A MODEL FOR THE PREDICTION OF JUROR DECISIONS*

A logit model approach designed to assign a probability whether prospective jurors will favor the defendant or plaintiff in a case, as a function of perceived individual juror characteristics, is described in the context of a case situation. The model, which has been programmed on a hand-held computer, is designed for implementation in courtroom settings to help defense attorneys evaluate and select jurors in order to minimize the likelihood of large jury awards. Empirical tests of the model are also described.     

SIMULATION EXPLORATION OF THE POWER OF MARGINAL REGRESSION ANALYSIS*

In either simple or multivariate regression models, measured coefficients represent a kind of average of co-variant relationships implicit in the data. For time-series data, if co-variant relationships follow no particular pattern over the sample period, use of this average in forecasting is rational. However, if relationships among independent and dependent variables is undergoing change over time, then measured regression coefficients are likely to be poor tools for prediction. This paper is an exploratory study of this general problem area. It develops implications as to the predictive impact of the problem and offers preliminary suggestions as to a method of developing predictions under such circumstances.     

CLASH OF CAREER AND FAMILY: FERTILITY DECISIONS AFTER JOB DISPLACEMENT

In this paper we investigate how career considerations may affect fertility decisions in the presence of a temporary employment shock. We compare the birth rates of women displaced by a plant closure with those of women unaffected by job loss after establishing the pre‐displacement comparability of these groups. Our results reveal that job displacement reduces average fertility by 5%–10%, and that these effects are largely explained by the response of women in more skilled occupations. We offer an explanation of our results based on career interruptions of women.     

AN EMPIRICAL EXAMINATION OF THE BIASING EFFECTS OF FRAMING ON BUSINESS DECISIONS*

This paper presents results of three studies testing whether decision bias due to framing, found in consumer purchase contexts, also would occur in a business setting. In the first study, business-student subjects displayed framing-induced decision bias just as subjects had in an earlier study by Tversky and Kahneman [12]. When the consumer problem was recast into a business decision in the second study, student subjects trained to handle the problem still revealed bias, although to a lesser extent. In the third study, experienced business managers revealed the same pattern of decision bias as the student subjects. Implications of this bias for business decision makers are discussed.     

THE RELEVANCE OF STATE-UNIVERSITY RETIREMENT PLANS IN JOB SELECTION

Documents provided newly hired faculty members were requested from the major statesupported university in each of the fifty states. Those documents revealed two basic types of retirement plans: (1) state-controlled, defined-benefit plans in which retirement benefits are a function of a formula rather than the contributions by the employee and employer, if any; and (2) third-party, defined-contribution, input-based plans in which the funds are controlled by a third party and the benefits depend upon the amount of contributions and the return earned by them. The analysis focused upon the defined-benefit formula plans and a standard was developed to facilitate analysis. The simulations undertaken reveal significant, decision-relevant biases—especially against faculty who elect to change employers as infrequently as once every ten years when different pension plans are involved! The risks attendant to each type of plan and possible career strategies are discussed.     

CONFLICT RESOLUTION IN INVESTMENT DECISIONS: IMPLEMENTATION OF GOAL PROGRAMMING METHODOLOGY FOR DUAL-PURPOSE FUNDS*

Goal programming (GP) is designed to resolve allocation problems with conflicting goals. Both goals and constraints are incorporated in the allocational decision, and the objective function is stated in a way that, upon solution, yields a result “as close as possible” to the priority-weighted goals. The present paper applies GP methodology to the investment decision of dual-purpose funds (DPFs), that are required by law to pursue allocational decisions with potentially conflicting objectives. It provides an empirical demonstration that DPF managers could have improved their investment selection and subsequent performance by the use of GP methodology. Finally the paper stresses the importance of sensitivity analysis to improve both the goal-ranking and target-selection aspects of the methodology and provides a limited but illuminating empirical demonstration of post-optimality analysis.     

EFFECTS OF USE OF COMPUTER IN FIRST-YEAR STATISTICS SEQUENCE ON STUDENT ATTITUDES*

The use of computers in teaching statistics can make an impact upon students in two ways. First, it can affect the amount and rate of learning of statistics. Second, it can affect student attitudes toward computers, statistics and quantitative methods. This note reports a study on the effects of the computer in teaching statistics on student attitudes. The data are based on pre- and post-course questionnaires administered in a first-year statistics sequence. The results partly confirm expectations, and partly raise questions for further investigation.     

COMPARATIVE PRODUCT TESTS IN THE CONSUMER DECISION PROCESS: CORRELATES OF USE AND IMPACT ON SATISFACTION *

Information inputs are an important determinant of the quality of outcomes in the consumer decision process. This paper reports on an exploratory study that examines the perceived use of Consumer Reports product tests in a specific durable-product purchase in a nationwide sample of Consumer Reports subscribers. Perceived use of Consumer Reports is found to be related to performance-oriented shopping, but other situation-related variables are less likely to be associated with use than the shopper's general confidence in the test reports as an information source. Perceived use of the test reports in this study is associated with higher satisfaction with the products purchased.     

THE INTERACTION OF TRANSPORTATION AND INVENTORY DECISIONS

The management of inventories and determination of transportation policy are normally carried out by different groups of people in an organization. These activities interact, however, when the transportation is used to replace inventory and when the transportation alternatives have different speed, reliability, and cost characteristics. This paper presents exact and heuristic procedures for jointly determining the inventory reorder points, order quantities, and transportation alternatives that provide minimum total transportation and inventory costs. The effectiveness of each procedure is demonstrated for a broad range of problems. The heuristic procedure was also applied to actual company data to determine the potential for cost savings. The results showed that savings were possible both from improving the current procedures and from applying the procedure developed in this study.     

