Background: Reading failure is far higher among disadvantaged students than among other students (National Center for Education Statistics, 2019). Many attempts to solve this have focused on schools, yet non-school factors have a substantial impact on learning (Clabaugh, 2008). One of these is uncorrected vision. Approximately 20-25% of school-age children need eyeglasses (Ferebee, 2004), yet disadvantaged and minority children are far less likely than others to obtain and wear eyeglasses (Ganz et al., 2007; Heslin et al., 2006; Qiu et al., 2014; Ruderman, 2016; Zhang, Cotch, et al., 2012; Zhang, Elliott, et al., 2012). One approach to solving the vision care access gap is school-based vision care. Purpose: Unmet vision needs are particularly important to address because of the association between vision and learning. While research supports this relationship (Dudovitz et al., 2016, 2020; Esteso et al., 2007; Hark et al., 2020; Pavithra et al., 2014), there is a lack of rigorous evidence supporting the connection between school-based programs and achievement. This experiment is one of the first in the U.S. to study the academic impact of a school-based vision intervention in an entire urban district. Setting: The analytic sample included 120 schools located in a large urban mid-Atlantic district. The included schools were mostly elementary (37%) and elementary/middle schools (56%), with a few middle schools (7%). Schools enrolled predominantly Black students (83%); most students qualified as low-income (89%). These schools included few English Language learners (ELL) (4%) and moderate numbers of students qualifying for special education (14%). Participants: The analytic sample included 2,304 students in 3rd-7th grades at the time of intervention. Students were mostly Black (78%), while 17% were Latinx and 19% were White. There were 18% of students in special education and 9% were ELL. Students were included in the analytic sample if they failed vision screenings, were prescribed glasses at the eye exam, had both a baseline and post-intervention test score, and opted into the research study. Intervention: The intervention included a vision screening, an eye examination, and eyeglasses, if needed. Vision screenings were conducted by the local health department for all students at each school. A mobile eye clinic visited each participating school, where eye examinations were conducted by licensed optometrists. Students who needed refractive correction selected eyeglasses at the time of examination. Eyeglasses were dispensed to students at school, and replacement eyeglasses were provided as needed. Program staff provided implementation support to schools. Research Design: This cluster randomized study assigned elementary and middle schools to one of three groups: intervention in the first, second, or third year starting in SY2016-17 (Cohort 1-3 respectively) (Figure 1). Data Collection: Academic outcomes were mathematics and English language arts (ELA) scores on the i-Ready, and mathematics and ELA scores on the state test, the Partnership for Assessment of Readiness for College and Careers (PARCC). Individual student test score data for SY 2015-16 through SY 2017-18 and student demographic data were obtained from the school district. Vision exam data were collected during the eye examinations. Data Analysis: Hierarchical linear modeling was used to assess the impact of the intervention on academic performance using an intent to treat approach. The models analyzed program effects on each outcome measure separately, and controlled for grade level and baseline achievement at the student level, as well as the stratifying variables used in randomization. Differential treatment effects for various student subgroups (gender, grade level, special education, and baseline achievement) were examined with interaction terms. Findings: The outcomes for all students are summarized in Table 1. On the i-Ready ELA (Figure 2), there was a positive impact at the end of the first year, with an effect size of +0.09 (p < 0.05). Impacts at the end of the second year were similar (ES = +0.12, p = 0.09) but not statistically significant. On the i-Ready Mathematics (Figure 3), outcomes were positive for students at the end of the second year, but not statistically significant (ES = +0.07, p = 0.07). Impacts on PARCC ELA (Figure 4) and Mathematics (Figure 5) were not significant. Differential effects were identified among students based on special education status and baseline achievement (Figures 6-7). For students receiving special education services, the effects on i-Ready ELA were large, and some were statistically significant (Year 1 ES = +0.25, p < 0.05, and Year 2 ES = +0.26, n.s.). In addition, for students in the Year 1 cohort performing in the lowest 25% at baseline, there was a significant positive impact of receiving eyeglasses on i-Ready ELA at the end of Year 1 (ES = +0.28, p < 0.05). Discussion: In this study, we found that, on average, students who received school-based vision care achieved higher scores on the i-Ready ELA when compared to controls. Students in special education and the lowest performers at baseline showed the greatest benefits. There was no significant impact overall of the school-based eye exam and eyeglasses on i-Ready Mathematics or PARCC tests. As a point of comparison, Figure 8 shows mean effect sizes for various widely-used educational interventions for reading (CREDO, 2015; Figlio et al., 2018; Neitzel et al., 2021). The modest overall effect size of +0.09 seen on i-Ready is larger than that for other common interventions except tutoring. The effects for students in special education (ES = +0.25, +0.24, +0.23) and low achievers (ES = +0.28) are comparable to those of tutoring. Limitations: Because of the different outcomes on different measures, the results should be seen as promising rather than conclusive. This study was done in the context of a real-world school-based program implementation encompassing over 30,000 students over a wide age range, and as such there may have been variations in screenings and exams that would not have occurred in a clinical trial setting. Conclusions: The findings of this study, if replicated, may provide schools with an important additional means of improving achievement on some types of outcomes for many students: providing eyeglasses to students who need them.