Purpose: The aim of this study was to determine the variability of visual acuity in a large eye clinic. Methods: A cross-sectional study using 50, consecutively presenting adult patients with visual acuity of at least 6/60 and aged between 18 and 75 years was performed. Measurements of visual acuity obtained under normal clinical conditions mere compared to measurements obtained using standard clinical research protocols. The variability of visual acuity was assessed by determining the 95% limits of agreement between test and retest measures. Results: There were no significant differences between test-retest measurements of visual acuity, either aided or unaided. Pearson r correlation coefficients between test and retest measurements mere high for both aided and unaided visual acuity. The 95% limits of agreement revealed repeatability of about ± 1.5 log-MAR or 1.5 lines on a standard log-MAR chart. Conclusion: In large eye clinics, in order to be confident that a real change in visual acuity has occurred between measurements, a difference of at least 0.15 log-MAR (8 letters on a standard log-MAR visual acuity chart) is required.