Abstract

Over the past decade there has been a radical shift in conceptualizations of learning, from viewing the learner as a passive recipient of information, to viewing the learner as an active processor of information. As a result, recent theories have attempted to identify personality characteristics and cognitive abilities that mediate learning outcome. However, there is little empirical support for these theories. The focus of the present study is to empirically test Biggs's (1985) theoretical model of the relationship between specific predictor and criterion variables. It examines the interrelationship among the predictor variables of Locus of Control, Approach to Learning, and Metacognition, and their relationship (singly and in combination) with the quality of learning outcome (structural complexity and depth of cognitive performance). Rotter's Locus of Control Scale (Rotter, 1966), Biggs's Study Process Questionnaire (SPQ) (Biggs, 1988b), and Tobias's Assessment of Metacognition (Tobias, Everson, Hartman and Gourgey, 1991) were administered to 177 college students from a state university to measure the predictor variables. Essay answers to open ended questions based on three text passages were evaluated using the Structure of the Observed Learning Outcome (SOLO) taxonomy (Biggs and Collis, 1982) to determine structural complexity. Finally a 50 item multiple choice test consisting of questions designed to measure the upper four levels of Bloom's taxonomy (application, analysis, synthesis and evaluation) was administered to determine depth of cognitive processing. The results of Pearson Product-Moment correlational analyses indicate a positive correlation between Internal Locus of Control and Deep (r =.23, p $<$ 0.01) and Achieving (r =.22, p $<$ 0.01) Approaches to learning, and Metacognition (r =.17, p $<$ 0.05). A negative correlation between Surface Approach and Metacognition (r = $-$.21, p $<$ 0.01) is indicated. Stepwise multiple regression analyses reveal that Metacognition is the only variable which predicts learning outcome (p $<$ 0.01). No relationship between any learning approach and learning outcome is supported. The present study also examines the psychometric properties of the SPQ and provides support for the factor structure identifying the three Approaches to learning, but questions the strategies/motives subscales.