AUD · Performing Further Procedures and Obtaining Evidence · 6 practice questions
Systematic vs. Block vs. Stratified Sampling
Systematic selection uses a random start and fixed interval; block takes consecutive items; stratified sampling selects separately within defined groups. Sort the alternative ways an auditor selects Cedar's invoices for a full-year credit approval test.
Try one first
Hint
Focus on the selection pattern after the first item is chosen, not just on the fact that the starting point was random.
Answer C. Systematic sampling selects items using a fixed interval after a random start (for example, every 25th invoice). The auditor's method of picking a random start and then selecting every nth item is the definition of systematic sampling; the assumption about no hidden pattern addresses the primary risk (periodicity) of this method.
Why not A: Although the auditor used a random starting point, simple random sampling would involve selecting items entirely by chance (e.g., using a random number generator) without applying a fixed interval. The presence of a fixed interval (every 25th) makes this systematic, not purely random, sampling.
Why not B: Block sampling involves selecting contiguous sequences or blocks of items (for example, a whole day's invoices or a consecutive run of records). Selecting every 25th invoice produces evenly spaced selections throughout the population, not contiguous blocks, so it is not block sampling.
Why not D: Haphazard sampling is an unstructured, nonstatistical method with no consistent selection pattern and often depends on auditor judgment. Because the auditor used a defined random start and a consistent interval, the selection is structured and therefore not haphazard.
Sort it
Select every nth item from one population list after a random start.
Select a consecutive run of items or all items from a short, continuous period.
Divide the population into meaningful groups and select a separate sample within each group.
| Item | Goes to |
|---|---|
| For Cedar's full-year credit approval test, use its complete list of 2,000 sales invoices. Randomly choose a start among the first 20, then select every 20th invoice. | Systematic selectionThe auditor uses a random start followed by a fixed selection interval. |
| Select all invoices issued during the first three business days of March because they are stored together. | Block selectionThe auditor takes one continuous time block. Easy retrieval does not establish full-year coverage. |
| Divide invoices into low, medium, and high dollar groups. Randomly select invoices separately from each group. | Stratified samplingDollar groups are formed before selection, and each group has its own sample. |
| Sort the full-year invoice list by dollar amount. Choose a random start among the first 50 invoices, then select every 50th invoice. | Systematic selectionSorting by amount does not create separate group samples. Selection still follows one fixed interval. |
| Randomly choose a starting invoice, then select that invoice and the next 49 invoices. | Block selectionA random starting point does not change a consecutive run into systematic selection. |
| Divide the year's invoices into quarters. Randomly select invoices separately within each quarter. | Stratified samplingEach quarter is a separate sampling group. Time-based grouping is not the same as taking one short block. |
| Arrange the full-year invoices by date. Choose a random start among the first 100 invoices, then select every 100th invoice. | Systematic selectionDate order does not make the sample a block. The auditor skips records at a fixed interval. |
| Select all invoices from one ordinary week in December to support the full-year conclusion. No unusual activity occurred that week. | Block selectionAn ordinary week is still a short block. Normal activity does not establish that it represents the entire year. |
| Divide invoices into dollar groups, then use a separate random start and fixed interval within each group. Classify the overall design. | Stratified samplingThe overall design is stratified because selection occurs separately by group. The selection method within each group is systematic. |
| Divide invoices by customer, then randomly select invoices separately within every customer's group. | Stratified samplingStrata need not use dollar amounts. Separate selection within the customer groups is the deciding fact. |
Key points
- Remember: Classify how items are selected, not how the list is sorted.
- A random starting point alone does not make a consecutive run systematic.
- A convenient block may miss conditions elsewhere in the year.
How the exam traps you
- Calling every selection from an ordered list systematic. Look for a random start and fixed interval, not a consecutive run.
- Calling a dollar-sorted list stratified. Require separate sampling within each dollar group.
- Treating an ordinary week as representative of the year. A narrow block can miss conditions elsewhere in the year, even without known unusual activity.
Question 2
Hint
Identify the feature that happens before selection: Is the auditor choosing a sequence, every nth item, or dividing the population into subgroups?
Answer A. This is stratified sampling because the auditor first divides the population into subgroups with similar characteristics (here by recorded dollar amount) and then samples within each subgroup. Stratification improves sampling efficiency and focuses attention on more significant items; the key feature is the intentional division of the population before selection.
Why not B: This distractor is tempting because the population is being separated into groups, but block sampling (block selection) typically refers to selecting a contiguous sequence of items (for example, consecutive invoices or transactions from a specific period). The auditor here divides by dollar amount rather than choosing a continuous block.
Why not C: Haphazard selection means choosing items without a structured technique and can introduce bias. Because the auditor deliberately divides the population into strata and plans selection within each, the approach is structured and not haphazard.
Why not D: Systematic selection means choosing every nth item after a random start. The stem describes dividing the population into subgroups before sampling, not interval-based selection, so it is not systematic.
