Proceedings

  • Over many years I have been attempting to improve statistical literacy in the population by changing the school curriculum. All such attempts have to be put in the general context of teaching, learning and assessing the subject. Ideally these should complement and reinforce each other. In practice they often conflict - in particular assessment can distort the learning process. In this talk I consider the nature of these conflicts and how they might be overcome in practice, giving examples from a lifetime's experience.

  • The development of a Profile of Statistical Understanding is aimed at providing a tool to assist educators to identify what 'can be' expected of students rather than what 'should be'. This profile needs to cover all basic areas of statistics in such a way that specific profiles will identify what to expect of graduating secondary students with respect to 'statistical understanding' for tertiary education or the work force. Responses from 13 to 18 year-olds to open-ended questions were analyzed using the SOLO model as a framework for an hierarchy. The profile is presented and responses from typical students are discussed to elaborate on the categories. Rasch analysis combined the rankings of students on different questions to produce a measure of statistical understanding for each student. A profile for an average student is discussed

  • Statistical literacy is analyzed from three different approaches: chance-based, fallacy-based and correlation-based. The three perspectives are evaluated in relation to the needs of employees, consumers, and citizens. A list of the top 35 statistically based trade books in the US is developed and used as a standard for what materials statistically literate people should be able to understand. The utility of each perspective is evaluated by reference to the best sellers within each category. Recommendations are made for what content should be included in pre-college and college statistical literacy textbooks from each kind of statistical literacy.

  • In 1985 the concept of a "DNA fingerprint" was introduced as a means of evaluating human identity and relatednes. (Jeffreys, Wilson, & Thein, 1985). The possible forensic and legal applications of DNA evidence were quickly appreciated and such data are now frequently presented in court cases involving serious crimes such as murder and rape. DNA evidence is also used in establishing paternity, in determining relatedness in immigration and inheritance disputes, and in identifying disaster victims. Such cases, especially those involving famous people, are widely reported in the media and are of interest to the general population. Also, many people will be called to serve on juries in cases where DNA evidence is presented. As statistical concepts are involved in evaluating such evidence, "DNA fingerprinting" as a topic can be used to introduce statistical analysis to undergraduates. If a non-mathematical approach is taken many concepts can be taught to secondary school children, extending their understanding of statistics while holding their interest with practical " real-life" examples.

  • In this paper I report two separate studies of congenital anomalies, one of which was generated by a media alarm, and the second that needed to disseminate research findings to the public through the mass media. There are many similarities between the two pieces of work and the process of disseminating results. Both show the importance of working with the media to share our work in an informed way to present a true evaluation of risks to the public. The first study, generated by media concern, was based on a cluster of four English babies born without a left hand. Although this original cluster on which the first part of this paper is based was several years ago, there are several important messages resulting from this work which are still as relevant today. The second study reports the findings of a collaborative study, known as EUROHAZCON, based in different locations in Europe, which tested whether living close to a landfill site was a risk factor for congenital anomalies. The media can have a huge impact both on what we do within our Congenital Anomaly Registers and in how we do it. It is useful to reflect on these studies and evaluate how the media influenced our work at the time. In particular, dealing with the media can be very time consuming whilst we are still trying to focus on the epidemiology.

  • Although secondary school students in many countries (like Ireland) get a limited and basic introduction to Statistics, it is often of a mechanical and tedious nature with little or no emphasis on data analysis and practical examples. In particular they, together with their teachers, rarely see the applicability and challenging nature of statistical thinking. Statisticians need to promote these aspects of statistics to the young, their teachers and the public at large. It is suggested that the use of examples of a local (often of a national) nature should be encouraged in an effort to emphasize the relevance of statistical thinking. With this in mind, several examples which have been successfully used in the Irish context are discussed.

  • In my paper 'Statistical Literacy - Statistics Long After School' presented at ICOTS V in Singapore, I discussed creating a course 'Citizens Statistics 101' and suggested what topics should be included in such a course. Unfortunately, the project has not happened, but I continue to think about it and present in this paper a practice exam that illustrates its content. The bottom line hope is that someday there will be no need for such a course as students will learn the statistics in school that will enable them to be statistically literate citizens.

  • Much has changed since the widespread introduction of statistics courses into the curriculum in the 1960s and 1970s, but the way introductory statistics courses are taught has not kept up with those changes. This paper discusses the changes, and the way the introductory syllabus should change to reflect them. In particular, seven ideas are discussed that every student who takes elementary statistics should learn and understand in order to be an educated citizen. Misunderstanding these topics leads to cynicism among the public at best, and misuse of study results by physicians and others at worst.

  • This paper develops insights into how to deal with the arithmetic mean in the initial phases of statistics instruction. It focuses on one of the various ways in which the mean can be used in statistics: as a normalized ratio. The paper analyzes individual interviews of 12 12-year-old students prior to their participation in a classroom teaching experiment. These interviews were used to test and develop instructional conjectures about how to support students' understanding of the mean and other normalized ratios. Three differences where detected in how students seemed to make sense of proportional comparison problems in which the use of the mean as a ratio is pertinent. The paper explains how these findings can be capitalized upon in designing and conducting instruction.

  • This paper focuses on 6-8 year-old children's thinking about randomness. It reports the findings of a study in which the children engaged with a game-like environment to construct for themselves spatial representations of sample space. The system was designed so that the rules governing the relationships between the selection of elements of the sample space and the outcomes of the game were available for inspection and reconstruction by the children. In response to a range of tasks, the children manipulated the sample space in ways that generated corresponding outcomes in the game. We present a case study of children's activities, which illustrates how the novel medium mediates the children's expression and understanding of chance events.

Pages

register