Young adults exhibit a subtle, group-level asymmetry of lateral spatial attention favouring the left hemispace over the right (pseudoneglect). We have recently shown that leftward biases are maintained in older adults aged ≥50 when measured using the line bisection task (Learmonth and Papadatou-Pastou 2021). Here we present a meta-analysis of spatial biases in children aged ≤16 years. Databases (PsycINFO, Web of Science, and Scopus) and pre-print servers (bioRxiv, medRxiv, and PsyArXiv) were searched up to 8th March 2021 for studies involving children aged ≤16, who were tested using the line bisection or landmark task, and who had not been recruited as an atypically developing group. Thirty-six datasets from 33 studies, involving 2515 children, were included. A small leftward bias was identified overall (Cohen’s d = -.18, 95% CI = -.36, -.01). No spatial bias was identified when the 33 line bisection studies were analysed separately (d = -.16, 95% CI = -.35, .04). Moderator analysis of the line bisection task found a strong influence of the hand that was used to bisect the line, with right-handed actions resulting in right-biased bisections, and left-handed actions in left-biased bisections (symmetrical neglect). Bisections were slightly more leftward in studies that included a higher percentage of boys relative to girls. Age, hand preference, and the control group status of the children did not moderate spatial biases and we found no evidence of small study publication bias. However, the number of studies included in each moderator analysis was small. This meta-analysis confirms that pseudoneglect is present in children, but its detection is dependent on the task that is used.