Forecasting: A hybrid approach

It has recently been shown that the Box-Jenkins approach to forecasting time series is superior to an econometric approach over a relatively short horizon. The results here support this contention. A combination of the two approaches, however, proves to be clearly superior to either one separately. By taking account of changes in economic and weather related variables in a time series model, improved forecasts are obtained.     

Corporate social responsibility: an assessment of the enlightened self-interest model

Much recent discussion of corporate social responsibility has concerned operationality. Many activities subsumed under corporate social responsibility can be shown to be public or partially public goods. The theory of public goods can clarify and explain some complex problems of operationalizing the social responsibility doctrine. An examination of philanthropy provides some behavioral applications.     

The Impact of Racism on Social Work Education

One of the most difficult problems in teaching social services has to do with the development of an analytic framework for an understanding of the delivery of social services within the community. This paper examines a particular systems framework that allows the student to conceptualize and evaluate what services are being delivered, what specific inputs and connections of agency with community act to produce this service, and what changes in the system may be possible. The model allows ready connection with practice, policy, and research components of the curriculum.     

The impact of non-bottleneck variation in a manufacturing cell

Protective capacity is the 'extra' capacity placed at non-bottleneck resources to absorb random disruptions in planned levels of performance so that the bottleneck resource continues to be effectively utilized. A full factorial experiment with a simulation model was conducted to explore issues associated with the quantity and location of processing variance in a five-station manufacturing cell. The cell's performance was measured using both mean flow time ( MFT ) and bottleneck shiftiness ( SHIFT ) for 3 patterns of variance for the non-bottlenecks at 5 different levels of variation. In order to investigate the importance of the quantity of added capacity on the variation both a low level of protective capacity (10%) and a high level (50%) were considered. The results indicate that having the higher variation work centres close to the bottleneck provides reduced MFT and SHIFT . The performance measures improved at both the low and high setting of protective capacity.     

Concepts and Precepts In Social Work Education

While a single elective course in human sexuality is a useful supplement to core social work courses in preparing students for the sex-related problems they will encounter, a more comprehensive sequence of learning experiences is necessary for those who wish to develop greater competence in this area.

This article describes such a sequence offered by the University of Hawaii School of Social Work which focuses on treating and preventing sex-related problems experienced by various populations, particularly the sexually oppressed These learning experiences include class and field activities, research projects, and guided experiences as sex educators at local agencies, as producers of a weekly radio program, and as leaders of small discussion groups for a basic sex education course.     

The estimation of derivatives of a nonparametric regression function when the data are correlated

A kernel estimator of a derivative of arbitrary order of a nonparametric average population curve is considered for a correlated-errors model with balanced replicate measurements at each design point. Asymptotic expansions of the mean squared error are derived for two classes of correlation functions in the model. Consistency, choice of smoothing parameter, and rates of convergence are examined for the important special cases of estimating the first and second derivatives.     

Empowering Community-Based Programs for Youth Development

This article discusses the need for community-based programs for youth and their families and urges the social work profession to resume its historical leadership in such programs while incorporating recent theory and practice. The authors describe the reasons why social workers abandoned neighborhood-based youth development services, the intended outcomes of such services, and current societal and professional trends that support a revival of social work’s involvement in youth development. They discuss one school of social work’s current participation in a youth agency partnership and examine implications of the suggested revival for social work education. Collaboration among families, schools, diverse community services, agencies, and educators is emphasized.     

Volunteerism and Social Work Educators: Findings from a National Survey

Abstract

This study reports the results of a national survey designed to identify the nature and levels of volunteer work that social work educators contribute to the social welfare sector. Findings reveal that nearly all social work educators engage in formal volunteer work and perform a variety of roles in human service agencies. The most common volunteer activities are advocacy, administration and consultation. The least common volunteer activities are evaluation, data management and emergency services. Study results also indicate that tenure status and primary teaching area are positively associated with the amount and type of volunteer work performed.     

Expected absolute departure of chi-square from its median

We develop a formula for the expected absolute departure of x² from its median.     

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.