Implicit Concept Mapping: A Computerized Tool for Knowledge Assessment in Undergraduate Psychology

The study continued validating the computerised method of implicit concept mapping (Aidman & Egan, 1998), while extending it from assessing the map's structural properties to content-based expert evaluation. The on-line concept mapping task (Aidman & Egan, 1998) was modified to elicit similarity / contrast judgements for a set of basic personality concepts, in a group of 65 introductory psychology students. The resulting individual concept proximity matrices were scored for complexity and internal consistency, as well as individually factor- and cluster analysed. Hierarchical cluster tree and un-rotated factorial representations were generated for each individual map. Students were asked to interpret their own cluster trees and factor plots by naming the clusters and factor axes (a brief statement accompanying the name was allowed). Three independent experts (lecturers in the subject) rated the clarity and accuracy of these interpretations, as well as the soundness of cluster trees and the factorial representations themselves. These data were compared with the overall grade the students had obtained in the Introductory Psychology course immediately prior to participating in the study. Internal consistency and complexity of students' implicit concept maps produced only crude associations with their course achievement. The expert ratings of cluster- and factorial representations of the maps, especially expert ratings of students' own interpretations of their maps, showed a more refined association with course achievement, thus confirming that structural assessment of the implicit maps needs to be complemented by their content-based evaluation in order to achieve a more accurate estimate of the learner's level of expertise in the subject. unknown publishedVersion