Moments and efficiency of the median and trimmed mean for finite populations

Exact formulas for the expected value and variance of the median and trimmed mean are found as functions of the elements of a finite population under simple random sampling. A simulation study is performed to compare the performance of the median and trimmed mean versus the mean when sampling from various simulated finite populations. Finally, the asymptotic performance of these estimators, when sampling from infinite populations, is compared with the finite populations results.     

Variance Reduction Using Nonlinear Controls and Transformations

Nonlinear regression-adjusted control variables are investigated for improving variance reduction in statistical and system simulations. To this end, simple control variables are piecewise sectioned and then transformed using linear and nonlinear transformations. Optimal parameters of these transformations are selected using linear or nonlinear least-squares regression algorithms. As an example, piecewise power-transformed variables are used in the estimation of the mean for the twovariable Anderson-Darling goodness-of-fit statistic W ₂ ². Substantial variance reduction over straightforward controls is obtained. These parametric transformations are compared against optimal, additive nonparametric transformations obtained by using the ACE algorithm and are shown, in comparison to the results from ACE, to be nearly optimal.     

A Two-Latent-Class Model for Smoking Cessation Data with Informative Dropouts

Non ignorable missing data is a common problem in longitudinal studies. Latent class models are attractive for simplifying the modeling of missing data when the data are subject to either a monotone or intermittent missing data pattern. In our study, we propose a new two-latent-class model for categorical data with informative dropouts, dividing the observed data into two latent classes; one class in which the outcomes are deterministic and a second one in which the outcomes can be modeled using logistic regression. In the model, the latent classes connect the longitudinal responses and the missingness process under the assumption of conditional independence. Parameters are estimated by the method of maximum likelihood estimation based on the above assumptions and the tetrachoric correlation between responses within the same subject. We compare the proposed method with the shared parameter model and the weighted GEE model using the areas under the ROC curves in the simulations and the application to the smoking cessation data set. The simulation results indicate that the proposed two-latent-class model performs well under different missing procedures. The application results show that our proposed method is better than the shared parameter model and the weighted GEE model.     

ELIZABETH BARRETT BROWNING'S "FAMILY DISEASE": ANOREXIA NERVOSA

What was Elizabeth Barrett Browning's mysterious and variously diagnosed “disease”? This study reviews biographical material in the light of modern family systems theory and develops a hypothesis that Elizabeth Barrett lived with her family of origin in “anorectic transaction”, and that chronic anorexia nervosa was embedded in the Victorian poet's lifestyle. The characteristics of anorectic families: enmeshment, overprotection, rigidity, and lack of conflict resolution, are shown to have been abundantly present in Barrett family patterns. Robert Browning is compared with a family therapist whose interventions restructure a system and free the anorectic from her dysfunctional symptom.     

Sociology and the Nation-State in an Era of Shifting Boundaries

Academic disciplines and nation-states resemble one another in the sense that both emerged worldwide as overarching categories for conferring collective identities about a century ago. During the past quarter century, however, the status of both has been challenged. Disciplines like sociology have been challenged by more salient intellectual boundaries organized around subdisciplinary, transdisciplinary, and supradisciplinary lines—a fact that helps account for sociology's response to external threats to the discipline with a sense of being in crisis. Nation-states have been similarly challenged by new jurisdictions organized around subnational, transnational, and supranational lines. Even so, these changes constitute a process of differentiation, not displacement; disciplines and nation-states are likely to persist in serving important functions. albeit reduced in scope.     

A unified theory for generalized cyclic designs

Properties of abelian groups are used to unify the construction of cyclic and generalized cyclic designs. Necessary and sufficient conditions are obtained for the generation of fractional sets of blocks, and also for the generation of resolvable designs. It is shown how to apply these results in conjunction with efficiency arguments to obtain useful designs.     

Limitations on the Usefulness of Risk Assessment

Quantitative risk assessment (QRA) is now regarded as an essential component in the analysis of risks arising from installations classified as major hazards. The purpose of this paper is to discuss the value of the results in decision-making in practical situations. The use made of QRA in three contrasting cases which came to extensive public attention in the U.K. is examined. The first concerned an extension of domestic development near a chemical factory; the second an extension to a large petrochemical complex; and the third to a proposal to build a pressurized water reactor. The two public inquiries concerned with the chemical industry accepted standards of individual risk which were comparable to the risks from everyday accidents; the evidence of societal risk that could arise from major accidents at the petrochemical complex was compared with that of a local natural hazard — flooding. Higher standards of individual safety were set in the inquiry into the PWR proposal, and the definition of societal risk was debated at length. The QRA results were analyzed to show that risks arising from accidents were lower than those from normal operations, but they were used explicitly as a check on the overall safety of the design and of the operational and licensing organization. Such qualitative examination will always be required in addition to QRA. All these inquiries were faced with considerable technical argument. There is a need for the full details of risk calculations to be clear. The usefulness of QRA as an input to decision-making would be much enhanced if the technical points at issue could be clarified outside a formal public inquiry. In addition, there are some technical questions which apply to many installations. There should be better mechanisms of technical debate to achieve a measure of agreement on the optimum methods of calculation in these cases, and some possibilities are explored.