Objectives To evaluate the psychometric property of the Insomnia Severity Index（ Chinese version） in students of a commercial school. Methods A total of 3 969 students of a commercial school were selected. Their sleep quality， insomnia， anxiety， daytime sleepiness， and depression were evaluated by the Pittsburgh sleep quality index（ PSQI）， the Insomnia Severity Index（ ISI）， Self-Rating Anxiety Scale（ SAS）， Epworth Sleepiness Scale（ ESS）， Beck Depression Inventory（ BDI）， respectively. Their general information was also gathered. After 9 months， 73% of the students（ n=2 915） were re-evaluated by the ISI. Results The score of ISI in the first test were 5.8±4.6（ 0-28）， in a skewed normal distribution. It had a single factor structure （interval variance value was 52%）； the item loading ranged from 0.62 to 0.80； KMO=0.87； Bartlett spherical test result was P ＜ 0.001； the correlation coefficient between the item and the total score was 0.63-0.79； Spearman-Brown semi-reliability and Cronbach alpha coefficients were 0.75 and 0.84（ P＜ 0.001）. There was a significant correlation between the first test ISI score and the ESS score， BDI score and SAS score， while the ISI score was not statistically related to the PSQI score. The effect size（ Cohen's d） of the total score and each item of ISI between the first test and the second test was less than 0.1， except for item 5（ Cohen's d=0.361）. Conclusions The ISI（ Chinese version） has good psychometric property， so it is suitable for evaluating insomnia among adolescents. Moreover， attention should be paid to the prevalence of insomnia in these young adults（ 30.2%）