Files in this item
|(no description provided)|
|Title:||A Critical Examination of Dollar Unit Sampling in the Audit of Accounts Receivable|
|Author(s):||Jenne, Stanley Earl|
|Department / Program:||Accountancy|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Business Administration, Accounting|
|Abstract:||This research examines three sample selection techniques used in conjunction with dollar unit sampling (DUS) and examines the behavior of five DUS bounds proposed in the accounting literature when applied to simulated accounts receivable populations. This research recognizes that large differences exist between the various populations which may be subjected to DUS and concentrates on accounts receivable populations. An analytical examination is made of the types of errors likely to occur in accounts receivable populations.
Prior published research regarding DUS has often employed the Poisson distribution to determine an upper limit on the number of errors in an auditing population. It has frequently been assumed that the use of the Poisson distribution was appropriate when random, systematic, or cell selection techniques have been employed. This research applies repeated sampling to four simulated accounts receivable populations and demonstrates that the achieved distributions of the number of errors discovered using systematic and cell selection are significantly different from a Poisson distribution under certain population assumptions.
Repeated sampling is employed to examine the behavior of five DUS bounds when applied to simulated accounts receivable populations. Test populations employed contain two different rates of error and five different distributions of relative error. Both the distribution of relative error and the population error rate are found to have an effect on the performance of each of the bounds examined. Two DUS bounds, introduced by Teitlebaum (1973) and McCray (1981), show consistently good performance in relation to the other bounds tested. A relatively small sample of size 50 is shown to produce upper bounds on monetary error which are extremely conservative. The magnitude of the effects of using a lower level of confidence is also illustrated.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1982.
|Date Available in IDEALS:||2014-12-16|