MANAGERIAL USE OF DYNAMIC STRUCTURAL MODELS*

Dynamic structural models were introduced as early as 1958 as “Industrial Dynamics, but there has been little managerial use and little response in the academic world. Yet, the basic modeling methods provide an important mode for examining the broad interacting effects of large systems. More recent work appears to make structural and dynamic models understandable and accessible to individuals not trained in the decision sciences. The nature of the modeling methods is such that managers and policy makers in public systems can be involved directly in the model building process. The authors hope that this survey paper may help rekindle interest.     

ON THE USE OF VERBAL PROTOCOLS IN CONJOINT ANALYSIS STUDIES

A recent study by Olshavsky and Acito helped gain insight into (a) the decision process used by respondents when asked to do a conjoint task, (b) the predictive validity of conjoint models compared to models derived from verbal protocols, and (c) the reasons for prediction errors. In this article, the author argues that the Olshavsky and Acito results do not lead one to believe that the compensatory conjoint models can be potentially more misleading than models that try to represent more accurately the underlying evaluation process with noncompensatory and compensatory rules. Hence, it does not seem necessary to try to identify noncompensatory rules in a conjoint study.     

TEACHING EFFECTIVENESS IN THE DECISION SCIENCES: THE USE OF STUDENT EVALUATIONS

This article presents the results of an AIDS-sponsored investigation of teaching effectiveness in the decision sciences. Two national samples were employed in the study: a sample of 247 business school deans and a sample of 3,292 students from 32 different business schools. Included among the findings are that 87 percent of the deans responding indicate they require some sort of faculty teaching effectiveness evaluation, and that great care must be exercised when interpreting student ratings of an individual instructor's teaching effectiveness.     

THE THEORY OF GRAPHS IN BEHAVIORAL SCIENCE*

In this tutorial the author shows how graph theory can be used to model the structural relationships in a behavioral theory. The baśic elements of a graph and its related matrices are defined and illustrated. A survey of applications of graph theory in behavioral science is followed by an extensive bibliography.     

THE SEU CALCULUS: EFFECTS OF RESPONSE MODE,SEX, AND SEX ROLE ON UNCERTAIN DECISIONS *

Subjects were instructed on how to use simple subjective probability and utility scales, and they were asked to actively role-play a decision maker in seven risk-dilemma situations. Each scenario provided subjects with specific subjective expected utility (SEU) information for both a certain and uncertain decision alternative, but left out one critical SEU component. Subjects supplied either the lowest probability or the lowest utility for success that they found necessary before they would select the uncertain over the certain alternative in each dilemma. Three experiments examined: (a) the degree to which Ss' estimations deviated from a pattern predicted by SEU models; (b) differences in choice patterns induced by response format variations (e.g., probability vs. utility estimation); (c) the effects of sex of S; and (d) the effects of the sex-role framing of the decision problems. Ss generally chose in accord with SEU maximization principles and did so with decreasing deviations from theoretical values as practice over situations increased (Experiments I, II and III). Decisions were initially more conservative on items requesting probability estimates (Experiment I), but this effect washed out over situations. Sex differences were revealed (Experiments I and III), but in limited fashion. Rather, a replicable (Experiments I, II and III) sex-by-sex role appropriateness by response format interaction was found, in which females responded “rationally” under both probability and utility estimation conditions and under both role sets (male and female). Males, however, responded extremely conservatively under female-framed, probability estimate conditions. Ss' choices were stable over a three-week interval (Experiment III).     

THE ANALYSIS OF OUTLYING DATA POINTS USING ROBUST REGRESSION: A MULTIVARIATE PROBLEM-BANK IDENTIFICATION MODEL*

Because the eight largest bank failures in United States history have occurred since 1973 [24], the development of early-warning problem-bank identification models is an important undertaking. It has been shown previously [3] [5] that M-estimator robust regression provides such a model. The present paper develops a similar model for the multivariate case using both a robustified Mahalanobis distance analysis [21] and principal components analysis [10]. In addition to providing a successful presumptive problem-bank identification model, combining the use of the M-estimator robust regression procedure and the robust Mahalanobis distance procedure with principal components analysis is also demonstrated to be a general method of outlier detection. The results from using these procedures are compared to some previously suggested procedures, and general conclusions are drawn.     

ELICITATION OF SUBJECTIVE PROBABILITIES IN THE CONTEXT OF DECISION-MAKING *

In this paper a framework is developed to seek information from the decision-maker about uncertain events, and to use this information for identifying a preferred alternative. It is shown how a preferred alternative can be identified with less than perfect knowledge of the probabilities.     

SOME CAUTIONARY NOTES ON THE USE OF CONJOINT MEASUREMENT FOR HUMAN JUDGMENT MODELING*

Conjoint measurement has been suggested as a methodology that might be useful in assisting research concerned with the identification of the structural form of a judge's model. This paper synthesizes the results of some recent research that examined the robustness of this methodology. This research suggests that conjoint measurement has three major weaknesses: (1) certain biases exist when diagnosing model structure, (2) model diagnosis is limited to a small set of potential models, and (3) error substantially compromises conjoint measurement's ability to diagnose model structure. An empirical example that demonstrates some of the difficulties of using this methodology with experimental data is also presented.