The provision of disability-related accommodations in a high-stakes testing environment raises important questions about the comparability of test scores between accommodated test takers (individuals with disabilities) and their counterparts who test under standard conditions, the validity and utility of test scores generated under accommodated conditions, and whether the accommodations themselves constitute an unfair advantage. Previous attempts to answer such questions have been primarily policy and judgment based. Sample size limitations and demographic and background differences between accommodated and non-accommodated test takers make quality empirical data difficult to gather and analyze. Drawing on a database of more than one million accommodated and non-accommodated test takers, this large-scale empirical study—the largest ever conducted to date—employed propensity score analysis to match 2,305 accommodated test takers with a comparable group of non-accommodated examinees. The key finding is that there were no meaningful or statistically significant differences in the distributions of Graduate Management Admission Test (GMAT) Quantitative, Verbal, Total, or writing scores for accommodated versus non-accommodated test takers when demographic and background characteristics were taken into account.