TOPIC 4: Statistics education/training and the workplace
Statistical thinking, methods and tools have become widespread requirements for many, if not all, areas of science, business, government, politics and community. Hence workers and professionals from all educational backgrounds are faced with the need to gather, process, analyze, interpret and report on data as part of their jobs. Many realize that they did not have the proper education and training required to perform these tasks with confidence and efficiency. In addition, statistical methods and tools are also changing fast.
The size of the statistical literature is vast, the impact of technology (communications and computing, in particular) on the methods and practices is substantial, and no single individual can aim to master all that is happening. Hence even statistically literate professionals that had their education in the past, will need to revise their training and update their practice to continually learn about new ideas, techniques and approaches that appear, which may question or improve upon previous behavior and methods. These two sets of circumstances pose a substantial challenge to those that need statistical thinking, methods and tools for their professional performance. Continued statistics education and training is a worthy goal to be pursued, and a substantial part of this must occur at the workplace.
The aim of this set of sessions is to bring together educators, professionals, practitioners and customers of statistical thinking, methods and tools to discuss ways in which effective statistical education and training can be developed at the workplace to help promote improved statistical practice in all avenues of professional life.
SESSION 4A: Statistics education and the engineering world
4A1: TEACHING ENGINEERING STATISTICS TO PRACTICING ENGINEERS
Jorge Luis Romeu
Syracuse University, United States
Teaching statistics to engineers is a challenging task. First, lacking space, many engineering curricula include few or no statistics courses, and these are often packed and highly theoretical. Thence, students don't perceive statistics as a part of their engineering toolkit, but as a nuisance to endure. On the other hand, engineering is a two-part endeavour. One consists in building or modifying systems. The second is measuring/assessing system performances, which are nothing but random variables. Therefore, there can be no engineering work without statistical analyses. In this paper we discuss and assess ways to enhance the insufficient statistical education that many engineers receive once they have left college. Such methods, designed for practicing professionals include (print and electronic) materials produced for self-study, short training courses and the development of industry-academe organizations to help practicing engineers by “looking over their shoulders.” Finally, a selection of related free Web Sites are presented.
4A2: ACHIEVING SUCCESS IN INDUSTRIAL TRAINING
Peter Martin
University of Ballarat, Australia
The teaching of statistical techniques to people in industry, as part of quality control or process improvement programs, can be a rewarding but somewhat daunting process for academics. Whilst the spin-offs for undergraduate teaching are numerous, and will be explored in a future paper, the biggest gains include the co-operation between the specific industry and the academic institution concerned. This paper explores some of the requirements and conditions that go towards enhancing successful industrial consultancies involving statistically based training programs provided by academics.
4A3: TEACHING MODERN ENGINEERING STATISTICS: THE CONTRIBUTION OF COLLABORATION AND SHARED VIEWS OF THE ROLES OF MATHEMATICAL STATISTICS IN ENGINEERING
Georg Lindgren, Lena Zetterqvist
Lund University, Sweden
The Mathematical statistics division at Lund University teaches 8 core and 13 elective statistics courses within 14 different engineering programmes leading to Master of Science in Engineering. This paper uses cases to analyse the combination of ingredients that seems to make the difference between success and otherwise in design of curriculum for engineering programmes. The key aspects underpinning the efforts seem to be collaborative curriculum development, and a joint view from both engineering and statistics of the role of mathematics and statistics in technology and engineering. This respect and high regard for mathematical statistics from the engineering side, often arising from research or other collaboration, changes them from clients to partners. This paper is an attempt to systematize what we see as the important success factors in an engineering statistics education.
4A4: ENGINEERING STATISTICAL NEEDS AND ENGINEERING CURRICULUM: AN ANALYSIS
Elena T. Fernández de Carrera
National Technological University/Litoral National University, Argentina
The importance of statistical science in engineering education is a subject that no longer merits discussion. The constructivist approach in its teaching, the need to teach using problem solving, and the contribution of modern technologies to optimize this learning are accepted facts by most statistics professors. In this paper we analyze the problems present in Argentina and other countries related to WHAT is the content of the subject?, To WHAT extent do they go?, and WHEN in the core curricula, is the content included? We also analyze whether to have one course with the same content for all the engineering majors or to adapt content to the specific needs in each of area. In this mixture of uncertainties, statistics is presented like an enigma to the students of engineering, especially if statistics is taught in the early stages of their program. This happens in Argentina where the roles of statistics in engineering are very difficult to demonstrate because students have only taken basic subjects like mathematics, physics or biology.
