The joint position reproduction (JPR) approach is commonly used to assess joint position sense (JPS), but its psychometric properties in children remain unexplored. This study aimed to assess the reliability and precision of a multi-joint JPR protocol for assessing lower-limb JPS in typically developing (TD) children. Ankle, knee, and hip JPS were assessed in TD children (aged 5-12 years), on two different days, by a single rater using a standardized JPR protocol. The mean and best error (joint reproduction error, °) between the target and reproduction angle were calculated from three-dimensional kinematics for each joint across three trials. Total, joint, and limb JRE scores were provided. For JPR reliability, the intraclass correlation coefficient (ICC, 2.1) was reported. For JPR precision, the standard error of measurement (SEM) and smallest detectable difference (SDD) were calculated. Across 270 JPR trials (15 children, 8.6 ± 1.2 years, 8 boys), the mean and best JRE were 3.7° and 2.5°, respectively. The ICC ranged from poor to fair (0.01-0.44) for mean JRE, and fair to very good (0.46-0.77) for best JRE. The SEM ranged from 0.8° to 1.8°. The SDD was less than 5°, ranging from 2.3° to 4.5°. Evaluating ankle, knee, and hip JPS in children using passive JPR is more reliable and precise when using the best JRE. This study highlights the need for a multi-joint JPR approach and provide joint- and limb-specific SEM and SDD values.