Stimuleren van metacognitie bij het oplossen van fysicaproblemen: de impact van topic-specifieke reflectie

One of the main objectives of science and engineering education is to guide students towards becoming expert-like problem solvers, as is testified by, for example, the many hours students spend in problem-solving sessions. To support students during their journey, instructors should ask themselves which gaps there are to bridge. In other words, one should wonder: what sets experts apart from novices in their problem-solving practices? A first important and expected difference between novices and experts is the structure and extensiveness of their knowledge base. Clearly, the numerous content-related courses in a science or engineering programme aim at enlarging students' knowledge base. A second difference between expert and novice problem solvers is their engagement in metacognitive activities during problem solving. However, this aspect is rarely explicitly targeted in science and engineering education. The research project presented in this dissertation presents a tool, the Disciplinary Learning Companion (DLC), that aims to support students' use of metacognitive activities during problem solving. In particular, the focus of this dissertation is physics problem solving. The DLC elicits reflection on the problem-solving process by asking students topic-specific reflection questions that are tailored to the problem at hand after solving a physics problem. The effectiveness of the DLC to improve first-year university students' metacognitive abilities, and thereby their academic achievement, is investigated. In a first quantitative study, the DLC was implemented in an introductory physics course of the bioscience engineering programme at KU Leuven in order to investigate the tool's impact on students' metacognitive abilities and overall academic achievement. With this objective in mind, we tracked students' interaction with the DLC during the semester and considered students' final exam score for the course as measure of their academic achievement. Additionally, we wanted to understand whether any potential increase in academic achievement could be attributed to an improvement in conceptual understanding of the learning materials, an enhanced approach to problem solving, or a combination of both. To this end, we studied students' written solutions to one of the exam problems in more detail. Based on this written solution, we assessed their conceptual understanding of the problem and their strategic approach to solve it. As students' metacognitive abilities are shown to be linked to their problem-solving practices, the latter served as a proxy for metacognitive abilities. The results of this first study demonstrated a positive relation between students' interaction with the DLC and their academic achievement, conceptual understanding, and strategic approach to problem solving. This suggests that providing topic-specific reflection questions is an effective approach to enhance students' metacognitive abilities and academic achievement. As designing topic-specific reflection questions that are tailored to a specific problem takes more effort than providing generic questions that could be applied to a wide variety of science problems, the goal of the second study was to compare the DLC's impact with the impact of a more generic reflection tool, referred to as the Learning Companion (LC). In a quasi-experimental study, one group of students could interact with the DLC during the semester, while the other group of students could access the LC. Similar to the first study, we investigated the relations between students' interaction with either the LC or DLC and their academic achievement, conceptual understanding, and strategic approach to problem solving. First, the results showed that students are more inclined to interact with the DLC than with the LC. Second, we found a significant difference between the impact of the DLC and LC on students' academic achievement and conceptual understanding, in favour of the DLC. Regarding students' strategic approach to problem solving, a significant difference between the LC and DLC could not be confirmed. In conclusion, topic-specific reflection seems to be more effective than generic reflection to enhance students' overall academic achievement and conceptual understanding. We could, however, not conclude from our study that this is also the case for improving students' metacognitive abilities for problem solving. Finally, to better understand how students interact with the DLC, we conducted a qualitative study in which we studied which (meta)cognitive activities are triggered by the DLC's reflection questions. Twelve volunteering students were invited to participate each in four individual think-aloud interviews. In each interview, the participant first solved a physics problem and then reflected on the solving process by interacting with the DLC's reflection questions. The student was encouraged to verbalize all thought processes out loud during the entire interview. In this study, we particularly focused on students' use of evaluation strategies as component of their metacognitive repertoire. We considered which evaluation strategies were spontaneously used by the students during problem solving, which strategies were only used after being triggered by the DLC to do so, and which of these strategies were transferred and applied spontaneously in one of the later think-aloud interviews. Analysis of the think-aloud interviews revealed that students spontaneously use some evaluation strategies to assess their solution to a numeric problem, but do this only rarely for a symbolic problem. Furthermore, it appeared that the DLC can trigger students' use of new evaluation strategies and can stimulate transfer of these strategies to new problems. However, this was only observed in a few occasions. We hypothesise that students might not feel the need to evaluate their solution or try out new evaluation strategies because of various reasons related to the perceived added value of engaging in (further) evaluation. Moreover, the discrepancy in type and difficulty of the problems discussed in the subsequent think-aloud interviews might have hindered the transfer of evaluation strategies. To summarise, the results imply that topic-specific reflection on the evaluation phase in problem solving can be an effective approach to enhance students' acquaintance with evaluation strategies, provided that they are given an incentive to start evaluating their answer. The research presented in this dissertation contributes to the literature by using various methods (quantitative, quasi-experimental, and qualitative study) to investigate the effectiveness of a topic-specific reflection tool for improving students' metacognitive abilities and academic achievement in an ecological educational setting. The latter enhances the relevance of the results for instructors, in particular for science and engineering programmes at KU Leuven. We concluded that designing topic-specific reflection questions is an appropriate approach, and more effective than providing generic questions, for improving students' academic achievement and conceptual understanding. Although it could not be confirmed that topic-specific reflection should also be preferred to enhance students' metacognitive abilities, we showed that students are more engaged in topic-specific than generic reflection, which might also motivate instructors to opt for topic-specific rather than generic reflection.