The Simple Location Heuristic is Better at Predicting Students' Changes in Error Rate over Time Compared to the Simple Temporal Heuristic.
Educational Data Mining(2011)
摘要
In a previous study on a physics dataset from the Andes tutor, we found that the simple location heuristic was better at making error attribution than the simple temporal heuristic when evaluated on the learning curve standard. In this study, we investigated the generality of performance of the simple location heuristic and the simple temporal heuristic in the math domain to see if previous results generalized to other Intelligent Tutoring System domains. In support of past results, we found that the simple location heuristic provided a better goodness of fit to the learning curve standard, that is, it was better at performing error attribution than the simple temporal heuristic. One observation is that for tutors where the knowledge components can be determined by the interface location in which an action appears, using the simple location heuristic is likely to show better results than the simple temporal heuristic. It is possible that the simple temporal heuristic is better in situations where the different problem subgoals can be associated with a single location. However, our prior results with a physics data set indicated that even in such situations the simple location heuristic may be better. Further research should explore this issue.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要