SESSION 4B: International cooperation in statistical training in the workplace
4B1: UTILIZING DISTANCE EDUCATION TO OFFER WEB-BASED PROFESSIONAL DEVELOPMENT IN STATISTICS EDUCATION TO TEACHERS ACROSS EUROPE
Stathis Mavrotheris, Maria Meletiou-Mavrotheris
Cyprus College, Cyprus
The paper provides an overview of a new program recently funded by the European Union that aims to enhance the teaching and learning of early statistical reasoning in European schools by utilizing distance education to offer high-quality professional development experiences to geographically-dispersed teachers across Europe. Acknowledging the fact that teachers are at the heart of any educational reform, the project will facilitate intercultural collaboration of European teachers using contemporary technological and educational tools and exemplary web-based materials and resources. Long-term sustainability will be assured through support of multilingual interfaces and online services for the accumulation of collective knowledge from teachers and teacher educators. An online knowledge base will offer access to usable and validated pedagogical models, didactic approaches, and innovative instructional materials, resulting in a complete and flexible teacher professional development program.
4B2: EXPERIENCES IN CONSULTING WITH RUSSIAN BIOLOGISTS ON APPLICATIONS OF STATISTICAL METHODS
Lyman L. McDonald
Western EcoSystems Technology, Inc., United States
I report on my experience in consulting with environmental biologists on applications of statistical methods while on a tour of State Nature Reserves and National Parks in Russia, 2003. The biologists that I worked with were well trained in mathematics in comparison to biologists trained in the United States. I also found some biologists with advanced graduate degrees to be well trained in analysis of time series data. However, training in many basic applied statistical methods was either absent, poor, or incorrect. This lack of knowledge of applied statistical methods greatly limits the ability of biologists to design observational studies or manipulate experiments and publish results of their work outside Russia. If there are individuals actively teaching applied statistics in Russia among the audience or among the readers of the proceedings of this conference, I apologize if I have offended anyone. The gap in knowledge of applied statistics is real, however, and I would appreciate communicating with you.
4B3: TOWARDS A MORE INTEGRATED INTERNATIONAL STATISTICAL SYSTEM: THE ROLE OF TRAINING
Morvarid Bagherzadeh, Enrico Giovannini
OECD Statistics Directorate, France
Over the last few years, several proposals have discussed how to strengthen the international statistical system. In particular, the Committee for Co-ordination of Statistical Activities has identified a medium-term strategy to improve the quality of statistics produced by international organisations, as well as the co-operation among them. Building on the experience of some organisations, including the OECD, some principles and recommended practices have been identified. One area that requires a special investment is training. The paper describes some experiences made at the OECD and highlights proposals for future investments in this field.
4B4: INTERNATIONAL STATISTICAL TRAINING – HOW BOTH LARGE AND SMALL ORGANISATIONS CAN BENEFIT FROM INTERNATIONAL COOPERATION
Todd Evans
Australian Bureau of Statistics, Australia
Teaching statistics raises many challenges. In the international context this can mean crossing cultural, language and socio-economic boundaries. With economic globalisation and the desire for statistical agencies to produce comparable, internationally consistent statistics, the need for international cooperation in statistical training in the workplace has never been greater. The Australian Bureau of Statistics (ABS) currently undertakes a range of international statistical training activities, encompassing providing technical assistance to other countries by ABS staff, particularly in the Asia-Pacific region; hosting overseas statistical staff for visits at the ABS offices; placing overseas staff in positions at the ABS from a few months to two years; supplying training materials to statistical organisations around the world; and visiting other statistical agencies to learn best practices. The benefits and costs of these activities, with particular emphasis on the cultural and language experiences that have been learnt, will be covered.
SESSION 4C: Statistics in medicine
4C1: STRENGTHENING RESEARCH AND STATISTICAL SKILLS OF MEDICAL DOCTORS THROUGH A HANDS-ON APPROACH: A CASE STUDY FROM IRAN
Abbas Bazargan
University of Tehran, Iran
Nasser Vallai
S.B. University of Medical Sciences, Iran
Students in medical and health-related fields as well as seasoned medical practitioners need to understand the issues related to the process of planning a scientific study, conducting the study, analysing data and reporting the findings. Recognizing this need, the Iranian medical education system (IMES) started to design and implement a series of workshops in research methodology and statistical data analysis during the past decade. The IMES includes more than 40 universities of medical sciences (UMS). Faculty members and students of the UMS participated in the aforementioned workshops. The workshops provided an opportunity for the participants to get hands-on experiences in the process of scientific research. Based on the above, in the first part of this paper the structure of the workshops, their content and instructional methods are described. Results from a survey to find out the effectiveness of one of the workshop are presented.
4C2: BIOSTATISTICS TEACHING IN THE NEW GENOME ERA
Júlia Pavan Soler
University of São Paulo, Brazil
Alexandre Pereira
University of São Paulo Medical School, Brazil
The objective of this article is to introduce a multidisciplinary biostatistics course program, targeting professionals and academics involved in researching genetics and genomics. The program is designed to adapt statistical concepts, uses and language to this field, which is at the forefront of knowledge. The idea underlying this initiative arose through a research project conducted by physicians and statisticians of the University of São Paulo, Brazil, whose purpose was to identify genetic determinants associated with cardiovascular risk factors in the Brazilian population.
