Predicting Physical Activity of Young Adults Based on Psychological Need Satisfaction in Exercise Using Explainable Decision Tree Model

Every individual is aware of the benefits of exercise and regular physical activity (PA) for maintaining a healthy lifestyle and a positive self-image. It is no longer a secret that engaging in regular physical activity boosts the immune system and reduces many health risks associated with a sedentary lifestyle. However, the challenge remains in the will and commitment to engage and sustain regular PA. In an effort to predict the PA of young adults and ascertain the psychological need satisfaction that drives their motivation to engage and sustain regular PA, the current study was undertaken. A total of 422 undergraduate students participated in the study and completed the Malay version of the psychological need satisfaction during exercise questionnaire. A decision tree model was developed to achieve the objective of the study with PA as the dependent variable and the psychological need variables that comprised autonomy, efficacy, competence, barriers, family, friends, relatedness, availability, benefits, and quality as the independent variables. The model accounted for an overall accuracy of 99% in predicting the PA of the participants. SHapley Additive exPlanations (SHAP) demonstrated that psychological need variables consisting of efficacy, competence, family, friends, relatedness, availability, benefits, and quality had positive effects on PA, whereas autonomy and barriers had negative effects on PA.