Models of individual trajectories in computer-assisted instruction for deaf students.

In this report we present a new approach to evaluation of curriculum. Many of us who have been engaged in curriculum reform efforts have been dissatisfied with the wait-and-see approach required when classical evaluation of a new curriculum is used. We have in mind evaluation by comparing pretests and posttests, with an analysis of posttest grade-placement distributions as a function of pretest distribution and exposure in some form to the new curriculum. In line with approaches used in other parts of science, it is natural to ask if a more predictive-control approach could be used and made an integral part of the curriculum to ensure greater benefits, especially for the disadvantaged or handicapped student. The approach discussed in this report is aimed precisely at this question. The strategy is to develop a theory of prediction for individual student progress This research was funded by Office of Education Grant No. OEG-0-70-4797(607), OE Project No. 142280.