Matt Crosslin begins by referencing a NY Times article on online cheating accur=sations at Dartmouth (I won't link to the paywalled article, but here's a summary and here's a good discussion). As Crosslin reports, 13 of the 17 accused are protesting their innocence. Now there are reasons for concern here but Crosslin takes it a step further by digging up a 2006 paper where "the overall goal is to predict which students are most likely to be cheating based on demographics and student perceptions." Now yes, that would be an issue, but it's not clear that anything of the sort happened in the Dartmouth case. The real problem here (as Crosslin later points out) is the use of clickstream data to identify cheaters because (as noted in several places) this data include automated requests. "Those pages can automatically generate activity data even when no one is looking at them, according to The Times's analysis and technology experts."
Today: 1 Total: 6 [Direct link] [Share]
