Promising outcomes are not the same as evidence of program impact. A first-phase mixed-methods evaluation of the AIM (Accomplish, Innovate, Motivate) Scholars project, presented last year, found higher GPA, retention, and graduation among scholars and identified mentoring, advising, peer networks, and career guidance as valued supports. This second-phase study asks a harder question: Do those advantages survive a counterfactual test? This question addresses two gaps in the S-STEM literature. National evidence about scholar experiences is dominated by four-year institutions and often relies on self-reported perceptions. In addition, evaluations commonly describe scholarship-plus-support programs as bundles, leaving limited evidence about whether these institutional support architectures change the academic processes that precede persistence. Rather than attributing success to individual services, this study stress-tests the combined AIM model and examines academic momentum as a potential pathway. Institutional longitudinal records from Fall 2022 through Fall 2025 were used to compare 98 AIM Scholars with 98 matched nonparticipants drawn from 549 eligible low-income STEM students. Propensity score matching constructed the comparison group, and stabilized inverse probability weighting achieved strong covariate balance across the longitudinal panel. AIM participation was associated with a 0.33-point increase in term GPA, a 7.7-percentage-point increase in credits completed relative to credits attempted, and more than twice the odds of term-to-term persistence. Weighted mediation analysis further indicated that improved credit completion partially transmitted the persistence advantage. The findings strengthen the evidentiary basis for coordinated institutional support at two-year colleges while identifying the next challenge: measuring service utilization, dosage, and implementation fidelity to determine which components, or combinations, produce the greatest return.