Abstract: To explore the influence factors of physical fitness and their interactions in preschool children based on decision tree model. 4 621 preschool children(36~83 months) were recruited in this study, the physical fitness level was measured and graded(qualified and unqualified) according to the Handbook of National Physical Fitness Measurement Standards(Preschool Children Part). Retrospective questionnaire was used to collect the data of birth information, rearing pattern, sedentary behavior(SB) and physical activity(PA), dietary nutrition, sleeping mode, and parents' information, the questionnaires were completed by the parents/legal guardians. The model of CHAID decision tree was established by using IBM SPSS modeler. The major factors influencing physical fitness of preschool children were as follows: PA at weekends, SB at weekends, gender, moderate to vigorous physical activity(MVPA) at weekends, PA at school days, calcium intake, SB at school days. At the root node of the decision tree, the qualification rate of physical fitness in preschool children with PA“>3 h/d”(95.35%) was significantly different from the children with“1~3 h/d”(91.43%) and“≤ 0.5 h/d”(85.43%)(P=0.000). The level 2 and 3 leaf nodes of PA“>3 h/d”were SB at weekends and PA at school days, respectively. The level 2 leaf nodes of“1~3 h/d”and“≤ 0.5 h/d”were both gender. The variables of level 3 to 5 leaf nodes of“1~3 h/d”were MVPA at weekends, MVPA at school days, SB at school days, calcium, etc.PA at weekends is the most important factor benefiting physical fitness in preschool children. PA at weekends is recommended to exceed the Physical Activity Recommendations(3 h/d) for preschool children, and they should reduce the SB at weekends and increase PA at school days. PA and SB, gender, and calcium intake are all important factors in promoting physical fitness for preschool children.