4C3: TRAINING CHIROPRACTORS FOR CAREERS IN CLINICAL RESEARCH
Cynthia R. Long
Palmer Center for Chiropractic Research, United States
A two-year clinical research curriculum offered in a graduate program at a U.S. chiropractic college was implemented in Fall 2003 and enrolls three to six chiropractors per year. The curriculum includes ten credit hours of required courses in biostatistics. Introductory courses in biostatistical thinking and reasoning and data management are offered the first term, followed by basic statistical methods, statistical graphics, and advanced topics over the next three terms. Trainees typically have little previous exposure to statistics, so program objectives move from developing critical appraisal skills to writing strong data-related sections in grant applications. As graduates will likely pursue careers at chiropractic colleges with little or no research infrastructure, nor necessarily a research culture, it is paramount they develop a strong foundation in research methods and become proficient users of statistical tools to succeed.
SESSION 4D: Statistics for future statisticians
4D1: TRAINING FOR APPLIED STATISTICIANS
Ernesto A. Rosa
National University of Tres de Febrero (UNTREF), Argentina
What are the skills a person who studies Statistics must know? Which kind of problems would these people have when they start working in groups with people of different disciplines in private or public institutions? Do they know how to solve those problems? Which is the correct way they would be trained so later they can transmit properly the knowledge they received in their professional studies? Finally: which skills would the statisticians really need to know if they want to solve properly the problems they can have in their daily work? This paper examines the answers to these and other similar questions, based on the experience made by the Tres de Febrero National University (Argentine), for those beginning a Statistics career.
4D2: INFLUENCE OF CONSULTING IN THE SELECTION OF TOPICS WHEN TEACHING STATISTICS
Edith Seier
East Tennessee State University, United States
Statistical consulting not only provides real life examples to mention in class; it provides a reality check that influences the way we teach and the choice of topics to teach or emphasize. The focus of this paper is on topics that are frequently omitted or not emphasized but which we consider important as a direct result of consulting experiences. The first four belong to introductory courses and among them are data issues, the test of hypothesis about proportions for small samples using the binomial distribution and some topics on categorical data analysis. The last two topics, one on statistical models and the other in time series, belong to upper division courses.
4D3: TEACHING FOR APPLIED STATISTICIANS
Lúcia Pereira Barroso
Universidade de São Paulo, Brazil
We consider the experience in teaching for applied statisticians of the undergraduate course of Statistics at Universidade de São Paulo, Brazil. Two disciplines are offered to the students which are linked to the CEA – Centro de Estatística Aplicada (Statistics Appied Centre).
4D4: EDUCATING FUTURE STATISTICIANS: AWARENESS, DIVERSITY, SERVICE
Julie Legler
St. Olaf College, United States
The groundwork for broadening a future statistician’s choice of and potential contribution to a workplace can be laid at the postsecondary level. Many quantitatively inclined entering undergraduates are unaware of statistics as a career, but even those with some awareness of the profession are unaware of the wide variety of venues in which statisticians can contribute and the great need for statisticians in many different settings. Our approach to attracting students to the field of statistics involves three primary principles: (1) Create an awareness among students of statistics as a career option; (2) Expose students to the diversity of uses of statistics; (3) Engage students in efforts where they can make meaningful contributions. Three different programs that have met with initial success at the undergraduate level are described: (1) a Center for Interdisciplinary Research; (2) a Practicum; and (3) a Biostatistics Interim in Geneva.
SESSION 4E: Statistics for Environmental Science
4E1: INTRODUCING DATA ANALYSIS IN A STATISTICS COURSE IN ENVIRONMENTAL SCIENCE STUDIES
Carmen Capilla
Technical University of Valencia, Spain
Education in methods of applied statistics is important for students who will be involved in management and decision-making processes. This paper discusses issues related to the teaching of statistics to students enrolled in an undergraduate environmental science degree course. The aim is to describe the teaching of graphical and numerical methods for summarizing and exploring data obtained in environmental studies. The application of descriptive and exploratory methods provides useful information regarding the distribution of the data at hand and of its patterns and associations. These methods are presented at the beginning of the course, following an introduction to the steps involved in the process of learning from data through the use of statistics. Students are instructed in the reading and interpretation of graphic and numeric data summary techniques. The importance of visualizing the main patterns and associations in the data is emphasized using environmental examples.
