In recent decades, informal project-based learning (PjBL) for science, technology, engineering, and mathematics (STEM) education has gained immense significance. Though some meta-analytical studies have reported the effectiveness of PjBL models, none of them has been explicit for informal PjBL models executed for school students in STEM education. Therefore, this meta-analytical study sought to investigate the effect of the informal PjBL model on students' learning gains when compared to the traditional classroom setting. The study also intends to explore the effect of potential moderator variables that might impact the effectiveness of the model. For this, a systematic review aligning to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology has been employed, by incorporating 26 peer-reviewed empirical studies, encompassing data from 3202 students from eight countries. The meta-analytical results revealed overall moderate effectiveness (d = 0.248, p < 0.001), indicating that informal PjBL has moderately significant effects on students' gains when compared to the traditional classroom setting. Correspondingly the moderator analysis revealed that the overall effectiveness of informal PjBL was influenced by the teaching model, assessment method, students' group size, and course duration. However, the participants' educational level and study location (country) did not emerge as being significant. This study contributes to the bulk of literature that can help academicians to design, develop and implement effective STEM-based informal PjBL models, with special consideration for influential variables (teaching method, assessment method, grade, location, course duration, group size, and subject area).