Ce travail sur la désimplication des salarìés propose un modèle qui intègre les dimensions financière et comportementale et fait appel à des formules mathématiques qui fournissent une estimation du coût global de la désimplica-tion pour l'entreprise. Pour nous différencier des tentatives antérieures de calcul d'un tel coût, nous avons essayé d'englober toute une série de comportements et d'attitudes de retrait, incluant le retrait psychologique, le freinage, les retards, absences et démissions. L'estimation du coût prend en compte des éléments directs et indirects, certains d'entre eux étant même traditionnellement considérés comme ne relevant pas de l'économie. Les équations reposent sur le fond théorique emprunté à la littérature sur le comportement comprenant l'établissement des normes, les interactions à l'intérieur de l'équipe de travail et l'éventuelle dérive d'attitudes et de comportements de retrait relativement discrets vers des formes plus accentuées. On utilise dans les équations des corrélations (corrigées empiriquement) entre les différentes formes de retrait. Ce modèle a été exploité pour évaluer les coûts de la désimplication pour l'année fiscale 1997 dans une société israelienne de dimension moyenne leader dans le domaine de la haute technologic. Il est apparu que le coût de la dé-simplícation était approximativement de 2,8 millions de dollars américains, ou encore de 16,5% du bénéfice avant impôts de l'entreprise. On discute aussi des consequences théoriques et pratiques de ces résultats.The present study proposes an integrative behavioral-financial model of employee withdrawal combined with mathematical formulas, allowing the estimation of the overall cost of withdrawal to the firm. In contrast to previous attempts to calculate such a cost, we aimed to cover an entire range of withdrawal attitudes and behaviors, including psychological withdrawal, withholding effort at work, lateness, absence, and turnover. Cost calculation involves direct and indirect components, including those that were traditionally considered non-economic. The equations are based on theoretical grounds of the behavioral literature, including norm determination, mutual interpersonal influences within the work teams, and the possible progression of relatively mild withdrawal attitudes and behaviors into more severe forms. Empirical mean corrected correlations between withdrawal forms are used in the equations. We applied the model by assessing the withdrawal costs for the fiscal year 1997 in a leading, medium-sized, Israeli high-tech company. It was found that the withdrawal cost was approximately 2.8 million US dollars, or 16.5 per cent of the company's before-tax income. Theoretical and applied implications of the results are discussed.

No abstract

This article highlights the virtue of integrating well‐being metrics (e.g., psychological well‐being, perceived meaning) into aspects of utility analysis for the purpose of enhancing human resource management strategies and worker performance. We present the reader with a review of conceptual and practical developments in this field and examples of utility analysis calculations, while we advocate for the necessity of including well‐being metrics in utility analysis for the 21st century. The basic thrust of this effort is to encourage the greater employment by managers of quantitative models that allow decision makers to generate all the factors needed to estimate long‐term financial gains and/or losses before any intervention strategy is implemented in the workplace. As indicated, the use of quantitative models to estimate the net financial gains of using particular intervention strategies, accompanied with the value estimation of certain types of employee states (e.g., psychological well‐being) and worker behaviors (e.g., employee turnover), can ultimately save companies from making gross tactical errors and, more positively, can assist management in promoting the organization's long‐term economic goals in conjunction with the enhanced well‐being of employees.

No abstract

Final ratings on twenty-five assessment dimensions were obtained for 382 candidates for high-level management positions in an Israeli corporation. They were determined by an assessment center process, two overall clinically derived assessment scores and an overall mechanically derived assessment score. For 49 of the assessees, scores were also available on a battery of paper-and-pencil cognitive tests administered at the preemployment stage. For all candidates, data were obtained on two criterion measures over a period of four years.In contrast to previous assessment center research, both the general intelligence rating formulated in the assessment center and the scores on the cognitive tests were found to be predictively invalid, while the personality measure proved valid. Despite the modest coefficients of predictive validity found for the other assessment center predictors, the economic utility analysis performed with an amended formula demonstrated a considerable cost-effectiveness for the assessment center program.Implications of the findings are discussed with a focus on the practical and economic value of the assessment center. ResumeTrois cent quatre-vingt-deux candidats a des postes de cadre superieur dans une societe isra£lienne ont £t6 soumis a une evaluation finale fondee sur 25 dimensions reliefs a l'emploi. Ces Evaluations ont ete produites a l'aide d'un processus utilise dans un centre devaluation, de deux notes globales obtenues cliniquement et d'une note globale calculEe mecaniquement. On disposait aussi des resultats d'une batterie de tests cognitifs papier-crayon qui avaient ete administres a 49 candidats avant leur embauchage. Dans le cas de tous les candidats, des donnees ont £t£ recueillies en mesurant deux crit&res sur une p^riode de quatre arts.Contrairement aux autres recherches sur les centres d'evaluation, 1'evaluarion de l'intelligence globale produite par le centre devaluation et les notes obtenues par les candidats aux tests cognitifs se sont reVe'lEes non valides sur le plan predictif, ce qui n'a pas 6te le cas de la mesure de la personnalite. Malgre les coefficients peu sieve's de la validite predictive, etablis

No abstract

No abstract

No abstract