Standard errors for the parameters of graphical Gaussian models

The comparison of an estimated parameter to its standard error, the Wald test, is a well known procedure of classical statistics. Here we discuss its application to graphical Gaussian model selection. First we derive the Fisher information matrix and its inverse about the parameters of any graphical Gaussian model. Both the covariance matrix and its inverse are considered and a comparative analysis of the asymptotic behaviour of their maximum likelihood estimators (m.l.e.s) is carried out. Then we give an example of model selection based on the standard errors. The method is shown to produce almost identical inference to likelihood ratio methods in the example considered.     

A CAUTIONARY NOTE ON ENVIRONMENTAL ECONOMICS

Environmental questions have become a permanent feature of the political landscape in contemporary Australia and now play a significant role in election campaigns at both state and federal levels. Indeed, one of the most significant policy debates in the 1996 Federal election focussed on the environmental policy proposed by the Liberal/National Party coalition, which sought to fund a host of environmental programs from the proceeds of a part privatisation of Telstra. The trade-offs involved in this policy have generated substantial opposition to it and served to underline the prominence of environmental questions in Australian politics. Despite the importance of ‘the environment’ in popular thought, public debate is still typically characterised by ‘feel good’ policy announcements by politicians, sensationalist media reporting and deceptive claims by some environmental interest groups. Whilst progress has undoubtedly been made, the result has often been a plethora of inchoate and frequently conflicting policies. Not surprisingly this ‘muddle through’ approach has produced numerous instances of government intervention not only failing to meet its planned objectives but also generating unintended consequences.     

Varieties of “Fundamentalism”: a Conceptual and Empirical Analysis of Two Protestant Denominations*

Many behavioral science studies of religious fundamentalism are seriously hampered by conceptual confusion. The major source of much of this confusion is the lack of a coherent sociological definition and theoretical context for the term fundamentalism. We propose a Parsonian definition of Fundamentalism, which is interpreted in the context of a theory of evolutionary social change. The methodological implications of this approach are identified and then applied in a survey research study of the lay leaders of two historically related denominations, the Disciples of Christ and the Church of Christ. Our factor analysis of 25 questionnaire items suggests there are three types of Fundamentalism. Socio-economic variables explain some of the variation in the Fundamentalism scores, but denominational affiliation explains more. Our findings suggest that future research should use denomination-specific indices of Fundamentalism.     

Careful The Tale You Tell Fairy tales,drama and moral education

SUMMARY: In this article, I examine theories which argue both for and against the educational potential of fairy tales, with special attention to the field of moral learning. I conclude that, in the contemporary world, stories have a particular importance for such learning but that the hidden moral values conveyed by many traditional tales may well be disquieting for a teacher. I propose that educational drama offers an appropriate pedagogy for exploring such values with children and describe in some detail a case study, in which I used a version of a traditional Hindu tale for these purposes. I offer a brief analysis on how the strategies employed by drama teachers can be harnessed for moral educational purposes and suggest that drama can offer children much needed opportunities to actively and creatively engage with stories and their values in a communal framework.     

A winning strategy for lotto games?

In lotto games, the distribution of k-tuples chosen by participants is not uniform, but the chance of any k-tuple being the winner is the same. The winning categories consist of matching exactly k — i numbers from the winning k-tuple for i = 0, 1, …, m for some m. The total prize pool for a category is divided equally among all the winning tickets in the category. Therefore the strategy of buying a ticket with a k-tuple consisting of unpopular numbers will increase the expected amount of the prize if this k-tuple is a winner in some category, because the prize pool is shared among fewer tickets. By modelling the distribution of 6-tuples chosen by participants of Lotto 6/49 in Canada, the expected return and standard deviation of return can be computed. It is shown that the expected return can be more than the amount spent when the carryover is large, but the large standard deviation means that it would take tens of thousands of years to millions of years for the strategy to have a high probability of yielding a profit.     

An international definition of social work for China

There are differing opinions on how far the international values and internationally understood practice of social work, as epitomised by the international definition on social work, can apply universally to all countries, notably China. The lack of capacity in the Chinese social work community to develop evidence-based and adaptable practice has hindered the effective learning and systematic 'indigenisation' of overseas social work knowledge. This article argues that the international definition can serve as an active working platform where Chinese social workers and the international social work community can search for commonalities and facilitate mutual understanding. Social workers can only effectively respond to the trends towards globalisation by trying to find out more about their common identity and similarities through interactions, while at the same time showing more acceptance of the differences between countries.