Investigating the Patterns of English Learner Students' Academic Writing Performance and Students' Use of Feedback

• In Event: Teaching and Learning of Academic Language and Literacy for English Learners

Abstract

Perspectives & Significance
The writing standards of the Common Core State Standards (CCSS) for English language arts (ELA) describe that students should be able to construct coherent and effective writings in various genres (e.g., argumentative, informational/explanatory, and narrative text types). For English learners (ELs) who are still developing their English language proficiency (ELP), meeting the high academic language demands in these standards is inarguably challenging (Hakuta, Santos, & Fang, 2013). To support the improvement of EL academic writing skills, it is essential to understand ELs’ writing performance and the types of feedback that can lead to success. While a considerable amount of research has been conducted on mainstream K-12 students and adult second language learners concerning writing performance and feedback (e.g., Ashwell, 2000; Ellis et al., 2008; Friginal et al., 2014), little empirical research is currently available on K-12 ELs’ academic writing. This study attempts to address this research gap.

Purpose
This study focuses on the argumentative writing of middle-school ELs. Specifically, this study addresses the following research questions:
(1) What lexical-, syntactic-, and discourse-level patterns are present in the argumentative writing of ELs at different ELP levels?
(2) What are the common types of errors found in ELs’ argumentative writing?
(3) How do ELs use feedback to improve their argumentative writing?

Methods & Data
A total of five teachers (four ESL and one mainstream ELA) and their EL and non-EL students participated in this study (135 total ELs; beginning = 62, intermediate = 30, and advanced = 19 students and 24 non-ELs). To utilize an automated tool for generating feedback, the Criterion® Online Writing Evaluation was employed. The students first wrote an argumentative writing response for a given prompt. The students then received immediate feedback from Criterion, followed by their teacher’s feedback a day later. The students revised their writing based on the feedback they received. The students’ pre- and post-feedback essays were analyzed based on a coding scheme (i.e., lexical, syntactic, and discourse patterns and errors). While a pair of researchers conducted human ratings on argument structures and content development, automated tools (e.g., e-rater® and TextEvaluator™) were used to create a detailed linguistic profile for each essay. The feedback types were also categorized based on a coding scheme. Descriptive statistics and multiple ANOVAs were conducted to compare pre- and post-essays between ELs at different ELP levels and non-ELs.

Results
Preliminary analyses resulted in noticeable patterns in the use of academic words, sentence structures, and the amount of text written among ELs at different ELP levels and for non-ELs (i.e., increased academic language command with the increase of ELP). On average, both ELs and non-ELs improved their writing in the post-feedback essays, with differing usages of the various types of feedback. ELs at intermediate and advanced ELP levels tended to address syntactic features from Criterion as well as teacher content feedback. In this presentation, we will report on the detailed analyses and make practical implications towards improving academic writing instruction for ELs.

Authors

• Mikyung Kim Wolf, ETS

• Yuan Wang, ETS

• Saerhim Oh, Teachers College, Columbia University

• Fred Tsutagawa, Teachers College, Columbia University