TOPIC 5: Statistics education and the wider society
There is an ever-increasing need for all of us to be statistically literate in order to be well informed and knowledgeable about the many important issues in the “Wider Society”. How can one properly debate issues of discrimination in the workplace, changes and trends in crime, inequalities in society, the impact of passive smoking, and the effect of greater life expectancy on pension plans without good data and statistical arguments? Statistical issues arise in understanding the validity of medical tests for detecting disease, but also on the impact of screening procedures on incidence rates. Issues related to inflation, consumer price indices, stock market trends and investment strategies also demand more sophisticated statistical reasoning and prudence. Knowledge of basic statistics is important in understanding the usefulness of surveys and polls, the influence of advertising and marketing, and making international comparisons in education, economics, employment and health. As educators in Statistics, how can we enhance and contribute more to the growing demand that most people obtain good skills in quantitative reasoning in order to be able to apply them to so many diverse areas? In particular, do we not have more responsibility than ever to emphasize both how statistical thinking can benefit society, and that it plays such a vital role in understanding many of the complex issues in our society?
We are aware that the rapid development of Statistics as a Science has historically been largely driven by challenging problems of a quantitative nature drawn from many facets of life. The more recent and continuing revolution in computing and technology has been also been accompanied by an increasing awareness of the need for statistical thinking in society at large! Database and spreadsheet technology, email and the internet, multimedia and sophisticated graphical methods, simulation tools, and in particular more powerful and often easy to use statistical software packages have also led to more people using, and being required to understand, statistical reasoning and concepts. Surely these advances in technology are leading to different approaches to statistical education and the teaching of Statistics?
In addressing statistics education and the wider society, Topic 5 will have a wide perspective and few limitations on subject matter! In particular, contributions relating statistical education to issues in law, commerce, business, economics, government, industry, journalism, the physical and biomedical sciences, medicine, demography, psychology, sociology, sport, engineering and mathematics education are welcome. Let this serve as a forum for asking and answering the challenging questions about how statistical education and teaching should move forward in order to cope with creating a more sophisticated and statistically literate society!
SESSION 5A: Statistical literacy: concepts, gaps, indicators
5A1: STATISTICAL ILLITERACY IN LATIN AMERICA: A CONSEQUENCE OF THE DIFFERENT VISIONS ABOUT THE MEANING OF STATISTICS
Carlos Araújo
Pontificia Universidad Católica de Chile, Chile
The lack of interest of statisticians to reach a consensual characterisation of Statistics is the main reason for the inefficacy in the efforts made until now to install or to improve statistical education in Latin American countries. It is therefore necessary to attain consensus within the international statistical community to produce a characterisation of Statistics that makes possible the establishment of a general common frame within which all the activities related to statistical development should be incorporated. Under this characterisation, some actions are suggested to encourage national and international Statistics associations to support the improvement of the teaching of Statistics and so set some sustainable bases for the establishment of a suitable statistical literacy.
5A2: WHAT KIND OF STATISTICAL CULTURE IS OFFERED BY THE MEXICAN EDUCATIONAL SYSTEM?
Olga Leticia Escudero, Fortino Escareño Soberanes
Ministry of Education, Mexico
Societies today are immersed in transformations and transitions filled with uncertainty. More treaties are signed every day with the purpose to open markets between different countries. There is a growth in consumer tendencies. Citizens can participate more and more in processes of decision making. The media makes use of statistics to promote commercial products and goods, to communicate the benefits that certain traditionally marginalized social groups have gained, and to design indices to show changes in the popularity of political candidates. While everybody can access this information, the processes involved in the calculation of numbers and indices is seldom explicit. It is thus important to ask what is being done in schools to enable young students and future citizens to understand and value statistical information. This talk will describe research in progress aiming to investigate the kind of statistical education offered at the elementary and middle levels in Mexico, through an analysis of educational materials and students’ knowledge.
