The effects of comprehensive written corrective feedback on the revision and acquisition of specific L2 forms

Perez Nunez, Antonio Blas

Permalink

https://hdl.handle.net/2142/88045 Copy

Description

Title

The effects of comprehensive written corrective feedback on the revision and acquisition of specific L2 forms

Author(s)

Perez Nunez, Antonio Blas

Issue Date

2015-07-15

Director of Research (if dissertation) or Advisor (if thesis)

Bowles, Melissa A.

Doctoral Committee Chair(s)

Bowles, Melissa

Committee Member(s)

• Montrul, Silvina A.
• Jegerski, Jill
• Polio, Charlene

Department of Study

Spanish and Portuguese

Discipline

Spanish

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• L2 writing
• Written Corrective Feedback
• Heritage Language Writing
• L2 acquisition
• Spanish as a Foreign Language
• Classroom Research

Abstract

The role of written corrective feedback (WCF) as an instrument to facilitate Second Language Acquisition (SLA) has attracted much attention over the past fifteen years. However, the results from existing research on the effects of WCF (e.g., Truscott, 2007; Van Beuningen et al., 2012) present contradictory findings due to important design limitations. First, studies on WCF have typically employed a pretest-posttest design and analyzed one or two new pieces of writing, which is clearly not sufficient when comparing the effects of a WCF treatment, as some authors have pointed out (e.g., Bruton, 2009a; Storch & Wigglesworth, 2010) since there are many instances in which the errors found in a new text bear no relation to the errors previously corrected. Consequently, their results can only assume that the students used the language knowledge gained from the feedback on a previous text (i.e., pretest) and applied it to a subsequent text (i.e., posttest). Similarly, the available research has not yet provided specific evidence on how grammar correction by means of WCF affects specific L2 forms and whether error revision actually leads to accuracy development or L2 acquisition. Furthermore, given that WCF research has solely focused on second language (L2) and foreign language (FL) learners, there is no information on whether heritage language (HL) learners can benefit from this type of focus-on-form intervention and if so, to what extent the effects are different for this population of learners. This study aims to fill these gaps by comparing for the first time the effects of error revision with and without comprehensive WCF (i.e., the most common type of error correction in writing classes, whereby multiple grammatical error types are corrected in each text) on the production, revision and accuracy development of specific linguistic features over time. Specifically, this investigation sought to answer the following questions: (1) Do students who do revision with and without comprehensive WCF differ in terms of error revision and accuracy development of specific linguistic forms over time? (2) Are there any differences between L2 and HL learners in terms of error revision and the accuracy development of specific linguistic forms over time? Additionally, Truscott’s claim (1996, 2007) that WCF could cause learners to focus on grammatical correctness at the expense of written fluency and complexity was also examined by comparing the participants on various measures of written complexity and fluency at the beginning and at the end of the WCF treatment. The participants were thirty-six learners (24 L2 and 12 HL) enrolled in a fifth-semester Spanish composition course at a public U.S. university. As part of the coursework, all learners completed a five-minute daily writing warm-up assignment three times per week at the beginning of each class session. The corpus for the study consisted of a total of 385 texts produced over the course of 11 sessions during a 4-week time span. The WCF group (N= 18) received indirect WCF (i.e., all errors were marked by the instructor by underlining the whole word) on these assignments. In contrast, the revision group (N=18) received no form corrections (as was standard practice in the course). Both groups were given five minutes at the beginning of each class to revise their writing from the previous class. The changes in the error rate of four structures - (a) regular/canonical gender marking, (b) irregular/non-canonical gender agreement, (c) omission of definite articles in obligatory contexts, and (d) the correct use of the present subjunctive - were measured and analyzed to compare the effects of the treatments on both L2 and HL learners. Overall, the WCF treatment led the students to make more accurate and varied morphosyntactic error revisions, whereas revision without WCF primarily led the students to make surface-level corrections (i.e., accent marks and spelling). The results also revealed that the group that received WCF significantly increased their accuracy in the correct production of definite articles in obligatory contexts; however, no differences were found in the accuracy development of the other three linguistic forms (canonical and non-canonical gender marking and the present subjunctive). Finally, a comparison between the WCF and no WCF conditions at the beginning and at the end of the treatment showed that error correction had no measurable effect on either the complexity (i.e., verbal density and lexical richness) or the fluency (i.e., number of words per minute) of their writing. This study makes various significant contributions to the WCF debate and the fields of SLA and L2 writing. The results and implications of this investigation provide the most exhaustive description of the factors that influence the efficacy of this teaching practice and explain them in terms of the acquisition process of each individual L2 form. It provides evidence of the limited effects of WCF and offers insight on how written production, error revision, and written feedback affect linguistic development at both group and individual levels by showing that for some forms, WCF can be effective not only to promote accurate revisions but also to increase learners’ written linguistic accuracy; however, it contextualizes the process in such a way as to show that WCF, as it is typically provided, is not a panacea that will fix any and all problems. Additionally, the data from L2 and HL learners show that differences in terms of proficiency and linguistic background may influence the extent to which these two populations could benefit from WCF and written production. Finally, the research methodology outlined in this study proposes a new framework that analyzes the role of WCF in production, revision and accuracy development, and also traces its effects on specific linguistic forms over time. This way, studies will not only be able to offer clearer evidence about the extent to which WCF can lead to accuracy development, but they will also serve to inform about and describe the true value of WCF in the SLA process.

Graduation Semester

2015-8

Type of Resource

text

Permalink

http://hdl.handle.net/2142/88045

Copyright and License Information

Copyright 2015 Antonio Perez Nunez

Owning Collections