 Proceedings
|
Invited TalksTopic 9 : Technology in statistics educationMost statisticians recognize the central role that technology can play in the education of our students at all levels. Visualization technologies can dramatically assist people to understand data. Virtual environments offer access to authentic scientific inquiry. Modern technical machinery enables educators to reinvent the statistics curriculum and provide alternatives to mathematics for exploring fundamental statistics concepts. The Internet provides access to data with rich structure that is of topical importance. Acquisition of computing skills is essential for statisticians to approach data problems creatively, to be partners in scientific endeavors, and to advance in the workplace. We must prepare out students to be full participants in the modern technological world. This session explores innovations in the way technology can impact how we teach, what we teach, and how students learn.Session 9A: New paradigms in teaching statistics through technology9A1: Learning to apply statistics using a virtual environmentStefan Steiner University of Waterloo, CanadaR Jock MacKay University of Waterloo, Canada We have designed a virtual environment, a manufacturing process that needs improvement, where we challenge students to come up with ideas for, design, execute and analyze a wide variety of studies. The motivation for performing these studies is easily understood. And it is all done on hugely abbreviated time scales with little cost. The purpose of this presentation is to demonstrate the environment. We also discuss some of our experiences that we hope apply to other virtual environments. One observation is that students need a framework to help to formulate, plan and execute statistical investigations A second is that students view the simulation as a game, enjoying the freedom and challenge of working in an environment where they have a well-defined goal and control over the entire project. Paper 9A2: Learning from the statistician’s lab notebookDeborah Nolan University of California at Berkeley, United StatesDuncan Temple Lang University of California at Davis, United States An essential component of statistics education is to provide first-hand experience with applications of statistics where students learn how to analyze data in the context of addressing a scientific question. The approach we present brings the work of statistics researchers and data analysts to the community of educators so that they can utilize their expertise, data, problems, and solutions in teaching statistics. We hope to accomplish this sharing of ideas and materials through a new type of “document”, an electronic lab notebook that captures the research process and acts as a database of the statistician’s activities and analysis. These documents can be explored in rich new ways: they can have interactive controls that allow students to modify computations, they can be projected into different views for different audiences, and they can contain different branches of analysis for exploration. Paper 9A3: Pupils reasoning with information and misinformationJim Ridgway University of Durham, United KingdomJames Nicholson University of Durham, United Kingdom Real-world problems are complex; the ability to engage with such problems should be an educational goal. We know little about the effects of exposure to information, or about the cycle of revising personal stories, beliefs and theories, in the light of information and misinformation (IAM). A second emerging educational goal is the ability to reason with IAM. We are working with mash-ups comprising interactive multivariate displays of survey data, and newspaper articles on a particular theme (e.g. smoking, alcohol abuse). We report on the ‘natural’ statistical skills that pupils display when engaging with the media linked to relevant complex data, and show that some ‘big statistical ideas’ – such as interaction, and effect size – can be acquired by young students. We argue that the statistics curriculum should be reformed to include qualitative descriptions of large scale multivariate data sets, and to include the critique of rival accounts. Paper Session 9B: Rethinking the statistics curriculum: computing skills our students need9B1: Developing introductory computing for stats undergraduatesPaul Murrell University of Auckland, New ZealandComputer literacy is an essential part of a modern education in statistics (or any field that involves working with data). Two statistical computing courses have been developed in the Department of Statistics at the University of Auckland. The first provides instruction in fundamentals of computer storage, writing computer code, and data structures. The second covers simple programming ideas, how to write larger pieces of code, and graphics. This paper describes the evolution of these courses, noting successes and failures both in terms of what content is delivered and in terms of how that content is delivered. Paper 9B2: Integrating computing and data technologies into the statistics curriculaDuncan Temple Lang University of California at Davis, United StatesDeborah Nolan University of California at Berkeley, United States It is increasingly clear that computing is becoming an essential skill for statisticians and anybody working with data. Computing is as important as