[quoting Tripp, 23 Oct 96.b] I think this kind of argument is not too interesting. It is usually a sign that you have no good research ideas. People in fields that are making good progress usually don't worry about philosophy too much.
What is the evidence for this progress, Steve? I went to a seminar in Atlanta yesterday. A major authoring system company was demonstrating the latest version of their software which allows CBT developed with their system to be automatically converted into HTML and Java files for distribution on the WWW. Cool!
But the actual CBT they were showing boiled down to essentially what we have been doing since the days of Skinner's teaching machine, i.e., show a little content, ask a question about it, judge the learner's response, and provide feedback. This type of CBT can be effective--but the design of it is hardly informed by the last five decades of media research. It is grounded in the principles of behavioral psychology, common sense, hopefully a dash of instructional analysis, enhancements in programming environments, evaluation, and a little creativity. Fifty years of media research helps the designers of this type of interactive learning system very little.
Consider the research on text density on CBT screens that was done years ago when researchers were focusing their investigations on 40 versus 80 character displays, and similar media elements. Of what use is that research now that we are dealing with interactive learning environments that allow users to set their own screen text densities within multiple windows and frames? The generalizations from your brand of media research are too weak to have practical value. This was the essence of Clark's original argument. He maintained that it was the instructional design (content, tasks, interactions, feedback, etc.) that make the real difference in learning, and that a good instructional design could, at least in theory, be implemented with any media. (Of course, costs, time, and other important factors would render the use of just any media impractical.) As I have written earlier, I agree with you that Clark's argument has many flaws, and I much prefer the position of someone like Bob Kozma. But I cannot support another round of media research studies of the type you propose.
Finally, as for your not finding this kind of argument too interesting, I would advise you think deeply before tossing these issues aside. The problem is Steve, that while you and members of a very small academic community, may recognize the limitations of your "print is better than audio" conclusion, most people don't bother finding out about all the caveats and compromises in your research. People with power can seize upon this type of simplistic conclusion and use it as a club to beat down innovation and creativity. Right now in my state, we have an elected school superintendent with a "far right" political agenda who is pushing "back to basics" curricula that have about as much research foundation as your "print > audio" assertion. These things have real consequences.