Robert Libby

Determinants of judgment performance in accounting settings: Ability, knowledge, motivation, and environment☆

Abstract The goal of this paper is to trace the development of our conceptions of the roles of ability, knowledge, motivation, and environment as determinants of decision performance in accounting settings, and provide a synthesis of the basic constructs, conceptual relations, and methodological guidelines that can be inferred from this diverse literature. We first outline the key characteristics of accounting settings and research principles appropriate for examining the characteristics. The primary emphasis is on identifying interactions between determinants of performance, specifying underlying cognitive processes, and abstraction based on theory and task analysis. We then demonstrate how selected studies which follow these basic principles have greatly enhanced our understanding of accounting-related decisions. Finally, suggestions for future research are presented.

Human information processing research in accounting: The state of the art in 1982

Awareness of the importance of human information processing research to accounting issues has increased dramatically since 1977. As a result, this literature has expanded in volume and addresses a larger spectrum of accounting problems. Further, it incorporates a wider variety of theories and methodologies. This paper draws upon the framework provided by Libby and Lewis (1977) to synthesize and evaluate accounting research conducted since 1977 using the lens model, probablistic judgment, predecisional behavior, and cognitive style approaches. In addition, the impact of the research on practice and some directions for future research are discussed. Along with the recognition that decision making is the focal point of the current practice of accounting, an extensive body of research which analyzes decision making in accounting settings has been developed. This research is usually referred to as human information processing (HIP) or behavioral decision making research. Accountants have shown particular interest in studies which investigate (1) the role of accounting information in user decisions (e.g. in commercial lending) and (2) the complex decisions required in the practice of accounting (e.g. in auditing). The evidence generated by this research serves a dual purpose. First, it may lead to improvements in these accounting decisions. Second, it can add to the basic knowledge of human decision processes. Four years ago, we provided a review of what was then an emerging research program (Libby & Lewis, 1977; LL-77 hereafter). Since this initial review, both the interest and research output in this area have grown at an increasing rate. In response to this activity, we have compiled a second state-of-the-art paper. As testimony to the growth in interest we note that this second review contains more than twice as many studies as did LL-77. As further evidence of the interest in human information processing research in accounting, we can point to the impact which such research has already had on accounting practice. Accepting consensus and consistency as measures of the quality of expert judgment, many audit firms have developed decision aids to increase the consistency of judgments. These notions derive directly from research in behavioral decision theory. Research exposing humans as poor intuitive statisti* The authors gratefully acknowledge the Paton Accounting Center for financial support and Garry Marchant for his assistance.

Modeling the determinants of audit expertise

Abstract Libby & Luft [ Accounting, Organizations and Society (1993) pp. 425–450] presented a model of the relations between experience, ability, knowledge, and performance in audit judgment. This paper extends the model by developing a framework for predicting the structure of these relations in different judgment in different judgment settings and provides an initial test of the predictions by analyzing data from Bonner & Lewiss [ Journal of Accounting Research (Supplement 1990) pp. 1–20] four tasks using LISREL. Key predictions were that problem-solving ability would directly affect performance only in unstructured tasks and would indirectly affect performance through its effect on knowledge acquisition where the learning environment was impoverished. The predictions were supported in most cases. In addition, the paper provides and tests a basis for predicting the associations between performance on different audit tasks. Construct measurement problems that need to be addressed in future research are also indicated.

The review process as a control for differential recall of evidence in auditor judgments

Abstract This experiment examines whether there are systematic offsetting differences in the manner in which initial decision makers and reviewers attend to information which ensure that evidence inconsistent with initial judgments is given adequate consideration. Differences in attention are proposed, which result in differential recall of evidence by the initial decision maker and reviewer and thus influence what knowledge initial decision makers and reviewers bring to their discussions and subsequent decisions. The results suggest that the review process can act as an effective control by increasing the chances that the implications of inconsistent evidence are considered.

