Whether catapulting gummy bears to demonstrate to his students the fundamentals of experimental design or chairing national committees on undergraduate education in his field, George Cobb is always focused on finding new and better ways to teach statistics. Beginning in the 1980s he was in the vanguard of those who radically altered courses in introductory statistics as computers liberated them to set their students to work with real data. More recently, Cobb turned to the content and pedagogy of more advanced classes. He successfully sought ways to engage students from widely diverse academic backgrounds in courses that simultaneously explore mathematics and statistics.
The fruits of his latest work are newly designed courses in Markov Chain Monte Carlo, linear statistical methods, and mathematical statistics. Expanded versions of Cobb's handouts and problem sets for each of these are destined for textbooks. That on Markov Chain Monte Carlo, supported by a National Science Foundation grant, is already in draft form. These upcoming books will join his other published texts, which include Introduction to Design and Analysis of Experiments (Springer Verlag, 1998), and Statistics in Action: Practical Principles for a World of Uncertainty (Key Curriculum Press, 2003), written with Richard L. Scheaffer and Ann E. Watkins.
Cobb's students found "his class environment is truly focused on learning." They regularly reported working very hard, but find themselves highly motivated to do so and "pushed to their maximum potential without extreme pressure." And they enjoyed it. One concluded, "I worked super hard, but it was all real brain power, and not tedious busywork."
Cobb's constantly probing explorations of what should be the content and pedagogy of statistics courses not only brought him student acclaim at Mount Holyoke, but also national and international prominence. He has written and spoken widely to varied audiences of statisticians, mathematicians, and educators, and has led major initiatives in statistical education. He was chair in 1990–1991 of the focus group on statistics of the Mathematical Association of America and from 1990 to 1999 of the Joint Committee on Undergraduate Statistics of the Mathematical Association of America and the American Statistical Association. He served on the committee that founded the Journal of Statistical Education in 1993 and then was its associate editor for five years. More recently he served on the Committee on Applied and Theoretical Statistics of the National Academy of Sciences, thereby becoming the first statistician from a liberal arts college to be part of that committee. Among Cobb's honors was an invitation to give an after-dinner address at the Quadrennial International Conference on Teaching Statistics held last summer in Cape Town, South Africa. In 1993 he was elected a fellow of the American Statistical Association and in June 2003 was elected to serve a three-year term as vice president of that organization.
In addition, over the past two decades Cobb has frequently served as an expert witness in lawsuits involving alleged employment discrimination and in 2000 was involved as a statistician in the federal trial of a nurse charged with (and later convicted of) serial murder.
At Mount Holyoke he served as department chair from 1980 through 1982, and as dean of studies from 1989 to 1992.
- "Professors Benfey, Cobb, Nicholson, and Savoy Honored with Faculty Awards," College Street Journal, April 25, 2003
- "Creative Classrooms: Innovative Ways of Teaching and Learning, Vista, summer 2000
- "Move Over, Bill Nye: George Cobb Makes Learning about Statistics Practical and Fun," College Street Journal, April 9, 1999