A Monte Carlo simulation assessed the relative power of 2 techniques that are commonly used to test for moderating effects. The authors drew 500 samples from simulation-based populations for each of 81 conditions in a design that varied sample size, the reliabilities of 2 predictor variables (1 of which was the moderator variable), and the magnitude of the moderating effect. They tested the null hypothesis of no interaction effect by using moderated multiple regression (MMR). They then successively polychotomized each sample into 2, 3, 4, 6, and 8 subgroups and tested the equality of the subgroup-based correlation coefficients (SCC). Results showed MMR to be more powerful than the SCC strategy for virtually all of the 81 conditions.

No abstract

A field study was conducted to examine the effect of timing of leader opinions on group problemsolving performance. University residence life teams ( N = 19) were asked to solve two typical residence hall problems. Teams were randomly assigned to one of two experimental conditions in which leaders either (a) stated their opinions immediately after the presentation of the problem or (b) withheld their opinions until later in the discussion. As predicted, groups with leaders who delayed stating their opinions generated significantly more alternative solutions to the problems, and these alternatives were rated more feasible and more likely to be adopted.

The results of moderated multiple regression (MMR) are highly affected by the unreliability of the predictor variables (regressors). Errors-in-variables regression (EIVR) may remedy this problem as it corrects for measurement error in the regressors, and thus provides less biased parameter estimates. However, little is known about the properties of the EIVR estimators in the moderator variable context. The present study used simulation methods to compare the moderator variable detection capabilities of MMR and EIVR. Specifically, the study examined the bias and mean squared error of the MMR and EIVR estimates under varying conditions of sample size, reliability of the predictor variables, and intercorrelations among the predictor variables. Findings showed that EIVR estimates are superior to MMR estimates when sample size is high (i.e., at least 250) and the reliabilities of the predictors are high (i.e., rij ≥ .65). However, MMR appears to be the better strategy when reliabilities or sample size are low.

The authors thank Maury Buster for providing valuable feedback on earlier drafts of this paper. We also thank three anonymous reviewers and section Editor James Smither for insight and comments that greatly improved the quality of this paper.Correspondence and requests for reprints should be addressed to Laura E. Baranowski,

The present study showed that researchers must consider underlying data structure when using hierarchical agglomerative cluster analysis to group jobs. Five cluster procedures were applied to four simulated data sets constructed to reflect common job analysis situations. The structures contained jobs varying in degree of task overlap, number of tasks performed, and relative number of people doing the jobs. Average linkage/distance was the most accurate procedure when jobs had highly positively correlated task profiles, a situation characteristic of jobs within a career family over a restricted range of levels. Average linkage/correlation was the most accurate for three other structures containing jobs whose profiles were not highly positively correlated. Such are characteristically found when analyzing (a) jobs in different functional units, (b) jobs over a wide range of hierarchical levels such as entry to advanced, and (c) jobs differing markedly in the number of incumbents.

No abstract

Education, training, experience, and licensuref Hunter, 1984). In fact, research has indicated that certification are all types of information that describe over the past 30 years there has been a growing rethe preparation needed before entering an occupa-liance on education for selecting employees (Monation. Education, training, and experience are com-han & Muchinsky, 1983). This is so because evalumonly recognized as mechanisms for acquiring gen-ations of education are relatively cheap and easily era1 knowledge and basic skills (Halpern, 1994), and accessible. licensure and certification are ways of demonstrating Changes in the U.S. workplace will likely lead to the acquisition of those knowledges and skills. There even greater emphasis on education, training, and exare many potential users of the O*NET who will be perience. Individuals who will enter the workforce in interested in having information on occupational the next 15 years have already been born, SO future Preparation for every job in the economY, and it is trends in labor force participation can be predicted only logical that such information should be collected with reasonable accuracy (Fullerton, 1985). The and available. many projected changes in the workforce will affect However, various issues must be addressed when how organizations manage their human resources measuring, collecting, and reporting data on OCCu-(Cascio & Zammuto, 1987). Perhaps the most im-Pational Preparation. What kind of information portant change is the decrease in the growth of the should be collected? HOW should it be collected? workforce. Fewer and fewer young people be How useful will this information be in informing us available for entry-level jobs (Fullerton, 1985). about how people should prepare for occupations? creasing numbers of the young people who will be In this chapter we hope to Provide Some initial available will lack the necessary skills for doing the about education, training, experience, and kensure/ increase in complexity with changes in technology, certification. we begin by briefly reviewing the ra-&ifts from manufacturing to service jobs, and in-