The findings from a comprehensive study on oral corrective feedback show that the success of such feedback as measured in a subsequent test is affected by its format, the type of error corrected, and certain learner characteristics. The most successful format of correction, both for the learners receiving the feedback and for their peers, is feedback successfully eliciting self-correction in practice situations. Among the least successful formats for both groups are recasts without further comments or repetition by the corrected learner. The type of error corrected most successfully differs for the two groups. Those corrected learn most from the correction of their grammatical errors and least from correction of pronunciation errors. Peers score best on pronunciation items and gain least from correction of lexical errors. Of the learner characteristics taken into consideration, verbal intelligence, relative proficiency (within levels at school or university), and the learners’ attitude towards correction proved to be most influential.