Local information and the design of sequential hypothesis tests

Various results on sequential hypotheses testing are reviewed. Optimal stopping rules are related to a local measure of statistical information. In some cases, local information can be approximated by L-numbers discovered by Lorden, and simple rules based on these approximations are asymptotically optimal to better order than the cost for a single observation.     

Personality preferences in laboratory economics experiments

Student volunteers at the U.S. Naval Academy (USNA) participated in one of the following one-shot games: a dictator game, an ultimatum game, a trust game, or a prisoner's dilemma game. We find limited support for the importance of personality type for explaining subjects’ decisions. With controls for personality preferences, we find little evidence of behavioral differences between males and females. Furthermore, we conclude that seniority breeds feelings of entitlement—seniors at USNA generally exhibited the least cooperative or other-regarding behavior.     

Design priorities and disciplinary perspectives: the case of the US National Children's Study

Summary.  The National Children's Study (NCS), which was undertaken in 2000 by collaboration between several US federal government agencies, is one of the largest and boldest longitudinal studies of children's health ever undertaken. One of the key design issues has been the nature of the NCS sample. The paper describes the nature of the choices and the reasons for the decision that the NCS be based on a national probability sample. Designed as a study of the environmental influences on children's health and development, the NCS is expected to identify, enrol and follow about 100000 children from their birth to the age of 21 years. A broad definition of relevant environments of interest, and a full partnership between government, university and medical scientists, introduces considerable challenges in the design of the study.     

Seat belt laws: why we should keep them

In the June 2007 issue of Significance , John Adams argued for the repeal of seat belt laws. They have never been demonstrated to save lives, he said; and they increase the number of pedestrians and cyclists killed on the roads. Richard Allsop, Oliver Carsten, Andrew Evans and Robert Gifford argue that Parliament should keep the seat belt laws on the statute book and that we should all keep wearing belts whenever we travel in a car, van or coach—and they contest John's statistics.     

A cross-national study of a method for researching non-profit organisational effectiveness

Are there criteria that define non-profit organisation effectiveness? If the criteria can be identified, is it possible to use them in a comparative study to assess whether and in what way the use of the criteria may vary between nations? This study addresses both questions. The study uses the factorial survey method and is based on the social constructionist conception that effectiveness is created as an ongoing and sometimes contested outcome of a social judgement process. We illustrate the method with samples from the US and UK and compare the results. The results establish the value of the factorial survey method. Analyses show that both US and UK respondents use percentage expenditures on fund-raising, advocacy emphasis, change in financial surplus and unit cost of services as effectiveness criteria. There is evidence that voluntarism is a more important criterion in the US and that advocacy is more important in the UK.     

Portfolio assessment in a collaborative program evaluation: the reliability and validity of a family literacy portfolio

Evaluations increasingly use open-ended assessments, such as portfolios, to gather information about program impact. Portfolios, however, present challenges in achieving acceptable levels of reliability and often lack evidence to support validity. This article describes the components of a portfolio assessment system developed in a collaborative evaluation of a family literacy program. Staff of a family literacy program and two evaluators collaborated to develop the assessment and the tools that frame the review of the portfolios. Rubrics, exemplar portfolios, portfolio logs, and lists of ‘look-fors’ were developed to guide raters in the scoring and to improve the reliability of scores and validity of decisions about the program. Inter-rater reliability for the portfolio assessment was 0.79 for holistic scores and 0.74 for analytic scores. With respect to validity, holistic and analytic scores correlated with criterion measures from 0.50 to 0.59, providing initial evidence of a common construct.     

The positive science of ethics in France: German influences onDe la division du travail social

What was Durkheim doing—in the sense of an intended social action—in writing De la Division du travail social? At least a part of the answer is that Durkheim's project was linguistic—i.e., he was attempting to replace an outworn vocabulary of Cartesian metaphysics with a more Germanic lexicon—one in which simplicity gave way to complexity, the abstract to the concrete, the ideal to the real, deduction to induction, rationalism to empiricism, and so on. To some extent, this was motivated by the superiority—widely acknowledged among intellectuals of the Third Republic—of German science and Protestant scientific education. But an additional motivation was Durkheim's belief that only a real, concrete entity—society as a “thing” (chose)—could provide an object worthy of the veneration of the “new man” of the Republic. Durkheim's attempt to construct a science of social facts was therefore itself subsidiary to another, “higher” purpose—i.e., the construction of a moral authority (real, concrete, complex) adequate to the needs of the Third French Republic. Rather than an end in itself, Durkheim's sociology should thus be seen as a means to other ends—i.e., the “construction” of a particular kind of “fact”—within a specific social and historical context.