2009 IEEE International Conference on
Systems, Man, and Cybernetics |
![]() |
Abstract
As Web-based Instruction (WBI) becomes increasingly popular, designers are faced with the challenge of identifying the varying preferences of learners and accommodating them in WBI programs. Cognitive style has been shown to have a significant effect on users' preferences for WBI. In particular, Witkin's Field Dependent/Independent has been widely studied in this area. It has been suggested that this cognitive style has conceptual links with the rarely studied Pask's Holist/Serialist dimension. Therefore, this study investigates the relationship between these two cognitive style dimensions by using a data mining approach that integrates feature selection and decision trees. The results show that in general Field Dependent users and Holists show the same preferences for WBI, as do Field Independent users and Serialists. However, this study also highlights the similarities in preference between Field Dependent users and Serialists, and Field Independent users and Holists. The findings of this study contribute towards the understanding of Field Dependent/Independent and Holist/Serialist users' preferences. Additionally, a novel data mining methodology is proposed.