Exploring Cognitive Difference in instructional outcomes using Text mining technology

Instructional outcome is an important part of instructional design theory which is hard to evaluate during education setting, especially, for attitude learning outcome. Traditionally, the most famous research method in attitude learning outcome is interview with open questionnaire. The drawback of interview method is the number of subjects is quite limited. This drawback can be improved by using new technology of text mining. Objective of this research is to find if there is any difference in cognitive learning for students from different backgrounds using text mining technology. The research is performed in three phases. First, documents of fifty four students concerning their thoughts after watching the film "Moon Child" were collected. High frequency terms were sifted from all terms to form the main thoughts of each and all documents. In the second phase, chaos theory was applied to divide these thoughts into two levels, basic concept level and attribute level. The main thoughts and each level of the main thoughts were visualized in network form and regarded as cognitive concept map. Finally, Average Path Length of cognitive concept maps were computed and analyzed. Experiment results indicate for students from different background, there is significant difference in their cognitive learning outcome. This result should be valuable for instructional design theory and can be served as an alternative in evaluating instructional outcome in the near future.