Annual production of crop residues has reached nearly 4 billion metric tons globally. Retention of this large amount of residues on agricultural land can be beneficial to soil C sequestration. Such potential impacts, however, may be offset if residue retention substantially increases soil emissions of N2O, a potent greenhouse gas and ozone depletion substance. Residue effects on soil N2O emissions have gained considerable attention since early 1990s; yet, it is still a great challenge to predict the magnitude and direction of soil N2O emissions following residue amendment. Here, we used a meta-analysis to assess residue impacts on soil N2O emissions in relation to soil and residue attributes, i.e., soil pH, soil texture, soil water content, residue C and N input, and residue C : N ratio. Residue effects were negatively associated with C : N ratios, but generally residue amendment could not reduce soil N2O emissions, even for C : N ratios well above ca. 30, the threshold for net N immobilization. Residue effects were also comparable to, if not greater than, those of synthetic N fertilizers. In addition, residue effects on soil N2O emissions were positively related to the amounts of residue C input as well as residue effects on soil CO2 respiration. Furthermore, most significant and stimulatory effects occurred at 60–90% soil water-filled pore space and soil pH 7.1–7.8. Stimulatory effects were also present for all soil textures except sand or clay content ≤10%. However, inhibitory effects were found for soils with >90% water-filled pore space. Altogether, our meta-analysis suggests that crop residues played roles beyond N supply for N2O production. Perhaps, by stimulating microbial respiration, crop residues enhanced oxygen depletion and therefore promoted anaerobic conditions for denitrification and N2O production. Our meta-analysis highlights the necessity to connect the quantity and quality of crop residues with soil properties for predicting soil N2O emissions.