Predicting Students' Attitudes Toward Collaboration: Evidence From Structural Equation Model Trees and Forests.

Area of Science:

• Educational Psychology
• Data Mining in Education

Background:

• Previous research often examined factors influencing collaborative attitudes in isolation.
• A data-driven approach is needed to identify the most significant predictors for targeted interventions.

Purpose of the Study:

• To identify key predictors of students' collaborative attitudes using advanced data mining techniques.
• To develop a concise model for educators to focus on high-impact factors for intervention.

Main Methods:

• Applied structural equation model trees (SEM trees) and structural equation model forests (SEM forests) to the Program for International Student Assessment 2015 dataset.
• Utilized data mining to construct multigroup common factor models of collaborative attitudes.

Main Results:

• Demographic predictors included home educational resources, home possessions, mother's education, and gender.
• Achievement motivation and sense of belonging at school emerged as the most significant predictors.
• SEM trees identified a 5-group model based on demographics and a 6-group model including psychological factors.

Conclusions:

• Achievement motivation and sense of belonging are crucial for fostering collaborative attitudes.
• The findings provide educators with a data-driven model for cost-effective interventions.
• This study demonstrates the value of data mining for uncovering insights in large-scale educational datasets.