EMPLOYER SEARCH, TRAINING, AND VACANCY DURATION

We develop a model of optimal employer search strategy when information about match quality is endogenous. The model is tested using four data sets, two of which have not previously been used. As theory predicts, we find that when filling positions requiring more training, employers search more intensively and extensively. Employers also search more extensively when hiring workers with more education and with prior experience. These findings provide strong evidence of systematic variation in search strategies based on the characteristics of the positions and job applicants. Factors that influence employer search also affect the duration of a vacancy.     

MANUFACTURING PROCESS TECHNOLOGY and SUPPORT STAFF COMPOSITION: AN EMPIRICAL VIEW OF INDUSTRY EVIDENCE

Despite the attention given to restructuring and trimming down manufacturing firms during the 198Os, little attention has been paid to the mix of skills they needed under different circumstances. We examined the patterns of employment by occupation in manufacturing industries utilizing different production technologies and the effect of establishment size on nonproduction employment. We found that a relationship exists between production technology and nonproduction employment per 100 production workers. Establishment size is found to be a moderator between nonproduction employment and production technology. Our findings imply two clear messages for managers. First, when considering major changes in production technology, managers should be aware that the supporting skills they will need from their nonproduction work force are likely to change greatly. Further, these changes involve technical and managerial workers as well as clerical and production support people. Second, they should restructure the functional or occupational mix of an organization in the context of the process technologies in place. Different process technologies require different structures.     

Unified Conditional Frequentist and Bayesian Testing of Composite Hypotheses

Testing of a composite null hypothesis versus a composite alternative is considered when both have a related invariance structure. The goal is to develop conditional frequentist tests that allow the reporting of data-dependent error probabilities, error probabilities that have a strict frequentist interpretation and that reflect the actual amount of evidence in the data. The resulting tests are also seen to be Bayesian tests, in the strong sense that the reported frequentist error probabilities are also the posterior probabilities of the hypotheses under default choices of the prior distribution. The new procedures are illustrated in a variety of applications to model selection and multivariate hypothesis testing.