A Radial Basis Function Neural Network Approach to Predict Preschool Teachers' Technology Acceptance Behavior.

Area of Science:

• Artificial Intelligence and Smart Computing
• Educational Technology
• Psychology

Background:

• The COVID-19 pandemic accelerated the adoption of online schooling, highlighting the need for effective technology integration in education.
• Quantitative approaches, particularly neural networks, offer efficient modeling tools for complex behavioral predictions without requiring intricate mathematical formulations.
• Understanding factors influencing technology adoption is crucial for successful implementation in educational settings.

Purpose of the Study:

• To predict preschool instructors' technology usage in classrooms using a radial basis function (RBF) neural network (NN) model.
• To identify key determinants of technology acceptance among preschool teachers based on an adapted Technology Acceptance Model (TAM).
• To explore the application of RBFNN for predicting psychological data in an interdisciplinary research context.

Main Methods:

• Utilized a radial basis function (RBF) neural network (NN) modeling technique.
• Adapted the Technology Acceptance Model (TAM) with eight dimensions: Perceived usefulness, Perceived ease of use, Perceived enjoyment, Intention to use, Actual use, Compatibility, Attitude, and Self-efficacy.
• Collected data from 182 preschool teachers via an online questionnaire.

Main Results:

• The RBFNN model achieved a sum of squares error of 37.5 in the training sample and 14.88 in the testing sample.
• The model demonstrated a significant prediction rate, with 63% of the classified data correctly assigned to the 'actual technology use' variable in the testing sample.
• Achieved a 43.3% incorrect prediction rate in the training sample and 37% in the testing sample.

Conclusions:

• The RBFNN model effectively predicts preschool teachers' technology usage, indicating the model's viability for analyzing psychological data.
• The study successfully applied RBFNN to psychological data, opening new avenues for interdisciplinary research in educational technology and AI.
• The findings support the adapted TAM, showing that perceived usefulness, ease of use, enjoyment, intention, compatibility, attitude, and self-efficacy significantly predict actual technology use.