mathematics in both statistical practice and research, yet it occupies a tiny portion of our curricula. We have an obligation to reform our upper-division and graduate curricula and integrate computing. We need to change our view of the role of computing in our programs, and teach computational fundamentals and reasoning, rather than ad hoc "tricks" or templates. Furthermore, we must broaden our notion of "statistical computing" to teach modern data technologies. The needs for statistical computing are different from computer science and we must teach this increasingly diverse topic within the statistics curricula. This requires us to fit more into our curricula and also for many of us to learn this material. Computing is important in its own right but can also greatly improve how students learn the traditional material and introduce them to a different aspect of statistics. Paper 9B3: Introducing undergraduates to probability using the open-source programming language RJane M Horgan Dublin City University, IrelandWe outline the role of the open-source statistical programming environment R (R Development Core Team, 2005) in teaching a first course in probability. We show how R, which is freely available and downloadable from the web, may be used not only as a tool for calculation and data analysis, but also to illustrate probability concepts, to simulate distributions, and to explore by experimentation different scenarios in decision making. Recognising that the student best understands definitions, generalisations, and abstractions after seeing their applications, almost all new ideas are introduced and illustrated with real examples, covering a wide range of applications in computer science. While we have addressed in the first instance undergraduate students of computing, the approach outlined could also be adapted for students from other disciplines. Paper Session 9C: Virtual environments and experimental learning in statistics education9C1: Setting up experiments in veterinary science: an example of virtual experimentationLuc Duchateau Ghent University, BelgiumLiesbeth Lievens Catholic University Leuven, Belgium Paul Darius Catholic University of Leuven, Belgium Experimental design and random allocation of treatments is essential in applied research to run efficient experiments from which valid conclusions can be drawn. The relevance of experimental design and randomization, however, is often not well understood by students and investigators, and therefore statisticians are only consulted for the statistical analysis. A virtual experimentation environment allows researchers to design trials in their own discipline, evaluate the consequences of their choices and recognize the importance of statistical concepts such as blocks, nesting and random assignment. The application presented here consists of a vaccination trial for the prevention of udder infection in dairy cows. Paper 9C2: A visual approach in the teaching of statistics and probabilityJorge Domínguez-Domínguez CIMAT Mathematics Research Center, MexicoAxel Dominguez-Lopez Conteck, Mexico From experience, we know there are several difficulties to convey different concepts in statistics and probability as well as being assimilated by the learner. Looking to contribute in the solution of this problem, a technology development, named CalEst, has been accomplished. It aims to provide a set of tools with an educational approach to assist the teaching/learning process. This development generates information in a visually-attractive manner, improving enormously the understanding of the concepts and motivating the learning of statistics and probability. Probability notions play a crucial role in the analysis and interpretation of statistical data. Accordingly, this project incorporates several animations to illustrate and experiment several probability concepts. Furthermore, CalEst assists in an animated approach to calculate probabilities; hence illustrating the concepts of density and cumulative distribution functions for a diverse number of distributions. Other concepts are also included, such as the hypothesis test using a biometric system as example. Paper 9C3: Statlab: learning DOE by doing!Koo Rijpkema Eindhoven University of Technology, The NetherlandsEmiel van Berkum Eindhoven University of Technology, The Netherlands Marko Boon Eindhoven University of Technology, The Netherlands Alessandro di Bucchianico Eindhoven University of Technology, The Netherlands Statlab is an interactive web-based tool for teaching Design of Experiments. It is freely available through www.win.tue.nl/statlab and can be used to introduce large groups of students to practical strategies for experimenting through virtual case studies. Starting point is an adaptive approach, in which the program flow is influenced by the student’s decisions throughout the DOE-process. Furthermore, Statlab forces students to think pro-actively about practical details since it hides options that students do not ask for. Once a design is generated, realistic experimental data can be simulated for further model estimation, evaluation and eventually optimisation using response surface methods. Furthermore, Statlab provides direct feedback with training exercises and contains an automatic grading system regarding the student's actions and decisions. Students generally are enthusiastic about Statlab