To document the difference between non-cycloplegic and cycloplegic refraction and explore its associated factors in Chinese young adults.A school-based study including 7971 undergraduates was conducted in Anyang, Henan Province, China. Cycloplegia was achieved with two drops of 1% cyclopentolate and 1 drop of Mydrin P (Tropicamide 0.5%, phenylephrine HCl 0.5%) with a 5 min interval. Non-cycloplegic and cycloplegic refractions were measured by an autorefractor. A paired-sample t-test and Spearman correlation analysis were used for analysis with data from only the right eyes included.  RESULTS: Of the 7971 students examined, 7793 (97.8%) with complete data were included, aging 20.2±1.5 years. Male students accounted for 36.8%. Overall, there was a significant difference between non-cycloplegic and cycloplegic SE (spherical equivalent) of 0.83±0.81D (p<0.01). The difference was 1.80±1.11D, 1.26±0.93D and 0.69±0.69D for those with cycloplegic hyperopia, emmetropia and myopia, respectively (p<0.01 for all). Those with a hyperopic shift less than 0.25D and 0.5D accounted for 11.1% and 34.1%, respectively. A significant relationship was found between difference in SE and cycloplegic refraction (r=0.33, b=0.11, p<0.01). Without cycloplegia, prevalence of hyperopia and emmetropia would be underestimated by 6.2% (1.0% vs 7.2%) and 5.7% (3.8% vs 9.5%), respectively, with prevalence of myopia and high myopia overestimated by 12.1% (95.3% vs 83.2%) and 6.1% (17.2% vs 11.1%).  CONCLUSION: Lack of cycloplegia will lead to significant misclassification of myopia, emmetropia and hyperopia in Chinese young adults. Cycloplegia is therefore essential for this age-group in epidemiological studies.© Author(s) (or their employer(s)) 2018. No commercial re-use. See rights and permissions. Published by BMJ.