BACKGROUND Professional writing is an essential outcome for engineering graduates and hence a vital part of engineering education. To provide a successful learning experience for students engaged in writing activities, timely feedback is necessary. Providing this feedback to increasing numbers of students poses a major challenge for instructors. New automated systems work towards providing both timely and appropriate writing feedback, but students' views on automated feedback, and feedback in general, are not well understood. PURPOSE (HYPOTHESIS) To contribute to a deeper understanding of students' conceptions of feedback from tutors and an automated system called Glosser, and how these conceptions are related to achievement. DESIGN/METHOD Students in an engineering course worked in pairs to write an engineering report on e-business. The design of the study involves in-depth interviews and the analysis employs an approach in which student conceptions of automated feedback are investigated in relation to related feedback from their tutor, perceptions of automated feedback in general, and their academic achievement. RESULTS Students' conceptions of feedback vary and can be grouped into cohesive and fragmented, which is consistent with other theoretical models. Close associations were found between more cohesive conceptions of feedback and better academic performance. CONCLUSIONS A student's conception of traditional and automated feedback is similar, being either cohesive or fragmented. Changing one may change the other. Deep learners see feedback as a way of learning about the topic whereas shallow learners see them as a way to improve the communication aspects of writing. Design considerations based on these results are discussed.
