This study investigated the effectiveness of a STEM-integrated Project-Based Learning (STEM-PjBL) algebra e-module in improving seventh-grade students’ algebra learning outcomes and examined students’ self-confidence following its implementation. The study was motivated by students’ difficulties in understanding algebraic concepts and their limited confidence in solving mathematical problems. A quasi-experimental design employing a nonequivalent control group was used, involving 44 students purposively selected from a population of 184 seventh-grade students at Al-Rifaie Gondanglegi Junior High School. The experimental group (n = 22) learned using the STEM-PjBL algebra e-module, whereas the control group (n = 22) used a conventional algebra e-module. The research instruments included an algebra achievement test, expert validation sheets, and a self-confidence questionnaire. Validation results indicated that the e-module demonstrated satisfactory content validity, with Aiken’s V coefficients of 0.78 for material validation and 0.83 for media validation, resulting in an overall coefficient of 0.80. The algebra achievement test showed excellent reliability (Cronbach’s α = 0.917). Following normality and homogeneity testing, learning outcome data were analyzed using the Mann–Whitney U test because the assumption of normality was not met. The results revealed a statistically significant difference in algebra learning outcomes between the experimental and control groups (U = 133.5, Z = −2.576, p = .010), with a moderate effect size (r = 0.39). Descriptive findings further indicated that most students in the experimental group demonstrated high to very high levels of self-confidence. These findings suggest that the STEM-PjBL algebra e-module may enhance students’ algebra learning outcomes and support the development of self-confidence in mathematics learning, although the results should be interpreted cautiously given the limited sample and the descriptive assessment of self-confidence.