Electronic waste, including in Indonesia, has become the world's most significant concern. The amount of electronic waste keeps increasing due to massive electronic device usage. Since refurbished smartphones are not popular among Indonesians, giving them away might be an alternative way to prolong their lifetime usage. Therefore, this research aims to predict Indonesian citizens' willingness to donate their used smartphones as an alternative to reduce electronic waste and support better education for students in rural areas. Three machine learning techniques, i.e., decision trees, random forests, and neural networks, are explored in this study. The data was collected from a survey with 416 respondents. According to prediction accuracy, the neural network model outperforms the other two. The result also shows that respondents with a solid spiritual life, serious social concerns, and those who have been taught the value of generosity from a young age are more likely to donate their used smartphones.