Quantile plots for mean effects in the presence of variance effects for 2k-p designs

In robust parameter design, variance effects and mean effects in a factorial experiment are modelled simultaneously. If variance effects are present in a model, correlations are induced among the naive estimators of the mean effects. A simple normal quantile plot of the mean effects may be misleading because the mean effects are no longer iid under the null hypothesis that they are zero. Adjusted quantiles are computed for the case when one variance effect is significant and examples of 8-run and 16-run fractional factorial designs are examined in detail. We find that the usual normal quantiles are similar to adjusted quantiles for all but the largest and smallest ordered effects for which they are conservative. Graphically, the qualitative difference between the two sets of quantiles is negligible (even in the presence of large variance effects) and we conclude that normal probability plots are robust in the presence of variance effects.     

Screening procedures using quadratic forms

In quality control, a performance variable having a two-sided specification limit is common for assessing lot quality. Sometimes it is difficult or impossible to measure the performance variable directly; for example, when testing is destructive, expensive, or when the performance variable is related to the lifetime of the product. However, it may happen that there are several concomitant variables which are easily measured and which correlate highly with the variable of interest. Thus, one may use measurements on these variables to select or screen product which will have a high conditional probability of meeting product specification. We consider this situation when all variables have a joint multivariate normal distribution and the specification limits on the performance variable are two-sided.     

Flexible Copula Density Estimation with Penalized Hierarchical B‐splines

The paper introduces a new method for flexible spline fitting for copula density estimation. Spline coefficients are penalized to achieve a smooth fit. To weaken the curse of dimensionality, instead of a full tensor spline basis, a reduced tensor product based on so called sparse grids (Notes Numer. Fluid Mech. Multidiscip. Des., 31, 1991, 241‐251) is used. To achieve uniform margins of the copula density, linear constraints are placed on the spline coefficients, and quadratic programming is used to fit the model. Simulations and practical examples accompany the presentation.     

Epidemic Mycoplasma Pneumoniae Infection at a College for the Deaf

Abstract

An outbreak of Mycoplasma pneumoniae infection affected up to 110 students and employees at a college for deaf students (Gallaudet). After the introduction of M. pneumoniae onto the campus in late August 1977, the disease spread slowly among the students; the last documented case occurred in mid-December. A thorough review of chart records, mail surveys, a dormitory survey, and a three-month surveillance of the student health service provided an intensive look at the epidemiology and clinical spectrum of a mild to moderately severe disease in the epidemic setting. A matched-pair analysis demonstrated a more serious clinical illness among patients with M. pneumoniae than other patients presenting with febrile respiratory illness. Appropriate antimicrobial therapy appeared to suppress a high antibody response to M. pnuemoniae. A follow-up serosurvey one year after this outbreak demonstrated diagnostic titers to M. pneumoniae in 11 of 30 patients with confirmed infection in the original epidemic.     

Applying the ‘comply-or-explain’ principle: discursive legitimacy tactics with regard to codes of corporate governance

The comply-or-explain principle is a central element of most codes of corporate governance. Originally put forward by the Cadbury Committee in the UK as a practical means of establishing a code of corporate governance whilst avoiding an inflexible “one size fits all” approach, it has since been incorporated into code regimes around the world. Companies can either comply with code provisions or may explain why they do not comply, i.e., why they deviate from a code provision. Despite its wide application very little is known about the ways in which comply-or-explain is used. In addressing this we employ legitimacy theory by which explanations for deviating can be understood as means of legitimizing the company’s actions. We analyzed the compliance statements and reports of 257 listed companies in the UK and Germany, producing some 715 records of deviation. From this we generated an empirically derived taxonomy of the explanations. In a second order analysis we examine the underlying logic and identify various legitimacy tactics. We discuss the consequences of these legitimacy tactics for code regimes and the implications for policy makers.     

Continuing Competence and Relicensure

Abstract

The purpose of licensing laws is to provide assurance of high quality service and to protect the public from incompetent practitioners. Safeguarding entrance to the professions by licensing examinations is one part of this responsibility which is entrusted to the professional boards. Assuring the continuing competence of practicing professionals is another part and a major challenge to every board of nursing throughout the United States.

A 1979 mail survey identified 18 states which have specific continuing education requirements for relicensure of nurses. In addition, three states have standards of practice incorporated in the rules and regulations governing the nursing profession.

Mandatory continuing education is a first step in the implementation of a system to assure the public of the continuing competency for practice of the professional nurse. It is acknowledgement, with the force of law, that continuous renewal of knowledge is a necessary condition for assuring competence. However, it is not sufficient in itself. Knowledge must be integrated in performance, which is the final test of competency. Thus, other system components that assure continuing competency must be developed and implemented. These include the identification of standards of acceptable performance, and the development and implementation of measures to evaluate practicing professionals and assure that they meet the performance standards. These measures include peer review, self-assessment, and recertification procedures. Successful implementation of a complete effective system will require the collaborative efforts of practicing nurses, their professional organizations, and the state boards of nursing.     

A two- sample partially sequential probability ratio test

A two-sample partially sequential probability ratio test (PSPRT) is considered for the two-sample location problem with one sample fixed and the other sequential. Observations are assumed to come from two normal poptilatlons with equal and known variances. Asymptotically in the fixed-sample size the PSPRT is a truncated Wald one sample sequential probability test. Brownian motion approximations for boundary-crossing probabilities and expected sequential sample size are obtained. These calculations are compared to values obtained by Monte Carlo simulation.     

Innovations in teaching social work research

This article reviews the innovations in teaching research in social work education, including conceptual material and specific examples. There is attention to the ideological differences inherent in research and its instruction, methodology/technology innovations, and integrative and group approaches, among other developments. Examination of three major models of research instruction and their assumptions about the purpose of research in social work is presented. Discussion about ethical codes and curricular guidelines for social work programs in the United States and the United Kingdom provides a framework to understand the importance of teaching and learning research for the social work profession.     

Unbiasedness of two-stage estimation and prediction procedures for mixed linear models

The traditional method for estimating or predicting linear combinations of the fixed effects and realized values of the random effects in mixed linear models is first to estimate the variance components and then to proceed as if the estimated values of the variance components were the true values. This two-stage procedure gives unbiased estimators or predictors of the linear combinations provided the data vector is symmetrically distributed about its expected value and provided the variance component estimators are translation-invariant and are even functions of the data vector. The standard procedures for estimating the variance components yield even, translation-invariant estimators.