Individual and Cooperative Group Learning with User-Controlled and Program-Controlled Mathematics Tutors

摘要

We used a 2x2 design to compare the performance of remedial subjects learning tosolve mathematics word problems. Subjects worked either alone or as a member of a dyadic cooperative group, and received up to 12 hours of instruction and practice, using either a 'problem-solving environment' over which the user exercised considerable control or a directive, sometimes intrusive tutor. Individual subjects who worked with the problem-solving environment showed the largest average improvement between pretest and posttest scores. Group members showed moderate improvements after working with the directive tutor. There were no significant changes between pretest and posttest for either group members who worked with the problem-solving environment or for individuals who worked with the directive tutor. (AN).