Home | Business News | Browse by Publication | I | International Advances in Economic Research

The effect of task experience on assessments of auditor expert potential.

Article Excerpt
Abstract

This paper proposes a model and methodology for studying the effect of experience on an auditor's expert potential. Expert potential is defined as an increase in the expected level of expertise at which an auditor will perform in an arbitrary future task. An auditor's behaviors during performance of a current task are treated as probes of that auditor's knowledge base. Prom an analysis of the responses thus obtained, inferences are made concerning the effects of task experience on the underlying properties of the knowledge driving task behaviors and on the probability of greater expert-like task behavior in the future. Application of the model is illustrated by evaluating the effects of experience on the expert potential of four first-year auditors who performed audit-related tasks in simulated auditing environments. (JEL M41)

Introduction

Managers, supervisors, and others in similar positions often must assess the potential for future growth in expertise of novices in their charge based on observations of the behavior of those individuals while performing tasks in the field. A recent paper by Russo [1999] provides an approach to an objective discharge of that responsibility by proposing a model and analytical methodology for measuring the effect of an auditor's experience in performing a task on the knowledge properties that determine the level of that auditor's expert-like task behavior. The metrics so provided are then interpreted as indicators of progress toward greater expertise. However, because the metrics provided by Russo's model are in part functions of the mix of task behaviors, they do not generalize beyond the specific task from which they are derived.

This paper builds upon Russo's research by proposing a model that focuses directly on those properties of an auditor's knowledge that generalize to an arbitrary future task and upon which observers may base assessments of the effects of task experience on that auditor's expert potential. For this purpose, expert potential is defined as an increase in the expected level of expertise at which an auditor will perform in an arbitrary future task. The research approach used in this paper treats an auditor's behaviors during performance of a specific task as probes of that auditor's knowledge. From an analysis of the responses thus obtained, inferences are made about the underlying properties of the knowledge driving those behaviors. Knowledge properties are defined in terms that are independent of the particular mix of behaviors employed during performance of a task. Measures are proposed that can be used for comparisons of the effects over time and across auditors of task experience on the probability of greater expert-like task behavior in the future. Application of the model is illustrated by evaluating the effects of experience on the expert potential of the auditors who participated in Russo's experiments. The model and findings presented here highlight a need to separate those effects of experience that are relevant to realized changes in expertise in a specific task from the more fundamental effects on knowledge that carry over to future tasks and, ultimately, to the realization of expert potential in the long-term.

The remainder of this paper is organized as follows. The first section summarizes basic concepts and definitions taken from Russo [1999] that form the foundation of the model presented in this paper. The next section presents a model of the effects of experience on the properties of knowledge and discusses how model metrics are interpreted in terms of realized and potential expert development. Several research hypotheses are proposed in the third section. The following section discusses the experimental data and analytical methodology. Findings are presented and discussed in the next section. Finally, the paper concludes with several comments about model assumptions and limitations and the role of the model in expertise research.

Conceptual Background

This section provides a very condensed summary of those aspects of Russo's [1999] paper that form the foundation for the model presented in the following sections. The original paper should be consulted for more complete exposition and discussion.

Task Behavior, Task Automaticity, Experience, and Expertise

Russo proposes that task behavior be modeled as a sequence of observable target behaviors (such as reading, inquiry, calculating, and writing) mediated by episodes of subconscious (automatic) and conscious (cognitive) mental activity. Utilizing task automaticity, a widely recognized indicator of task expertise (for example, Alba and Hutchinson [1987]; Anderson [1982, 1987]; Mayer [1992, p.305]; Davis & Solomon [1989]; Bedard [1989]), it is also proposed that the level of an auditor's expertise be measured by the proportion of mediating episodes during performance of a task that are automatic. The effect of experience performing a task on expert-like behavior can then be assessed by measuring the resulting change in task automaticity. (1) For this purpose, experience is operationalized as repetition in performing the various target behaviors.

The analytical methodology for measuring the effect of experience separates the task behavior sequence chronologically by target behavior, and within target behavior, by semi-frequency, producing two groups of equal frequency for each target behavior. Behaviors in the below-median group are the inexperienced or naive instances, and those in the above-median group are the experienced instances. The effect of experience on task automaticity is measured by the task learning ratio, defined as the ratio of the automaticity of the experienced group to that of the inexperienced group. (2)

Knowledge-Driven Behavior

An auditor's intentional and purposeful task behavior is assumed to be knowledge-driven. Russo defines and examines three properties of that knowledge: organization, content, and availability, and shows that changes in these properties, measured by a set of serially dependent properties learning ratios, are directly related to changes in the automaticity of task behavior. (3) Therefore, the process by which an individual auditor progresses toward more expert-like task behavior can be explained by measuring the effect of experience during performance of a task on the properties of that auditor's knowledge.

Knowledge Base Probes and Response Sampling

Each instance of a behavior performed serves as a probe of an...

View this article FREE - Now for a Limited Time, try Goliath Business News
Free for 3 Days!