propensity score (PS) is a common method to assess the causal effects of interventions by balancing the covariate distribution between treatment and control groups to control confounding bias. Among many PS applications, inverse probability of treatment weighting (IPTW) is the most commonly used, and its target effect size is the overall average treatment effect (ATE). However, when individual PS is close to 0 or 1, IPTW tends to produce extreme weights, which causes the variance of effect estimation to expand and the robustness to decrease, especially in small sample studies. Therefore, the propensity score overlap weighting (PSOW) is proposed under the framework of balanced weighting, and its target effect is the average treatment effect in the overlap population (ATO).