This study aims to examine the effectiveness of online problem-based learning (PBL) tasks on problem-solving ability (PSA) and programming skills of secondary school students. A quantitative research design of a quasi-experimental research was adopted in this study. A purposive non-random sampling method was used to select two schools. Then, a clustered random sampling technique was carried out to select four classes, consisting of 120 students. The classes were divided equally into online PBL and non-online PBL. A problem-solving inventory questionnaire was used to measure PSA before and after the treatment, and pre-post performance tests were used to assess students' performance. The students underwent 8 weeks of learning tasks by using online PBL via Facebook groups. The findings from this study proved that online PBL improved students' PSA, and their higher-order thinking skills, and helped them to perform better in learning programming by using Facebook for interaction and participation in the online discussion with a facilitator and peers.