AN EMPIRICAL STUDY OF JUDGMENTS ON INTERNAL CONTROL by SHANE PADRIAC COOK B.S., University of Oregon, 1970 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN BUSINESS ADMINISTRATION in The Faculty of Graduate Studies (Commerce and Business Administration) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March, 1980 © Shane Padriac Cook In presenting this thesis i s p a r t i a l f u l f i l l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted to the Head of my Division or his representatives. It i s understood that copying or publication of th i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Division of Accounting and Management Information Systems Faculty' of Commerce and Business Administration The University of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, B.C. Canada, V6T 1W5 Dated this Abstract A review of the l i t e r a t u r e in the area of audit judgment and cognitive s t y l e research forms the basis for the research study reported in t h i s paper. T h i r t y - s i x independent auditors were required to complete a psycho-l o g i c a l test structured to measure cognitive complexity. They were also asked to judge weakness of internal control in a hypothetical company as a function of f i v e yes/no questions representing p a r t i a l judgments according to a f u l l f r a c t i o n a l f a c t o r i a l design. The a n a l y s i s - o f - v a r i a n c e version of the Brunswik lens model was used to analyze the subject responses. The results of the experiment indicated that the subjects' models of judgment i s highly l i k e l y to stem from the fact that most of the subjects were concrete (cognitive simple) in their judgments. The usefulness of f r a c t i o n a l f a c t o r i a l studies in human judgment research was also appraised, and i t was found to be adequate in the experiment performed. Suggestions for further research applying the cue u t i l i z a t i o n methodology in au d i t and management information systems research were made. In addition, suggestions for a more n a t u r a l i s t i c approach to audit research were proposed. Thesis Supervisor i i ACKLOWLEDGEMENTS I would l i k e to thank my advisers, Dr. Albert S Dexter, Dr. Michael Gibbins, and Dr. Hartmut J. W i l l (Chair man) for the numerous suggestions and constructive c r i t i c isms. I would also l i k e to thank Dr. Izak Benbasat for hi suggestions during the planning stage of thi s thesis and Mr Kenneth P. Johnson, of Coopers and Lybrand, New York, NY for his commentary on public accounting prac t i c e . Thei contributions to thi s thesis are g r a t e f u l l y acknowledged. i i i TABLE OF CONTENTS Page CHAPTER I: INTRODUCTION Objectives Of The Thesis 1 Organization Of The Thesis 1 Research In Audit Judgments 2 Internal Control Studies 3 Audit Program Planning Study 6 Cognitive Psychology 7 Cognitive Styles 9 Cognitive Complexity 14 Summary 18 CHAPTER II : METHODOLOGY OF THE EXPERIMENT Introduction 28 The Experimental Task 28 Experimental Design 30 Strengths And Weaknesses Of The Experiment 32 Brunswik Lens Model 32 Fractional F a c t o r i a l Designs 37 Paragraph Completion Test 38 Administration Of The Experiment 40 Summary 41 CHAPTER I I I : RESULTS OF THE EXPERIMENT Introduction Judgment Description Importance Of Main Effects And Interactions Effects Of Firm A f f i l i a t i o n And Level Of Experience Judgment Insight Subjective Weights Insight Index Judgment Consistency Cognitive Complexity . Summary CHAPTER IV: CONCLUSIONS Summary Of The Problem 6 7 Summary Of The Experimental Results 68 Implications Of The Experiment's Results 70 Implications for Further Research 71 BIBLIOGRAPHY 74 APPENDIX 1 CASE STUDY 79 APPENDIX 2 INDIVIDUAL SUBJECT OMEGA-SQUARED RESULTS 81 APPENDIX 3 INDIVIDUAL SUBJECT CONSENSUS AND INSIGHT SCORES 90 48 48 49 50 51 51 52 53 54 i v LIST OF TABLES Page Table I Ashton Questions 19 Table II Hamilton And Wright Questions 20 Table III Ashton And Brown Questions 21 Table IV Joyce Variables 22 Table V Questions Used In Current Study 43 Table VI Example Of Internal Control Case Used In Exper-iment 44 Table. VII Relative Importance Of Main EffectsAnd Inter-actions To The Auditors ( f u l l Model) 57 Table VIII Relative Importance Of Main Effects And Two Way Interactions To The Auditors (One Half Frac-t i o n a l Model) 58 Table IX Relative