Testing Intuitively for the Statistical Independence of Two Categorical Variables

There has been a recent emphasis on making the teaching of introductory statistics for business students more intuitive. Members of Making Statistics more Effective in Schools of Business, for example, have been active for more than 15 years promoting, among other things, a de-emphasis of mathematical formulation and a focus, instead, on performing practical data analysis and on getting students more actively engaged in learning. At a recent INFORMS conference in San Jose we discussed a new method of teaching hypothesis testing, using graphs and p-values, which is much more intuitive than the traditional approach followed by virtually every textbook. This method has been objectively tested in the classroom with excellent results. Here we summarize our discussion in San Jose, present the results we had with our students using the new approach, and introduce a new example based on the test for the independence of two categorical variables.