Background: STEM disciplines run a deficit of majors needed for national workforce needs, and students of color, first-generation college students, and women are systemically underrepresented. This deficit is often attributed to entrenched instructor-centered teaching approaches and noninclusive classroom environments, suggesting the need to explore the impact of alternative STEM pedagogies, including varieties of experiential learning, on these outcomes. Community-based learning, in which students help communities solve real-world problems, may be particularly valuable for these populations’ cultural and personal values. Purpose: We apply Astin's Input-Environment-Output model to a systematic review of the literature to investigate the impact of CBL in STEM interventions on historically underrepresented populations’ STEM major, minor, or career intent and their STEM-affiliated self-beliefs. Method: Following PRISMA reporting guidelines and JBI methodologies, we identified nine studies published between 1999 and 2024 and analyzed them for themes and strength of evidence. Findings: Only two compared the effect of CBL to that of other pedagogies, two compared outcomes across racial or ethnic groups, two compared outcomes for first versus continuing generation college students, and four compared outcomes across genders. Implications: Future studies should compare student outputs across demographic groups that are well-represented versus underrepresented in STEM (inputs) for CBL interventions compared to other STEM pedagogies (environments).