We describe how the use of a games environment combined with technology supports upper primary children in engaging with a concept traditionally considered too advanced for the primary classes: *The Law of Large Numbers*.

Statistical tasks that can be solved in a variety of ways provide rich sites for classroom discourse. Orchestrating such discourse requires careful planning and execution. Five specific practices can help teachers do so. The five practices can be used to structure conversations so that coherent classroom narratives about solutions to tasks may be formed. In this manuscript, two classroom examples that illustrate the five practices are offered. It is argued that employing the five practices can lead to higher quality classroom discussion than some commonly used arrangements.

This article introduces a fun, hands-on activity for comparing hand water displacements.

This article describes a bivariate data set that is interesting to students. Indeed, this particular data set, which involves twins and IQ, has sparked more student interest than any other set that I have presented. Specific uses of the data set are presented.

In this paper, the dice game *Unders and Overs* is described and presented as an active learning exercise to introduce basic probability concepts. The implementation of the exercise is outlined and the resulting presentation of various probability concepts are described.

As the number of independent tosses of a fair coin grows, the rates of heads and tails tend to equality. This is misinterpreted by many students as being true also for the absolute numbers of the two outcomes, which, conversely, depart unboundedly from each other in the process. Eradicating that misconception, as by coin-tossing experiments, should be incorporated early on into learning the law of large numbers.

Students often are confused about the differences between bar graphs and histograms. The authors discuss some reasons behind this confusion and offer suggestions that help clarify thinking.

The aprioristic (classical, naïve and symmetric) and frequentist interpretations of probability are commonly known. Bayesian or subjective interpretation of probability is receiving increasing attention. This paper describes an activity to help students differentiate between the three types of probability interpretations.

The standard deviation is related to the mean by virtue of the coefficient of variation. Teachers of statistics courses can make use of that fact to make the standard deviation more comprehensible for statistics students.

The Tukey mean-difference plot, also called the Bland–Altman plot, is a recognized graphical tool in the exploration of biometrical data. We show that this technique deserves a place on an introductory statistics course by encouraging students to think about the kind of graph they wish to create, rather than just creating the default graph for the variables types they have. This graphical technique is described, and two examples are presented: one dealing with official agricultural data of Poland and the other one with an experiment on anorexia. Our opinion is that the plot is so easy and yet efficient in visualizing paired data that it should be included in statistics courses to support understanding and interpretation of data and their analysis. © 2014 The Authors. Teaching Statistics © 2014 Teaching Statistics Trust

The number of increases a particular stock makes over a fixed period follows a Poisson distribution. This article discusses using this easily-found data as an opportunity to let students become involved in the data collection and analysis process.

Students have intuitive notions of the meaning of variability; some may see variability as how values in a set vary from each other. This article provides a measure of variability that is based on that conception. We introduce this new measure and a method for calculating it. Finally, we prove that this measure is equivalent to the population variance.