Analysts' Reactions to Earnings Preannouncement Strategies

Preannouncements of earnings tend to overstate negative or understate positive news, which decreases the chance of a negative surprise when actual earnings is announced. We conduct an experiment to investigate how experienced sell-side analysts’ earnings forecasts are affected by preannouncements that either understate, accurately state, or overstate the magnitude of positive or negative total earnings news, holding total earnings news constant. We find that firms with negative (positive) total news receive the highest post-earnings announcement forecasts of future earnings when the earlier preannouncement overstates (understates) the magnitude of the news. These forecasts are consistent with the analysts’ perceptions about the firms’ future prospects, but not their perceptions of management. While analysts expect preannouncements to be lower than actual earnings, they do not adjust their forecasts for these beliefs. These insights into analysts’ responses have implications both for managers and analysts.

Incentives, Effort, and the Cognitive Processes Involved in Accounting-Related Judgments

In this paper, we investigate how the performance-related incentive effects of monetary payments depend on the cognitive processes involved in accounting judgment tasks. We argue that this dependency exists because the processes involved in encoding, retrieving, and combining information require varying effort levels and thus are differentially sensitive to effort increases resulting from the incentive. Considering this interaction of incentive effects and cognitive processes will help paint a more complete picture of cognitive performance in these tasks and will aid both evaluations of the generalizability of existing experimental results and efficient planning of the use of monetary incentives in future experiments.

Performance of a composite as a function of the number of judges

Abstract In decision situations where relevant variables cannot be easily measured, mathematical aggregation of individual judgments may prove to be a useful decision aid. A wide variety of studies indicate that composite judgments formed by equal-weighting aggregation models outperform the average individual judge making up the composite. However, the use of these composite judgments in actual decision making situations has not been suggested because of inefficiencies caused by the need to include the judgments of a large number of individuals in the formation of the composite. If composites formed by pooling judgments of fewer decision makers produce similar incremental performance, these inefficiencies would be substantially reduced, making use of such a technique more practical. This study empirically tested the effects of group size upon the incremental accuracy of an equal-weighted composite judge in three different judgment tasks. The results indicate that on average the majority of the increment gained by aggregating large numbers of judges can be obtained by aggregating three judges.

Man versus model of man: some conflicting evidence

Abstract This study searched for “bootstrapping” behavior in a financial analysis task where 43 professional loan officers predicted business failure from five-ratio financial profiles. As in previous studies, a linear model proved to be an excellent predictor of subject responses. But, where in previous studies the models of the judges outperformed the judges themselves, in the present study the judges significantly outperformed their linear models ( p ≤.01). The present study differed from previous studies in that the subjects had a greater amount of task-related expertise, the criterion was more well defined and reliably measured, and the cue distributions were more highly skewed. It is also suggested that the most accurate judges tend to outperform their models.

Do Investors Overrely on Old Elements of the Earnings Time Series

This paper reports an experiment demonstrating that MBA students overrely on old earnings performance when predicting future earnings performance in a laboratory setting. In the experiment, MBA students relied too heavily on old annual ROE information to predict future annual ROE. The experiment shows how a common cognitive error (overreliance on unreliable information) interacts with the structure of the earnings time series to create particular patterns of prediction errors. The results also suggest directions for research on two well†known anomalies, long†run overreactions (De Bondt and Thaler 1985, 1987) and post†earnings†announcement drift (Bernard and Thomas 1990).

Audit category knowledge as a precondition to learning from experience

Abstract Prior research indicates that inexperienced auditors lack knowledge of basic auditing categories (e.g. transaction cycles, audit objectives), instead developing this knowledge over time. As a consequence, learning from early experiences may be hampered because these experiences are not stored with respect to the category structures that are needed for important audit decisions. We performed an experiment which demonstrates that: 1. (1) providing transaction cycle and audit objective category knowledge through instruction prior to experience facilitates one particular type of subsequent learning from experience learning of category-level error frequencies), and 2. (2) this learning advantage cannot be duplicated by providing listings and explanations of category memberships after experience. In addition, actual experience consequently has a greater influence on later audit decisions when category knowledge is acquired prior to experience.