Proceedings

  • Teaching future applied statisticians requires the teaching of consultation skills since the student must learn to interact with research workers, learn to abstract the statistical aspects of substantive problems, to provide appropriate technical assistance, and to effectively communicate statistical results. The approach of the Department of Biostatistics at the University of North Carolina at Chapel Hill is to provide a dual training that includes classroom work, but also involves a 'real' practicum. The objective of this paper is to present various modalities of the experience in training future consultants. These are evaluated by former students of the Department of Biostatistics that are currently involved in consultation and in training in their respective countries. The success of the training is seen through subsequent consultations in worldwide settings.

  • An analysis of the steps involved in forming the idea of an empirical sampling distribution and the nature of the methods and/or images used in most computer based strategies to teach this idea suggest that this way of using the computer adds little insight to the usual text based explanations that they are designed to complement. This analysis suggests reasons why a more recent approach which uses the computer to model and dynamically display the processes that underlie the idea is more likely to be successful. (orig.)

  • Students in an introductory statistics course are often preoccupied with learning the computational routines of specific summary statistics and thereby fail to develop an understanding of the meaning of those statistics or their conceptual basis. To help students develop a better understanding of the meaning of three frequently used statistics, this document presents HyperCard programs and a Lotus 1-2-3 spreadsheet for use by teachers in classroom simulation demonstrations of Pearson's correlation coefficients, t-test of two independent population means, and one-way analysis of variance. The instructional contents and typical computer screen outputs for each demonstration are discussed in association with their respective HyperCard programs, which are included in the appendix. (JJK)

  • The Pearson product moment (P.M.) correlation ("r") and four of its most widely used variations--the phi, the rho, the biserial, and the point-biserial coefficients--are reviewed. Using small data sets between one and nine, the conditions under which the various forms are restricted in power and robustness are explored. Seven sample data sets were constructed to illustrate the effects of the following conditions: (1) perfect correlation; (2) restriction of range; (3) measurement error (one variable); (4) measurement error (two variables); (5) extreme scores (one outlier); (6) extreme scores (two outliers); and (7) heterogeneity of sample distribution (one variable). Conceptual and algebraic linkages among the coefficients, and teaching practices that will facilitate the assimilation of these concepts and that will allow a student to predict the practical effects of the conditions are discussed. Two data tables are included. Appendix A contains definitions, formulas, and examples of the coefficients, which are presented as reproductions of a Macintosh computer hypercard stack designed for instructional purposes. Appendix B contains sample spreadsheet calculations for the seven conditions studied. (SLD)

  • Generally, when empirical statistical distributions are discussed in basic text books, a stress is given to the type of the data (quantitative, qualitative, ..). Rarely the problem of the meaning of the data and of their quality is treated. Only in the text books of sampling techniques and in some books of applied statistics this problem is considered, essentially with reference to the collection of the data. We think, instead, that it is very important to explain also the meaning of the data and the fact that they are one of the possible images of the phenomena. A summarized exposition of the problem is presented, emphasizing some particular aspects. (orig.)

  • The author presents guidelines for the selection of statistical analysis software given to graduate students to work independently. Criteria for a good teaching program are delineated. Several software programs are evaluated: STATMASTER, Statistics and Probability, Monte Carlo Simulations, Survey Sampling, KEYSTAT, CAPSAS: Computer Assisted Program for the Selection of Appropriate Statistics, EDA: Exploratory Data Analysis, INTROSTAT, Statistics With Finesse, STATPAC, GANOVA: Generalized Analysis of Variance, Speedstat, Micro-DSS/Analysis, and Micro-TSP. Several statistics packages are briefly reviewed, including STATPRO, A-STAT, Computer Models for Management Science, Multiple Factor Analysis, a General Correlation Program, and Test Construction Package. An appendix lists criteria for evaluating software. Demonstration pages prepared with a TEXPRINT printer interface card in the APPLE II+ computer are included. (DWH)

  • This paper describes the development of curriculum materials for teaching the Sum of Squared Errors (SSE) to one class of 25 eighth graders in Hawaii. Microcomputers were used in class. Prior to explicit introduction of the SSE students were given repeated contact with a data base of various statistics collected from members of their class. Then a computer program was introduced which randomly selected values using the data base; students tried different guessing strategies. Worksheets were used in connection with this and other content; they are included in the paper. Interactive game-like situations were also used. The bridge to standard deviation was then described. Misconceptions that students may bring to the discussion of probability are described, with ways the developed materials sought to clarify the concepts. Finally, the application of SSE to linear regression is discussed. (MNS)

  • The enclosed summaries were provided by working group facilitators and placed in the public domain in an unedited form to inform dialouge about and contribute to the improvement of assessment practices in statistics education. A formal conference report is in the planning stage.

  • Episodes recounted in this paper illustrate the evolution of statistics as a discipline and as a profession. The two are indissolubly linked, and it is useful to remember this as we contemplate present-day development.

  • Statistics is taught widely in the UK as an element of mathematics up to age 16. Over a quarter of a million children taking examinations involving statistics or mathematics with statistical elements. At the age of 18 the examinations involving the use of statistics show a different pattern. Some 50 thousand pupils take statistics or mathematics with statistical elements, but over 100 thousand take examinations in geography, biology, psychology, etc. involving significant elements of statistics. Thus the prime interest in statistics comes not from those with a specialist interest but from the "rest of humanity" who need statistics to support their other interests. The aim of this paper is to explore some of the implications of this fact.

Pages

register