Background Early and effective ocular screening may help to eliminate treatable eye disorders. The Lhasa Childhood Eye Study (LCES) revealed the particular prevalence of refractive error and visual impairment in grade one schoolchildren (starting age of 6 years old) in Lhasa. Methods This is a cross-sectional part of school-based cohort study. One thousand nine hundred forty-three children were enrolled (median age, 6.78 years, range, 5.89 to 10.32). Each child underwent general and ocular examinations, including logarithm of the minimum angle of resolution (logMAR) visual acuity, cycloplegic autorefraction, and slit-lamp biomicroscopy evaluation. Multivariate and correlation analyses were performed to evaluate the association between refractive error with gender and ethnics. Results The prevalence of visual impairment (logMAR visual acuity >= 0.3 in the better-seeing eye) of uncorrected, presenting and best-corrected visual acuity (BCVA) was 12.2, 11.7 and 2.7%, respectively. Refractive error presented in 177 (78.0%) out of 227 children with bilateral visual impairment. Myopia (spherical equivalent refractor [SER] <= - 0.50 diopter [D] in either eye) was present in 4.7% children when measured after cycloplegic autorefraction. Hyperopia (SER >= + 2.00 D) affected 12.1% children. Hyperopia was significantly associated with female gender (P<0.001). Astigmatism (cylinder value <= - 0.75 D) was present in 44.8% children. In multivariate regression and correlation analysis, SER had no significant difference between ethnic groups. Conclusion The Lhasa Childhood Eye Study is the first school-based cohort study to reveal the prevalence and pattern of refractive error and visual impairment in Lhasa. Effective strategies such as corrective spectacles should be considered to alleviate treatable visual impairment.