Importance Of Main Effects And Inter-actions To The Auditors By Firm ( f u l l Model) 59 Table X Relative Importance Of Main Effects And Two Way Interactions To The Auditors By Level Of Exper-ience ( f u l l Model) 60 Table XI Values Of Subjective Weights By Firm And Exper-ience Level 61 Table XII Insight Index By Firm A f f i l i a t i o n And Level Of Experience 62 Table XIII Judgment Consensus Among Auditors In Same And Different Firms 63 Table XIV Judgment Consensus Among A u d i t o r s In Same Experience Levels 64 v LIST OF FIGURES Page Figure 1 General Relationship Between Environmental And Behavioral Complexity 23 Figure 2 Relationship Between Environmental And Behav-i o r a l Complexity For Different Levels Of Per sonality Structure 23 Figure 3 Brunswik Lens Model 45 vi 1 CHAPTER I INTRODUCTION Objectives Of The Thesis The objective of thi s thesis i s to study the consis-tency of audit judgments 'among auditors. A secondary objective i s to determine whether differences exist due to t r a i n i n g and experience, in the s u b j e c t s ' a b i l i t y to d i f f e r e n t i a t e and integrate information. A further objec-tive of t h i s thesis i s to be found in an evaluation of the usefulness of f r a c t i o n a l f a c t o r i a l designs in human judgment research. Organization Of The Thesis The remaining sections of t h i s chapter w i l l discuss l i t e r a t u r e relevant to the research. These sections w i l l include reviews of research in audit judgments, and research in cognitive psychology. The second chapter w i l l discuss the methodoloqy of the research experiment performed. Among the topics to be discussed are the experimental task, the s p e c i f i c experi-mental design used, strengths and l i m i t a t i o n s of the experiment, and d e t a i l s concerning the administration of the experiment. Chapter III presents of the re s u l t s of the experiment. A f i n a l chapter summarizes the experimental findings and 2 presents further implications of the findings with sugges-tions for further research. Research In Audit Judgments This section presents a review of auditing research considered as a behavioural view of auditing and management information systems (MIS) and a preview of the type of research to be conducted. A comprehensive review of two authors' works w i l l be given as the techniques used by them form the basic methodology of t h i s experiment. Auditor's judgments have seldom been the subject of systematic research. Yet the importance of auditor judg-ments i s accepted almost without auestion. "A t y p i c a l rationale for t h i s s i t u a t i o n i s the AICPA's statement that 'judgment i s the most important factor in the making of any audit, but in many situations i t i s p r a c t i c a l l y impossible to write out in s p e c i f i c language how the auditor applies judgment. 1" 1 This thesis w i l l review two areas of study on the judgmental processes of auditors. The f i r s t area deals with judgment formation i n the e v a l u a t i o n of i n t e r n a l control and the second area focuses on the audit planning stage. Both of these studies are examples of "cue u t i l i z a -t i o n " or "analysis-of-variance" techniques. Through the systematic manipulation of a predetermined set of information cues, the researcher can produce a quasi-experimental manipulation of the cues within each subject's set of judgment situ a t i o n s . The r e s u l t s for each subject 3 can then be analyzed using several s t a t i s t i c a l techniaues that provide descriptions of which cues were important to the s u b j e c t ' s judgments, how cues i n t e r a c t e d , and how 2 consistent a subject was. S p e c i f i c advantages and disad-vantages of t h i s technique w i l l be discussed in chapter II. Internal Control Studies 3 Ashton was the f i r s t to use cue u t i l i z a t i o n in the study of auditor judgments. This f i r s t experiment conducted by Ashton studied judgments on internal control over payroll systems by 63 practicing auditors in the Minneapolis/St Paul area. The auditors were presented with 32 sets of hypothet-i c a l internal control cases, each consisting of six ques-tions (cues) that had been pre-answered "yes" or "no." The six questions used are presented in Table I. The thirty-two sets were chosen according to