The identification of logistic regression models with errors in the variables

The simple logistic regression model with normal measurement error and normal regressor is shown to be identifiable without any extra information about the measurement error. The multiple logistic regression model with more than one regressor variable measured with error is not identifiable. If the covariance matrix of the measurement error is known up to a scalar factor, the model is identified. Further we discuss why in spite of the identifiability the models cannot be estimated in a reasonable way without extra information about the measurement error.     

On social work and what gender has got to do with it

Living in gendered societies makes it necessary to take into account the specific ways in which social problems and social issues affect the lives of women and men, girls and boys. Not only are social problems and social issues very often gender biased; this is also true for the professions dealing with these problems and issues, and for the training programmes for professions like social work. Sociological analysis of society and philosophical understanding of a good life are permeated by gender-specific perceptions -- often without those involved being aware of them. Furthermore, social policies reinforce gender-specific biases because they are based on gender-specific (i.e. male) standards. Gender as an analytical category has been brought into scientific and public discourse by feminist theoreticians and activists making women -- and thus also men -- visible with regard to structure and subject issues. The development of gender theory and research highlights the significance of gender arrangements to understand social phenomena and to support gender-equality ways of thinking and policies. At the same time, men and women are not only confronted with gendered structures, but via socialization processes and personal development they also reproduce these gender differences in their daily lives by 'doing gender'.     

Editorial

Ohne Zusammenfassung

---

Editorial

     

The state of social capital: Bringing back in power,politics, and history

Theory and Society -     

Changing times in pharmaceutical statistics: 2000‐2020

In a previous paper we considered how pharmaceutical statistics had changed between 1980 and 2000. In this paper we go on to consider some of the likely influences and changes in the world of pharmaceutical statistics over the next twenty years. Statistical research by pharmaceutical companies is supported, although its direct value in boosting company profits is challenged. The duality between fraud and conflict of interest is explored, particularly with regard to data and safety monitoring boards. Political correctness within clinical trials and (allegedly) better patient‐oriented outcomes is questioned and the whole notion of intention‐to‐treat and per‐protocol analyses is rejected in favour of pragmatic trials and explanatory trials. The dominance of SAS software within the industry is explored and the possibility for change considered. Finally, whilst the future of pharmaceutical companies is not addressed, the impact of future mergers, collaborations and take‐overs between clinical research organizations is considered. Copyright © 2002 John Wiley & Sons, Ltd.     

Führung durch Selbstführung

This paper outlines the conceptual bases, strategies, and psychological attributes of leadership by self-leadership. In addition, relations to other theories of leadership are drawn. Dimensions and correlates of self-leadership competence are described and discussed. Various approaches for developing self-leadership competence are discussed as well.