 Proceedings
|
Invited TalksTopic 1 : Data and context in statistics education: towards an evidence-based societyThis topic emphasizes the main theme of the conference: “Data and context in statistics education: towards an evidence-based society”. With the increasing amount of information now available, this has become crucial in decision-making in all areas of life. In this topic special attention will be given to evidence based decisions in areas including medicine, statistics education research, public policy, business and industry, practice, statistical literacy, the international statistical literacy project and for the use of context by statistical educators.Session 1A: Evidence-based medicine1A1: Evidence-generating research and evidence-based medicineGlenn Jones McMaster University & Credit Valley Hospital, CanadaEvidence-Based Medicine acknowledges Evidence-Generating Research that informs clinical practice but does not wholly determine patient decisions. As it matures, it will be widely applied, and with further development it may become a complete Theory of Medical Choice. Experience with Evidence-Based Medicine can inform the broader pursuit of an Evidence-Influenced Society. Paper 1A2: Divergent needs of learners in evidence based medicineKaren Smith University of Leicester, United KingdomThose who are actively engaged in health research expand the evidence base which, it is hoped, is used by all. In contrast, end-users of that evidence base must appropriate it, with some critical appraisal and interpretation for patients. Therefore, the statistical needs of researchers and users of evidence are not always the same. But the statistical education all too often is. Initial statistics teaching is typically focused on the demands of undergraduate degree courses. The mismatch of this education and later needs is a problem that grows in significance during a career, with many healthcare researchers having limited opportunity to undertake further training in statistics, and practitioners getting none. The diverse statistical needs of these two groups, and the challenges to how and what we might teach are discussed, drawing on experiences teaching undergraduate students, providing training to practitioners, and active involvement in health research. Paper 1A3: Drip-feed education: statistics notes in the British Medical JournalMartin Bland University of York, United KingdomDouglas Altman University of Oxford, United Kingdom Evidence-based medicine depends on decision makers, such as clinicians, being able to understand evidence. Statistical understanding is essential for this. Although most clinicians include some statistics in their training, it is often forgotten or confused by the time they want to apply it in practice. Statistics Notes, published in the British Medical Journal, help to overcome this. In this paper we discuss how we came to write these short pieces, what they are like, and how they have been received and cited. We discuss the value of such series, both to medicine and to ourselves, and how the process is repeated in other journals. We recommend that similar series be undertaken in journals in other fields and suggest that laboratory research using animals would be a good focus for other authors to try. Paper Session 1B: Evidence-based policy making1B1: Evidence-based policy making: the Pensions Commission and beyondJoanna Littlechild Department for Work and Pensions, United KingdomBav Hirani Department for Work and Pensions, United Kingdom The Pensions Commission first report (2004) set out the evidence base in relation to pension provision in the UK drawing on national surveys, academic and social research, modelling and administrative data. Its recommendations (2005) were similarly evidence-based. Since then the Department for Work and Pensions (DWP) has worked to introduce pension reforms through the 2007 Pension Act, focused on a fairer State Pension system and the 2008 Pensions Act and the Workplace Pensions Reforms Regulations that aim to increase private pension saving in the UK. This paper will focus on the Workplace Pensions Reforms. Evidence-based impact assessments and implementation plans have been published. Discussions are ongoing to develop an Evaluation Strategy. Liaison continues between DWP and other colleagues to build on existing surveys, such as those produced by the Office of National Statistics to develop the evidence base for the future. Paper 1B3: The evidence gap and its impact on public policy and decision-making in developing countriesBen Kiregyera African Centre for Statistics, UgandaIt is now recognized that statistics are no longer just a technical issue but also a development issue. They constitute an essential element in improving the ability of government to develop appropriate policies, manage the economic and social development processes, monitor improvements in the living standards of the people and report back this progress to the public. Statistics are also needed by organizations other than governments and the public. The new focus on Managing for Development Results in developing countries has resulted in unprecedented latent demand for data. This has put pressure on already fragile, under-resourced and under-performing national statistical systems, leading to an evidence gap. As a result, policy-making in some countries can be purely subjective or ideological and the monitoring of policy outcomes can be constrained. The nature of the evidence gap is explored and a case is made for bridging this gap. The paper presents some