a h a l f - f r a c t i o n a l f a c t o r i a l design. The subjects were required to evaluate the adequacy of the internal control using a six-point scale. The data was analyzed lisinq the analysis-of-variance technique. This design allowed the estimation of the signficance of main and two-factor interaction e f f e c t s . A l l other interactions were included in the error term. To estimate the importance of each question and two-factor interactions, omega-souared 4 s t a t i s t i c s were computed. This s t a t i s t i c i s a measure of the strength of association and can be used as an estimate of the r e l a t i v e proportion of variance that i s accounted for 4 by each main e f f e c t and interaction in an analysis-of-variance design. As a second portion to his study, Ashton investigated two types of judgment consistency. One type of judgment consistency i s consensus, or the consistency of the subjects at one point in time. Consensus was determined by c o r r e l a t -ing the internal control ratings given with those of the other subjects. A second type of consistency i s s t a b i l i t y , which i s the consistency over time for the same subject. Ashton determined s t a b i l i t y by having the subjects complete the experimental instrument at a l a t e r dates (6 to 13 weeks later) and correlating the two sets of internal control ratings for each auditor. In addition to this work, Ashton evaluated judgment insight by having each subject a l l o c a t e 100 points among the six questions accordina to the r e l a t i v e importance the subject gave the question. Insight was measured by co r r e l a t i n g the omega-squared s t a t i s t i c s with the subjective weights given. The results of the experiment indicated that the main e f f e c t s accounted for, on average, 80.22% of the variance, and that the f i f t e e n two-way interactions only 6.4%, leaving the error factor to account for 13.38%. Two questions on separation of duties accounted for, on average, 51.4% of the variance. This tends to support the hypothesis that the auditor judgments varied systematically with the answer to the question. There was some minor support for the hypothe-s i s that the judgments were influenced by the interaction of 5 the answers. The average c o r r e l a t i o n s f o r consensus, s t a b i l i t y , and insight were .70, .81 and .89 respectively. Ashton's experiment has been r e p l i c a t e d by three a d d i t i o n a l s t u d i e s : Hamilton and Wright^, Ashton and 6 7 Kramer , and Ashton and Brown . 5 Hamilton and Wright used a f u l l r e p l i c a t i o n of a 2 f a c t o r i a l design using the f i v e questions shown in Table IT. The average proportion of variance explained by the f i v e main ef f e c t s was 82%. The average co r r e l a t i o n for consensus and insight were .66 and .87 respectively. Hamilton and Wright divided t h e i r 17 subjects in two groups - those with more than three years of experience and those with three or less years of experience. The more experienced subjects had a higher average proportion of variance explained by the main e f f e c t s than the less experienced group (.85 versus .80), a higher correlation for insight measurement (.93 versus .84), and higher c o r r e l a t i o n for judgment consensus (.78 versus .62 ). Ashton and Kramer replicated Ashton's o r i g i n a l experi-ment using 30 students enrolled in an undergraduate auditing course at the University of Texas at Austin. The six main ef f e c t s accounted for, on average, 65.6% of the judgment variance and the two-factor interactions accounted for, on average, 8.4% of the variance compared to the o r i g i n a l experiments averages of 80.2% and 6.4%, respectively. The correlations for consensus and insight were .66 and .77 6 respectively, compared to .70 and .89 for Ashton's o r i g i n a l work. Ashton and Brown replicated the previously described work using a one-half r e p l i c a t i o n of a 2 f a c t o r i a l design using the eight questions shown in Table III. This design allowed for the estimation of a l l main e f f e c t s , and a l l two-factor and three-factor interactions. The subjects were 31 p r a c t i c i n g a u d i t o r s employed by seven i n t e r n a t i o n a l public accounting firms in the United States. The eight main e f f e c t s