The effect of the teams games tournament (TGT) cooperative learning model assisted by mathcards on students’ mathematical problem-solving ability

This study examined whether a Mathcard-assisted Teams Games Tournament (TGT) model develops students’ mathematical problem-solving ability more effectively than direct instruction, and which components of problem solving it reaches. A quasi-experimental pretest–posttest control group design was used with Grade X students at SMK Yayasan Pendidikan 17 Magelang. Cluster random sampling by lottery selected two intact classes totaling 27 students: 14 in Grade X AKL 1 as the experimental group and 13 in Grade X AKL 2 as the control group. The intervention ran for four meetings of 2 × 45 minutes on sequences and series. Because intact classes were used, initial equivalence was tested through normality, homogeneity and independent samples t-tests on pretest scores. Data were collected through interviews, observations and a mathematical problem-solving test scored on Polya’s four phases. Classical mastery was tested with a proportion test (individual score ≥ 70; classical criterion ≥ 80% of students) and group differences with an independent samples t-test at the 5% level. The experimental class reached 100% classical mastery, whereas the control class reached only 53.85%. The experimental group scored significantly higher (t = 2.144 > 1.708, p < 0.05), with a mean difference of 9.78 points and a large effect size (Cohen’s d = 0.83). Disaggregated by phase, the groups differed only marginally in understanding the problem (98% versus 97%) and devising a plan (100% versus 99%), differed substantially in carrying out the plan (95% versus 79%), and both remained weak in looking back (48% versus 36%). Read through Polya’s phases together with Schoenfeld’s framework, the Mathcard-assisted TGT environment was associated particularly with stronger enactment of solution strategies, while producing minimal differences in problem interpretation and planning and leaving substantial difficulty in reflective evaluation. Cooperative, game-based structures alone appear insufficient for developing metacognitive problem-solving practices.