Important Factors Discriminating Between Problem-Solving Experts and Novices: A Data Mining Approach
Digital problem-solving competence is widely recognized as one of the core skills of the 21st century. A number of important factors influence this competence; some are task-specific pertaining to the problem-solving processes while others are non-task-specific related to knowledge, skills, attitudes and beliefs of the problem solvers, as well as the student learning environment. This study sought to determine important factors that classify student problem-solver as “high-performing expert” versus “low-performing novice”, using computer-generated log files of an exemplary digital problem task assessed in Organization for Economic Co-operation and Development (OECD)’s Programme for International Student Assessment (PISA) 2012 Study. The participants comprise 11,599 fifteen-year-old students from 42 economies. Apart from multilevel logistic regression of problem-solving process and student questionnaire data, the secondary data analysis employed was a data-mining approach involving classification and regression trees. Five important factors were identified that are key to the discrimination of the “expert vs novice” dichotomy.
JIN, Song Li; Cheung, Kwok Cheung; and Sit, Pou Seong
"Important Factors Discriminating Between Problem-Solving Experts and Novices: A Data Mining Approach,"
Chinese/English Journal of Educational Measurement and Evaluation | 教育测量与评估双语季刊: Vol. 3:
2, Article 3.
Available at: https://www.ce-jeme.org/journal/vol3/iss2/3