This study examined how peer comments were reflected in students’ second drafts in a university general English course. The corpus comprised 1,116 comments received by 188 students, who were grouped into three levels according to their final course achievement. Each comment was traced across both drafts and coded into one of nine response categories, including direct implementation, broad uptake, possible broad uptake, and non-implementation. Content/logic comments were the most frequent comment type in all groups, and after correction for multiple comparisons, no pairwise between-group difference in the number of comments received was significant.
Implementation, including instances that students extended through self-initiated revision, accounted for 62.0% of all comments and 71.1% of implementable comments, whereas broad and possible broad uptake together accounted for 8.3% of all comments.
Second drafts were significantly longer than the corresponding first drafts. None of the uptake indicators differed significantly across achievement groups, although the confidence intervals were wide. An exploratory review of 66 traceable cases showed that some accurate comments were passed over and some inaccurate ones were declined; however, this subset could not indicate how often either pattern occurred. Broad uptake was concentrated in comments addressing content and organization. Because broad uptake can be attributed to specific comments only loosely, it should be reported separately from direct implementation.