requirements for this to happen. Paper Session 1C: Evidence-based management1C1: Diagnosis, provision and assessment of quantitative skills for managers in local governmentShirley Coleman University of Newcastle, United KingdomRichard Browne Newcastle University, United Kingdom Basic numerical and graphical skills give enormous returns and yet are sorely lacking in otherwise highly skilled personnel. Many local government managers find themselves unable to understand or utilise basic tools, such as aggregated indices, run charts, bar charts and histograms. ISRU has developed a sensitive and innovative training programme for Local Government Organisations (LGOs). It is presented as skills updating to avoid confrontation and includes relevant computing (spreadsheet) skills. The programme is aligned to competences recognised for LGOs and this encourages take-up by managers. The need for improved mathematical skills links sector skills councils, improvement charities and education. The Mathematics in Education and Industry (MEI) charity has developed a diagnostic tool which will be piloted in the LGO arena. Training programmes work best with realistic data and examples placed in context. Feedback is vital for continuous improvement and has played a significant role in the development of the programme. Paper 1C2: Information quality for process improvementOriol Camps-Lorente Technical University of Catalonia, SpainXavier Tort-Martorell Technical University of Catalonia, Spain As George E. P. Box once stated, every system executing a process is like a radio transmitter except that it transmits information instead of electromagnetic waves. Information flows from processes to management decisions, through systems involving people, technologies and operating procedures. However, quite frequently that informational chain gets disconnected and decisions are not properly based on evidence from processes. Errors in information capture, coding, representation, storage, recovery, analysis, communication or interpretation may cause such disconnections. A framework for analyzing the quality of information flows involved in managing particular processes is proposed and exemplified with a case study. The framework has interest for statisticians and its implications for statistical education are discussed, together with a case of pedagogical innovation. Paper 1C3: The evidence-based management of learning: diagnosis and development of conceptual thinking with Meaning Equivalence Reusable Learning Objects (MERLO)Masha Etkind Ryerson University, CanadaUri Shafrir University of Toronto, Canada Ron Kenett KPA Ltd, Italy This paper discusses the management of learning for achieving deep understanding of concepts and techniques for the analysis of conceptual situations. Our goal is to to teach not only “how” but also “why”. Conceptual understanding is a critical component of evidence-based management. The paper presents Meaning Equivalence Reusable Learning Objects (MERLO), a novel methodology for evaluating and enhancing deep comprehension of the essence of multidimensional, complex conceptual situations, often embedded in mixed data-sets. We take a broad view of conceptual understanding and use training of applied statisticians, architecture students, and Continuous Medical Education programs as examples. The next sections provide background to the evidence based management of learning; an introduction to MERLO; a concept mapping of information quality (InfoQ); and discussion and directions for future research. Paper 1C4: Enhanced TESF methodology for course excellenceStefano Barone Chalmers University of Technology, SwedenEva Lo Franco University of Palermo, Italy The need to achieve excellence in services provided by a University has been the object of study and research for several decades. Often, the focus is not only on the education service, but on several satellite services provided by a university. This article focuses on teaching where, according to the authors, improvement actions of an academic institution should always begin. The work describes a tool for measuring student perceptions of selected aspects of a course. This tool is part of the Teaching Experiment and Student Feedback (TESF) methodology previously developed by the authors. Here, the steps to build a student satisfaction coefficient are fully described and data collected in one specific course is analysed. The TESF methodology is in line with evidence-based management principles and helps statistics teachers make practical data driven decisions. Paper Session 1D: The researcher/practitioner gap1D3: Bridging the researcher-practitioner gap: views from different fieldsGabriella Belli Virginia Polytechnic Institute and State University, United StatesIssues dealing with a researcher-practitioner gap are neither unique to one area nor are they a recent phenomenon. These issues are prevalent in fields such as computer science, education, healthcare, information systems, library and information science, management, nursing, political science, psychology, sociology, and social work. Given the fact that such a researcher-practitioner gap is prevalent, both across disciplines and time, how have professionals and academicians dealt with this issue? What can statistics education learn from what others have tried? This paper will provide an overview of how the researcher-practitioner gap has been addressed in a variety of fields, what attempts have been made to lessen the gap, and what strategies might apply to statistics programs. The challenges and opportunities