Date of Original Version

Fall 2015


Response or Comment

Rights Management

© 2015 Robert E. Kass

Abstract or Description

The teaching of statistics has evolved more slowly than statistical practice. In diagnosing the problem with undergraduate statistics education, Cobb returns to Leo Breiman’s “two cultures” article and makes some important points. I completely agree with him when, consistently with Breiman’s earlier sentiment, Cobb warns against ceding to others “all methods of analysis that do not rely on a probability model.” Tukey’s profoundly important emphasis on the distinction between exploratory and confirmatory (inferential) methods, including the corruption of operating characteristics due to exploratory preprocessing, remains central to modern statistics. Furthermore, Cobb rightly suggests that computation should play a big role throughout the curriculum.


Comment on "Mere renovaton is too little too late: we need to re-think our undergraduate curriculum from the ground up" by George Cobb, The American Statistician, 69. Online Discussion: Special Issue on Statistics and the Undergraduate Curriculum.