Objective: To identify latent class characteristic of negative life events among college and middle school students and to examine their associations with cigarette, e-cigarette, and dual use experiences, thereby providing evidence for precision prevention for tobacco use among young students. Methods: Using multistage cluster sampling from 2022 to 2023, 2 347 college and middle school students were recruited from 13 cities, including Beijing, Taiyuan, Longkou, Shanghai, Suzhou, Hangzhou, Yuyao, Guangzhou, Shenzhen, Zhongshan, Chongqing, Chengdu and Chuxiong, covering northern, eastern, southern, and southwestern China. And a questionnaire was applied to collect data on demographics, family characteristics, and tobacco use experiences. Negative life events of students were measured by Adolescent Self rating Life Events Checklist. Latent class analysis (LCA) was conducted to identify distinct subgroups of negative life events, and multivariable generalized estimating equation (GEE) model was used to assess associations between negative life events of different classes and tobacco use experiences among college and middle school students. Results: Three latent classes for negative life events of interviewed students were identified by LCA: a low risk group ( n = 1 606, 68.4%), an interpersonal-academic stress group ( n =704, 30.0%), and a multiple high risk group ( n =37, 1.6 %). Multivariable GEE model revealed that, compared with the low risk group, the interpersonal-academic stress group had higher risk of cigarette use ( OR=1.78, 95%CI =1.25-2.53), e-cigarette use ( OR=2.33, 95%CI =1.58-3.43), and dual use ( OR= 1.99, 95%CI =1.28-3.08) (all P <0.01). The multiple high risk group showed substantially elevated risk for cigarette use ( OR=5.98, 95%CI =2.56-13.98), e-cigarette use ( OR=7.94, 95%CI =3.35-18.83), and dual use ( OR=4.85, 95%CI =1.81-12.99) (all P < 0.01). Conclusion Negative life events among college and middle school students exhibit marked heterogeneity, with a cumulative risk gradient for tobacco use risk.