Summary: Intelligent Tutoring for high-stakes
Ivon Arroyo, Carole Beal, Ari Bergman, MikeLindenmuth, David Marshall, Beverly P. Woolf
Universityof Massachusetts, Amherst
Abstract. We describe our on-going work in the creation of a web-based Intelligent
Tutoring System (ITS) for the Math section of the Scholastic Aptitude Test (SAT).
Wayang Oupost has several distinctive features: heavy web-based multimedia on the
client; decisions about problem and help selection made on a remote web server;
modelling of students' cognitive abilities; alternative teaching strategies. Difficulties in
developing ITS for high-stakes achievement tests are analyzed.
High stakes achievement tests have become increasingly relevant in the past years in the
United States, as a student's performance on them can have a significant impact on students'
access to future educational opportunities. At the same time, concern is growing that the
use of such tests simply exacerbates existing group differences, and puts female students
and those from traditionally underrepresented minority groups at a disadvantage. New
approaches are required to help all students perform to the best of their ability on high
stakes tests. This paper describes our ongoing work in creating "Wayang Outpost", an
Intelligent Tutoring System to prepare students for the mathematics section of the
Scholastic Aptitude Test (SAT). The intention is that Wayang does not only tutor every