Proceedings

  • Multivariate data in ecological applications most often occur in the form of counts of species abundances in assemblages, where each species is a variable. These data do not generally conform to traditional statistical assumptions, and so special approaches and methods are needed in this context. Statisticians need to be informed about these special problems with ecological data. In addition, the rationale for complex experimental designs that is a trademark of most ecological studies needs to be well understood by applied statisticians in this area. On the other hand, successful approaches for teaching ecologists about the use of multivariate statistics include sticking to the conceptual, rather than the mathematical. I provide here an overview of the methods that have helped teaching across these two disciplines, including a general approach for the use of novel non-parametric methods in the analysis of ecological community data.

  • The U.S. Federal statistical agencies have long faced the challenge of creating an expert, stable workforce with skills in census and survey methods relevant to the production of official statistics - interdisciplinary skills not available in a single academic department. Numerous statistical commissions identified this need, but no ameliorative action was taken until the fiscal year 1991 Boskin Initiative for Economic Statistics promoted funding. The Joint Program in Survey Methodology was selected through a process administered by the National Science Foundation. Funding now comes through a contract with participation from twelve statistical agencies. Present and future agency employees participate in academic and short courses and seminars and are enrolled in citation, certificate, and graduate degree programs. Survey problems are addressed in courses that allow transfer of knowledge between the work environment and the academic learning experience. This partnership has substantially enhanced the capabilities of the federal statistical workforce. The paper reviews challenges and successes through this partnership.

  • Much of the material in the graduate survey sampling course is tedious to teach and learn. The classroom is enlivened and students are better able to use the concepts taught in the course when they have some experience applying it in real populations. This paper discusses some activities or mini-projects that can be used in the classroom to give students this experience without introducing come of the complexities associated with a full-scale project.

  • The use of significance tests in science has been debated from the invention of these tests until the present time. Apart from theoretical critiques on their appropriateness for evaluating scientific hypotheses, significance tests also receive criticism for inviting misinterpretations. Although these misinterpretations are well documented, until now there has been little research on pedagogical methods to remove them. Rather, they are considered "hard facts" that are impervious to correction. We discuss the roots of these misinterpretations and propose a pedagogical concept to teach significance tests, which involves explaining the meaning of statistical significance in an appropriate way. The present contribution is based on Krauss and Wassner (2001) and Haller and Krauss (in press).

  • For many years null hypothesis testing (NHT) has been the dominant form of statistical analysis in psychology. It has also been subject to periodic criticisms from within the field of psychology. In the past decade these occasional criticisms have turned into a more or less steady stream which have lead some to call for an outright ban on NHT in psychology, while others have called for greater use of alternative procedures. The solution lies neither in banning NHT nor in relying solely on alternative procedures, but in "reforming" NHT, replacing a-theoretical null hypotheses with theoretically meaningful hypotheses. Such reform requires that training of researchers emphasize parameter estimation and the testing of theoretical models, an approach that exists in some areas of psychology and appears to be common in other sciences. Such an emphasis will help ensure that the statistical hypothesis being tested matches the substantive hypothesis of interest. I will discuss the changes that are occurring in psychology and propose further changes that are still needed.

  • Applied researchers are often interested in obtaining confidence intervals for key nonlinear model parameters so as to answer important research questions, and the usual "plus and minus 2 SE's" confidence interval leads easily into the usual Wald hypothesis test covered in most introductory statistics courses. However, since information about a specific parameter is often asymmetric, a skewed confidence interval is often more appropriate and reasonable in practice. This leads to the use of likelihood-based tests, typically introduced in intermediate undergraduate and basic graduate course. This paper argues that the superiority (in terms of for example increased power) of likelihood-based and score hypothesis tests over the Wald test is most easily conveyed and appreciated by first providing a reasonable motivation (as well as examples) using confidence intervals, and then exploiting the confidence interval-hypothesis test equivalence.

  • In this paper I consider the characteristics of a statistically literate (a "statisticate") person. I suggest that a statisticate person should be able to read and understand statistical arguments of moderate complexity, and to carry out statistical analyses to some degree. Significantly, the truly statisticate person should also have developed the habit of thinking quantitatively. Furthermore, he or she does not rely on rigid rules to make statistical decisions, but uses informed judgment. In particular, he or she should understand the concepts of modelling and selection between models, and recognise their importance. Consideration is given to one of the major barriers to developing statistacy: the vocabulary used, in particular, the common use of two words that should only be used with the greatest of care (if used at all). These words are "prove" and "true". An important illustration of the way that vocabulary hinders the development of understanding is the case of hypothesis testing, a vital statistical tool that is widely misunderstood. It represents a mode of thought that is fundamental to statistical analysis, and so belongs in the kit bag of any statisticate person.

  • The widespread use of rating scales in applied research fields implies need of statistical methods applicable to various types of studies involving ordinal response data. The aim of this paper is to present a teaching model of joining research courses in rating scale data analysis for statisticians and applied scientists together in order to stimulate inter-disciplinary communication. The participants experienced the complexity of applied research problems that involve subjective assessments on scales and also some of the possibilities and limitations of novel and classical statistical methods of analysis.

  • At the National Statistical Institute (NSI) a Programme of Technical Cooperation with Spanish-speaking countries of America was started up some twenty years ago (1978). The aim of the programme was to exchange experiences and offer further training to professionals working at National Statistical Offices, Central Banks, Ministries of Planning and Customs, and the main institutions producing the official statistics of the countries in the Region. The training activities consist of courses, seminars and workshops, which are held both in Spain and other countries. These are also supplemented with "étages" in Spain and the sending of "experts" to the countries requiring advice on new projects. The training activities are predominantly practical, although they also have a theoretical module of statistical methodology and a conceptual framework of the different projects. The purpose of this Programme is to train personnel in the production of official statistics, pass on the new technologies from Spain and the European Union, and supply them with the material and documentation used to carry out the different projects.

  • This paper, after considering the reasons and aims that gave origin to the International Conference on Teaching Statistics (ICOTS), traces the line of thought along which the Conference developed from 1982 to 2002. This is done by applying textual data analysis to the titles of the papers published in the Proceedings of the first five Conferences, and to the titles that were on the International Programme Committee Web site on October 27, 2001. Knowing past and present enables one to present suggestions about possible future Conference scientific developments.

Pages