accounted f o r , on average, 7.1.9% of the judgment variance, and the two and three cue interactions accounted for 6.1% of the variance. The insight measurement co r r e l a t i o n was .86 and the test of s t a b i l i t y used showed a c o r r e l a t i o n of .91. The consensus correlation was .67. The four studies reviewed here have been remarkably c o n s i s t e n t . The average values f o r judgment i n s i q h t , consensus and s t a b i l i t y were found to be high. The lower amount of variance of the main e f f e c t s explained in the Ashton and Kramer study and higher amount of judgment variance of the main e f f e c t s for the more experienced auditors can be explained by the amount of training and p r a c t i c a l experience of the respective subjects. Audit Program Planning Study g Joyce perfomed a cue u t i l i z a t i o n study evaluating auditor judgments on audit program planning (audit time estimation) based on information cues about preliminary data 7 c o l l e c t i o n on accounts receivable. The independent and dependent variables are shown in Table IV. The subjects were 35 p r a c t i c i n g auditors from four large public account-ing firms in the Chicago area. Joyce conducted two experi-ments, the f i r s t using a one-half r e p l i c a t i o n of a 2^ f a c t o r i a l design and the second using a f u l l r e p l i c a t i o n of 5 a 2 f a c t o r i a l design. The f i v e main ef f e c t s accounted for 74.7% of the judgment variance. No estimates of the interactions were made. Joyce's r e s u l t s indicated a high variance among the importance of the d i f f e r e n t cues given to the subjects. This i s consistent with p r i o r studies which analyzed sample size estimations by auditors, which the author believes i s highly correlated to the amount of audit work that would be 9 required to be done. Joyce did find that there was - with one exception - high c o r r e l a t i o n of the r e s u l t s among the auditors within the firms studied. Joyce used the same method as Ashton to obtain a measurement of s e l f - i n s i g h t . His r e s u l t s showed a c o r r e l a t i o n of .449 versus .89 for Ashton's o r i g i n a l work. Cognitive Psychology Research in management information systems and cogni-tive psychology w i l l be reviewed here because auditors form s p e c i a l i z e d group of users of management information systems. Cognitive research i s important to t h i s thesis as the judgement processes of the subjects are being studied. 8 Mason and M i t r o f f l u discussed the implications of psycho-l o g i c a l type for researchers and p r a c t i t i o n e r s in MIS. The a r t i c l e has stimulated t h i s author's i n t e r e s t in the application of psychology to MIS. The research in MIS and accounting has used concepts from the f i e l d of cognitive psychology. This focus on cognitive psychology i s a t t r i -butable in part to the research in human information processing and human decision making. Cognition i s a general term covering a l l the various models of knowing — perceiving, remembering, conceiving, judging, reasoning. Researchers in the accounting and MIS f i e l d s have tended to l i m i t t h e i r interest to d e c i s i o n -making applications of cognitive psychology. Accountants' primary interest in studying d e c i s i o n -making i s the p o t e n t i a l f o r improving the q u a l i t y of d e c i s i o n s by improving the information provided. The understanding of human information processing i s thought to aid in c l a r i f y i n g the needs of decision makers.^ MIS concerns are s i m i l a r . Two c r i t i c a l research needs identifed are the development of methods for determing appropriate content of an information system, and investigating the c h a r a c t e r i s t i c s of d e c i s i o n makers which a f f e c t MIS design. Cognitive factors are only a subset of the factors influencing accounting and MIS design. Mason and M i t r o f f i d e n t i f i e d six key variables which comprise a management 9 information system (person,psychological type, organization-1 3 al context, evidence, mode of presentation). Benbasat and 1 4 Taylor referenced three studies which discussed influences on MIS design: Chervany, et al studied the functional r o l e ; Adams and Schroeder surveyed desired