4E2: VISUALIZATION TOOLS TO AID IN THE UNDERSTANDING OF GEOSTATISTICS
William V. Harper
Otterbein College, United States
Isobel Clark
Geostokos (Ecosse) Limited, United Kingdom
Geostatistics is sometimes a difficult leap for even those individuals well versed in classical statistics. The impact of data location in spatial statistics may be only vaguely understood initially. Visualization tools that allow the student or practitioner to see the impact of moving data, adding additional data, deleting data, adding fault lines, changing search radiuses, and so forth aid the learning of geostatistical concepts. Due to page limitations only a few items are briefly illustrated. This visualization software called the Kriging Game is available free at http://geoecosse.bizland.com/softwares/ . This site also has other free geostatistical software and tutorials.
4E3: TEACHING STATISTICS TO ENVIRONMENTAL SCIENTISTS
Jennifer Brown
University of Canterbury, New Zealand
David Borchers
University of St. Andrews, Scotland
Teaching statistics to scientists and managers who work in environmental science can be stimulating and rewarding. Environmental science is data-orientated and there is a growing need for better statistical understanding amongst environmental scientist and managers. The Biomathematics Research Centre at University of Canterbury and the Centre for Research Unit for Wildlife Population Assessment (RUWPA) at St. Andrews offer a workshop covering some aspects of environmental statistics. We discuss some of our experiences from these.
4E4: COMBINING INTERACTIVE VISUALIZATION AND STATISTICS TO DETECT PATTERNS IN ENVIRONMENTAL DATA
Jean-Yves Hervé, Qing Liu
University of Rhode Island – Kingston, United States
Matthew Nicholson
U. S. Environmental Protection Agency, United States
Liliana González, Thomas Mather
University of Rhode Island – Kingston, United States
One of the biggest challenges statisticians face when working with non-statisticians on applied problems is to be able to effectively communicate the statistical results. In this paper we discuss the use of interactive visualization as a tool to present the relationship between a binary response and a set of explanatory variables. The visualization system we present allows users to "manipulate" directly, dynamically, and interactively their data set. At a first level, this allows to integrate visualization with a classical statistical analysis by providing interactive 3D views of the data set. Beyond its potential use as a straightforward visualization tool, this new system opens up interesting possibilities for exploring data visually, by its better exploitation of the human visual system. The paper presents an example of exploring visual relationships between environmental variables and the presence/absence of Lyme disease in Rhode Island.
SESSION 4F: Improving the statistical literacy of users through their professional activities
4F1: THE USE OF ADVANCED VISUALISATION TOOLS FOR COMMUNICATING EUROPEAN DATA ON EARNINGS TO THE CITIZEN
Werner Grünewald, Hans-Joachim Mittag
European Commission, Luxembourg
This paper illustrates by means of an important European survey on earnings how user-friendly interactive visualisation tools can be applied to communicate results of official statistics and to connect official statistics to the world of statistics education. The visualisation tools presented are self-contained with built-in methodological comments. They can be used offline on a CD-ROM or as part of dynamic PowerPoint presentations. They might be likewise used online on the Web sites of statistical offices, possibly embedded in virtual libraries, or as an integral part of electronic publications.
4F2: CHANGING APPROACHES AND PERCEPTIONS: BIOSTATISTICS AND ITS ROLE IN TEACHING THE STELLENBOSCH DOCTOR
Paul Mostert
Stellenbosch University, South Africa
The role of Biostatistics in Medicine and Health Care is sometimes only fully understood and appreciated by medical practitioners once they are fully qualified. Biostatistics and Epidemiology are also subjects in the medical curriculum that are disliked by most undergraduate students. The ‘Profile of the Stellenbosch doctor’ is a set of professional characteristics that every successful graduate possess, once they have qualified. The curriculum is therefore designed around this profile, where the Department of Statistics and Actuarial Science at Stellenbosch University in South Africa was one of the role players, especially with the introduction of a ‘golden thread’. The negativity of our medical students towards Biostatistics is real and a lot of that blame can be assigned to the way we teach and what biostatisticians think should be covered in a syllabus. Students normally struggle with probability and its applications and by introducing Bayesian approaches to the analyses will further complicate their understanding thereof.
4F3: SAMPLE SURVEY STATISTICS TEACHING: AN ALMOST WORLDWIDE PROBLEM ON TEACHING AGRICULTURAL SURVEY METHODS
Álvaro González Villalobos
National Academy of Sciences of Buenos Aires, Argentina
Most national or large-scale multiple-purpose Agricultural Surveys, because of their complex nature, should be based on multiple frame survey designs that use samples selected from an area frame, list frames of holdings and housing units, and point sampling methods. The problem is that, although such Agricultural Surveys imply important human and economic resources as well as time for their implementation and maintenance, the area and multiple frame survey methods already mentioned are simply not taught at most universities and statistical teaching institutions. In other words, it has been almost a worldwide tradition to teach in an inappropriate way agricultural survey methods. However those methods are widely used and involve specialized statistical knowledge for their implementation and analysis. In this paper we shall refer to the reasons of such lack of appropriate teaching as well as possible solutions.