Performance models of American indian students on computer-assisted instruction in elementary mathematics

This investigation applied experimentally the use of predictive-control models integrated into computer-assisted instruction (CAI) as discussed earlier by Suppes, Fletcher, and Zanotti (1973). Many of those who are engaged in curriculum reform efforts have been dissatisfied with classical evaluations that simply compare the pre- and post-treatment achievement of experimental and control groups. It is natural to seek a more predictivecontrol approach that can be used as an integral part of the curriculum in order to ensure greater benefits, especially for students who are educationally disadvantaged or handicapped and for whom global performance models derived from standard populations are inappropriate. Such an approach was discussed by Suppes, Fletcher, and Zanotti (1973), who developed a theory by which the amount of time a student spends on a curriculum is a function of his progress, and his achievements in given course objectives, which are individually set for each student, are expressed as post-treatment grade placement (GP). Using the approach described by Suppes et al. (1973), we were able to achieve precise individualization of instruction both in the amount of instruction for each student and in the goal set for him. Further, the approach separates global features of the curriculum described by a simple differential equation from parameters that are characteristic of individual students.