Abstract While there is a significant amount of compelling evidence to support written corrective feedback (error correction), it is also acknowledged that research findings may not be applicable or conclusive enough given the great variability among studies. Nevertheless, no systematic attempt has been made to review and synthesize the extensive amount of literature to identify the sources that lead to such variation. This study aims to identify such variables. Results indicate that in a research base of 76 relevant publications, variations can be explained based on 11 main sources. Pursuant to these findings, this study sketches the main variance-related problems and outlines design recommendations to further expand L2 research and practice of error correction in Second Language Acquisition.