No paper available
5A3: HOW STATISTICALLY LITERATE ARE ADULTS? RESULTS FROM TWO INTERNATIONAL COMPARATIVE STUDIES
Iddo Gal
University of Haifa, Israel
Numerous researchers and educators lament the poor state of statistical literacy and statistical skills in the population. However, few authors base their arguments on actual results from comparative or large-scale studies which provide stable, population-level estimates of performance. This talk will present and discuss selected results regarding statistical skills of adults in multiple countries, based on analysis of two data sources. Secondary analyses of data from the International Adult Literacy Survey (IALS), conducted in 1994-1998 in over 20 countries, were performed with regard to items assessing understanding of bar graphs and pie charts. Implications will be discussed in terms of the need for teaching statistics with a focus on interpretation and communication and on using real-life contexts and tasks in instruction. Conclusions will also be presented regarding the importance of conducting analytic assessments which shed light on multiple types and levels of statistical skills.
No paper available
5A4: PROBING COLLEGE STATISTICS TEACHERS’ INSTRUCTIONAL GOALS AND CLASSROOM PRACTICES WITHIN A STATISTICAL LITERACY FRAMEWORK
Enriqueta D. Reston
University of San Carlos, Philippines
Liza Lorena C. Jala
University of Cebu, Philippines
Timoteo P. Edullantes, Jr.
Cebu Institute of Technology, Philippines
In line with the reform efforts in statistics education that emphasized the development of statistical literacy skills, we explored instructional goals and classroom practices of 12 college statistics teachers and analyzed them using Gal’s statistical literacy model. Through focus group discussions, we find that their goals and practices in the teaching-learning-assessment cycle are primarily based on mathematical and statistical knowledge and only three displayed evidence of literacy and context knowledge, and the capability for critical questions. While these teachers indicated positive attitude towards the promotion of statistical literacy, there are gaps between attitude and classroom implementations. Aside from the need for alignment of instruction and assessment with this goal, there is need for concerted efforts towards equipping statistics teachers with the knowledge and resources necessary in the promotion of statistical literacy.
SESSION 5B: Statistics education and the Statistics profession
5B1: SOME EXPERIENCES IN TRAINING STATISTICIANS IN OFFICIAL STATISTICAL OFFICES
Len Cook
International Statistical Institute, New Zealand
The paper looks at the scale of change statistical offices now experience, and what is expected of their statistical methodologists. The paper discusses the ways in which the preparation by universities needs to supplemented by training on the job, in house education, planned work experience, and how far personal and group experiences contribute. The experiences of the United Kingdom Office for National Statistics, and of Statistics New Zealand are drawn on, with examples from several other countries.
5B2: RECOGNISING AND DEVELOPING GOOD STATISTICS TEACHERS
Peter Petocz
Macquarie University, Australia
Sue Gordon
The University of Sydney, Australia
Anna Reid
Macquarie University, Australia
In this paper, we examine some results from a series of interviews carried out by e-mail with an international group of statistics educators, all of them members of the International Association for Statistical Education (IASE). We asked for their ideas on the qualities of ‘good’ statistics teachers and ways in which they could develop as statistics educators. Follow-up questions explored their answers in depth. The responses highlight the diversity of views about recognising and developing good statistics teachers at tertiary level, an important consideration for any discussion on professional practice and certification in statistics education.
5B3: STATISTICS EDUCATION AND THE COMMUNICATION OF STATISTICS: A REPORT ON THE IASE/ISI SATELLITE MEETING, 2005
Brian Phillips
Swinburne University of Technology, Australia
The International Association for Statistical Education, IASE, has a very strong commitment to the general development of statistical educators including research and the sharing of ideas and experiences. In particular the IASE/ISI satellite meetings are always very friendly, informative affairs as was the one in Sydney April 4-5, 2005. The theme was Statistics Education and the Communication of Statistics, it was jointly organised by the IASE and the Victorian Branch of the Statistical Society of Australia and immediately preceded the ISI session in Sydney. The theme of the conference was chosen because some IASE members felt that the communication of statistical methods and results needed more emphasis in our courses and wished to draw the attention of statistics educators to this area to encourage development and innovation. This paper summarises the main points the speakers made and the main issues that emerged.
