Developing a Dual Approach, Convergent, Self-Validating and Self-Correcting Scheme for the Assessment of Affective Domain Attributes

Abstract

Since affective attributes, such as motivation, curiosity, honesty, etc., are vital for effective graduates; hence gauging these attributes effectively at various stages of a degree program can allow an institute to perform continuous quality improvement (CQI) of its learning environment and teaching activities. Purely self-report-based assessment provides a single approach of measurement, which may be prone to issues e.g. social desirability bias. The current effort attempts to develop a novel survey framework comprising two approaches, which are self-reporting and scenario-based questioning, respectively based on iceberg model of systems thinking and Process-Person-Context-Time (PPCT) model of Bronfenbrenner’s bioecological systems theory. For a sound implementation of this framework, both approaches largely need to converge (for individual students, within an allowed tolerance) to a single measurement of the latent affective attribute, thus lowering the probability of a bias-contaminated final evaluation. An initial test of the proposed framework was performed via developing a pilot survey instrument administered to a small number (N=22) of undergraduate students of electrical engineering degree program. Subsequent factor analysis of resultant data indicated an overall convergence for the dual approaches; also Bland-Altman plots (and relevant numbers) were used to detect outliers in the data, which were then resolved. In general, the dual approach scheme presented herein may hold the potential to enhance confidence in the affective attribute evaluation of a student cohort, hence subsequent CQI measures can be more helpful for the graduates as well as the society at large.