The overall objective of this project is to evaluate the feasibility of a new automated reading comprehension diagnostic system called ARCADE (Automated Reading Comprehension and Diagnostic Evaluation). The ARCADE system provides detailed assessments and recommendations of reading comprehension directly to classroom teachers on the basis of student free response data (e.g., answers to short essay questions). The ARCADE system generates a unique customized reading profile for each individual student and then using a specialized clustering algorithm, groups students with similar reading profiles. Ultimately, the system would provide suggestions to the classroom teacher regarding the optimal instructional method for each such student cluster generated by the ARCADE system. The specific objectives of this project are: (1) develop a very preliminary prototype of the ARCADE system, and (2) evaluate the mathematical algorithms used by the ARCADE system using human response data.
|
