Sampling is a cornerstone methodology in auditing that students prepping for the US CPA exams need to grasp thoroughly. Simply put, sampling is the practice of reviewing a subset of transactions or data points from a much larger set to draw general conclusions about the entire dataset. This technique is critical because auditing every single transaction or data point is often impractical or cost-prohibitive. Sampling is covered extensively in the Auditing and Attestation (AUD) section of the CPA exams.

There are several types of sampling methods that CPA candidates should be familiar with, including but not limited to:

1. Random Sampling: Each item has an equal chance of being chosen.

2. Haphazard Sampling: A non-random method, where items are chosen arbitrarily.

3. Stratified Sampling: The population is divided into subgroups, and samples are taken from each.

4. Block Sampling: Consecutive groups of items are tested.

Example: Imagine you are auditing a large retail store with 20,000 transactions per month. Instead of reviewing all 20,000 transactions (a daunting and inefficient task), you opt for random sampling and select 400 transactions to inspect for irregularities. Based on the results, you can form a qualified opinion about the accuracy of the entire month's transactions. If two out of the 400 transactions contain errors, you might project that roughly 0.5% of all transactions could be erroneous, which would then inform your audit procedures moving forward.