http://bit.ly/1qnv6gy October 8, 2009
I have been interested in John Sweller and Cognitive Load Theory (http://bit.ly/1lSmG0f) since reading Ray Lister's ACE keynote paper from a couple years back (http://bit.ly/1wPYrkU). I assigned several papers on the topic (see the papers in the References) to my educational technology class. Those papers have been influencing my thinking about how we teach computing.
I see some inconsistency in the arguments used in the discussion. From one side, we teach students reading by reading, but on the other side, some people question when we teach problem-solving by solving problems. Reading is a general skill and competency, basic and used in mastering other skills. Program-texts are greek for most beginners in programming, they should first learn "vocabulary" of programs, that is basic structures/constractions used in programs and, more important, they should learn meanings of this constructions. Simple reading programs does not help very much and teaches nothing if the context is not considered (parameters, instructions which precede and follow etc.).
Today we should also remember that visual programming (using graphical blocks), as the first step in programming, changes the meaning of "reading" programs. In fact, the Scratch environment is designed to use and modify other programs when building our own programs. Therefore, "reading" and "writing" such programs is a continuous integrated process of reading and writing, Then when students change their visual programming environment for textual, the same approch remains - they make use of subroutine libraries or programs found in the net to write their own programs
Maciej M. Syslo
I'm a big believer in empirical evidence. When it comes to learning, we need to test our models with experimentation and not rely on thought experiments. For example, students don't learn to read by reading. They learn to read by being immersed in a textual culture, by having parents read to them, and contrary to some expectation, by writing -- see this report: http://www.readingrockets.org/articles/researchbytopic/36871. The empirical evidence suggests that reading problems can help develop problem-solving skills more than just solving more problems, but at a certain level of experience -- see the work on expertise reversal effect for where it's more important to problem-solve than to read worked examples https://en.wikipedia.org/wiki/Expertise_reversal_effect.
Thus statements like "Simple reading programs does not help very much and teaches nothing if the context is not considered" need to be tested. We cannot assume that they are true on face value or by thought experiments.
Displaying all 2 comments