Employment and the Risk of Domestic Abuse Among Low‐Income Women

This paper uses data from 2 randomized evaluations of welfare‐to‐work programs—the Minnesota Family Investment Program and the National Evaluation of Welfare‐to‐Work Strategies—to estimate the effect of employment on domestic abuse among low‐income single mothers. Unique to our analysis is the application of a 2‐stage least squares method, in which random assignment enables us to control for omitted characteristics that might otherwise confound the association between employment and domestic abuse. We find that increased maternal employment decreases subsequent reports of domestic abuse in both studies. In the Minnesota Family Investment Program—a program with an enhanced income disregard that allowed welfare mothers to keep a portion of their welfare income as earnings rose—an increase in household incomes appears to have contributed to reductions in reports of domestic abuse.     

A scientist's viewpoint on promoting effective use of experimental design: Ten things a scientist wants to know about experimental design

This paper tackles the practical application of statistical tools. Chemical development is a rich area for collaboration between statisticians and chemists, but most statisticians encounter difficulties in promoting the tools to this client group. This paper, presented from a chemist's perspective, is intended to help bridge the gap between the two professions. The paper explores differences in uptake and attitudes towards adopting these techniques and suggests what has hindered or helped the process of effective application of experimental design. Copyright © 2003 John Wiley & Sons, Ltd.     

Allocation of research and development funds: A zero-one goal programming approach

The research and development project selection process is one of the most difficult and important problems faced by management. It is typically complicated by indivisibility of projects and multiple and conflicting objectives, in addition to limitations on funding, facilities, and qualified researchers. In this paper a case example involving a high technology electrical equipment manufacturer is developed to illustrate this problem using zero—one goal programming to accommodate indivisibility of projects in addition to multiple and conflicting goals. The model presented is an attempt to provide managers with a robust tool for allocating scarce resources among research and development projects.     

Change,clarity and clausewitz

The framework of laws which in the past has governed the interaction of technology and business is proving inadequate as the scale and interconnectedness of economic, social and technological problems accelerate. The author evinces the need for new systems of control wherein both the macro and micro levels of perception of and response to large scale, multifaceted problems can be harmonised.     

Deprivation and lifestyles in rural Wales. — I. Towards a cultural dimension

This paper reports on the preliminary findings of research into deprivation in mid-Wales. It argues that the current debate on rural deprivation has left many important questions unanswered. In particular, we still know relatively little about the extent of deprivation, the specific social groups affected by it, the particular experiences of these groups, and the importance of cultural dimensions to that experience. Indeed, the very term deprivation has become somewhat chaotic, and may subsume a number of different ways in which individuals and groups can become marginalised in rural areas. The two case studies in rural Wales discussed here give some indication of the extent and anatomy of marginalisation within particular communities. Perhaps more importantly they highlight the need to consider the importance of cultural marginalisation within a wider understanding of the problems of living in rural areas.     

Cancer Risk Assessment with Intermittent Exposure

Applications of methods for carcinogenic risk assessment often focus on estimating lifetime cancer risk. With intermittent or time-dependent exposures, lifetime risk is often approximated on the basis of a lifetime average daily dose (LADD). In this article, we show that there exists a lifetime equivalent constant dose (LECD) which leads to the same lifetime risk as the actual time-dependent exposure pattern. The ratio C = LECD/LADD then provides a measure of accuracy of risk estimates based on the LADD, as well as a basis for correcting such estimates. Theoretical results derived under the classical multistage model and the two-stage birth-death-mutation model suggest that the maximum value of C, which represents the factor by which the LADD may lead to underestimates of risk, will often lie in the range of 2- to 5-fold. The practical application of these results is illustrated in the case of astronauts subjected to relatively short-term exposure to volatile organics in a closed space station environment, and in the case of the ingestion of pesticide residues in food where consumption patterns vary with age.     

COURSEWORK AND SELF-STUDY IN THE FAMILY TREATMENT OF ALCOHOL AND DRUG ABUSE: EXPANDING HEATH AND ATKINSON'S CURRICULUM

In designing a course on the family treatment of alcohol and drug problems, a number of important issues and topics need to be addressed. A design by Heath and Atkinson (1988) provides an excellent, innovative model for such coverage. These authors, however, had to contend with certain constraints that have had a bearing on the course content and how it is taught. The present paper offers some revisions and alternatives to Heath and Atkinson's model in order to make it generalizable to a greater number of contexts. For clinicians who are less interested in teaching such a course, the two papers, combined, could serve as a guide for self-study in this area.     

Multinomial and conditional logit discrete-choice models in demography

Although discrete-choice statistical techniques have been used with increasing regularity in demographic analyses, McFadden's conditional logit model is less well known and seldom used. Conditional logit models are appropriate when the choice among alternatives is modeled as a function of the characteristics of the alternatives, rather than (or in addition to) the characteristics of the individual making the choice. We argue that this feature of conditional logit makes it more appropriate for estimating behavioral models. In this article, the conditional logit model is presented and compared with the more familiar multinomial logit model. The difference between the two techniques is illustrated with an analysis of the choice of marital and welfare status by divorced or separated women.