This project develops and tests a new approach for improving the effectiveness of personnel selection and classification decisions There are three main data characteristics that cause difficulty for existing classification methods: (1) unbalanced group sizes, (2) unequal misclassification costs, and (3) non-normal data Preliminary research suggests a hybrid method incorporating data envelopment analysis (DEA) and linear programming discriminant analysis (DEA/DA) is effective in this difficult situation and outperforms other methods Research is conducted to fully develop and test this promising methodology. Results suggest ways DEA/DA may help alleviate long-standing selection and classification problems.

Bringing together several key elements needed to identify the most promising themes for future research in selection and classification, this book's underlying aim is to improve job performance by selecting the right persons and matching them most effectively with the right jobs. An emphasis is placed on current, innovative research approaches which in some cases depart substantially from traditional approaches. The contributors -- consisting of professionals in measurement, personnel research, and applied and military psychology -- discuss where the quantum advances of the last decade should take us further. Comprehensive coverage of the selection and classification domain is provided, including a broad range of topics in each of the following areas: performance conceptualization and measurement, individual differences, and selection and classification decision models. The presentations in each of these areas are integrated into a set of coherent themes. This integration was the product of structured group discussions which also resulted in a further evolution of some of the ideas presented.

Beginning in the early 1980s and continuing through the middle 1990s, the U.S. Army Research Institute for the Behavioral and Social Sciences (ARI) sponsored a comprehensive research and development program to evaluate and enhance the Army's personnel selection and classification procedures. This was a set of interrelated efforts, collectively known as Project A. Project A had a number of basic and applied research objectives pertaining to selection and classification decision making. It focused on the entire selection and classification system for Army enlisted personnel and addressed research questions that can be generalized to other personnel systems. It involved the development and evaluation of a comprehensive array of predictor and criterion measures using samples of tens of thousands of individuals in a broad range of jobs. The research included a longitudinal sample--from which data were collected at organizational entry--following training, after 1-2 years on the job and after 3-4 years on the job. This book provides a concise and readable description of the entire Project A research program. The editors share the problems, strategies, experiences, findings, lessons learned, and some of the excitement that resulted from conducting the type of project that comes along once in a lifetime for an industrial/organizational psychologist. This book is of interest to industrial/organizational psychologists, including experienced researchers, consultants, graduate students, and anyone interested in personnel selection and classification research.