How much feedback is enough? Error treatment in second language writing
Abstract
How to respond to errors in student writing has long represented a topic of controversy in second language writing research. While some scholars have argued that error treatment does not contribute to accuracy in student writing and may, in fact, impede second language acquisition (i.e. Krashen, 1982; Truscott, 2007 & 2008), many empirical studies have found evidence in support of it (i.e. Lalande, 1982; Robb, Ross, & Shortreed, 1986; Fatham & Whalley, 1990; Ashwell, 2000; Ferris & Roberts, 2001; Chandler, 2003; Sheen, 2007; Bitchener & Knoch, 2009a & 2009b; Van Beuningen, De Jong, & Kuiken; 2009). Among those scholars who do believe that error treatment is effective, most advocate that instructors take a selective approach to marking errors (i.e. Bates, Lane, & Lange, 1993; Ferris & Hedgcock, 2005; Ellis, Sheen, Murakami, and Takashima, 2008; Lee, 2011). That is, rather than mark every error that occurs in a student text, instructors should mark only those that most impede understanding. The benefits of such an approach are that it saves an instructor time, allows students to see patterns in their errors, and facilitates independent editing skills. However, because advice from literature regarding which and how many errors to mark can be nebulous and even contradictory, some instructors can have difficulty implementing a selective approach. In addition, surveys of student attitudes toward error treatment found that they overwhelmingly prefer that errors are marked comprehensively (i.e. Leki, 1991b; Oladejo, 1993; Lee, 2005). The purpose of this qualitative study was therefore to investigate (1) to what extent instructors of second language writing marked errors in student work and why and (2) student attitudes toward selective and comprehensive error treatment. The participants included three instructors and 19 students of First Year Composition of International Students at Purdue University. Interviews revealed that the three instructor participants each differed in how much feedback they provide, but that their approaches were flexible and context-dependent. Reflecting previous studies, the student participants in this investigation also preferred comprehensive error treatment but reported being satisfied with the approach of an instructor who used a selective approach. Additional findings show that there are discrepancies in how instructors and students of the same class describe the instructor's approach to error treatment and that students rely overwhelmingly on instructor feedback when editing. Pedagogical implications are included.
Degree
Ph.D.
Advisors
Silva, Purdue University.
Subject Area
English as a Second Language|Rhetoric
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.