This is a report on SIGCSE 2002 (an ACM conference on Computer Science Education). I was mostly interested in introductory computer programming and this is reflected in the talks I attended. Generally there were talks related to introductory programming during every session which meant there was little time to see anything else.
So what's new in the intro to programming world. The big thing this year was pair programming.
A few educators decided to try the idea for new students to programming, as it seemed to combine peer learning with the successful strategy in industry. Charlie McDowell presented a paper and maintains a list of papers related to pair programming. He seems to have missed this interesting paper on student's perceptions of pair programming
The main benefits of pair programming are typical of those that you get both in industry and in peer learning environments:
Naturally there will be problems ensuring that the students share the work fairly. But the potential benefits are large.
Another big issue was the discovery that many introductory courses are not actually successfully teaching students to program. These results are an extension of the "Report by the ITiCSE 2001 Working Group on Assessment of Programming Skills of First-year CS Students", "A multi-national, multi-institutional study of assessment of programming skills of first-year CS students". There were a number of working groups trying to identify the problems associated with intorductory programming courses.