

Sampling refers to the examination of less than 100% of a population & using the results as a basis for drawing a general conclusion on the entire population.

Question 2

Nonsampling Risk

vs.

Sampling Risk (2)

Answer 2

Nonsampling Risk - The risk that human error will lead to drawing the wrong conclusion. (Human error, misinterpreting audit test results, not recognizing misstatments in documents audited)

Sampling Risk - The risk of drawing the wrong conclusion based on a bad sample. Two types of sampling risks:

1. Type I - Efficiency Error
   
   • Population is OKAY, but sample is BAD
     
     • Underrely on I/C = High RMM assessment
     • Incorrectly Reject an accont balance for substantive testing purposes.
2. Type II - Less Effectiveness
   
   • Population is BAD, but sample is OKAY
     
     • Overrely on I/C = Low RMM assessment
     • Incorrectly Accept an account balance for substantive testing

Question 3

There are two times when audit sampling are used, when are they?

Answer 3

1. During the internal control phase of the audit, the auditor will perform test of controls on a sample basis to determine the operating effectiveness of controls they plan to rely on. (called Attribute Sampling)
2. During the substantive testing phase of the audit, auditor will perform test of details details (I-CORRIIA) of transactions , accounts & disclosures on a sample basis to obtain sufficient appropriate audit evidence to support managenent assertions. (called Variable Estimation Sampling)

Question 4

What are two ways sample size is determined?

_{Statistical & Non-statistical (Judgemental)}

Answer 4

Statistical Sampling - refers to the use of quantitative measures of the auditor s taking in the use of sampling. Formulas** are **used** to **determine** the **sample size necessary.

Nonstatistical Sampling - (Judgemental Sampling) a method of sampling under which the auditor applies judgement to determine sample size & to interpret sample results.

• Auditors using non-statistical sampling tends to overestimate the needed sample size. Therefore, audit tends to be inefficent despite being effective.

Question 5

Sampling Risk Summary

(Internal Ctrl vs. Substantive Testing)

Answer 5

Internal Control: Attribute Sampling (Characteristics)

• Assessing RMM HIGH (under rely) = Inefficient
• Assessing RMM LOW (over rely) = Ineffective

Substantive Testing: Variable Sampling ($ Amounts)

• Incorrect Rejection = Inefficient
• Incorrect Acceptance = Ineffective

Question 6

Attribute Sampling

(two risks,sample size,evaluation)

Answer 6

• Attribute sampling is for Internal Controls
• Two Risks
  
  • Assessing RMM too Low = Audit ineffective (T2)
  • Assessing RMM too High = Audit inefficient (T1)
• Factors affecting Sample Size (Tested)
  
  • Tolerable Rate - Inverse effect
  • Expected Error Rate (Deviation) - Direct effect
  • Acceptable Risk (Allowance) = Inverse effect
  • Population Size = Little or No effect
• Evaluation of Sample Results
  
  • Calculate Sample Deviation Rate
    
    • Deviation Rate = # of errors / sample size
  • Calculate Maximum Deviation Rate (MDR)
    
    • MDR = Sample Deviation + Allowance
  • Compare Maximum Deviation Rate vs. Tolerable Rate
    
    • Maximum Deviation Rate must NOT exceed Tolerable rate or else Auditor must modify planned reliance

Question 7

Attribute Sampling

What are the two risks associated with Attribute Sampling?

Answer 7

• Attribute sampling is for Internal Controls
• Two Risks
  
  • Assessing RMM too Low = Audit ineffective (T2)
  • Assessing RMM too High = Audit inefficient (T1)

Question 8

Attribute Sampling

What are factors affecting sample size? (TEA)

Answer 8

• Attribute sampling is for Internal Controls

​

• Factors affecting Sample Size (Tested)
  
  • Tolerable Rate - Inverse effect
  • Expected Error Rate (Deviation) - Direct effect
  • Acceptable Risk (Allowance) = Inverse effect
  • Population Size = Little or No effect

Question 9

Attribute Sampling

How to evaluate sample results?