SESSION 5C: Statistics and journalism
5C1: CHALLENGES IN MODELING AND COMMUNICATING RARE AND EMERGING INFECTIOUS DISEASES RELATED TO BLOOD SAFETY
Susie ElSaadany
Centre for Infectious Disease Prevention and Control, Canada
The Public Health Agency of Canada (PHAC) has unique experiences with new and emerging disease issues in blood Canadian crisis of Hepatitis C virus (HCV) transmission through blood transfusion. To protect the Canadian public from risk, the government enacted restrictions on blood donation. Statisticians must provide an interpretation of the risk to decision makers that is within the realm of comprehension, so that they may effectively communicate risk information to the public. For the HCV, there was no information available for incidence and prevalence during the 1980s/early 1990s for the Canadian population. Estimating the blood borne infection of Canadians with HCV was crucial to the government’s decision regarding victim compensation. As this was a much publicized issue, it was Public Health Agency of Canada’s responsibility to meet the challenge of producing accurate figures in the face of little knowledge at the time.
No paper available
5C2: WORKING TOGETHER WITH THE MEDIA – EXPERIENCES OF STATISTICS FINLAND
Reija Helenius
Statistics Finland, Finland
The media is a key partner and customer group for national statistical agencies. The media largely creates the public image of a national statistical organisation. What types of statistics are presented in the media, and how widely and in what way, are the essential questions. It is a challenge to the services of an NS0 to answer these questions. The keys to answering them are co-operation, development of electronic services and organisation of training for journalists. Understandability and ease of use are also crucial for widespread use of statistics. The presentation is a summary of how we at Statistics Finland have developed our services to the media and how we continue to study and monitor our public image.
5C3: BUILDING HOUSES ON STRAW POLLS: A BRIEF LOOK AT SOME CURRENT STATISTICAL REPORTING BY JOURNALISTS IN NEW ZEALAND
Warren Palmer, Anthony Crawford
University of Otago, New Zealand
This paper considers how New Zealand journalists report political polls. Two recent newspaper articles are featured. Perhaps not surprisingly we have detected a tendency for journalists to focus on sample size, to misunderstand the concept of margins of error, and to have little idea as to whether a result is generalisable. We also consider the importance of non–respondents. We wonder if journalists question the validity of survey results they have been given. We ask the question: could a “non–random” convenience survey have as much validity as a more formal survey conducted by a specialist research company?
SESSION 5D: Statistics education and the world of life and health sciences
5D1: TEACHING STATISTICS BASICS TO BIOLOGY AND HEALTH PROFESSIONALS THROUGH BAYESIAN IDEAS
J. Andrés Christen
CIMAT, Mexico
Romulo Lins
UNESP, Brazil
Bayesian statistics is usually viewed as an 'advanced' course; Bayesian methods are either seen as an optional extension or its founding ideas are seen as difficult to grasp. There are, however, two good reasons for changing this. First, the recent strong increase in interest in Bayesian methods, in several areas. Second, the fact that Bayesian methods are particularly relevant for the kinds of decisions health related professionals have to make. Here we will argue that the Bayesian approach makes a quite convenient platform for introducing key ideas of probability and statistics to biology and health professionals; we will also outline how such a course can be organized around everyday, easy to grasp, examples, exploring how those simple situations involve a complexity that requires a deeper discussion of the concepts and ideas underlying the modeling process that can lead to acceptable and adequate answers to them.
5D2: THE ROLE OF STATISTICAL METHODS IN COMPUTER SCIENCE AND BIOINFORMATICS
Irina Arhipova
Latvia University of Agriculture, Latvia
This article discusses the links between computer science, statistics and biology education on the basis of research at the Latvia University of Agriculture. Bioinformatics study is considered from two aspects – as one for biologists learning Information Technologies (IT) to use within their speciality, or for IT specialists learning biology so they can apply their skills to biological problems. The different computer science technologies and statistical methods in bioinformatics are considered. The multidisciplinary approach facilitates the understanding of interrelations between computer science technologies, statistical methods and bioinformatics applications and improves the education at an agriculturally-based university. Therefore the teaching of Information Technology at agriculture-based university should exploit the close relationship with mathematics, statistics and biology.
