Modern Statistical Computing and Graphics in Academia

Topic: Teaching statistical computing and graphics

The teaching of statistical computing and graphcs tends to be done in an ad-hoc fashion as part of more classical statistics courses. With more thought, we might design more appropriate and more stimulating classes.

  • Curriculum Often the focus of statistical computing classes is on learning the tools of the computer rather than ideas. One reason for the lack of more focussed courses is the small number of people that are prepared to teach them.

    It would be valuable to understand what topics might be usefully included in a statistical computing course and what we would like students to learn. Should we teach them LaTeX or Word, algorithms for linear algebra, programming in C or Java, or focus on environments such as database technology?

  • Encouraging New Students In order for the field of statistical computing to flourish, we must encourage students to study the subject and consider it to be a valid, valuable and sustainable research area for them. Some people dissuade students from engaging in statistical computing research because of the difficulties of being rewarded for it.

    We also have to present it as an interesting and exciting area. This will most likely involve collaboration with other disciplines like computer science and biology. An increasing number of students are studying statistics from inside other departments and thus we are ``losing'' students to more attractive presentation of very similar material. While this may in fact be a good thing, we should discuss our response to it: try to arrest it, or embrace it?


Posted April 2001