Groups can make a difference: voting power measures extended

The voting power of a voter—the extent to which she can affect the outcome of a collective decision—is often quantified in terms of the probability that she is critical. This measure is extended to a series of power measures of different ranks. The measures quantify the extent to which a voter can be part of a group that can jointly make a difference as to whether a bill passes or not. It is argued that the series of these measures allow for a more appropriate assessment of voting power, particularly of a posteriori voting power in case the votes are stochastically dependent. Also, the new measures discriminate between voting games that cannot be distinguished in terms of the probability of only criticality.     

现代突厥语的内部分类及其历史意义(二)

本文旨在建立一种现代突厥语的语言内部分类模式.该模式基于共时资料,但像以前的模式,它试图整合历时资料.除语音资料外,这一模式还利用了形态学、词汇学和句法学方面的资料.根据历时和地理体因素,对这些特征的分布模式进行了相互比较.由此,我们可以划分出突厥语不同层级的内部语组.而且,这些子语组可以有不同的来源.其中的一些语组是由共同的语言特征维系在一起的,在这个群的语言内部它们似乎具有可从一个时间层面传递到另一个时间层面,并构成他们共同的发生学遗传的语言特征.笔者将这类群称为发生学的(子)分支,如鸟古斯语支、克普恰克语语支、布尔加尔突厥语支等.显然,其他突厥语的子语群是由不同分支的语言组成,它们已经形成了新的区域单位,进入区域互动.诸如此类的区域群有西伯利亚突厥语.在文中第二部分,笔者从历史事实和发展的时序关系方面,对发生学分支进入子分支的内部分类过程和区域群的形成及分解过程进行了阐述.     

Verfassungen als Spiegel globaler Normen?

The founding of the United Nations permitted to define the discrimination of women as a problem concerning world society at large. Today, equal rights are considered as a global norm by which states and organisations are measured. The article focuses on the questions of how far global norms of equal rights govern national constitutions and on which kind of factors their national designs depend. For answering these questions we use a quantitative content analysis of family rights and equal rights in national constitutions (n = 164). In contrast to neo-institutionalist assumptions we found three clearly distinguished models of equal rights. They serve as dependent variables in a regression analysis which shows that equal rights evolve independently of the national degree of development. To the contrary, the cultural and religious tradition of a country, its inclusion into world society and the power of its civil society are far more decisive for their evolution.     

Lord Moser

Three prime Ministers—Wilson, Health, Callaghan—listened to him when he was head of the Central Statistical Office. As advisor on social policy, and especially on higher education policy, he has shaped our universities. Now he is a member of the House of Lords. Claus Moser spoke to Julian Champkin .     

Employee performance management across borders: A review of relevant academic literature

The purpose of this paper is to survey the emerging academic literature on employee performance management (PM) from an international perspective. The primary information source was a search of electronic databases. Two criteria were used to determine which studies to include in the literature survey: (1) those published in an English-language blind refereed academic journal, and (2) those focused on employee PM or performance appraisal (PA) from an international/global perspective. Sixty-four articles, published between 1985 and 2005, met the criteria. A preliminary scheme was developed for classifying the existing academic research into conceptual and empirical articles. The empirical articles were further classified by their foci and themes. In addition to classifying the articles, the paper presents a summary of what can be learned from the major findings of the empirical studies. The major conclusion is that the academic literature on cross-border PM is relatively atheoretical and exploratory in nature and that the design and substance of the research studies are weak. With regard to themes, there are a number of foci in the literature with only scant attention paid to the central concern with the PM of expatriates as expressed by firms. The empirical articles look at the disparate components of PA and are interested mainly in cultural differences as an intervening variable. Several recommendations are presented for researchers to help focus future research on cross-border PM.     

Children as insurance

This paper presents a dynamic model of fertility decisions in which children serve as an incomplete insurance good. The model incorporates uncertainty about future income and the survival of children as well as a discrete representation of the number of children. It contributes to the understanding of the negative relation between fertility and education, shows why parents may demand children even if the return is negative, and explains why fertility might rise with increasing income when income is low and decrease when income is high. Furthermore, the model can account for the decline in fertility when the risk of infant and child mortality decreases. Finally, the implications for empirical tests of the demand for children are also examined. Received: 8 September 1998/Accepted: 9 June 1999     

A simple method for finding a legal configuration in complex Bayesian networks

This paper deals with an important problem with large and complex Bayesian networks. Exact inference in these networks is simply not feasible owing to the huge storage requirements of exact methods. Markov chain Monte Carlo methods, however, are able to deal with these large networks but to do this they require an initial legal configuration to set off the sampler. So far nondeterministic methods such as forward sampling have often been used for this, even though the forward sampler may take an eternity to come up with a legal configuration. In this paper a novel algorithm will be presented that allows a legal configuration in a general Bayesian network to be found in polynomial time in almost all cases. The algorithm will not be proved deterministic but empirical results will demonstrate that this holds in most cases. Also, the algorithm will be justified by its simplicity and ease of implementation.     

Weighted Logrank Permutation Tests for Randomly Right Censored Life Science Data

In biomedical research, weighted logrank tests are frequently applied to compare two samples of randomly right censored survival times. We address the question how to combine a number of weighted logrank statistics to achieve good power of the corresponding survival test for a whole linear space or cone of alternatives, which are given by hazard rates. This leads to a new class of semiparametric projection tests that are motivated by likelihood ratio tests for an asymptotic model. We show that these tests can be carried out as permutation tests and discuss their asymptotic properties. A simulation study together with the analysis of a classical data set illustrates the advantages.