Webinars

Many ideas and recommendations for meeting the GAISE guidelines at the college level have targeted relatively small class sizes. This webinar will provide an overview of a suite of twelve simulation activities that were designed to develop student conceptual understanding of inference in large lecture classes using personal response systems (clickers) to collect data. Details will be provided for three of the activities, in which each student performs a simulation once using a calculator and the results are collected via clickers. The activities allow students to experience statistical concepts such as distributions or models, variability, and the Central Limit Theorem. The large class, therefore, becomes a learning asset, rather than a liability.

Teaching several semesters of classes where all students in the class have a learning disability has offered me a unique perspective on how some LD students learn statistics. I have found that some students seem to "see" statistics problems differently than the average student. In this webinar, I will share with you some tips on how to show your LD students how to read statistics problems more effectively to help them overcome their learning disability.

This presentation focuses on how to incorporate a service learning component into introductory statistics. Service-learning is a concrete application of statistical methods using real data with the analysis and interpretation that is useful to a community agency. Discussion will include how to locate an organization, the selection of appropriate content for the project with focus on understanding what questions need to be answered and how to do so, the grading rubric for the presentations or posters and the time line of formative evaluation as the project proceeds.

The Statistics Education Research Journal (SERJ) publishes high quality research related to the teaching and learning of statistics. Bob delMas, co-Editor of SERJ, will present characteristics of manuscripts that tend to result in published articles, as well as point out critical flaws that can keep a manuscript from being published in SERJ. Ample time will be provided for the audience to ask questions of the co-Editor.

In the early 1990's, the National Science Foundation funded many research projects for improving statistical education. Many of these stressed the need for classroom activities that illustrate important issues of designing experiments, generating quality data, fitting models, and performing statistical tests. This webinar describes such an activity on logistic regression that is useful in second applied statistics courses. The activity involves students attempting to toss a ball into a trash can from various distances. The outcome is whether or not students are successful in tossing the ball into the trash can. This activity and the adjoining homework assignments illustrate the binary nature of a response variable, fitting and interpreting simple and multiple logistic regression models, and the use of odds and odds ratios. Trashball activity website

When a course begins, students may not arrive with abundant background knowledge (and certainly haven't yet done any assigned reading in the textbook), but do arrive with some (mis)conceptions about the course and the discipline. Based largely on the Lesser & Kephart paper in the November 2011 issue of Journal of Statistics Education www.amstat.org/publications/jse/v19n3/lesser.pdf (which you are welcome but not required to read in advance), this webinar gives concrete class-tested activities and process (with rationale) of how instructors can go beyond calling roll and discussing the syllabus and set the tone on a course's opening day. We will also discuss how the process may be applied to other days of the course, to various types of courses, to classes of varying sizes, to meeting times of varying lengths, etc.

After teaching the concepts of statistics and statistical reasoning for almost twenty-five years I became convinced that my lecture-recitation format was inefficient and maybe even counter-productive with respect to student learning. Throw in an excruciating self reflection focused on "what do my students really need me for anyway?" and it quickly became clear that my style and my classroom needed some kind of substantive change. The result was the development of an inverted classroom environment where traditional lecture material is off-loaded as mp4 files, the classroom is used for discovery and discussion, and the recitations are better tailored to the deductive abilities of new TAs. In this presentation we will demonstrate some of what we are doing here at the University of Kentucky in a course that serves approximately 4200 students in a calendar year. We will be sure to point out the things that may not be working that well, in addition to those that are. Questions to Think About Assuming you teach an introductory conceptual statistics course in a lecture/recitation format with TAs in charge of the recitations: Do you use first-year TAs in your recitations? If so, do they have difficulties with appropriately handling conceptual questions and demonstrations? Have you ever thought about what things you say and do in the "lecture" that are truly essential for you to say and do? Are these things that reflect the depth of your knowledge and experience in the field of statistics?

We discuss how and why we now use simulation methods (bootstrapping and randomization) to introduce fundamental topics of inference (intervals and tests) in an introductory statistics course. We describe ways to make these methods accessible early in the course, demonstrate new user-friendly applets for teaching and using these methods, and discuss some of our experiences with using this approach.

This webinar is a short visual, narrative journey from the vibrating boxplot imagery of the author's 2009 USCOTS Plenary and Wild et al. (2011) to visualisations of bootstrap confidence intervals and, if time permits, randomisation tests. Software under development will be used and made available as-is to any brave souls willing to live on the edge. References: www.stat.auckland.ac.nz/~wild/09.USCOTSTalk.html Wild, C.J., M. Pfannkuch, M., Regan, M. and Horton, N.J. (2011). Towards more accessible conceptions of statistical Inference (with Discussion). Journal of the Royal Statistical Society A, 174, 247-295.

This webinar will discuss the presenters' experiences conducting research with collaborators at several different institutions. Both presenters have in the CAUSE research cluster program for more than two years. This NSF funded program brings together statistics educators earlier in their career for the purposes of research. We will discuss some literature that suggests collaborations have a better chance of publication, the difficulties and the rewards of cross institutional research based on our own experiences and how clusters' view of the research process has changed over these past two years.