## Files in this item

FilesDescriptionFormat

application/pdf

8908762.pdf (10MB)
(no description provided)PDF

## Description

 Title: The Effects of Clustering and Response Mode Correspondence on the Evaluation of Internal Accounting Controls Author(s): Macur, Kenneth Michael Department / Program: Accountancy Discipline: Accountancy Degree Granting Institution: University of Illinois at Urbana-Champaign Degree: Ph.D. Genre: Dissertation Subject(s): Business Administration, Accounting Abstract: This dissertation studied the effects of clustering and response mode correspondence on the accuracy and consistency of the evaluation of internal accounting controls. 'Accuracy' was operationalized as the means of subjects' evaluations. 'Consistency' was operationalized as the variance of those evaluations. 'Clustering' relates to the physical organization of the facts or evidence concerning the problem or task. The 'response mode' is an instance of framing, and relates to the form of the presentation of the decision task. Correspondence between the response mode and clustering scheme was defined as a 'structured environment,' and a lack of correspondence defined an 'unstructured environment.'The study used undergraduate students as subjects in a split-plot factorial (3-2,2) design. Each of the three treatment groups received one of three forms of cue-clustering (labelled Irregularity-Based, Transactions-Based, and Random-Based). All subjects provided evaluations to a block of six Irregularity-Based and seven Transactions-Based questions for each of two independent cases. In addition, a panel of seven experts were used to provide a benchmark for the measurement of accuracy.The study found that clustering and response mode correspondence did affect the accuracy and consistency. In one test of accuracy, subjects' evaluations between treatment groups were statistically significantly different for 6 of the 26 questions. In another test of accuracy, using the experts' responses, responses by subjects in structured environments were more accurate than responses by subjects in unstructured environments (F = 21.87, $\alpha \leq .001$). In a test of consistency, subjects in structured environments had the smallest variance for 17 of 26 questions. Issue Date: 1988 Type: Text Description: 307 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1988. URI: http://hdl.handle.net/2142/71413 Other Identifier(s): (UMI)AAI8908762 Date Available in IDEALS: 2014-12-16 Date Deposited: 1988
﻿