In our LPSS project we had a section called 'Automated Competency Detection and Recognition'. The idea was to use analytics to study expert performance and identify using actual evidence (as opposed to anecdfote) to determine what actually characterized an expert's expertise. This would then feed into the content recommenders, etc. Unfortunately, there was nobody in LPSS interested in doing this, and it languished. But here we have some work that demonstrates really well what that sort of feature could look like, and this makes it important. This study (37 page PDF) of expert and novice teachers found significant differences in what "what they notice and how they organize, represent, and interpret information." The authors suggest designers "can use information generated from studies that employ eye-tracking methodology to facilitate more efficient inspection." But more importantly, I think, the novice teachers can learn from seeing exactly what it is that the experts do.
Today: 4 Total: 86 [Direct link] [Share]
