you asked if gains in interactive teaching are due to extra time spent in
teaching. It is quite true that deep learning takes time, perhaps more than is
usually expected. I am using collaborative discussions (close to Peer
Instruction) and research based material in teaching high school physics
(nterantional Baccalaureate syllabus). I have not noticed a dramatic difference
between the time I spent when I taught almost explusively problem solving. I
can fairly confidentally state that now my students learn better in terms of
conceptual understanding ( I have data to support this, see Savinainen (2001)) .
I think that my students do better in problem solving as well but I do not
have data to support this claim.
There is a very good article which compares interactive and tratiditional
teaching using traditional exams (basically problem solving): Gautreau and
Novemsky (1997). One group was taught using "Concepts first" approach developed
by A. Van Heuvelen. The idea was to create meanings first using graphical,
verbal and diagrammatic representations before mathematical poroblem solving.
These represenational tools were then then used as an aid in problem solving.
Three other groups were taught conventionally. If I remember correctly (I don't
have the article here at school) the groups were randomized to makes ure that
all the groups were as similar as possible. Teaching time and syllabus were
identical.
The final exam was set by convetional group professors; Gautreau did not even
know what the questions were. The interactive teaching group had superior
results. It was not good news for experienced professors who taught using
traditional lecturing (the article describes their reactions, it is fun to read
:-)).
Eric Mazur (1997) reports that his students in Harvard did better in a problem
solving test after Peer Instruction than his earlier group which received
traditional lecturing. Peer Instruction improved results in Mechanics Baseline
Test as well. This test involves some problem solving even though some feel
that it is essentially conceptual test.
It seems that problem solving and conceptual understanding are not always
correlated. Mel Sabella's disseratation (Sabella 1999) gives an interesting
analysis on this topic. It could be stated that
1) Traditional problem solving do not significantly enhance conceptual
understanding. This is well supported by Hake (1998).
2) Focusing on concepts do not necessarily enhance problem solving. Both
aspects has to be practised and integrated together, just like Gautreau did.
Sabella's work gives some support for this.
Gautreau, R. and Novemsky, L. (1997). Concepts first - A small group approach
to physics learning , Am. J. Phys., 65 (5), 418-428.
Hake, R. (1998a). Interactive-engagement vs traditional methods: A
six-thousand-student survey of mechanics test data for introductory physics
courses, Am. J. Phys. 66, 64- 74. Online at <http://www.physics.indiana.
edu/~sdi>.
Mazur, E. (1997). Peer Instruction: A User's Manual. Prentice Hall.
Sabella, M. (1999). Using the context of physics problem solving to
evaluate the coherence of student knowledge. Dissertation, University
of Maryland.
Savinainen, A. (2001). An Evaluation of Interactive Teaching Methods in
Mechanics: Using the FCI to Monitor Student Learning. In Ahtee, M., Björkqvist,
O., Pehkonen, E. and Vatanen, V. (Eds.) (2001). Research on Mathematics and
Science Education - From Beliefs to Cognition, from Problem Solving to
Understanding, Institute for Educational Research, University of Jyväskylä.
Online at <http://kotisivu.mtv3.fi/oma/physics/>