information charac-t e r i s t i c s such as accuracy, use and summarization f o r d i f f e r e n t organizational l e v e l s ; Ghymn and King showed that the importance of economic, legal and so c i a l information was di f f e r e n t for various managerial l e v e l s . Cognitive Styles Cognitive style i s a hypothetical construct developed to explain the mediation between stimuli and responses. The term cognitive style refers to the c h a r a c t e r i s t i c ways in which individuals organize conceptually t h e i r perceptions of the i r environment. Several psychologists have noted that cognitive style i s a process of information transformation. If i t were not for these cognitive representations which modify the one-to-one r e l a t i o n s h i p between s t i m u l i and response, an individual would be expected to respond to stim u l i in completely predictable fashions. Common to a l l theories and research in cognitive style is an emphasis on the structure, rather than the content of the thought. Structure refers to how cognition i s organ-ized; content refers to what information i s available. Numerous approaches to cognitive s t y l e have been developed and several authors have attempted to c l a s s i f y various 10 approaches. Despite the elements of common interest in the study of c o g n i t i v e s t y l e , r e s e a r c h e r s d i s a g r e e on the d e t a i l s of, methodology and the use of measuring instruments. The r e s u l t i s a large body of l i t e r a t u r e from which i t i s d i f f i c u l t , i f not impossible, to extract general p r i n c i p l e s . Two d i f f e r e n t interpretations of cognitive s t y l e are now presented, those proposed for use by accountants and those developed by MIS researchers. These demonstrate in part the variety of constructs applied in the f i e l d , and s p e c i f i c a l l y i d e n t i f y the d e f i n i t i o n s often used in account-ing and management information systems research. The Committee on Human Information Processing of the American Accounting Association stated that "Cognitive s t y l e r e f l e c t s how decision makers prefer to approach and solve various tasks." 1^ MIS researchers, on the other hand, have taken a s l i g h t l y broader view of cognitive s t y l e s . In attempting to bring order to information processing a c t i v i t i e s , i n d i v i -duals are viewed as tending to act consistently despite information v a r i a t i o n . These r e l a t i v e l y fixed patterns are called cognitive s t y l e s . 1 ^ Related to the cognitive s t y l e s issues i s the area of cognitive complexity. Cognitive complexity r e f l e c t s an individual's capacity to d i f f e r -entiate and integrate s t i m u l i . Cognitive complexity w i l l be discussed in more d e t a i l l a t e r in t h i s chapter. The c o g n i t i v e s t y l e c o n s t r u c t i m p l i e s i n d i v i d u a l differences in cognitive processes and that such differences 11 are r e l a t i v e l y s t a b l e over time and s i t u a t i o n s . The following i s a discussion of the two cognitive styles which have received the most recognition in the MIS l i t e r a t u r e , f i e l d independence-dependence and a n a l y t i c - h e u r i s t i c s t y l e s . The f i e l d independence-dependence cognitive s t y l e has been extensively studied by W i t k i n . 1 7 F i e l d independence-dependence has been defined by Witkin as " a n a l y t i c a l " as opposed to a "global" way of perceiving, and hypothesized that i t e n t a i l s a tendency to experience items as discrete from t h e i r backgrounds — i . e . , the a b i l i t y to overcome the influence of an embedding context. The Embedded Figures Test, which requires subjects to perceive a simple figure within a larger complex figure, i s often used to measure thi s cognitive s t y l e . This c l a s s i f i c a t i o n has been used by 18 19 Doktor and Hamilton and by Lusk to study the r e l a t i o n -ships between the acceptance of managerial reports and t h i s cognitive s t y l e . The a n a l y t i c - h e u r i s t i c reasoning types were developed by Huysmans for an experiment designed.to test the impact of 21 cognitive styles on managerial recommendations. Analy-t i c a l reasoning refers to a tendency to reduce problem situations to a core set of underlying causal r e l a t i o n s h i p s . H e u r i s t i c reasoning r e f e r s to an emphasis on workable solutions to t o t a l problem sit u a t i o n s . The search i s for analogies with familiar,previously solved problems, rather than a system of underlying causal r e l a t i o n s h i p s . 