Estimation of Error and Bias in Bayesian Monte Carlo Decision Analysis Using the Bootstrap

Bayesian Monte Carlo (BMC) decision analysis adopts a sampling procedure to estimate likelihoods and distributions of outcomes, and then uses that information to calculate the expected performance of alternative strategies, the value of information, and the value of including uncertainty. These decision analysis outputs are therefore subject to sample error. The standard error of each estimate and its bias, if any, can be estimated by the bootstrap procedure. The bootstrap operates by resampling (with replacement) from the original BMC sample, and redoing the decision analysis. Repeating this procedure yields a distribution of decision analysis outputs. The bootstrap approach to estimating the effect of sample error upon BMC analysis is illustrated with a simple value-of-information calculation along with an analysis of a proposed control structure for Lake Erie. The examples show that the outputs of BMC decision analysis can have high levels of sample error and bias.     

Analysis of the Bovine Spongiform Encephalopathy Maternal Cohort Study: the Discordant Case–Control Pairs

The progeny of the affected and non-affected dams in the bovine spongiform encephalopathy (BSE) cohort study were paired in terms of their farm of origin and approximate date of birth. This paper presents an exact logistic regression analysis that avoids modelling the differences between the pairs and so analyses only the discordant pairs. There is some evidence that the pairs in which the affected animal was born to the affected dam tend to be those in which the onset of BSE in the affected dam preceded or occurred soon after the calving. This suggests some occurrence of maternal transmission.     

A comparative analysis of public sector restructuring in the U.S., Canada, Mexico, and the Caribbean

We examine public sector restructuring in North America and selected Commonwealth Caribbean nations. Although all the countries studied experienced significant restruc-turing in response to public debt pressures, there were major differences across coun-tries in the magnitude, pace, form, and the manner in which restructuring decisions were made. These differences reflect the state of economic development and institu-tional characteristics, e.g., the role of the state and the industrial relations system. In developing countries, international lending institutions played a major role in trans-forming the role of the state. In developed countries, the inherent stability of the eco-nomic systems and institutional pressures led to a gradualist approach to restructuring.     

A Translation and Validation Study of the Life Orientation Test Revised in the Greek Speaking Population of Nurses among Three Hospitals in Athens and Ioannina

The life orientation test-revised (LOT-R) (Scheier et al. in Journal of Personality and Social Psychology 67:1063–1078, 1994) is a brief measure for assessing dispositional optimism. The aim of this study was to develop a Greek language version of the LOT-R and to assess the instrument’s psychometric properties. The LOT-R was translated and culturally adopted in Greek language, and the final version was administered, along with a questionnaire consisting socioeconomic characteristics and a single item measuring optimism, to 276 Greek speaking, hospital nurses (222 female, 54 male), aged 22–65 years (mean 37.8, SD 8.3). Results showed that the LOT-R has good internal consistency (Cronbach’s α = .71 and item total correlation coefficients from .27 to .73, a unitary structure, and stability over a 3-months period (r = .66). Moreover, the Greek version of the scale exhibited good convergent validity with single-item optimism scale (r = .73). Principal components analysis revealed a two-factor structure representing the constructs of optimism and pessimism. The Greek life orientation test revised appears to be a valid tool in assessing dispositional optimism in Greek speaking people and is expected to facilitate the examination of optimism in Greek speaking populations.     

Accountability in research using persons with mental illness

Although medical research involving the use of persons with mental illness is critically important, in order for the research to be ethical and legal there are certain considerations and restrictions which should be immediately readdressed in order to insure that the welfare of these vulnerable research subjects is protected, and their best interests are assured. A brief historical examination of medical research codes, guidelines, recommendations and Federal Regulations reveals the various considerations and restrictions on informed consent and accountability applicable to the use of persons with mental illness in medical research. Several concerns are raised about how these considerations and restrictions have been interpreted, and specific recommendations are offered to improve them immediately by means of representation from consumers and/or their families, and organizations, e.g., NAMI members.     

De-aliasing effects using semifoldover techniques

Semifoldover designs, obtained by semifolding a regular two-level factorial design, have been discussed recently in the literature. In this article, with the use of indicator functions, we investigate various semifoldover designs that are obtained from a general two-level factorial design. We discuss when a main factor or a two-factor interaction can be de-aliased from their aliased two-factor interactions, and extend some of the existing results from regular designs to non-regular designs. Finally, we present some examples to illustrate the results developed here.