Facebook/Meta usage in higher education: A deep learning-based dual-stage SEM-ANN analysis

Area of Science:

• Educational Technology
• Social Media in Education
• Human-Computer Interaction

Background:

• Social media platforms like Facebook/Meta are increasingly explored for academic purposes.
• Understanding student adoption factors is crucial for effective virtual classroom integration.
• Existing research often lacks advanced predictive modeling for social media in education.

Purpose of the Study:

• To investigate and predict key factors influencing students' behavioral intentions towards using Facebook/Meta as a virtual classroom.
• To analyze the adoption level, purpose, and educational usage patterns of Facebook/Meta.
• To identify significant predictors of Facebook/Meta adoption in higher education.

Main Methods:

• A novel two-phase analysis approach was employed.
• Structural Equation Modeling (SEM) was used to identify initial predictors.
• Deep learning, specifically Artificial Neural Network (ANN) analysis, was utilized for advanced prediction.
• The study focused on higher education students' academic usage of Facebook/Meta.

Main Results:

• Perceived task-technology fit emerged as the most significant positive predictor of Facebook/Meta usage.
• Facilitating conditions, collaboration, subjective norms, and perceived ease of use demonstrated strong influences.
• The deep learning approach successfully identified key drivers of adoption.
• The findings highlight the importance of technological and social factors in platform integration.

Conclusions:

• The study provides valuable insights for enhancing social media tool integration in teaching and learning.
• Theoretical implications relate to technology adoption models in educational contexts.
• Practical implications include strategies for optimizing Facebook/Meta as a virtual learning environment.
• Future research can build upon the predictive modeling techniques used.