The effect of metacognitive training on the mathematical word problem solving of Singapore 11-12 year olds in a computer environment

This study aims to establish the extent to which metacognitive training plays a part in Singapore primary students' word problem solving in a computer environment. The study involved 142 Singapore 11 to 12-year-old students from two primary schools. The study adopts a two-phase design, combining a quasi-experimental design and a case study design. For the quasi-experimental design, analysis of students' mathematical achievement test data is used to investigate the relationship between metacognitive training, students' level of mathematical achievement and their mathematical word problem solving performance. For the case study design, analysis of the think aloud protocol data during word problem solving of eight pairs of students is used to explore the role of metacognition in mathematical word problem solving in a computer environment. In addition, student questionnaire and teacher interview data provide descriptive accounts of students' metacognitive knowledge during mathematical word problem solving. The findings from the analysis of mathematical achievement test and think aloud protocol data reveal that metacognitive training results in improvement in mathematical word problem solving performance, and that lower achievers appear to show the full benefit from metacognitive training only after a period of time. The findings of the think aloud protocol data also reveal that i) generating metacognitive behaviours, and knowing when and how to use them during word problem solving are important determinants for successful word problem solving, and ii) students have distinctive progressions of word problem solving activity which can be represented by five types of cognitive-metacognitive word problem solving models. These progressions of word problem solving activity seem to relate to students' success in word problem solving. It is also proposed that there is a relationship between affect, students' ability to develop metacognitive awareness, and word problem solving. In addition, effective pair collaboration is influenced by students' mathematical beliefs, and how students are paired according to their metacognitive knowledge. The educational and pedagogical implications of these findings are discussed, particularly in relation to the Singaporean context.