5D3: TEACHING BAYESIAN METHODS IN BIO-MEDICAL RESEARCH
Keith R. Abrams
University of Leicester, United Kingdom
This paper considers experiences of teaching Bayesian statistical methods within a bio-medical research setting to both statisticians and non-statisticians at postgraduate level. In particular, it considers topics covered, level of mathematical exposition, software and texts.
5D4: WHEN TEACHING STATISTICS TO NON-STATISTICS STUDENTS
Jenny Freeman, Kevin Smith, David Staniforth, Steve Collier
University of Sheffield, United Kingdom
Statistics is important and relevant to many college subjects. Coincidently, it is common for students to dislike and under-perform in modules involving mathematics, numeracy or statistics. In order to address these issues the project brought together a statistician, a clinician and educational experts to redesign the way that statistics is taught to medical undergraduates, making it more relevant to them. It re-conceptualized the syllabus, focusing on developing different methods of delivery, including a web-based learning environment which allowed students to access materials at levels consistent with their current understanding. The new teaching materials, including videos, animations and the online resource, placed greater emphasis on applying statistics and interpreting data. This innovative approach enabled the project to make the teaching, learning and understanding of statistics more people-centred and relevant.
SESSION 5E: Statistics in the social sciences: Psychology, Sociology
5E1: A NON-STANDARD APPROACH TO TEACHING AN INTRODUCTORY STATISTICS COURSE TO SOCIAL SCIENCE STUDENTS
Patrick Murphy
University College Dublin, Ireland
This paper describes a course that was developed to teach statistics to students majoring in Psychology and Politics. There were several interesting aspects to this course. Firstly each lecture contained between 550 and 800 students. Secondly those students were almost uniformly negatively disposed to Statistics prior to the beginning of the course. Thirdly we were required to provide an introduction to Statistics in just 12 lectures, each of 50 minutes duration. Constrained, we were forced to think deeply about what we want to provide to students in an Introductory Statistics course. Making use of simulations and the internet, we chose to emphasise concepts and critical thinking and supported these with examples which had direct relevance to our students. Restricted to 12 lectures, we learned to make optimum use of each lecture. Can a short course like this act as a useful pre-cursor to the standard Introductory Statistics course?
5E2: INTENSIVE USE OF FACTORIAL CORRESPONDENCE ANALYSIS FOR TEXT MINING: APPLICATION WITH STATISTICAL EDUCATION PUBLICATIONS
Annie Morin
IRISA, Université de Rennes, France
Textual data are found in any survey or study and can be easily transform in frequency tables. Any method working on contingency tables can be used to process them. Besides, with the important amount of available textual data, we need to find convenient ways to process the data and to get invaluable information. It appears that the use of factorial correspondence analysis(CA) allows to get most of the information included in the data. CA produces a visual representation of the relationships between the row categories and the column categories in the same space. But there are several problems: the first one is the interpretation of the results. And even after the data processing, we still have a big amount of material and we need visualization tools to display it. In this paper, we present some methods to process the data and to get invaluable information. We also show how to use correspondence analysis in a sensible way and we give results of studies of publications dealing with statistical education.
5E3: ASSESSING PSYCHOLOGY STUDENTS’ DIFFICULTIES WITH CONDITIONAL PROBABILITY AND BAYESIAN REASONING
Carmen Díaz, Inmaculada de la Fuente
University of Granada, Spain
Conditional probability and Bayesian reasoning are important to psychology students because they are involved in the understanding of classical and Bayesian inference, regression and correlation, linear models, multivariate analysis and other statistical procedures that are often used in psychological research. A study of previous literature showed that there is considerable research on this topic, but no comprehensive questionnaires have been developed to globally assess students' understanding and misconceptions on these topics. At the University of Granada we started building a questionnaire, which takes into account the content of conditional probability taught in the Spanish universities to psychology students, as well as the biases and misconceptions described in the literature. In this work we will describe the process of developing the questionnaire and will report the results from a sample of 206 psychology students.