12 The following section w i l l review two research studies i n v o l v i n g c o g n i t i v e s t y l e s and MIS. Benbasat and 22 Schroeder reported that analytics who used decision aids (e.g., forecasting models) used fewer reports than the h e u r i s t i c s who used decision aids, but for subjects without decision aids the reverse was true. This indicates a reliance on the use of models by analytics and that the models have more information value f o r them than the h e u r i s t i c s who r e l y more on the reporting system even when they have the models available. 2 3 Benbasat and Dexter reported s e v e r a l important experimental r e s u l t s with implications of MIS design. The r e s u l t s indicated that decision making style had a strong impact in determing which of the two information generation approaches would be more suitable in the given decision setting. Their finding showed that structured reports with aggregated data are better' suited for a n a l y t i c s , while the database inquiry system i s better suited for h e u r i s t i c s . In terms of p r o f i t performance h e u r i s t i c s who used struc-tured/aggregated reports did the poorest. The i n a b i l i t y of h e u r i s t i c s to "break up" the structured report to understand the impact of the individual elements would lead to poorer performance with the structured/aggregated r e p o r t s . Analytics performed equally well in terms of p r o f i t with either system? th e i r performance l e v e l was matched by the h e u r i s t i c s who had the database inquiry system. This 24 finding i s consistent with B a r i f f and Lusk's study which 13 reported that h e u r i s t i c s preferred disaggregated reports. Although analytics did not d i f f e r as to p r o f i t performance, those using the structured/aggregated reports had a s i g n i f i -cantly lower decision making time. There was no such difference for h e u r i s t i c s . Furthermore, analytics requested l e s s h i s t o r i c a l information when given the s t r u c t u r e d reports; however, with the database system, they requested approximately twice as much data than did other groups. 25 Many additional studies have been published on the application of the cognitive style constructs to MIS because of the simularity of the re s u l t s with the papers discussed, the i r results w i l l not be included here. Cognitive s t y l e 2 6 issues have been addressed some what by Driver and Mock but t h e i r major thesis dealt with the more s p e c i f i c matter of cognitive complexity than with a general cognitive s t y l e issue. Cognitive style research in accounting has tended to focus on the acceptance of accounting reports by management. 2 7 In Lusk's study, analytics selected an investment repre-sented by a r e l a t i v e l y d i f f e r e n t i a b l e , nonglobal, a r t i c u l a t -ed annual report (a detailed annual report). Heuristics selected an investment represented by a global, u n a r t i c u l a t -ed annual r e p o r t (a very general r e p o r t with minimal f i n a n c i a l information and b r i e f discussions of operations and other matters). These re s u l t s are consistent with those of Benbast and Dexter. 14 Cognitive Complexity As stated e a r l i e r , cognitive complexity i s viewed as the processing of input stimuli as two related but separate a c t i v i t i e s : d i f f e r e n t i a t i o n and integration. D i f f e r e n t i a -tion refers to the individual's a b i l i t y to locate s t i m u l i among dimensions; integration refers to the a b i l i t y to u t i l i z e complex rules to combine t h e i r dimensions. The i n d i v i d u a l who i s low in d i f f e r e n t i a t i o n and integrative a b i l i t y i s called concrete. The individual who i s high in th i s a b i l i t i e s i s ca l l e d abstract. A l l i n d i v i -duals may be c l a s s i f i e d along a continuum from concrete to abstract. The individual's a b i l i t y to d i f f e r e n t i a t e and integrate mediates