In auditing, nonstatistical sampling can be acceptable under which condition?

In auditing, nonstatistical sampling can be acceptable under which condition?

• A. Nonstatistical sampling is appropriate when the objective is to quantify sampling risk.
• B. Nonstatistical sampling is appropriate when the population is homogeneous and the auditor uses formal sampling plans.
• C. Nonstatistical sampling is appropriate when the objective is to form impressions rather than quantify sampling risk.
• D. Nonstatistical sampling is never appropriate.

Answer: (C)

The main idea is that nonstatistical sampling is used when you don’t need to quantify sampling risk or project misstatements to the whole population. If the goal is to form an overall impression about the population—such as judging whether controls appear to be operating effectively or whether the account balance looks reasonable—professional judgment guides the sampling and you don’t rely on statistical calculations to measure precision. In these cases, you’re not trying to estimate an exact misstatement or attach a numerical confidence interval to your conclusion, so the statistical framework isn’t required. If the objective were to quantify sampling risk or to project misstatements, statistical methods would be appropriate. Saying there’s never a use for nonstatistical sampling ignores legitimate scenarios where qualitative conclusions are sufficient.

The main idea is that nonstatistical sampling is used when you don’t need to quantify sampling risk or project misstatements to the whole population. If the goal is to form an overall impression about the population—such as judging whether controls appear to be operating effectively or whether the account balance looks reasonable—professional judgment guides the sampling and you don’t rely on statistical calculations to measure precision. In these cases, you’re not trying to estimate an exact misstatement or attach a numerical confidence interval to your conclusion, so the statistical framework isn’t required. If the objective were to quantify sampling risk or to project misstatements, statistical methods would be appropriate. Saying there’s never a use for nonstatistical sampling ignores legitimate scenarios where qualitative conclusions are sufficient.