To analyze the factors associated with high-level fear of negative evaluation (FNE) among junior high school students and to construct a nomogram risk prediction model, so as to provide scientific tools for psychological health intervention for junior high school students.

A convenience sampling combined with cluster random sampling method was used to select 5 485 junior high school students from 4 cities (Wuhan, Huanggang, Xianning and Xiaogan) for an online questionnaire survey in March 2025. The total sample was randomly split into a training set (n=3 839) and a validation set (n=1 646). Univariate analysis was performed in the training set using Chi-square test and t-test. Variables with statistical significance were subsequently included in multivariate Logistic regression to identify independent predictors and to construct a nomogram-based risk prediction model. The discriminative ability and clinical utility of the model were evaluated in the validation set using the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA).

There were 1 649 junior high school students with low-level FNE and 2 190 with high-level FNE in the training set. The self-control ability of junior high school students with low-level and high-level FNE showed a statistically significant difference (23.96±3.96, 21.48±3.37, t=25.15, P < 0.01). Statistically significant differences in the detection rate of high-level FNE were observed among junior high school students with different genders, family types, parenting styles, academic rankings, psychological flexibility, mobile phone addiction tendencies, emotional management training, exercise frequency, left-behind experiences, and places of origin (χ²=82.01-1 126.68, all P < 0.01). The results of Logistic regression analysis revealed that, the following factors were identified as significant factors influencing high-level FNE among junior high school students: exercise frequency (OR=0.21, 95%CI=0.17-0.26); parenting style (OR=0.48, 95%CI=0.40-0.58); left-behind experience (OR=3.88, 95%CI=3.27-4.61); smartphone addiction proneness (OR=2.19, 95%CI=0.89-0.93); self-control ability (OR=0.91, 95%CI=0.89-0.93); and psychological flexibility (OR=0.16, 95%CI=0.10-0.28) (all P < 0.05). The AUC for the training and validation set were 0.88 (95%CI=0.87-0.89) and 0.87 (95%CI=0.85-0.89), respectively. The Hosmer-Lemeshow goodness-of-fit test yielded χ²=8.57, 15.20 (both P>0.05).

The risk prediction model with high-level FNE demonstrates good accuracy and can assist educators and parents in timely screening of junior high school students with high-level FNE, thereby providing a basis for implementing targeted interventions.