Analysis of Questionnaire about Self-Evaluate Achievement using Data Mining Technique for Improving Lectures

Kanazawa Institute of Technology requires all the students to take engineering design classes in their freshman years. During these classes the students are asked to rate their own engineering design abilities to self-assess their acquisition of engineering and social skills. The authors employed clustering and text-mining techniques to analyze survey data collected over time from more than 1,100 students in order to gain insight for instructional improvement. The students were found to be clustered into nine groups with meaningful characteristics. Text mining techniques were employed on the results of the free response section to extract 41 keywords and determine their frequency of appearance. These results will lead to finding the common motivations or hindrances that underlie high and low achieving students.