In other classes, we have administered the other problems
discussed in this article. The results, summarized in Table IV,
are mixed. Perhaps not surprisingly, the success rate on the
connected-spools problem is especially high. Performance on
the part of the unconnected-spools problem that involves predicting
the spools’ motions, already relatively high before
tutorial instruction, seems to have decreased slightly. We
believe that some students may confuse this experiment with
the one they viewed in tutorial. Additional data are being collected
to examine this possibility. Performance on the part of
this problem that involves comparing forces is higher than in
all but one of the 31 classes that had this problem prior to tutorial
instruction. It is possible that student ability to interpret
this experiment in terms of forces is slightly improved.
Performance on the incline problem is improved to a statistically
significant, but smaller degree. The role of student ideas
about frictional forces in this and other, related problems is
discussed in detail in another article