Automated written corrective feedback: how well can students make use of it?

Automated written corrective feedback (AWCF) has qualities that distinguish it from teacher-provided WCF and potentially undermine claims about its value for L2 student writers, including disparities in the amounts of useful information it provides across error types and the fact that inaccuracies in error-flagging must be anticipated. It remains unclear how these factors influence different students' ability to use AWCF, partly because classroom research to date has been based on within-group designs and writing tasks in which students can choose other ways to respond to feedback besides correction. To investigate these issues, the present study used an AWCF-based error-correction task in which student response type was controlled, as were the explicitness and accuracy of the feedback. Eighty-two ESL students in two, sequenced developmental writing courses participated. Generic feedback was found to result in fewer successful error corrections than specific feedback as well as higher perceptions of mental-effort expenditure and lower ratings of clarity and helpfulness. Course level was found not to be a significant factor in most analyses. Implications for L2 classroom applications of AWCF as well as the design of AWCF tools are discussed.