Background Childhood allergic rhinitis (AR) is clinically heterogenous. We aimed to identify distinct phenotypes among children with AR using data-driven techniques and to ascertain their association with patterns of symptoms, allergic sensitization, and comorbidities. Methods We recruited 510 children with physician-diagnosed AR, of whom 205 (40%) had asthma. Latent class analysis (LCA) was performed to identify latent structure within the data set using 17 variables (allergic conjunctivitis, eczema, asthma, family history of asthma, family history of allergic rhinitis, skin sensitization to 8 common allergens, tonsillectomy, adenoidectomy). Results A four-class solution was selected as the optimal model based on statistical fit. We labeled latent classes as: (1) AR with grass mono-sensitization and conjunctivitis (n = 361, 70.8%); (2) AR with house dust mite sensitization and asthma (n = 75, 14.7%); (3) AR with pet and grass polysensitization and conjunctivitis (n = 35, 6.9%); and (4) AR among children with tonsils and adenoids removed (n = 39, 7.6%). Perennial AR was significantly more common among children in Class 2 (OR 5.83, 95% CI 3.42-9.94, p < .001) and Class 3 (OR 2.88, 95% CI 1.36-6.13, p = .006). Mild and intermittent AR symptoms were significantly more common in children in Class 2 compared to those in Class 1. AR was more severe in Class 1 compared to other 3 classes, indicating that upper respiratory symptoms are more severe among children with isolated seasonal rhinitis, than in those with rhinitis and coexisting asthma. Conclusion We have identified 4 phenotypes in school-age children with AR, which were associated with different patterns of clinical symptoms and comorbidities.