Abstract
OBJECTIVE: This study aimed to develop age- and sex-specific percentile reference curves and evaluation criteria for balance ability in preschool children using the Generalized Additive Models for Location, Scale, and Shape (GAMLSS) model. It also sought to analyze the influencing factors of balance ability through network analysis, providing evidence to support strategies for improving balance development in early childhood. METHODS: A cross-sectional study was conducted from April to July 2023, involving 5,559 preschool children aged 3 to 6 years from 12 districts (cities and counties) in Weifang City, Shandong Province, China. Participants were selected using a stratified, randomized, whole-cluster sampling method. Physical fitness tests and questionnaires on physical activity participation were administered. The GAMLSS model was used to generate balance ability percentile curves. Analysis of variance (ANOVA) and other statistical methods were employed to examine differences by age, sex, and urban-rural residence. Network analysis was conducted to identify key factors influencing balance ability. RESULTS: Significant age-related differences in balance ability were observed (p < 0.05). Among 5-year-olds, girls demonstrated significantly better balance performance than boys (p < 0.01). Urban children outperformed rural peers in all age groups (p < 0.01). Balance beam completion time was significantly negatively correlated with height, weight, grip strength, and standing long jump performance (p < 0.01), while it was positively correlated with 10-meter shuttle run, 15-meter obstacle run, and two-legged continuous jump (p < 0.01). CONCLUSIONS: Girls exhibited earlier development in balance ability compared to boys, and urban children demonstrated superior balance performance relative to rural children. Physical morphology was closely related to balance ability, and various physical fitness components showed compensatory interactions, collectively influencing balance development in preschool children.