This study investigated the effect of computer-mediated corrective feedback types in an English as a foreign language (EFL) intact class over time. The participants were 64 English majors who were assigned randomly into three treatment conditions that gave and received computer-mediated corrective feedback while writing (track changes, word processor, and track changes and word processor) and one control group that neither gave nor received writing corrective feedback. Students sat to a pre-test (week 1), immediate post-test (week 8) and delayed post-test (week 12) in writing. Results showed decrease in mean error and mean feedback in students' writing performance related to correcting 11 major error types on immediate and delayed post-tests, indicating that there was a significant effect for feedback type in favour of the group that used a combination of track changes and word processor. There was also a significant effect for the computer-mediated corrective feedback over the control group. Moreover, results showed that both mean error and mean feedback decreased significantly from pre-test to post-test (week 1-week 8), indicating that overall, there was a significant change over time. (Contains 7 tables and 2 figures.)