OBJECTIVE Single optotype tests of visual acuity are widely used for preschool vision screening in order to optimise cooperation with testing. These tests may, however, underestimate the visual acuity deficit in amblyopia because they lack visual crowding. This study assessed the resultant negative predictive value (NPV) for amblyopia. METHODS Cohort study of 936 children in the Cambridge Health District selected by date of birth. The presence of amblyopia among children who had passed preschool vision screening was determined using Snellen line acuity as the reference test. Preschool vision screening was conducted at 3.5 years of age by community orthoptists. The screening assessment comprised Sheridan-Gardiner single optotype test of visual acuity (referral criterion 6/9 or worse in either eye), cover test, ocular movements, 20(Delta) prism test, and TNO stereotest. RESULTS The overall NPV of preschool vision screening for amblyopia was 100% (95% CI 99.4% to 100%). Most children with amblyopia were detected by the Sheridan-Gardiner single optotype test of visual acuity, but the other screening tests were necessary to prevent any false negatives. In isolation, the Sheridan-Gardiner single optotype test of visual acuity has a NPV for amblyopia of only 99.6% (95% CI 98.7% to 99.9%). CONCLUSIONS Preschool vision screening using a single optotype test of visual acuity does achieve a high NPV for amblyopia, but only under certain conditions. These comprise a low threshold for referral (6/9 or worse in either eye) and the inclusion of a cover test and tests of binocular function in the screening assessment.