5E4: SHOULD PSYCHOLOGY ABANDON p VALUES AND TEACH CIs INSTEAD? EVIDENCE-BASED REFORMS IN STATISICS EDUCATION
Fiona Fidler
La Trobe University, Australia
Several editorial and institutional interventions in psychology have aimed to improve statistical reporting in journals. These efforts have sought to de-emphasise statistical significance and encourage alternative analyses, especially effect sizes and confidence intervals (CIs), but the interventions to date have had short-lived and superficial impact—if any impact at all. I review some of these interventions in psychology and discuss possible reasons for lack of success. I give an inter-disciplinary context by discussing reform efforts in medicine—in which useful reform has already been achieved—and ecology. I then identify statistics education as the next major challenge for reformers, and report data on students’ understanding of CIs, and difficulties they have making appropriate interpretation of CIs. I explain the need for further evidence on which to base improved statistics education in psychology.
SESSION 5F: Statistics in sport
5F1: SPORTS TEAM QUALITY AS A CONTEXT FOR UNDERSTANDING VARIABILITY
K. Laurence Weldon
Simon Fraser University, Canada
When instruction in statistical concepts can be tied to practical sports issues, students are motivated to understand the statistical concepts. In this paper we describe an issue that would be relevant to discussions of many different sports leagues, and would also be a vehicle for teaching statistical concepts such as simulation, graphical displays, illusions of randomness, measurement of variability, and the logic of hypothesis testing. In addition to motivating a keen interest in the effects of random variation, these examples provide students with a way to verbalize what they learn in statistics classes to their lay acquaintances. Moreover, examples like these have the potential for engaging instructors who have been focused on more traditional approaches. Programs in the software language R are provided and their use with introductory classes is discussed.
5F2: TEACHING AN INTRODUCTORY STATISTICS CLASS BASED ON SPORTS EXAMPLES
Robin H. Lock
St. Lawrence University, United States
We discuss our initial experience with offering a version of our standard introductory statistics course that focused primarily on sports related examples, rather than a more traditional selection of applications. This special sports section was offered in parallel with a regular section, covering the same statistical topics, with the same instructor, at the same pace. We examine how the students enrolling in the sports section might differ from the regular, illustrate how we converted material from the regular section to the sports equivalent, compare the performance of students between the two sections and reflect on the effectiveness of the sports-based approach.
5F3: TEACHING THE FUNDAMENTALS OF STATISTICS WITH SPORTS DATA: SHOULD TEAMS WALK OR PITCH TO BARRY BONDS?
Jerome P. Reiter
Duke University, United States
Sports data are commonly used to present topics from introductory statistics, such as exploratory data analysis and probability. They also can illustrate more subtle and complex statistical issues, such as selecting appropriate variables, making casual inferences from observational data, and specifying appropriate inferential populations. In this paper, I discuss how sports data can be used to engage students on such fundamental aspects of data analysis. I frame the discussion around the question posed in the title, a question which has generated much debate among baseball enthusiasts.
SESSION 5G: Statistics education in the financial and actuarial world
5G1: STATISTICAL METHODS IN GENERAL INSURANCE
Philip J. Boland
National University of Ireland, Dublin
The Actuarial profession appeals to many with excellent quantitative skills who aspire to “make financial sense of the future.” However the road to qualification as an Actuary through the Institute or Faculty of Actuaries in the UK (or the Society of Actuaries or Casualty Actuary Society in the United States) is not an easy one, and a series of very challenging exams must be passed to qualify as a Fellow Actuary. These exams test many skills, and in particular demand a good knowledge of probability and statistics. The main areas of work for actuaries are traditionally life assurance, actuarial consultancy, general insurance, and investment. Although statistical skills are required in all of these areas, they are particularly important in general insurance. In this paper we discuss the basic tools and techniques in probability and statistics that are essential for an actuary who intends to work in general insurance.
