Abstract:
Finding the most effective ways to teach and learn mathematics at university is
becoming a critical problem in higher education as rapid advances in related fields demand
graduates with advanced mathematics knowledge and skills. In this study, an instructional
innovation was designed, developed, implemented and evaluated in a large university
classroom setting involving novel assessment tasks – Knowledge-Organisers. The tasks
comprised prompts for students to generate examples/non-examples and construct concept
maps of the key concepts in the course. The initiative’s design was based on the current
understanding of human cognitive architecture and cognitive science research. Concept
mapping is a visual way of presenting a group of related abstract concepts and identifying
relationships between them by connecting related concepts with directed arrows that specify
relationships.
This research project had three broad aims: (1) to investigate ways to evaluate students’
concept mapping activity, (2) to examine the role of concept mapping as an assessment
component in comparison to traditional assessments, and (3) to evaluate the intervention
overall by reporting on its feasibility and students’ perceptions. Multiple linear regression was
used to identify the best rubric for assessing student concept mapping, with the Ratio method
proving to be the best way to assess student work accurately. The relationship between concept
mapping performance and two major outcome variables, academic achievement and selfefficacy, was investigated to compare concept mapping with other forms of assessment. The
results demonstrated that concept mapping performance is a significant predictor of both the
final exam score and the Emotional Regulation factor of assessment self-efficacy. Moreover,
using hierarchical multiple regression, it was shown that the addition of concept map scores significantly improves the models’ ability to predict the outcome variables above and beyond
that which was accounted for by other assessments. Lastly, by utilising a mixed-methods
approach and triangulation of the findings from qualitative and quantitative analyses, critical
aspects pertaining to the feasibility of implementation and evaluation of learners’ perceptions
were discerned. Students’ performance on concept mapping is positively correlated with their
perceptions of the novel tasks and the time spent to complete them. Qualitative analysis showed
that students’ perceptions are demonstrably insightful about the key mechanisms that
supposedly make the novel tasks beneficial to their learning. Based on the results of the data
analyses and their theoretical interpretations, implications for practice and future research are
formulated.