Pre-service student teachers' metacognition in an online learning community: An epistemic network analysis

Abstract

Pre-service student teachers (PSSTs) should improve their metacognition in order to support their long-term development in their future complex teaching and learning lives. Although previous studies have explored the role of metacognition in learners' learning effectiveness, the interrelationships pattern of metacognition elements in the process of metacognition and the metacognition characteristics of different genders and learning performance groups still remain unclear. This study clarified the intrinsic structure of metacognition, and gender and learning performance differences among PSSTs in an online learning community. We collected 33 PSSTs' self-reported reflection data on six learning topics. A total of 1,188 reflection paragraphs were coded and 2,335 metacognitive utterances were presented based on the metacognition model. Subsequently, this study identified metacognitive components among different groups of PSSTs by comparing epistemic networks of self-reported reflections via epistemic network analysis. Results show that metacognitive components of PSSTs in an online learning community were mainly metacognitive experiences (36.32%) and actions (or strategies) (28.01%). Female PSSTs used metacognitive knowledge (21.4%) and goals (or tasks) (15.8%) more frequently, and the relationship between their metacognitive knowledge and goals (or tasks) was closer compared with male PSSTs. Specifically, the ENA results showed that the associations between metacognitive knowledge and goals (or tasks) (connection coefficient: 0.24) of females were stronger than those of males. In addition, the high learning performance group used metacognitive knowledge (18.7%) and goals (or tasks) (15.9%) more frequently, and they also had a closer connection between metacognitive knowledge and goals (or tasks) compared with the high learning performance group. Specifically, the ENA results showed that compared with the low learning performance PSSTs, high learning performance PSSTs' metacognitive knowledge and goals (or tasks) had a higher connection coefficient (0.32). The findings provide broad and important implications for analyzing metacognition of PSSTs in the network environment, and provide new evidence for metacognition differences between different groups (gender and learning performance).