5G2: STATISTICS AND FINANCE: LIVING ON THE “HEDGE”
Ricardo Gimeno
Banco de España, Spain
Ruth Mateos de Cabo
Universidad San Pablo –CEU, Spain
Statistics plays a leading role in finance. The explosive development of increasingly complex markets makes it more and more difficult for practitioners to correctly value financial asset. Statistical analysis has become a powerful tool for a better market valuation, taking a leading role in the development of new financial products that try to hedge the increasing amount of risks that an investor has to take. Statistics knowledge demand is steadily increasing in Hedge Funds, Investment Banking and Financial Institutions in general, where statistics students could developed a professional career. Finance can be seen as a way to motivate students on the applications of almost any statistical tool we would like to teach them, since we could always find an example where these techniques are put into practice.
5G3: STATISTICS EDUCATION FOR ACTUARIES: THE SYLLABUS FRAME
Juan Manuel López Zafra, Sonia de Paz Cobo
Universidad Complutense de Madrid, Spain
Access to the actuarial profession varies widely across countries. This causes actuaries difficulties in being recognized outside their country of origin, showing a paradox in the frame of the European Union that bans the barriers in the labor market. With the aim of solving the problem, the Groupe Consultatif Européen, a European actuarial association, has developed a Core Syllabus that states the basics of what an actuary must know wherever its origin. What we discuss here is the importance of statistics in the education of actuaries. The paper is organized as follows. After an introduction where the actuarial profession is defined and the different ways of access to it are shown, we will go through the description of the Core Syllabus, paying special attention to the contents of the statistics topics. We will go then into a description of the present situation in Spain, just before the concluding remarks.
SESSION 5H: Psychological and social issues in the teaching of Statistics
5H1: USING EXAMPLES RELATED TO HIV/AIDS TO ENHANCE UNDERSTANDING OF STATISTICAL THEORY AND OF SOCIAL IMPLICATIONS OF STATISTICS
Jacky Galpin
University of the Witwatersrand, South Africa
Without practical applications of theory, students often do not understand why they are studying the techniques and what statistics is about. Service course students are often antagonistic to and afraid of the statistics courses. Some progress can be made by noting that anyone who crosses a road and survive, is an expert in probability. Relating techniques to current social issues can further assist in breaking down barriers, increasing awareness of the usefulness of statistics and enhancing academic citizenship. Specifically using HIV/AIDS examples, in a country where this is a major problem, helps students to understand the issues and how to deal with the risk. Since many are the first from their home areas to attend university, they have some standing in the community, and communicate with others. This paper outlines some of the issues and methods and examples used in both major stream and service courses by the author and by colleagues.
5H2: A COMPARATIVE STUDY OF THE EFFECTS OF MOTIVATIONAL AND ATTITUDINAL FACTORS ON STUDYING STATISTICS
Monique Bijker, Gabriëlle Wynants, Hans van Buuren
Open Universiteit Nederland, The Netherlands
This study focuses on the differential effects of motivations and attitudes on Critical Thinking and Self-Regulation. Two different samples of Psychology students of Open Universiteit Nederland have been studied. The first sample is composed of students, confronted with an integrated, research-based teaching and learning design; the second sample is composed of students, exposed to classic service courses in statistics and research methods. Results reveal that motivations significantly affect Critical Thinking and Self-Regulation and that motivations and attitudes affect learning processes and achievement in their own characteristic way. Moreover the crucial role of an integrated teaching and learning design as a significant contributor to optimal motivations, more favourable attitudes and more adequate learning strategies for studying statistics and research methods has been endorsed.
5H3: A WAY OF INTEGRATING CIVIC LEARNING INTO STATISTICS COURSES: MEDIA REPORTS
Engin A. Sungur, Benjamin S. Winchester, Jon E. Anderson, Jong-Min Kim
University of Minnesota, United States
In this paper, we present a brief history of our efforts to incorporate civic learning into our statistics curriculum, highlighting our most recent approach, media reports. We discuss implementation issues, educational objectives, and give examples of student projects. Learning objectives, expected outcomes, and our assessment process are also given. An important aspect of this effort is the use of technology in report generation and dissemination. We discuss the development of these tools and how they have been used. We conclude with remarks on sustainability and possible future directions.