BACKGROUND AND PURPOSE: Transcranial direct current stimulation (tDCS) has shown mixed results in post-stroke motor recovery, possibly because of tDCS dose differences. The purpose of this meta-analysis was to explore whether the outcome has a dose-response relationship with various dose-related parameters. METHODS: The literature was searched for double-blind, randomized, sham-controlled clinical trials investigating the role of tDCS (≥5 sessions) in post-stroke motor recovery as measured by the Fugl-Meyer Upper Extremity (FM-UE) scale. Improvements in FM-UE scores were compared between active and sham groups by calculating standardized mean differences (Hedge's g) to derive a summary effect size. Inverse-variance-weighted linear meta-regression across individual studies was performed between various tDCS parameters and Hedge's g to test for dose-response relationships. RESULTS: Eight studies with total of 213 stroke subjects were included. Summary Hedge's g was statistically significant in favor of the active group (Hedge's g = 0.61, p = 0.02) suggesting moderate effect. Specifically, studies that used bihemispheric tDCS montage (Hedge's g = 1.30, p = 0.08) or that recruited chronic stroke patients (Hedge's g = 1.23, p = 0.02) showed large improvements in the active group. A positive dose-response relationship was found with current density (p = 0.017) and charge density (p = 0.004), but not with current amplitude. Moreover, a negative dose-response relationship was found with electrode size (p < 0.001, smaller electrodes were more effective). CONCLUSIONS: Our meta-analysis and meta-regression results suggest superior motor recovery in the active group when compared to the sham group and dose-response relationships relating to electrode size, charge density and current density. These results need to be confirmed in future dedicated studies.