An Evaluation of the Factor Structure and Internal Consistency of the ‘Conceptions of Learning’ and ‘Preferences for Teaching’ Measures in American Occupational Therapy Students

• Main Authors: Tore Bonsaksen, Adele Breen-Franklin
• Format: Article
• Language: English
• Published: Eastern Kentucky University 2020-01-01
• Series: Journal of Occupational Therapy Education
• Subjects: factor analysis; higher education; occupational therapy; psychometrics; scale reliability; students
• Online Access: https://doi.org/10.26681/jote.2020.040104
• ISSN: 2573-1378; 2573-1378

When planning to use measurement scales in new samples and contexts, examining the scales’ psychometric properties is an important initial step. This study examined the factor structure and internal consistency of two measures that are part of the <em>Approaches and Study Skills Inventory for Students</em> (ASSIST) – the <em>Conceptions of learning</em> and <em>Preferences for teaching and courses</em> – in a sample of American occupational therapy students. The students (<em>n</em> = 115) completed the measures and provided basic sociodemographic information. Scale structure was examined with Principal Components Analysis (PCA), while consistency between scale items was assessed with mean inter-item correlations. For the <em>Conceptions of learning</em> measure, one item was removed due to cross-loading between factors. The subsequent analysis revealed two factors, representing deep and surface conceptions of learning, on which the items – with one exception – loaded in line with theory. For the <em>Preferences for teaching and courses</em> measure, two factors were found, representing preferences denoted in theory as supporting understanding and transmitting information, respectively. The items showed good fit with the two theoretically proposed factors. The scales’ mean inter-item correlations were satisfactory, ranging 0.27-0.36. One item on the <em>Conceptions of learning</em> measure appears to be problematic due to cross-loading, and another may be interpreted in a different way than originally proposed. After removing the problematic item, all scales showed satisfactory psychometric properties for assessing conceptions of learning and preferences for teaching.