Enhancing conflict resolution skills through artificial intelligence-based problem-based learning in civic education at Indonesian secondary schools

Traditional civic education in Indonesia does not sufficiently develop practical conflict resolution skills among adolescents. This gap contributes to ongoing social tensions in diverse communities, where democratic skills are essential for social cohesion. This quasi-experimental study examined whether Artificial Intelligence–enhanced Problem-Based Learning (AI-PBL) could improve conflict resolution skills among 90 eleventh-grade students in three urban schools in Yogyakarta, Indonesia. The intervention used a natural language processing system to present culturally relevant conflict scenarios, provide real-time adaptive feedback, and create personalized learning pathways across six structured sessions. Conflict resolution skills were measured in three areas—empathy, peaceful negotiation, and ethical reasoning—using a 24-item instrument with good reliability (Cronbach’s α = .86). The results showed significant improvements in empathy (Cohen’s d = 1.67), negotiation (d = 1.72), and ethical reasoning (d = 1.63), all with p < .001. Bayesian analysis gave strong evidence of effectiveness (BF₁₀ > 1000), and regression analysis showed the largest benefits for students who initially performed at lower levels. However, because this was a single-group design, stronger causal conclusions require future randomized controlled trials.