and consider it to be a stimulating teaching environment. Paper Session 9D: Advancing statistics education through visualization technologies9D1: Using visualisation to teaching data analysis and programmingHadley Wickham Rice University, United StatesModern data analysis demands computing skills that most potential statisticians lack. This paper discusses my approach to teaching data analysis and programming focused around the potential of visualization to engage students with the data and give them a flexible toolbox with which to attack many potential problems. Paper 9D2: Statistical cartoons: the role of graphics in understanding statisticsAdrian Bowman University of Glasgow, United KingdomAnimation is a graphical device which can be used to considerable effect in a teaching and learning context but which in the past has required considerable effort to achieve. A variety of tools are now available to assist. Several of these are mentioned, although the focus here is on the rpanel package in the statistical computing system R. More importantly, illustrations are given of the nature and design of animations, referred to here as cartoons, ranging from very elementary concepts to more sophisticated ones and focusing on the display of models as well as data. Paper Session 9E: e-learning tools: evaluation and the role of the instructor9E1: Using blended learning environments in teaching introductory statistics to a strong diversity of students: the role of background factorsDirk Tempelaar Maastricht University, The NetherlandsBart Rienties Maastricht University School of Business & Economics, The Netherlands Sybrand van der Loeff Maastricht University School of Business & Economics, The Netherlands Bas Giesbers Maastricht University School of Business & Economics, The Netherlands In the teaching of introductory statistics, the Maastricht University uses a blended learning environment that allows students to attune their use of available learning tools to personal preferences. The blended learning environment consists of small-group tutorials designed according to problem-based learning principles, a sequence of overview lectures and seminars, independent learning based on learning goals set in tutorial sessions, and an electronic learning environment: the adaptive e-tutorial ALEKS. Participation in tutorial sessions is required; the usage of other components can be set according to individual preferences. In this contribution, we will focus on student background characteristics that influence the intensity of the use of the e-tool, using data of 3000 students. We conclude that the adaptive e-tutorial not only supports students with weaker statistical background, but also less academically prepared students. Paper 9E2: Using simulations for active learning: the query-first method in practiceCamille Peres University Houston Clear Lake, United StatesDavid Lane , United States Katherine R Griggs University of Houston-Clear Lake, United States Although simulations are designed to make students active learners, simply presenting a simulation as a demonstration in class may lead students to be passive observers rather than active participants. One way to engage students is to ask them to anticipate the outcome of a simulation before seeing the simulation. We call this approach the “query-first method.” This query–first method has been incorporated into the design of the simulations on the website “Online Statistics Education: An Interactive Multimedia Course of Study.” Although this website can be used as a stand-alone resource, we have found the query-first method with its simulations is also a powerful learning tool in the classroom. In this paper, we describe examples of how the query-first method has been incorporated into lectures, students’ reactions to the method, and alternate ways this method could be used in the classroom. Paper 9E3: KNOU mobile learning for innovation in statistics educationTae Rim Lee Korea National Open University, KoreaMobile learning is novel in that it facilitates delivery of learning to the right person, at the right time, in the right place using portable electronic devices. In the near future, m-Learning will be a normal part of lifelong education and self-directed learning. From 2008 KNOU kick off the mobile learning system with KT. In this paper the mobile learning for statistics education will be introduced. This paper describes the new paradigm of Statistics Education with the e-learning contents, Mobile learning and ubiquitous learning system for statistical education that anyone who wants to study could study anywhere, anytime with the internet and multimedia system. In the future society with rapid change of educational circumstance and globalization, distance education using ICT technology will satisfy educational desires in various classes of learners. Paper 9E4: Creating active learning in a large introductory statistics class using clicker technologyZenaida Mateo University of Manitoba, CanadaOver the years, various teaching strategies have been developed to energize teaching and learning specifically for large classroom enrollment. Large classes do not lend themselves to student participation or inquiry. The use of clicker technology provides one way to address a part of this problem. It is a handheld device commonly called “clickers” or “key-pads” used by students