Objective: To explore the prevalence and influencing factors of depressive symptoms among university students in some universities in Shandong Province, so as to provide a theoretical reference for implementing corresponding intervention measures. Methods: A cluster sampling method was used to select 8 079 university students studying in universities in Shandong Province from March 2023 to May 2024 as the research subjects. They were randomly divided into a training set (6 463) and a validation set (1 616) according to the 8/2 ratio. The influencing factors of depression among university students were analyzed, and a risk prediction model for depression among university students was constructed and validated. Results: In the training set of university students, the detection rate of depression was 35.09%(2 268/6 463), with 1 632 cases (71.96%) of mild depression, 545 cases (24.03%) of moderate depression, and 91 cases (4.01%) of severe depression. In the validation set of university students, the detection rate of depression was 33.97% (549/1 616), with 384 cases (69.95%) of mild depression, 127 cases (23.13%) of moderate depression, and 38 cases (6.92%) of severe depression. In the training set, the proportions of those who surfed the Internet for more than 3 h/d, occasionally or did not participate in physical exercise, had average or poor relationships with classmates, often drank sugary drinks, occasionally or did not have breakfast, had unsatisfactory academic performance, had an average monthly living expense of less than 1 500 yuan on campus, and had divorced or widowed parents in the depression-detected group were all higher than those in the undetected group( χ 2=1 193.85,1 584.41, 1 115.10 ,826.00,1 424.05,924.58,803.68,797.65, P <0.05). The scores of the Chinese version of the Childhood Trauma Questionnaire (CTQ-SF) in the depression detected group were also higher than those in the undetected group( t =98.48, P <0.05). The results of Logistic regression analysis showed that physical exercise, classmate relationships, academic performance, average monthly living expenses on campus, and CTQ-SF scores were influencing factors for depression among university students( OR =3.87, 4.82, 3.63, 3.75, 4.39, P <0.05). The sensitivity of the model in the training set for predicting depression among university students was 89.0%(95% CI =87.6%-90.2%), the specificity was 93.0% (95% CI =92.2%-93.7%), and the area under the curve was 0.9(95% CI =0.8-1.0); the sensitivity of the model in the validation set for predicting depression among university students was 87.6%(95% CI =84.5%-90.1%), the specificity was 91.3%(95% CI =89.4%-92.9%), and the area under the curve was 0.9(95% CI =0.8-1.0). Conclusions The high detection rate of depressive symptoms among university students in some universities in Shandong Province warrants attention. Constructing a risk prediction model is helpful for early identifying the risk of depression among university students.