Understanding students' problem-solving patterns: Evidence from an allotted response time in a PISA 2012 item

Area of Science:

• Educational Psychology
• Cognitive Science
• Learning Analytics

Background:

• Student-centered learning design requires understanding individual learning characteristics, especially in problem-solving.
• Current analysis of student learning primarily uses final answers, limiting insight into cognitive processes.
• Investigating thinking processes is crucial for developing effective adaptive learning strategies.

Purpose of the Study:

• To propose and validate an approach for analyzing students' cognitive behavior in problem-solving using response time data.
• To identify distinct problem-solving patterns among students based on their response times.
• To demonstrate the applicability of this method to both correct and incorrect answers.

Main Methods:

• Analysis of log data from the Programme for International Student Assessment (PISA) 2012 Creative Problem Solving (CPS) item CP038q1 (Ticket).
• Utilized response time data from 30,453 students across 42 countries.
• Employed clustering techniques to categorize students based on problem-solving patterns derived from response times.

Main Results:

• Four distinguishable clusters of problem-solving patterns were identified among students who answered correctly.
• The response time analysis approach demonstrated internal validity, successfully classifying students with incorrect answers into similar clusters.
• Response time data effectively differentiates various problem-solving strategies.

Conclusions:

• Response time analysis offers a valuable method for uncovering diverse cognitive behaviors in problem-solving.
• The identified patterns underscore the importance of tailored adaptive learning and feedback mechanisms for students.
• This approach provides deeper insights into student cognition beyond final answer evaluation.