Abstract
Current audit standards sanction the use of both statistical and nonstatistical sample-selection methods, and suggest that the method chosen should depend on relative cost and effectiveness. Anecdotal evidence and results of a recent survey of practicing auditors indicate that the majority of audit samples are nonstatistical, with haphazard sampling being the method of choice in most circumstances. However, two recent research studies confirmed that haphazard sampling is susceptible to selection bias and therefore may not yield representative samples. The tendency of haphazard sampling to yield biased selections appears to result from subconscious human behavior in the areas of visual perception and performance of tasks requiring physical effort. Given the closer scrutiny that auditors are experiencing in the new regulatory environment, and the apparent difficulty individuals have selecting unbiased haphazard samples, auditors seeking a representative sample are advised to consider the use of random selection techniques.