between stimuli and responses. The input s t i m u l i are f i l t e r e d or processed by the mediating structure. This allows the i n d i v i d u a l to d i f f e r e n t i a t e environmental elements and integrate then to provide behavioral outputs. The r e l a t i o n s h i p to management information systems should be obvious. Managers need data to be f i l t e r e d or manipulated to become useful as information. The mediating structures involved in cognitive complexity i f they could be adequately understood could be p o t e n t i a l l y useful in helping managers f i l t e r data i n t o the p a t t e r n s that they are accustomed to using. Schroder, Driver and Steufert developed a theory of human information processing based on cognitive complexity. They proposed three hypotheses: 15 (1) "Information processing by 'people in general 1 ( i n d i v i -dual differences disregarded) reaches a maximum l e v e l o f s t r u c t u r a l complexity at some optimal l e v e l of environmental complexity (point X i n Figure 1 ) . Increasing or decreasing environmental complexity (points Z and Y) from the optimal point (X) lowers the conceptual l e v e l , as indictated by a reduction in the le v e l of information processing involved in behavior 2 8 (from A to B in Figure 1)." This has been referred to as the "U-Curve Hypothesis." (2) The second hypothesis states that individual d i f f e r -ences w i l l be greater around the optimal point than at the extreme points of the environment. This i s shown in Figure 2. (3) The t h i r d hypothesis proposed was "that f o r more complex structures (Curve A) (a) i s always higher (generates more i n t e g r a t i v e l y complex in f o r m a t i o n -processsing behavior) over the midranges of environ-mental complexity and eoual at the extreme ranges of environmental complexity; and (b) reaches i t s optimal point at higher l e v e l s of environmental complexity. Consequently, the maximal differences between behav-i o r a l complexity for any two curves expressing d i f f e r -ences in str u c t u r a l l e v e l would occur at the optimal 2 9 l e v e l for the more abstract group." Compare E-F with E-G in Figure 2. 1 6 A l l of these hypotheses were supported in the subse-quent empirical studies performed. Driver and Mock"^ presented a review of behavioral research in accounting. Driver and Mock performed an experiment varying the information structures so that they would be optimal for the various decision styles that they had proposed. This study i s s i g n i f i c a n t in that i t attempt-ed to relate decision style theory based on human informa-t i o n processing concepts i n v o l v i n g c o g n i t i v e complexity concepts to preferences to information and amounts of decision time. However, the i r experimental results were inconclusive as were the res u l t s of two sim i l a r research 31 3 2 experiments (Savich , McGhee, Shields and Birnberg ). Complexity theory implies that individuals reauire an optimal l e v e l of information load. Individuals with low lev e l s of d i f f e r e n t i a t i o n w i l l be able to handle only a few l e v e l s of integration and individuals with low l e v e l s of integration w i l l be e a s i l y overloaded when too much informa-tion i s presented. This supports Mason and Mit r o f f ' s argument that the information system should be matched with 3 3 the psychological type of i t s users. C a r l i s l e in his Ph.D. d i s s e r t a t i o n studied the r e l a t i o n s h i p s between cognitive complexity and the complexity of user interfaces for time-sharing terminals. His res u l t s upheld Schroder, Driver and Steufert's hypotheses. What i s more important, however, are the two i n t e r r e -lated components of conceptual complexity already mentioned: 17 d i f f e r e n t i a t i o n , i . e . , the amount or number of attr i b u t e s of information used by the individual in his thinking and integration, i . e . , the number of concepts formed by the i n d i v i d u a l . Although some authors continue to d e f i n e cognitive complexity only in terms of d i f f e r e n t i a t i o n (e.g., 3 4 B i e r i ), the complexity of integration has become the primary focus of inte r e s t . The a b i l i t y to form a vari e t y of concepts on the basis of a given informational array appears