Studies on peer collaboration, and on peer review in particular, have flourished in the last two decades in the field of second language writing. Although peer review enjoys sound theoretical support from both Sociocultural Theory (Vygotsky, 1978) and the Interaction Hypothesis (Long, 1996), its effectiveness in foreign language classrooms has been hotly debated. Whereas some studies found positive effects for peer revision, as measured by the number of corrections students made based on peer feedback (Mendonca & Johnson, 1994; Schmid, 1999; Villamil & de Guerrero, 1998), other researchers reported concerns regarding the learners' ability to detect errors and provide quality feedback (Connor & Asenavage, 1994; Hyland, 2000; Kasanga, 2001; Leki, 1990; Lockhart & Ng 1993; Nelson & Murphy, 1993). Moreover, numerous studies have suggested that students' perceptions of their interlocutors are key in determining the effectiveness of peer review in foreign language classrooms, as learners may be reluctant to accept suggestions from peers perceived as less proficient in the target language (Amores, 1997; Saito, 1994; Tsui & Ng, 2000; Zhang, 1995).
This talk will present results from a research study investigating: (a) what proportion of errors learners detect during peer editing, (b) what proportion of detected errors is corrected in a targetlike way, (c) what proportion of peer corrections is incorporated in subsequent individual writing tasks, and (d) the students' perceptions of their peers' linguistic abilities and helpfulness. The learners' audio-recorded interactions and the narratives produced by the learners were analyzed in conjunction with data from an attitudinal questionnaire completed after the peer-editing session.