Threshold concepts and metalearning capacity

This study operationalises the empowering concept of metalearning in the specific context of engagement with a threshold concept. An experience of metalearning was constituted in two parts. First students’ awareness of themselves as learners is prompted by, and focuses on, a learning profile that is generated online through the completion of the Reflections on Learning Inventory (RoLI). Second, students are given an opportunity to interpret their respective profiles and write a short and undirected reflective account of their interpretation.The second part of the experience focuses not only on students’ awareness but also on their capacity to control their future learning on the basis of their heightened awareness. An initial metalearning experience was provided early in a microeconomics module encouraging students to reflect on their learning in the context of their prior learning of microeconomics. A second metalearning experience was provided later in the module when the RoLI response context was shifted specifically to the learning of the threshold concept in question. This metalearning experience intervention yielded matched sets of quantitative data (the RoLI responses used to generate the learning profiles) and qualitative data (the reflective accounts based on the interpretation of the learning profiles). Data analysed here emanate from a research project involving three UK universities, two of which are represented in the present study with one foregrounded in terms of a relatively large sample of c. 300 participating students. The findings indicate that a metalearning experience can be successfully constituted and encapsulated within the learning of a threshold concept for a majority of students.There is however variation in the success (or not) of the metalearning experience, the detail of which reveals much about the dynamics of changed or changing metalearning capacity in relation to the threshold concept considered. Threshold concepts and metalearning capacity 135 It is acknowledged at the outset that these datasets represent students who have persisted in their studies up to the point of completed data gathering, and the findings reported here are strictly within this constraint. Resultant analyses of these datasets may therefore be biased in that they do not reflect all the students who entered the module(s) in question but dropped out before data gathering commenced, nor those who, for whatever reason, either failed to persist in contributing data, or failed to complete the module(s) in question. The remainder of the present study is presented in four sections.The first provides a selected commentary on the quantitative analyses of the largest dataset from the first university.The second further illuminates this quantitative analysis in terms of a summary of the findings from the corresponding qualitative analyses, and also findings yielded by a smaller dataset from the second university for which outcomes measures were also available. In the third section a summary of students’ reflections on their metalearning experience is presented, and we finally conclude with a discussion and indications for future work. The quantitative analyses In order to generate a self-reported learning profile as part of a metalearning experience each student completed the Reflections on Learning Inventory (RoLI) via an online portal (www.RoLIsps.com) especially designed for this purpose.The psychometric development of the RoLI is summarised in Meyer (2004) and its domain, in the generic version used in the present study, is defined in terms of 16 observables, each represented by a subscale of five items scored in terms of a mixed metric that includes a Likert-type response format. It is the subscale scores that are the discrete conceptual components of the learning profile and that constitute the data for the quantitative analyses. Location differences The two matched sets of response data (n=354 individual students) comprise 16 derived subscale scores representing the16 observables in the RoLI domain.The first set (referred to as the A dataset) comprises responses from students within an economics module just prior to the introduction of the learning segment containing the procedural threshold concept of elasticity. In completing the RoLI students were asked to respond in terms of their experiences of learning economics within the module up to that point.The second set of responses (referred to as the B dataset) was obtained after the concept of ‘elasticity’ and its application had been dealt with, and the same students were asked to respond specifically in terms of their learning engagement with this concept. International Review of Economics Education 134 specific threshold concept in economics.The two largest datasets (representing two of these universities) respectively represent a metalearning experience involving a specific threshold concept, and these data form the basis of the present study.The first threshold concept, that of the ISLM model, is referred to as a discipline threshold concept by Davies and Mangan (2008) while the second, that of elasticity, is referred to as a procedural threshold concept.The distinction being made here by Davies and Mangan is essentially that the ‘ISLM model’ depicts the interaction between markets, while ‘elasticity’ (and other threshold concepts) enable such modelling in procedural terms in the ‘ISLM model’ and more generally. Students in each university were invited, on two occasions, to (a) generate a self-reported and contextualised learning profile of themselves (via completion of an online learning inventory), (b) interpret this profile with the aid of a non-evaluative guide written for this purpose and, (c) write a short (500 word) reflective essay on their profile as self-interpreted. Sandwiched in between these two ‘A: first’ and ‘B: second’ metalearning experiences was the learning engagement of the designated threshold concept.The first metalearning experience was contextualised in terms of the learning in general of the microeconomics content up to that point, and the second experience was contextualised in terms of the ensuing learning engagement of a designated threshold concept. It is emphasised that, although the focus here is on economics threshold concepts, the architecture of the metalearning experience as described here is transferable to threshold concepts in other disciplines.