The second workshop on technology-enhanced learning at ANU was by Paul Francis on "data mining information from student quizzes and assignments". He used a simple questionnaire for physics students and analysed the words in the responses and then correlated these with their results for the course. The results for this were interesting. One expected result was that introverted "geeks" do do better at physics. Some unexpected results were that those with full time jobs do better than those who don't have jobs. Overseas students do better than local Australian students. Students from outside the ACT do better than ACT students. Paul emphasised these were very preliminary results but they are interesting non the less, as is the technique. Paul was also able to look at the relationship between the answers to different questions to see if there was a lack of understanding around particular topics in the course.He also said that ethical approval had been obtained for the study. Ian Barnes had previously done some work on "Personality Type and Software Development" at the Department of Computer Science.
Having full-time jobs: I wonder if these are just zero-order correlations, or controlling for age: perhaps most of the students with F/T jobs are also "returning" students.
ReplyDeleteI've found that the academic performance of "returning students" tends to be bi-modal: they're either very good (due to maturity, discipline, and the very conscious decision to attend uni), or barely scraping by (because they haven't sufficiently re-arranged their lives to compensate for uni responsibilities, or because there was a very good reason they didn't attend uni in the first place [i.e. not academically gifted]).
--GG