Effects of teaching problem-solving through cooperative learning methods on student mathematics achievement, attitudes toward mathematics, mathematics self-efficacy, and metacognition

Date of Completion

January 1997


Education, Mathematics|Education, Educational Psychology




Recently, an emphasis has been placed on teaching mathematics in cooperative learning settings and through a problem solving approach (NCTM, 1989). Although numerous research studies have been conducted on the effects of cooperative learning on mathematics achievement, attitudes, and self-efficacy, no study was found that addresses the use of cooperative learning while teaching mathematical problem solving and heuristic strategies and its effect on metacognition. The purpose of this study was to assess seventh and eighth grade male and female students' metacognition, self-efficacy, attitudes toward mathematics, and achievement after participating in a two-week course on problem solving. Problem solving is an important area of inquiry, as previous research indicates that females demonstrate lower performance in solving non-routine problems.^ This experimental study used a pretest-posttest control-group design in which students were randomly assigned to one of two experimental groups or a control group during a summer enrichment program offered to talented students in a southern state. All three groups received instruction in probability and statistics through a mathematical problem solving approach using heuristic strategies. The two experimental groups were taught through cooperative learning methods. In the first experimental group, students worked in mixed-gender groups and, in the second one, in single-gender groups. The control group was taught using whole-group instruction in which competition and individual work was stressed. Students' achievement in probability and statistics, self-efficacy, and attitudes toward mathematics was assessed at the beginning and end of instruction. Data from the assessment of these three variables were analyzed using a multiple analysis of covariance and a follow-up discriminant function analysis. Students' metacognitive episodes were assessed using content analysis procedures.^ Although the literature suggests that cooperative learning settings, particularly single-gender groups, are more beneficial for females, no statistical differences in achievement or self-efficacy were found among the groups. Statistically significant differences in attitudes toward mathematics were found favoring students in the whole group instruction, competitive setting. Also, the highest achieving male and female students exhibited more metacognitive episodes in the competitive setting than students in the other groups. Lower achieving male and female students, however, exhibited fewer metacognitive episodes in this type of setting. ^