In the practice of peer feedback, it has been observed that writingrnteachers are not confident about the ability of their students inrnpracticing peer feedback on their peers' essays. It has also beenrncomplained that the peers themselves are reluctant to comment onrneach other's writing. Others, however, argue that students canrnsuccessfully practice peer feedback if they get appropriate training andrnguidance in how to exercise peer feedback in writing classes. Thernobjective of this study was therefore to investigate the effect ofrntraining students in giving and receiving peer feedback on the revisionrntypes and their writing quality .rnThe study focused on the case of ten second yea r students who wererntaking sophomore English language course. To gather information orrndata four essays of the succulents with two drafts each were mainlyrnused. These were to evaluate the actual change in the revision typesrnand writing quality. Questionnaires were also designed andrnadministered to both the instructors and the students to identify therncustoms of feedback practice on the students' writing and to check thernperception the students had before and after the training. The studyrntook a month through the four essays together with two drafts each. The result of the data presentation and analysis shows that sign ofrnimprovements were observed in the types of revision made on t heirrnclassmate's essays and the qualities of these essays were also realisedrnsince the writers used most of the feedback given by t heir peers onrnthe higher concerns of their writings. It can be concluded that this wasrnthe result of the tra ining and the guidance given by the teacher.rnSo, if students get proper t raining and continuous guidance in thernpractice and use of peer feedback or revision, they may be reliablernreviewers of each other's drafts of t heir essays. To avoid the wrongrnperception the students may have towards their classmates' commentsrnon their writing they might be briefed of the benefits of peer feedback.rnThe result of this study could also be more reliable if other relatedrnstudies could be made. Therefore we could suggest further studyrnwhich may substantiate the current finding .