The objective of this paper is to make known some activities in Stochastic Education in the Ibero-American countries. To achieve this objective I have collected information about working groups; Conferences on Statistics Education, Conferences on Statistics and Mathematics Education with papers about Stochastic Education; Journals that devote special issues to the Statistics Education and resources on Internet. Below I summarize this information.
We describe our project to develop curricular materials for a course that introduces students at the post-calculus level to statistical concepts, methods, and theory. This course provides a more balanced introduction to the discipline of statistics than the standard sequence in probability and mathematical statistics. The materials incorporate many features of successful statistics education projects that target less mathematically prepared students. The student audiences targeted by this project are particularly important because they have been overlooked by previous curricular reform projects. Most importantly, the proposed audience includes prospective teachers of statistics, introducing them to content and pedagogy that prepare them for implementing NCTM Standards with regard to statistics and probability and for teaching the Advanced Placement course in Statistics.
This paper will describe a content-pedagogy course designed to prepare elementary and middle school teachers to teach statistics in the schools. The course is organized around the newly revised content standards developed by the National Council of Teachers of Mathematics. A central objective is to encourage teachers to see statistics as a problem solving process. The course has been implemented as one component of the "Learning Math" Project. Produced by WGBH with funding from the Annenberg/Corporation for Public Broadcasting Math and Science Project, "Learning Math" is developing a series of five college-level courses designed to teach mathematics content to elementary and middle schools teachers. In the statistics course, nine video sessions follow an actual class of teachers through content classes, with footage edited to highlight critical statistical concepts. An on-line course, which parallels the nine videos, is also being developed.
Whatever the debates about the relation between mathematics and statistics as disciplines, the latter is typically offered within school mathematics curricula. This relatively new inclusion has enhanced the opportunity for learners to experience a greater relevance of mathematics curricula to their own lives, and hence also created the imperative to better understand how best to organise teaching and learning toward such goals. Not surprisingly, teacher education has had to take on such challenges and in so doing brought a focus also on what happens within the halls of tertiary institutions. The question this paper addresses is how best do we prepare teachers to connect mathematics and statistics education to learners' own realities. If project work, within a broad social, cultural political approach, is one means for forging such links then there is a need to analyse and better understand the kinds of teacher education pedagogies that may be engaged to build the necessary knowledge, skills, attitudes and values among teachers.
A common concern in the professional development of teachers is to provide them with appropriate technical support and training in the use of information and communication technology (ICT). The need for such training is a current concern in, amongst other places, the UK and the Northern Territory of Australia, and this has provided the first motivation for the development of an appropriate training course emphasizing ICT tools such as the web, email and MS(tm) Excel. The second motivation is through a desire to help the development of teaching data handling in schools. Consequently, the Royal Statistical Society (RSS) Centre for Statistical Education, UK, together with Dr Ian Roberts of Northern Territory University, Australia, developed an ICT-type training course that is centred on data provided by the UK-based CensusAtSchool project (http://www.censusatschool.ntu.ac.uk).
In their first and often only statistics course, health-care professionals are taught Bayes' theorem in the context of diagnostic testing. They learn the concepts of sensitivity/specificity and predictive value positive/negative and how Bayes' theorem can assist in diagnostic decision-making. Then the class moves on often spending weeks on tests of significance. This paper will argue for changing this practice, and instead focusing such courses on statistics for decision-making beyond diagnostic testing. It will argue that such changes will make our health-care professionals better consumers of statistical information and better decision makers.
In this paper we will explore the challenge of making statistics more meaningful to future nurses. In the fast moving undergraduate student world the expectations we place upon nursing students are considerable. Typically they experience high class-contact hours in addition to their clinical placements. Compounding the problem, undergraduate nursing students have diverse mathematical backgrounds and seldom perceive statistics as being relevant for them. Given these constraints we have adopted the relatively modest aim of producing informed and discriminating consumers of statistics and research, rather than skilled statistical practitioners or researchers. With a focus on computer output rather than by-hand calculations, we have made use of strategic examples, appropriate journal articles and an historical hypothetical. This approach has both relieved the anxiety and distraction associated with calculations and increased students' engagement in the learning process.
Each year, a new crop of physicians enters residency training programs in medical teaching institutions worldwide. Second and subsequent year residents continue with the programs in which they have participated in a prior year. The educational curriculum may include a biostatistical component, where the instructor is presented with an opportunity to focus on biostatistical issues bearing on various aspects of medical practice and research. This paper describes such a presentation in a university medical school residency training program. The training session centered on research findings published recently in the medical literature. Issues regarding topic, journal, and article selection, teaching aids, approaches to illustrating aspects of study design, power analysis, statistical methodology, and interpretation of results, promoting contact with a biostatistical consultant, and feedback from lecture attendees are described.
Teaching of statistics involves developing and adapting robust procedures for understanding statistical concepts, and for the management and analysis of statistical data. The field of statistics is constantly challenged problems that arise from science, industry and business. Traditionally, the statistics curriculum deals with data often collected to answer specific questions. However, in the modern 'information' age, vast amounts of data are collected, often automatically, with the advent of powerful computers. Data Mining is the process of extracting knowledge from large volumes of data. Since 'computation' plays a major role in this process, computer scientists have a significant claim over the ownership of data mining. Nevertheless, data mining techniques, in general, have a statistical base; and statisticians are beginning to show a significant interest in the area, including offering tertiary courses in 'statistical' data mining.
In this paper, two examples of multilevel modeling as part of the analysis of data from HIV evaluation studies are presented. Strategies for teaching multilevel models for each type of data are discussed. The first, a panel study, uses multiple linear regression models to show how a hierarchical linear model can be developed. The second, a repeated cross-sectional design, uses simple analysis of variance models to show how a random coefficients model can be fit to the data. Complex multilevel models may be easier to understand and apply when broken down into these more familiar strategies. Analyses are presented using the HLM program and SAS.