1.Small dam removals are increasing on a global scale; yet general predictions of organism response to dam removal are constrained by heterogeneity of study designs, implementation strategies, geographies, and characteristics of dams and their removals. 2.Macroinvertebrate data extracted from 29 studies including 34 small dam removals over a broad geographical range was re-analysed utilizing dam removal effect sizes (a quantified change from before to after removal). Effect sizes of 10 metrics of community structure were calculated to investigate the spatiotemporal extent of small dam removal effects and if responses differ with characteristics of the dam and environmental settings. 3.We found that dam removal had initial negative effects on total macroinvertebrate density and EPT (Ephemeroptera, Plecoptera, and Trichoptera) density, both downstream and upstream; however recovery to pre-removal values was reached and exceeded after ca 15-20 months. Mean annual discharge, land use in the catchment, and distance from the dam affected the magnitude and direction of responses of four community metrics: total density, EPT density, %EPT density, and Family Biotic Index. 4.Synthesis and applications. Our study provides evidence that macroinvertebrate community recovery from dam removal is mediated by catchment characteristics and system size, which may correlate with sediment flushing efficiency. Negative impacts were observed in smaller systems or those with a high percentage of undisturbed catchment areas, conditions that may benefit from sediment management prior to dam removal. Significant responses in reaches upstream of the impoundment clearly indicate that caution be applied to interpretations of response in sampling designs that utilize upstream sites for reference condition.