Purpose: To identify clinical vision measures that are associated with the driving performance of glaucoma patients who have visual field loss and visual acuity better than 20/100 and to compare the driving performance of glaucoma patients with the driving performance of a group of age- and sex-equivalent individuals without eye disease. Patients: Forty patients with glaucoma and 17 normally sighted control subjects participated in this study, Methods: Clinical vision data, consisting of visual acuity, letter contrast sensitivity, and visual fields, were collected. Driving performance was assessed by (1) an interactive driving simulator that measured 7 indices of performance (including number of accidents) and (2) the self-reported accident involvement for the past 5 years. Main Outcome Measures: Driving simulator performance and real-world, self-reported accident involvement. Results: The number of accidents as measured on the driving simulator in the glaucoma group was significantly correlated with three Goldmann visual field measures: combined horizontal extent (p= -0.47, P= 0.01), total horizontal extent (p = -0.49,P=0.007), and total peripheral extent (p = -0.55, P = 0.002). There were no statistically significant correlations between the driving performance of the glaucoma group and the visual acuity or contrast sensitivity measures. When compared with the control group, a significantly greater proportion of the glaucoma group reported having at least one real-world accident within the past 5 years (Fisher exact test, P 0.005). Conclusions: Visual field reduced to less than 100 degrees of horizontal extent may place patients with peripheral field loss at greater accident risk. A higher incidence of real-world and simulator accidents was found for the group with glaucoma.