to be the most adaptive facet of human i n t e l l i g e n c e . Cognitive complexity i s important to audit judgment research because auditors are required to integrate various p i e c e s of information i n making d e c i s i o n s concerning internal control systems,, substantive testing procedures, s u f f i c i e n c y of e v i d e n t i a l matter, fairness of statement presentation, and so on. Cognitive complexity and cognitive style concepts have not yet been applied to auditor judgment research. This thesis w i l l study the differences between cognitive complex and cognitive simple subjects in the context of internal control evaluations which w i l l be analysed to determine auditor consensus. Summary This chapter has presented a review of the l i t e r a t u r e on audit judgment and cognitive s t y l e s research. This l i t e r a t u r e forms the foundation for the research to be carried out. Although much has been written in the area of 18 cognitive s t y l e s , l i t t l e research has been done in applying cognitive s t y l e research to the accounting and MIS f i e l d s . To the author's knowledge, no research has been published on the a p p l i c a t i o n of c o g n i t i v e s t y l e s to a u d i t judgment research. 19 Table I — Ashton's Questions' 3 3 Are the tasks of both timekeeping and payment of employees adequately separated from the task of payroll preparation? Are the tasks of both payr o l l preparation and payment of employees adequately separated from the task of pay r o l l bank account r e c o n c i l i a t i o n ? Are the names on the pa y r o l l checked p e r i o d i c a l l y against the active employee f i l e of the personnel department? Are formal procedures established for changing names on the p a y r o l l , pay rates, and deductions? Is the p a y r o l l audited p e r i o d i c a l l y by i n t e r n a l aud i t o r s ? Was the internal control over payr o l l found to be sat i s f a c t o r y during the previous audit? 2 0 Table II - Hamilton and Wright Questions J b Is the task of timekeeping adecruately separated from the task of payroll preparation? Is the task of payro l l preparation adequately separated from the task of payro l l bank account r e c o n c i l i a t i o n ? Is the task of payment of employees adequately separat-ed from the tasks of timekeeping and payro l l bank account r e c o n c i l i a t i o n ? Are formal procedures established for changing names on the p a y r o l l , pay rates and deductions? Did internal auditors audit the p a r o l l during the audit period? 21 Table III - Ashton and Brown Questions 3 7 Are the tasks of both timekeeping and payment of employees adequately separated from th task of payr o l l preparation? Are the tasks of both payroll preparation and payment of employees adequately separated from the task of p a y r o l l bank account r e c o n c i l i a t i o n ? Are the names on the p a y r o l l checked p e r i o d i c a l l y against the active employee f i l e of the personnel department? Are the duties of theose preparing the payroll rotated? In h i r i n g new employees, i s an inquiry made as to his/her background and former employers? Are formal procedures established for changing names on the p a y r o l l , pay rates, and deductions? Is the p a y r o l l audited p e r i o d i c a l l y by i n t e r n a l aud itors? Was the internal control over payroll found to be sa t i s f a c t o r y during the previous audit? 22 Table IV - Joyce variables' 3 0 Independent Variables 1 . Sales approval 2. Bad Debt Expense/Sales 3. Write-off Approval 4. Separation of B i l l i n g Function and Subsidiary Ledger Maintenance 5. Sales/Average Accounts Receivable Dependent Variables (man-hours to complete audit procedures) 1 . Confirmation of accounts receivable 2. Review of accounts written o f f as u n c o l l e c t i b l e 3. Review of cash c o l l e c t i o n s of accoutnts receivable subsequent to balance sheet date 4. Determination of adequacy of allowance f o r uncol-l e c t i b l e accounts 5. Review of year end sales cutoff 23 Figure 1 - General Relationship between Environmental and 39 Behavioral Complexity High low y x Environmental Complexity High Figure 2 - Relationship between Environmental and Behavioral 40 Complexity for Different Levels of Personality Structure High Low j Curve A