The aim of this study was to gain insight into the sleep quality of college students and related factors from a new perspective by using Latent Class Analysis (LCA). A total of 1,288 college students from four universities in Wuhu city participated in the study. LCA was used to identify the classes of sleep behaviors. Differences in class membership related to selected research factors were examined using multinomial logistic regression analysis.Four distinct classes of behaviors were identified: (1) good sleep (Class 1, 31.8%), (2) prolonged sleep latency (Class 2, 49.1%), (3) sleep disturbances and daytime dysfunction (Class 3, 6.8%), (4) multiple poor sleep behavior (Class 4, 12.3%). The latent classes of sleep behavior were correlated with the DBAS-16 total score (rs = -0.109, P < 0.001). Learning pressure and mental state during the day could affect overall sleep (Class 2, Class 3 and Class 4), and female students were at higher risk of severe sleep problems (Class 3 and Class 4), while bedtime exercised could improve mild sleep problems (Class 2). The sleep behavior of college students in Wuhu city has obvious class heterogeneity, and different influencingfactors may affect sleep to varying degrees. In addition, our research provides a basis for targeted intervnetion in college student's sleep. .
Keywords: College students; LCA; sleep behaviors; sleep related cognitions.