Is the Computer Appropriate for Teaching Physics?

Edward F. Redish

Department of Physics, University of Maryland, College Park, MD 20742-4111

(Published in Computers in Physics, 7(6), 613 (1993))

When the personal computer first came out in 1981 I was elated. As a physics teacher at a research university and as a researcher in a field that uses computers heavily, I was concerned that our undergraduates weren't learning to do physics with computers. My attempts to use our mainframe in class were frustrated by the difficulty of teaching my students our job control language and the difficulty of their getting computer time. The personal computer seemed like the answer. I bought my first PC in December 1981.

A dozen years later I'm now the proud owner of three machines -- home, office, and laptop. My department has two microcomputer labs and a laboratory with microcomputers. My campus has computer workrooms in the dorms, library, and parking garage. Computers in education are clearly here to stay. Now seems like a good time to ask: Have we figured out the right way to use the computer in teaching physics?

One of the first computer education applications I saw was at a meeting of the AAPT in the early '80s. A cart ran down an inclined plane. The computer recorded its one hundred times each second, fit the data with the textbook formula, and, in less than a minute, printed out a report showing the values of the parameters extracted and graphs of the data and theory. This is what I call a "the computer gets an A the student gets an F" experiment. If you didn't understand it before you saw it, you wouldn't learn much from it.

While the revolution in computer availability has been taking place, a second thread has been running in parallel: a new focus on the student through research in physics education. Educational physics researchers ask: What do our students know and how do they respond to what we teach them? This allows us to model our students as well as to assess them. It treats their learning as a scientific problem to be solved. Two guiding principles that designers of educational computer materials should keep in mind are:

Students are not blank slates. What they learn depends strongly on what they know -- or think they know.

Students learn best through active engagement -- but "hands-on" activities are not enough; it must be "brains-on" as well.

I classify computer activities designed to promote learning as being of two types: constructive and video game. The former help the student understand through active mental engagement. In the latter, students learn by "calibration" -- they do something and see what happens. Though some learning takes place, it's not the conscious, precise, and insightful kind that we need in physics.

By now there are a number of excellent constructive computer applications, many of which have won awards from this journal. Our M.U.P.P.E.T. programs allow even freshman physics majors to begin self-guided independent research. Thornton and Sokoloff have shown that using the computer in a "guided discovery mode" to take and display data in the lab can help students solidly build concepts that are otherwise difficult to learn. Laws has demonstrated the value of giving beginning students a variety of powerful general purpose computer tools -- graphers, data accumulators, and spreadsheets. Sherwood and his collaborators have developed a number of excellent programs that are powerful and flexible but still provide students with sharply focused activities that allow them to correct and build on what they know.

The programs I have listed above all aim at introductory students. The CUPS project, which is developing materials for upperclass physics majors, has won its first award in this issue. Software for advanced courses is inherently more difficult for us to develop successfully. Above the introductory level (and often there as well), we tend to focus our attention on the course's content and forget the students and their response. Computers offer professional physicists new ways of approaching topics they may have considered boring. But presentations that enthrall the expert may bewilder the novice.

So after more than a dozen years of working on computers in education. do I use computers in my classes? Yes, but cautiously. When I teach majors I start them on learning to do physics with computers. When I teach service courses my uses are more limited. Each program I add has to provide a large fraction of my students with important learning experiences that are hard to get in other ways. I carefully watch what students do and make improvements for next time. The same program, used in different ways can produce highly positive effects -- or none.

So what's the answer to the question in my title? Yes, the computer can help your students learn physics -- but not in all cases and not in general. The crucial question is: What physics to which students?

Dennis Donnely says in the article describing this year's awards: "If software developers would get their products into the hands of potential users and watch in silence, they would learn much valuable information." That's a good start. But I would like to add: If software developers asked users what they thought was happening in the program and what they were learning, and then listened carefully to the responses, they would take a long step toward of developing computer software that would actually promote learning.


Edward F. Redish
Department of Physics, University of Maryland
College Park, MD 20742-4111
Phone: (301) 405-6120 Fax: (301) 405-6114
redish@umd.edu