Question 3
Hint
Separate errors caused by chance in a sample from errors caused by defining the wrong population or using the wrong sampling frame.
Answer A. This is a nonsampling risk because the problem arises from using an incomplete sampling frame that omitted items that should have been included. Sampling risk refers to uncertainty from chance selection when testing a correctly defined population; here the population itself was wrong. The root cause is a frame/selection error (nonsampling), not chance variation in the sample.
Why not B: Tempting because candidates often equate any unrepresentative result with sampling risk; however sampling risk presumes the population was correctly defined and the issue is chance variability in the sample. The core problem here is an incomplete population (a frame error), which is nonsampling risk.
Why not C: This describes a plausible consequence, the auditor might overrely on the control, but it is a consequence, not the classification of the error. The primary classification is nonsampling risk due to the incomplete frame.
Why not D: This is incorrect both on likely effect and on classification: if omitted items have more deviations the practical effect would be underestimating control deficiencies (assessing control risk too low), not too high. Regardless, the fundamental issue is the incomplete sampling frame (nonsampling).
Question 4
Hint
Focus on which method is most likely to overrepresent one part of the population instead of the entire year.
Answer C. This is block selection because the sample is concentrated in a short period rather than spread across the population. A one-week block near year-end may not represent how the control operated during the rest of the year, so it is the least appropriate choice when the auditor wants evidence for the full-year population.
Why not A: Random selection helps avoid selection bias and gives items throughout the year a chance of selection; it supports a conclusion about the full-year population better than a concentrated block.
Why not B: With a random start, systematic selection typically spreads selections across the population (unless an unusual periodic pattern exists), making it more representative of the year than selecting a single week.
Why not D: Although less formal, haphazard selection can be acceptable in nonstatistical sampling if the auditor consciously avoids bias and ensures coverage across the full year; it is still preferable to concentrating on one week.
Question 5
Hint
Focus on whether the auditor selected items throughout the population or pulled a consecutive group from one part of it.
Answer D. Selecting a group of consecutive items from the same time period is block sampling (a form of convenience sampling). Block samples can be efficient to retrieve, but they may not be representative of the full population because adjacent items often share timing-related or operational characteristics. For that reason, auditors prefer random or properly designed systematic selection when representativeness is important.
Why not A: Random sampling requires that each item has a known, usually equal, chance of selection via a random process. Simply choosing the first three business days in advance does not give all invoices a known or equal chance of selection and therefore is not random sampling.
Why not B: Systematic sampling involves selecting every nth item across the whole population, typically after a random start. Choosing all items from a short consecutive period is a block selection, not true systematic sampling.
Why not C: Haphazard sampling is an unstructured, nonrandom selection without a deliberate pattern. Deliberately choosing the first few business days is a patterned choice (a block), so the issue is method rather than intent.
Question 6
Hint
Separate two issues: what is being tested, and whether sampling risk is being quantified.
Answer A. Because the auditor is testing a control, attribute sampling is the appropriate type of sampling. Although the sample was selected systematically (random start and fixed interval), the approach is nonstatistical because the auditor is not using probability-based sample-size determination or statistical evaluation (such as computing an upper deviation rate) to quantify sampling risk.
Why not B: This is tempting because of the random start, fixed interval, and the mention of expected and tolerable deviation rates. However, statistical sampling requires using probability theory to determine sample size and to evaluate results quantitatively; the engagement team explicitly is not doing that.
Why not C: Interval selection can suggest monetary‑unit (PPS) sampling, but MU sampling is a substantive, dollar‑based technique focused on monetary misstatement. The stem specifies a control test (attribute sampling), not a substantive dollar‑amount test.
Why not D: Block sampling selects contiguous groups of items (e.g., invoices 501-560). Selecting every 62nd item after a random start is systematic selection, not block sampling; a sequential listing alone does not make the method a block sample.
Common questions
What is the difference between systematic and block sampling?
Systematic selection takes every nth item after a random start. Block selection takes consecutive items, even if its starting item was chosen randomly.
Is sorting invoices by dollar amount stratified sampling?
No. Stratification requires dividing invoices into groups and selecting separately within each group; one fixed-interval selection from a dollar-sorted list is still systematic.
Is sampling invoices from each quarter block or stratified sampling?
It is stratified when each quarter is a separate group and the auditor selects a sample within each. Taking all invoices from one short period is block selection.
Watch it solved
A full CPA AUD task-based simulation on Sampling Techniques, worked step by step.
AUD Simulation: Attribute Sampling, Deviations and Upper Limit on YouTube
Practice AUD like the real exam
The free ChatCPA simulator: real exam layout, timed testlets, starting with a question on this topic. No account needed to start.
Open the free simulator →More on Sampling Techniques
Questions from the ChatCPA bank of 17,658 CPA exam questions, each with a written reason for every wrong answer. ChatCPA is built by Nicholas Miller, CPA (Oregon #14907). How these pages are made. Spot an error? Tell us.