Abstract

Pre-service mathematics teachers need opportunities to interpret evidence, explain statistical reasoning, and communicate conclusions, providing a rationale for examining structured cooperative approaches in undergraduate statistics. This study compared statistics achievement in two intact classes after one class received Learning Together (LT) with digital and artificial intelligence-supported resources and the other received conventional instruction. A pretest-posttest comparison involved 54 second-year pre-service mathematics teachers (27 per class) at a Thai Rajabhat University. The two classes were allocated to conditions by drawing lots. Both classes studied the same content with the same instructor and completed the same 30-item test. ANCOVA compared posttest scores while adjusting for pretest achievement. The LT-based class had a higher adjusted posttest mean (20.22, 95% CI [19.11, 21.33]) than the comparison class (17.37, 95% CI [16.27, 18.48]), F(1, 51) = 13.34, p < .001, partial η² = .207; the adjusted difference was 2.85 points (95% CI [1.28, 4.41]). The integrated condition was associated with higher achievement; because each condition was represented by one class, the result is exploratory and cannot separate the contributions of LT, digital tools, AI-supported resources, and class membership. The findings suggest that university teacher-preparation courses can combine structured cooperative tasks with guided verification of AI-supported materials as a practical design for undergraduate statistics.

This work is licensed under a Creative Commons Attribution 4.0 License.