A Concept- and Corpus-Based Instruction to Promote Effective Communication in Disciplinary Research Writing: Development of Sentence-Level Linguistic Knowledge in ESL/EFL Graduate Students
Restricted (Penn State Only)
- Author:
- Qiu, Xixin
- Graduate Program:
- Applied Linguistics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 30, 2024
- Committee Members:
- Tommaso Milani, Program Head/Chair
Meredith Doran, Major Field Member
Catherine Berdanier, Outside Unit & Field Member
Kevin McManus, Major Field Member
Xiaofei Lu, Chair & Dissertation Advisor - Keywords:
- concept-based language instruction
data-driven learning
engineering writing
L2 writing pedagogy
multilingual graduate students
subordination - Abstract:
- Effective written communication is paramount for modern engineers (National Academy of Engineering, 2020). With engineering continuing to enjoy the top place of choice for first-time international doctoral enrollment in the U.S. (Zhou & Gao, 2021), addressing the unique needs of multilingual doctoral students has become increasingly important. This study investigates the grammatical concept of subordination in Mechanical Engineering (ME) research writing using a two-phase study design: (1) identifying key sentence-level linguistic patterns (2) and assessing a concept- and corpus-based teaching intervention for multilingual ME graduate students. Phase one uses a three-pronged needs analysis, examining engineering textbooks, instructor feedback on student manuscripts, and a corpus-based comparison of subordination features between student and published ME research writing. Data were examined using content analysis to investigate both instructor feedback and writing textbooks on engineering writing. Additionally, corpus-based syntactic and discourse-functional comparisons were made using independent t-tests in order to find out differences among various linguistic features. Findings from this phase have brought forward four concepts that were taught in sentence-level linguistic features to five L1 Chinese ME graduate students. Data was collected during the span of five-week remote tutoring and two months post-tutoring which included Likert-scale ratings, concept definitions, in-class activity performance recordings, exit interviews, and post-intervention self-directed revisions complemented by stimulated recalls. These data have been analyzed using Five-Element Concept Analysis (FECA), Materialized Action Profile (MAP) as well as Tool-Mediated Verbalization Analysis (TMVA) through problem spaces (Yoon, 2016) to unravel the complex process of L2 conceptual development. The study discussed the effectiveness of integrating corpus analysis both as an indirect tool in developing instructional materials and as a direct component of the teaching intervention. Moreover, it investigates the impact of this integrated pedagogical strategy on developing conceptual knowledge of discipline-appropriate sentence structures and enhancing sentence-level written communication skills among multilingual graduate students in advanced disciplinary contexts. This assessment is conducted through a Vygotskian genetic approach, offering insights into the developmental progress of participants’ ability to appropriate the concepts intentionally in self-directed revisions.