BackgroundThe outbreak of the new coronavirus-2019 (COVID-19) has had a significant impact on people's mental and physical health. Meanwhile, people's perceptions of risk may influence their emotional states and preventative behavior during an epidemic. Previous research have revealed the diversity and uniqueness of risk perception, and college students may have a different perspective on risk perception. The objective of this study was to describe the subtypes of risk perception for COVID-19 among college students in China, identify the subtypes' traits, and investigate their affecting variables.MethodsCollege students from 10 Chinese provinces participated in a cross-sectional study (n = 2,000) that from January 16 to 30, 2022. The latent profiles and influencing factors for risk perception were investigated using latent profile analysis, one-way analysis of variance, and multinomial logistical regression.ResultsThe sample group of this survey was 1,946 students, and the response rate was 97.3%. The best model was suggested to consist of three profiles: “neutral risk perception” (20.3%), “perception seriously without susceptible” (52.8%), and “low risk perception” (26.9%). Risk perception of COVID-19 was positively associated with attention to negation information (r = 0.372, p < 0.01), anxiety (r = 0.232, p < 0.01), and depression (r = 0.241, p < 0.01), and negatively associated with perceived social support (r = −0.151, p < 0.01). Logistic-regressions analyses mainly revealed that the risk perception of three profiles related to having chronic diseases (OR = 2.704, p < 0.01), medical major (OR = 0.595, p < 0.01; OR = 0.614, p < 0.05), without having COVID-19 confirmed cases around (OR = 0.539, p < 0.01), attention to negative information (OR = 1.073, p < 0.001; OR = 1.092, p < 0.001), and perceived social support (OR = 0.0.975, p < 0.01).ConclusionsThe level of risk perception for COVID-19 among Chinese college students was unsatisfactory, and the risk perception of COVID-19 had significant group characteristics and heterogeneity. Colleges and public health practitioners could have a theoretical and empirical basis to implement risk perception intervention efforts by identifying latent subgroups during the COVID-19 epidemic.