will be discussed, as well as what may be learned from past efforts at bridging the researcher-practitioner gap. Paper Session 1F: Creating an evidence-based society1F1: Wikis, dynamic charts, videos and other innovative tools to transform statistics into knowledgeEnrico Giovannini National Statistical Office, ItalyThe development of web 2.0 and other Information and Communication Technologies (ICT) are creating a revolution in the way information is produced and shared among different interest groups and individuals. Concepts like “collective intelligence”, “crowdsourcing” and “prosumers” are at the basis of successful initiatives like Wikipedia, Innocentive, Facebook and other platforms used to develop both free and fee products and services widely appreciated, especially by new generations. How are these trends affecting the statistical world? Can “official” data providers continue to play their role just by introducing new ICT tools without changing their business model? The purpose of this paper is to discuss where the valued added of statistics comes from and the way in which information is spread in society. Some experiences about the use of innovative communication tools/approaches are also highlighted. Paper 1F2: What we know and what we should know; examples of ways of helping real users of statistical informationMaria Manuel da Silva Nascimento University of Trás-os-Montes and Alto Douro, PortugalJosé Alexandre Martins Polytechnic Institute of Guarda, Portugal With and within our classes we would like to effectively contribute to create an evidence-based society. From our past experiences with the students in various services courses, the project work approach seems to have the main components to motivate teachers and students to develop a cooperative work mainly aiming the students to be acquainted with the statistical methods. Since our students (future Engineers and Oenologists) will be users of statistical information they developed a survey to study the use(s) of statistics in their own areas. With the project approach we used the data to explore statistical concepts and to use statistical information to make some decisions in their one professional area of (future) work and even in their own lives. Paper 1F3: Analysis of clustered measurements: a comparison of the performance of foundation year students, 1994 cohort, with those of direct students, 1995 cohort, at the University of Limpopo, South AfricaMaseka Lesaoana University of Limpopo, South AfricaMaupi Letsoalo Tshwane University of Technology, South Africa In applied sciences, one is often confronted with the collection of correlated data. Dependence between observations from the same study subject renders invalid the usual chi-square tests of independence and inflates the variance of parameter estimates. Disaggregated approaches such as hierarchical linear models which are able to adjust for individual level covariates are suitable in the analysis of such data. We compare the performances of Ex-UNIFY (Foundation Year) students and direct students at first year of Bachelor of Science programme at the University of Limpopo, South Africa. Both aggregated and disaggregated analyses are performed. The results show that Ex-UNIFY students perform better than direct students in their main examination of first year in their Bachelor of Science programme. Disaggregated analysis is able to adjust for individual level covariates, therefore gaining power over aggregated analysis. Paper Session 1G: Lies, damn lies, statistics: lessons from past and present for the future1G1: The “compleat” applied statisticianDonald L Bentley Pomona College, United StatesIn March of 2007, Discovery Channel released the controversial documentary, “The Lost Tomb of Jesus,” which asserts that a first century tomb in Jerusalem contained the remains of Jesus of Nazareth and members of his family. Included in the documentary is an interview with a statistician who calculated the odds in support of this assertion at 600 to one. As skepticism intensified among biblical scholars, archaeologists, and the Christian community, support of the assertion relied more heavily upon the statistical results. Using the above example, this paper discusses the responsibilities of an applied statistician on a research team: becoming familiar with the subject of the research, scrutinizing every assumption used in the analyses, validating all data, and being sensitive to consequences the statistical results might have on society. Paper 1G2: Unintentional lies in the media: don’t blame journalists for what we don’t teachJessica Utts University of California, Irvine, United StatesIt’s easy to find misleading and even harmful reporting of statistical results. For example, a 2008 study titled “You Are What Your Mother Eats,” asserted that children born to mothers who eat breakfast cereal are more likely to be boys than are children born to mothers who do not eat breakfast cereal. A 2009 analysis by statistician Stan Young and colleagues showed that the result was almost surely a false positive, but by then the study had gained widespread media attention. Many students who take introductory statistics come away from the course able to compute a standard deviation, yet unable to spot an egregious example of poor statistical reporting such as the one illustrated by this example. We are doing an inadequate job of educating the next generation of medical researchers, journal referees, policy-makers, journalists, and so on. I will discuss some ways we can do a better job. Paper |