Predicting Upper Secondary School Students’ Programming Self-efficacy in Tobacco Growing Areas of Southwest China Using Decision Tree Analysis

Abstract

Background: In the field of artificial intelligence, programming self-efficacy plays an indispensable role in the success of programming learning. However, how to predict the level of students’ programming self-efficacy has not been addressed. Objective: To predict the level of programming self-efficacy among upper secondary school students in tobacco growing areas of Southwest China, this study used survey data to develop a decision tree model. Methods: First, a total of 512 questionnaires were collected by using the Academic Achievement Test, Creative Style Scale, Programming Learning Attitude Questionnaire, Motivation Scale, Higher-order Thinking Preferences Scale, and Programming Self-efficacy Scale. Secondly, a decision tree model was constructed by SPSS modeler 18.0. Results: The results showed that academic achievement, creativity style, programming learning, motivation, and higher-order thinking propensity were highly predictive of programming self-efficacy. Conclusions: This is the first study in the direction of educational technology and it represents a novel approach to predicting programming self-efficacy among upper secondary school students. The experimental analysis demonstrate that the encouraging results prove the practical feasibility of the approach.