in a large classroom to transmit their answers by pressing the clicker buttons. In this study, some initial results will be discussed on how clickers are used for teaching and learning of students in a large Introductory Statistics lecture environment. As this technology will be implemented for the first time in our department, the major challenges that the faculty and students will encounter when using this technology are also explored. Paper Session 9F: Sharing data for educational purposes (standards, databases, case studies)9F2: Promoting statistical literacy: a European pilot project to bring official statistics into university and secondary school classroomsHans-Joachim Mittag Distance University of Hagen, GermanyStatistical literacy is part of the knowledge and information that represents a driving force in today’s fast-moving globalised economies. For young people it is recognised as a key qualification for employability. In the academic world, the growing demand for an education in statistics is nowadays met by combining traditional ways of learning with web-based communication environments. Statistical Offices ought to play an important role in promoting statistical literacy, but most of them still have not considered this as part of their central function. This paper presents an EU-funded project that aims at promoting statistical literacy amongst young people by providing an innovative e-course with interactive and dynamic components. The project involved the National Statistical Offices of Malta and Finland and a German State University. The main aims were to encourage international cooperation between statistical agencies and educational institutions and to promote shared use of the project output. Paper 9F3: On and off-line dynamic data interrogationKate Richards RSS Centre for Statistical Education, United KingdomNeville Davies RSS Centre for Statistical Education, United Kingdom Mark Crowley Nottingham Trent University, United Kingdom In this paper we describe an interactive Flash®-based data interrogation tool which interacts with large datasets of real data collected from learners around the world. The dynamic tool can be used online to access in real-time databases stored on the Royal Statistical Society Centre for Statistical Education web server. It can also be used offline as a stand-alone facility to access local databases stored on the user’s PC. The tool has been developed to help improve learning and understanding of statistics and to demonstrate the importance of looking at data to help make evidence based decisions. We shall illustrate with examples of data interrogation aimed at helping students make trustworthy decisions from those data. Paper Session 9G: Effective online educational materials9G1: Online learning materials: are they put to different uses by online and on campus students?Glenda Francis Swinburne University of Technology, AustraliaAs part of the redesign of our first year introductory statistics course, students were provided with two additional interactive resources: a Java applet which simulates a sampling activity and an extensive series of diagnostic self tests. This paper explores the extent to which online and on campus students made use of these materials and describes the benefits students perceived. Differences in the patterns of use between online and on campus students are also described. Paper 9G2: In search of the “perfect” blend between an instructor and an online course for teaching introductory statisticsCandace Thille Carnegie Mellon University, United StatesMarsha Lovette Carnegie Mellon University, United States Oded Meyer Carnegie Mellon University, United States As part of the Open Learning Initiative (OLI) project, Carnegie Mellon University was funded to develop a web-based introductory statistics course, openly and freely available to individual learners online and designed so that students can learn effectively without an instructor. In addition, the course is often used by instructors in the hybrid form, to support and complement face-to-face classroom instruction. This paper documents two studies where we investigated the OLI-statistics courses’ effectiveness in the hybrid instructional model. We describe the design, results and limitations of the studies and discuss the implication of the results for finding the “perfect” blend between an instructor and an online course for teaching introductory statistics. Paper 9G3: Improving lectures with CAST appletsDoug Stirling Massey University, New ZealandMost statistical concepts and methods can be explained and illustrated using diagrams and in most places where a diagram might be used, a dynamic diagram is more effective than a static one. A few such diagrams per lecture quickly build to hundreds in an introductory statistics course. The collection of CAST applets for lectures contains over 400 including simulations to demonstrate the properties of sampling distributions and inference and many others that are uniquely designed to illustrate individual statistical concepts. Use of these diagrams helps to explain statistical ideas and methods, demonstrates calculations, breaks up the reading of static slides and adds some entertainment value. The paper also discusses the differences between applets designed for selfstudy and for use by a lecturer in both large and small classes. Paper |