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…

This article addresses an important problem of graphing quantitative data: should one include zero on the scale showing magnitude? Based on a real time series example, the problem is discussed and some recommendations are proposed.

Asterisks should not be used to indicate if the result of a hypothesis test is significant.

This article suggests how to explain a problem of small sample size when considering correlation between two Normal variables. Two techniques are shown: one based on graphs and the other on simulation. (Contains 3 figures and 1 table.)

Interpretation of correlation is often based on rules of thumb in which some boundary values are given to help decide whether correlation is non-important, weak, strong or very strong. This article shows that such rules of thumb may do more harm than good, and instead of supporting interpretation of correlation--which is their aim--they teach a…