Prior-likelihood factorization and missing data

Marginal posterior distributions, when not available ana­lytically, can be at present numerically inaccessible if the number of parameters for intergration exeeeds 7 to 10. For the normal multivariate regression model, with data absent (missing)in a monotone pattern, some integrations have been accomplished analytically (Guttman and Menzefricke, 1983; Bartlett, 1983; for example).

In this note we show how monotely missing data support an extended prior-likelihood factorization and the needed posterior extended prior-likelihood factorization and the can be obtained directly using standard results.     

Research Notes the Relationship Between Admissions Criteria and Academic Performance in an MSW Program

The authors report on an investigation of the correlation between admissions criteria and academic performance at one state university. The admissions data used in this study were collected from 654 MSW students who attended the school from 1985 to 1992. Test scores on a comprehensive exam taken by these MSW students prior to graduation were used as the measure of academic performance. The authors found that two admissions criteria, undergraduate GPA and GRE score, and two applicant characteristics, gender and race, were significantly correlated with academic performance.     

Integrating Environmental Issues into Social Work Practice: Lessons Learnt from Domestic Energy Auditing

Abstract

This study of domestic energy audit practice draws on current literature and focuses on a program at Kildonan UnitingCare. It explores and highlights the relevance of natural environmental issues for social work practice and includes practical ideas for addressing financial, social, and environmental issues within case work practice and advocacy. The overlap between energy audit and social work practice is said to apply to both service user issues and the professional skill base of social workers. Finally, a transformation in social work thinking practice and education is suggested, within a vision that is community focused and holistically integrated.     

Vocational rehabilitation in multiple sclerosis (MS): a profile of clients seeking services

Although a number of innovative vocational rehabilitation (VR) demonstration projects have been completed with MS populations, there remains a lack of clarity as to a profile of these VR participants. An effort is made in this article to carefully describe the demographic and occupational characteristics of people with MS seeking vocational services at intake, as well as their self-perceived psychosocial functioning across a range of pertinent measures. The issue of "program dropouts" in VR services for people with MS is also addressed. Implications for better understanding participants with MS within the VR process and improved nuances in service delivery are then discussed.     

Community,the private and the individual

Attempts are being made to associate a supposed ‘return to the private’ with a parallel phenomenon of a ‘return to community’. Both terms, private and community, are central to sociological theory, but are surrounded with confusion and a profusion of historical referents which confirms their importance but obscures their significance. This review of the literature and of the various problematics for which the ‘private sphere’ is a key concept concludes that it is civil society which determines the modalities of discussion as well as the actual degree of autonomy of the ‘private’, and thus that an approach from the side of the ‘private’ (or of ‘community’) can be of only restricted value.     

Fibre analysis and tangent models

Available methods for deriving tests and confidence regions for multiparameter models seem seriously limited in their feasibility. This paper gives preliminary survey to some developing techniques that use conditioning as a primary device to derive tests and confidence regions. The techniques usefibre analysis which involves the conditional determination of OLS (observed level of significance); the techniques are available initally for models with pivotal quantities and for continuous exponential models. The techniques can then be extended to more general models using the notion of tangent model.     

Views on conditional and marginal methods of statistical inference

Conditional and marginal likelihood analysis has a long history of development. Some recent methods using exact and approximate density and distribution functions lead to more sharply defined likelihoods and to accurate observed levels of significance for a wide range of problems including nonnormal regression and exponential linear models. These developments will be surveyed.     

PARADOX AND ORTHODOX: FOLIE À DEUX?*

In the recent family therapy literature, therapists have been cautioned to use paradoxical looking techniques as either a last resort, or else not to use such actions at all with specific problems. The paper argues that these stances arise from a rigid adherence to a set of orthodox views from which these actions appear contradictory. Similar to family systems, when such orthodox therapy systems do not question or alter their own premises in the face of contradiction, they tend to attribute either badness or madness to the action and either try to exorcise the badness, or banish the action. The literature which either prescribes “paradox as a last resort,” or which proscribes it altogether is reviewed and critiqued. Walter Buckley's process/adaptive system model is recommended as a positive alternative view from which such previously pardoxical actions may often be elected as a logical and effective first choice.