Answer 9

• Attribute sampling is for Internal Controls
• Evaluation of Sample Results
  • Calculate Sample Deviation Rate
    
    • Deviation Rate = # of errors / sample size
  • Calculate Maximum Deviation Rate (MDR)
    
    • MDR = Sample Deviation + Allowance
  • Compare Maximum Deviation Rate vs. Tolerable Rate
    
    • Maximum Deviation Rate must NOT exceed Tolerable rate or else Auditor must modify planned reliance

Question 10

Attribute Sampling

Sampling Methologies (4)

_{(Random Number,Systemic,Haphazzard,Block)}

Answer 10

A sample, by definition, should be representative of the population under consideration. For a sampling method to be valid, all items in the population should have an equal opportunity to be selected. Such methods include:

1. Random-Number Sampling - numbered documents or transactions are selected through the use of random number tables or computer software.
2. Systemic Sampling - every “Nth” item is selected from a randomly-distributed population from a randomly-selected starting point.
3. Haphazard Sampling - a sample consisting of units selected without any conscious bias - again assuming the random distribution of the population.
4. Block Sampling - a sample consisting of contiguous units, example: a selection of three blocks of ten vouchers each.

Question 11

Variable Sampling

What are the two risks?

What affects Sample Size? TAPES

Answer 11

• Substantive Testing - Test of Details
• Two Risks:
  
  • Incorrect Acceptance = Audit Ineffective
  • Incorrect Rejection = Audit Inefficient
• Factors affecting Sample Size (TESTED)
  
  • Tolerable misstatement - Inverse
  • Accetance Risk - Inverse
  • Population Size - Direct
  • Expected Misstatement - Direct
  • Standard Deviation - Direct
  • RMM - Direct

Question 12

Variable Sampling & Sample Size Formula

What makes sample size increase or deacrease? TAPES

TESTED

Answer 12

To be more Precise, Sample Size (SS) will need to increase.

• Tolerable Rate = Inverse
• Acceptable Risk = Inverse
• Population Size = Direct (very little)
• Expected Misstatements/Errors = Direct
• Standard Deviation = Direct
• Risk of Material Misstatements = Direct

SS = (Std Deviation + Reliability + Population) / Allowance

Question 13

Types of Classical Variables Sampling (4)

_{(MPU,difference estimation,ratio estimation,stratified)}

Answer 13

• Mean-per-Unit
  
  • MPU = Average Actual Audit Results x Total Population
• Difference Estimation
  
  • DE = ((Actual - Sample) x Total Population) + Base
• Ratio Estimation
  
  • RE = Avg Ratio x Population
  • The use of the ratio estimation sampling technique is most effective when the calculated audit amounts are approximately proportional to the client’s book amounts.
• Stratified Sampling
  
  • ​Population is divided into groups
  • Sample size is drawn from each group

Question 14

Probability Proportional to Size Sampling (PPS)

Answer 14

PPS - A sampling plan under which items exceeding a certain dollar amount (sampling interval) are always included in a sample &, for remaining items, the greater the amount, the greater the likelihood of being included in the sample.

Calculations Needed:

1. Sampling Interval (SI): 2 Ways to Calculate
   
   • SI = Tolerable Misstmt / Reliability Factor (from Table)
   • SI = Population Amount / Sample Size
2. Sample Size = Population Amount / Sampling Interval

To Determine Projected Misstatements (PM):

• Misstatement = Book (Recorded) Amt - Audited Amt
• Tainting Factor (TF%) = Misstatement / Book Amount
• Projected Misstatements (PM) - Audit Adjustment
  
  • If SI > Book (Recorded) Amount
    
    • PM = TF% * SI
  • If SI < Book (Recorded) Amount
    
    • PM = Book Amount - Audited Amount

Question 15

PPS Advantages (4)

vs.

PPS Disadvantages (2)

Answer 15

Advantages of PPS over classical variable are:

• Standard deviation is NOT needed
• Stratifies sample automatically
• Smaller sample usually results in few errors expected
• Able to start sampling without having entire population available or finished

Disadvantages of PPS over classical variable are:

1. NOT Useful to detect understatement
2. Zero & negative balances require special handling