Don N. Kleinmuntz

Determinants of the justifiability of performance in ill-structured audit tasks

This paper examines determinants of justifiability of performance in an ill-structured audit task. Justifiability is the extent to which task performance can be defended and an ill-structured audit task is a nonroutine task that lacks definitive applicable authoritative guidance, has multiple alternative courses of action, and requires the auditor to exercise significant judgment (Abdolmohammadi and Wright [1987]). Performance evaluation for these tasks generally rests upon the perceived justifiability of either the decision process or actual choice (Davis and Solomon [1989] and Waller and Felix [1984]), so that understanding expert performance on these audit tasks requires understanding the determinants of justifiability (Solomon and Shields [1995]).

The value of action and inference in dynamic decision making

Abstract The use of action- versus judgment-oriented decision strategies in a dynamic decision task is investigated. Subjects engaged in a simulated medical decision-making task, where the goal is to select treatments in order to cure patients suffering from an unknown disease. The experiment manipulated two task factors that were predicted to influence the effectiveness and efficiency of action- and judgment-oriented strategies: (1) the availability of a decision aid that promotes the interpretability of outcome feedback in the task and (2) the level of risk associated with treatment choices, which influences the opportunity for corrective actions to compensate for earlier decision errors. Results indicated that these manipulations did in fact influence performance. However, most subjects did not use action-oriented strategies even when they would have led to superior performance. Possible explanations for this finding are discussed and the implications for theoretical accounts of strategy acquisition and selection are considered.

The Reliability of Subjective Probabilities Obtained Through Decomposition

The use of decomposition as a procedure for improving the consistency of subjective probability encoding is discussed. Using a psychometric measurement model, an expression is developed that describes the random error associated with decomposition estimates as a function of characteristics of the component assessments. Decomposition is compared to direct assessment in terms of the percent change in measurement error that can be attributed to the use of decomposition. Potential benefits of decomposition are specified and recommendations made on how to utilize decomposition as an approach for error control.

Human decision processes: Heuristics and task structure

Publisher Summary This chapter discusses some key findings and methods of behavioral decision theory. The study of human decision processes is currently an active and exciting research domain. Recent progress promises a more complete understanding of information processing strategies in judgment and choice and is building important links to other aspects of human cognition. The identification of cognitive mechanisms has not been limited to the area of judgments under uncertainty. A parallel stream of research focuses on the information processing strategies used for evaluative judgments and choices. Researchers have identified a large number of rules that decision makers use, including compensatory rules (additive utility and additive difference) and non-compensatory rules (conjunctive, disjunctive, and elimination by aspects). Theories that encompass the task-contingent nature of decision making can help in the design and implementation of decision aiding systems and, ultimately, improve the effectiveness of decision making in important problems.

An Experimental Investigation of Forward-Looking Non-Financial Performance Disclosures

The AICPAs Special Committee on Financial Reporting has urged disclosure of relevant forward-looking and non-financial information to supplement conventional financial statements. We conduct an experiment consisting of 20 laboratory security markets with eight participants each to assess the effects of such disclosures on capital allocation decisions. We observe four markets in each of five accounting information conditions: a baseline condition with an income statement and balance sheet only and four conditions that combine this baseline with supplemental disclosures of proved reserves (a best estimate), total reserves (an upper bound), and minimum reserves (a lower bound). We find first that proved reserve disclosures improve capital allocation decisions, even though these disclosures are redundant with information in the primary financial statements. Second, disclosures of the upper bound (total reserves) in the absence of lower bound (minimum reserves) has the potential to bias security prices upwards, while disclosures of both total and minimum reserves remove this bias. Third, a comparison of individual price predictions to actual market prices reveals both a systematic prediction error and a differential effect of supplemental disclosures on security prices, suggesting that experimental investigations of capital allocation decisions should include market settings. The paper concludes with a discussion of implications for accounting standard setters.