Evaluations of Drug-Testing Programs

In three experiments undergraduate students at an urban university evaluated drug-testing programs described in scenarios. Evaluations were inversely related to the false positive rate and were more favorable when a positive test result led to rehabilitation than when it led to termination of employment. Inconsistent effects on evaluations were observed for ability of the test to distinguish between previous drug use and current drug-based impairment, false negative rate, and implications of drug-impaired performance for public safety. There was also mixed evidence concerning the relation between evaluations and respondent belief in a just world. These results provide some support for Crant and Bateman's (1989) justice-based model of reactions to drug-testing and have implications for organizational drug-testing policies.     

国内外石油公司财务风险实证研究

根据我国实际修正Z值模型,研究国内外多家石油公司财务数据的财务风险问题,得出我国石油上市公司财务风险逐年上升的结论,由此研究我国石油上市公司财务风险控制弱于国际顶尖石油公司的原因,并从资本运动角度提出了一些合理化建议。     

Some model-restricted shrinkage estimators for contingency tables with missing data

For contingency tables with extensive missing data, the unrestricted MLE under the saturated model, computed by the EM algorithm, is generally unsatisfactory. In this case, it may be better to fit a simpler model by imposing some restrictions on the parameter space. Perlman and Wu (1999) propose lattice conditional independence (LCI) models for contingency tables with arbitrary missing data patterns. When this LCI model fits well, the restricted MLE under the LCI model is more accurate than the unrestricted MLE under the saturated model, but not in general. Here we propose certain empirical Bayes (EB) estimators that adaptively combine the best features of the restricted and unrestricted MLEs. These EB estimators appear to be especially useful when the observed data is sparse, even in cases where the suitability of the LCI model is uncertain. We also study a restricted EM algorithm (called the ER algorithm) with similar desirable features. Received: July 1999     

Comparisons of Bayesian estimation procedures for two-way contingency tables without interaction

Bayesian and empirical Bayesian estimation methods are reviewed and proposed for the row and column parameters in two-way Contingency tables without interaction. Rasch's multiplicative Poisson model for misreadings is discussed in an example. The case is treated where assumptions of exchangeability are reasonable a priori for the unknown parameters. Two different types of prior distributions are compared, It appears that gamma priors yield more tractable results than lognormal priors.     

Multiple and Partial Correlation Coefficients of Fuzzy Sets

In many applications, multiple correlation and partial correlation for three or more fuzzy sets are very important, but Chiang and Lin (1999, Fuzzy Sets and Systems 102: 221–226) do not solve this problem. Here, we propose a method to calculate the multiple correlation and partial correlation for fuzzy data, by adopting the concepts from the multivariate correlation model. In order to fit into normal framework, we use empirical logit transform (see, Agresti, [1990, Categorical Data Analysis. New York: Wiley]; Johnson and Wichern, [1992, Applied Multivariate Statistical Analysis 3rd edn. Engelwood Cliffs; Prentice-Hall.]) for membership function grades to achieve this.     

Workplace Substance Testing—An Exploratory Study

This study, using 1993 survey data taken from large Ontario, Canada, headquartered organizations (n = 127), compares workplace substance testing versus nontesting organizations based on a number of external and internal factors. Results showed that few Canadian organizations conduct workplace substance testing. Variables such as sector, location of operations, risk sensitivity and organizational structure appear to be not related to the workplace substance testing decision. Partial support was found for age and size of organization (measured in sales volume) as differentiating workplace substance testing organizations from non-testers. Public policy implications based on these findings are provided.     

Confronting the issue of substance abuse in the workplace: When organized labor and management work together

The focus of this report is to describe the efforts of management (Midwest Coca-Cola) and organized labor (Local 792, International Brotherhood of Teamsters) in attempting to confront the issue of substance abuse (drugs and alcohol) by employees covered by a collective bargaining agreement. The conclusion is reached that the experience at Midwest Coca-Cola's bottling facility is an excellent example of management and organized labor recognizing a common threat, arguing to deal with the threat, and attempting through joint action to confront the menace of substance abuse in the workplace.     

Idiot's Bayes—Not So Stupid After All?

Folklore has it that a very simple supervised classification rule, based on the typically false assumption that the predictor variables are independent, can be highly effective, and often more effective than sophisticated rules. We examine the evidence for this, both empirical, as observed in real data applications, and theoretical, summarising explanations for why this simple rule might be effective.     

A model of employee responses to drug-testing programs

Testing employees for drug use is an increasingly widespread organizational response to the problem of employee drug abuse. Despite this, little attention has been given to its effects on employee attitudes and behavior. This paper reviews the issue of drug testing in industry, provides a theoretical model of employee perceptions of and responses to drug-testing programs, and offers an agenda for future research directions.