• Two issues often arise in the teaching of statistical models and methods to large groups of<br>students in non-statistics programmes. Firstly, such students struggle with the relevance and the<br>applicability of the material to their own discipline. Secondly, these students struggle with the<br>notion of randomness, how to model it and how to account for it in their analyses. One approach<br>to dealing with the second issue is to expose students to as much data as possible, for both single<br>and multiple contexts, by providing individual data sets to each student for tutorial and<br>assignment work. This has been the approach taken in developing DOTS, Directed Online<br>Tutorials for Statistics, for students in engineering and the health sciences at the University of<br>Queensland. The current version of DOTS will be presented and discussed, as well as future<br>directions.

  • Early attempts to define statistical thinking revolved around discussions of the need for data,<br>understanding the nature of variability and how statisticians go about solving statistical<br>problems. More recently, Wild and Pfannkuch (1999) have proposed a framework in which they<br>identify five types of thinking they perceive as being fundamental to thinking statistically. We<br>understand these types of thinking as a mapping onto the statistical problem solving cycle (real<br>world problem, statistical problem, statistical solution, real world solution) and as providing us<br>with a beginning definition for statistical thinking. Wild and Pfannkuch (2004), in their paper on<br>understanding statistical thinking discussed the historical development of statistical thinking,<br>emerging from this discussion was a broader view of statistical thinking, namely, statistical<br>thinking is a way of making sense of the world, a particular world view.<br>We believe understanding statistical thinking as a world view may provide additional insights<br>into how we, as educators, can recognise, develop and assess statistical thinking within our<br>students. We explore the links between these constructs and what we observed when we<br>introduced a new teaching and learning strategy into a statistics design and analysis subject.

  • In this paper we report on the results of an experiment conducted in the unit Introduction to<br>Statistics at the University of Canberra. A variety of strategies, referred to as games, sets and<br>matches, were employed, many of which emanate from the teaching of foreign languages. These<br>included personal strategies for the lecturer such as recording lectures, individual strategies for<br>the students such as the use of Hot Potatoes software, and group strategies for the students such<br>as vocabulary cards and in-class activities and discussions. The aim of the experiment was firstly<br>to improve students' use of statistical language, and secondly to see if we were indeed still<br>teaching statistics, and improving overall student performance in the unit..

  • Examples have long been an integral component of statistics teachers' instructional repertoires<br>but tend to be in the background of pedagogical knowledge. We explore the diverse ways that<br>university statistics educators use examples, drawing on data from recent research (Gordon, Reid<br>&amp; Petocz, 2007). Three overlapping categories are proposed: examples are developed and<br>presented by educators in basic instruction, examples are generated by students, under teacher<br>direction, to aid learning and examples connect statistics with students' future professional work.<br>Expressions in the second category were sparse suggesting an opportunity for statistics educators<br>to develop teaching. We review models of exemplification in mathematics education and relate<br>these to the empirical findings to begin the development of a framework for characterising<br>examples in statistics education. We conclude that examples help promote statistical literacy.

  • Prior beliefs and attitudes of students can have a significant impact on their learning experience<br>but it is usually difficult to engage with student beliefs in depth when dealing with large classes.<br>As part of an ALTC Associate Fellowship project, we have developed technologies and strategies<br>for facilitating connections between staff and student beliefs through the embedding of student<br>reflective writing in statistics courses. Students were free to write whatever they liked in their<br>journals but weekly themes were also provided to give them a starting point if needed. The aim of<br>this paper is to give an overview and analysis of entries around the themes that were particularly<br>related to beliefs about statistics, as well as to demonstrate the use of text mining tools in this<br>context.

  • Many tertiary institutions now include 'Data Mining' as a topic in their Statistics curriculum,<br>both at undergraduate and postgraduate levels. The choice of software for learning the topic of<br>Data Mining is an interesting issue to think about. There is a wide range of such software<br>available, from commercially popular ones such as SAS/Enterprise Miner, Statistica/Data Miner<br>and S-plus/Insightful Miner to free ones such as R and Weka. The main aim of this paper is to<br>discuss the pros and cons of such software, including their capabilities to handle and manipulate<br>large volumes data, all from a teaching/learning point of view at both introductory and more<br>advanced levels.

  • Nowadays, statistics is taught to university students in many non-statistics disciplines, such as<br>chiropractic and health. In this study we examined and compared chiropractic students'<br>experiences of using two statistical software programs, one web based and another Excel based,<br>to identify the strength and weakness of each program that might influence statistical learning.<br>Data were collected through a survey of students, and analysed using the chi-squared and the<br>McNemar tests. The web based program appeared to have a good potential due to its particular<br>features and functions, flexibility and easy accessibility. Although the web based program has<br>better graphical displays, broader range of analytic capabilities and remote access, students<br>found the Excel based program relatively easier to use than the web based one. This study<br>suggests that more online instructions and explanations are needed in the web based program for<br>non-statistics students, to aid their statistical learning.

  • The development of safe, efficacious and cost effective medicines and medical devices requires<br>dedicated teams of people with diverse backgrounds including all manner of the sciences,<br>regulatory affairs, marketing, health economics and so on. Statistics and statisticians also have<br>an integral role in the drug development industry. In fact, it is probably one of the few industries<br>where there is a regulatory requirement that statisticians MUST be involved in each project from<br>the design of an experiment through to the report. The life cycle of a medicinal product, focusing<br>on the statistical needs of various team members and others exposed to the product will be<br>described to understand the range of statistical skills required by the various disciplines within<br>the pharmaceutical, medical device and biotechnology sectors.

  • The training of industrial or business personnel in using various statistical tools to enhance<br>quality control programs presents a challenge to all concerned. The trainer needs to be familiar<br>with the features underlying adult learning as well as the workplace context within which the<br>training will apply. The training material needs to be relevant to participants' work practices if<br>commitment is to be achieved. In addition, there seems to be an increasing interest in assessing<br>the knowledge and skills of the trainees participating in such programs. As such, the traditional<br>forms of assessment, such as written assignments and examinations, are of little practical use in<br>settings where the primary focus is upon using the tools to improve processes to save dollars. This<br>presentation will describe the procedures used to assess trainees who recently participated in<br>such a training program, specifically with respect to their participation, knowledge gained and<br>application.