# Teaching by teaching or teaching by doing?

Mark Guzdial at Georgia Tech writes about: How we Teach Introductory Computer Science is Wrong. I think it clearly applies to science, statistics, mathematics, physics and other basic and difficult skills.

The premise is that for *introductory level learning* (but not more advanced students) having the students learn by solving problems is much less effective than having the students learn by reading a set of worked problems. Basically, without the fundamental foundation of information, doing problems on your own is just frustrating and unproductive. Introductory classes should be about *telling* students the information they need to get to the intermediate level where they can start doing.

It's a fascinating set of research, and I'm not sure how generally it applies. But unlike many teaching theories, this is at least based on **data.**

(thanks to my wife for the link).