Classroom demonstrations: education or entertainment?

Classroom demonstrations have two important purposes: to increase student understanding of physical concepts and to enhance interest in the subject matter. Do demonstrations in fact achieve the first of these goals? Does the manner of presentation determine the effectiveness of demonstrations as teaching tools? To answer these questions, we presented several demonstrations to different sections of an introductory physics course in different ways: (1) students were asked to predict the outcome before the demonstration, (2) students were shown the demonstration and told how it works (traditional style), or (3) no demonstration was shown. After the course, students took a free-response test asking them to predict and explain the outcome of physical situations similar or identical to the demonstrations. Analysis of the performance of different student groups shows that students who do not see a demonstration fare no worse than students who are shown the demonstration in the traditional way. The data suggest a small improvement in performance when students have to predict the outcome of a demonstration before seeing it.