Objectives: Idiopathic short stature (ISS) is a heterogeneous in etiology and response to growth hormone treatment. This study was focused on the person-centered approaches to understanding the characteristics of idiopathic short stature. Subjects and Methods: Data from 218 children (127 boys and 91 girls) aged 4–15 years was obtained from “LG Growth Study”. Growth hormone dose, first-year height standard deviation score (ΔHtSDS), mid-parental HtSDS, and initial bone age were inputted into the model and latent profile analysis of ISS was conducted. Results: Three-class model was selected as the final model because the differences among the groups were explained graphically and statistically optimal. The group names were determined as follows: younger good responder (n=56), older good responder (n=111), older poor responder (n=51). The mean growth hormone dose were 0.64, 0.72, and 1.64 IU/kg/week and the mean ΔHtSDS were 0.85, 0.60, and 0.53, respectively. The mid-parental HtSDS, initial body mass index SDS and proportion of prepuberty were higher in younger good responder. The pretreatment height velocity (3.99 vs. 4.83 cm/yr), peak growth hormone level (8.72 vs. 11.65 ng/mL), initial insulin-like growth factor-I (IGF-I) SDS (-1.01 vs. -0.78), ΔIGF-I SDS (1.10 vs. 1.66) were lower in older poor responder compared to those of older good responder. Conclusions: The latent profile analysis revealed three distinct ISS phenotypes. This study offers an interest multidimensional approach for identifying ISS phenotypes that exhibit variable outcome to growth hormone treatment.