This report examines the consequences of increasing the Navy depot's role in the logistics system by directing its resources toward the day-to-day needs of the fleet. Using a simulation that examined whether mission capability could be improved during a 90-day war through some combination of responsive stock management, proactive use of depot repair capabilities, and shortened transportation pipelines between carriers and depots, the authors found that priority repair at the depot can make on important difference in mission capability, that shortened pipelines can have large effects on mission capability, and that constructing an aviation consolidated allowance list (AVCAL) based on aircraft availability goals may offer promise for maximizing aircraft availability per dollar spent. The study also concluded that data synthesis is a missing ingredient in the Naval aviation logistics management system that inhibits the depot's ability to react quickly in support of sudden demand peaks.

A Data Management and Statistical Services (DMSS) core unit within RAND's Population Research Center performs common functions including: database management review, specialized programming, support for data acquisition, and statistical consulting. In fitting the DMSS unit's services to RAND's organizational setting and particular project needs, a model of this unit's functions has evolved that both enhances scientific quality and promotes further efficiencies. This model--and lessons learned in implementing it--may interest those who direct or administer other population research centers. This Note describes the model's rationale and key features; reviews the problems in doing research with large datasets; identifies the different types of researchers typically involved in a data-intensive research project and the analytic steps each type performs; describes an efficient allocation of research talent across these tasks; and considers how the Center's DMSS unit fosters that allocation.

The authors conducted a workforce study of eye care providers (ophthalmologists, non-ophthalmic physicians, and optometrists) in the United States. Using data from a RAND survey of U.S. ophthalmologists and from publicly available datasets, the authors determined the supply of available providers, and estimated the present and future public health need for eye providers and the current level of demand. The authors then reconciled the supply of providers with present demand and need for services with future need through the year 2010. The study found an overall large surplus of eye care providers in the United States. Which type of provider will have the greatest surplus depends on whether the health care delivery system allocates patients to optometrists or to general ophthalmologists for primary care.

How much of the world's oil and energy supply will the non-OPEC less-developed countries (NOLDCs) demand in the next decade? How will their economic growth affect world demand? To answer these questions, the authors have tried to develop some reasonable forecasts of NOLDC energy demands in the next ten years. Although the focus is mainly on demand for oil, some attention is given to the total commercial energy requirements of these countries. The data used in fitting the models cover 77 NOLDCs, which, in 1976, accounted for 79 percent of total oil consumption by all 124 NOLDCs. The uncertainties associated with the forecasts are spelled out, as are the income and price elasticities on which the forecasts are based. Finally, the authors consider the forecasts in terms of their implications for U.S. energy policies concerning the NOLDCs and suggest areas of future research on NOLDC energy issues.

Because censored sampling is often unavoidable in much sociological data analysis, computationally simple corrections of censoring bias would be useful. Heckman's correction is simple to compute, widely used, and proven asymptotically correct under certain assumptions, but its limitations in practical situations are not well known in sociology. Here, we overview prior criticisms of Heckman's estimator, and we consider the case in which its normality assumptions are satisfied, censoring rates are high, and sample sizes are small. Results of 14,400 analyses of computer-generated simulation data suggest that Heckman's method performs well under certain circumstances, but that it very frequently worsens estimates, especially under conditions that are likely to be present in sociological data. Thus, the technique is probably not a general cure for censoring bias in sociology, except perhaps where strong theory permits certain strong assumptions. We reconsider censored sampling correction strategies in the context of statistical analysis as a theory-building tool, with emphasis on research strategy in the presence of irremediable censoring bias.