Dufour, Dave (2017) Teaching Grammar Using a Parallel Concordancer: A feasibility study. Masters thesis, Concordia University.
Preview |
Text (application/pdf)
3MBDufour_MA_F2017.pdf - Accepted Version Available under License Spectrum Terms of Access. |
Abstract
Studies have illustrated several benefits of Data-driven learning (DDL), when students are researchers of the language using authentic language data (corpora), for learning grammar, including linking adverbials and phrasal verbs. However, few have examined the use of parallel concordancers, a tool that displays the same source-text in two languages side by side, although recent research points to potential benefits in helping learners notice L1/L2 differences. This study examined the feasibility of implementing a DDL approach to teaching L2 English grammar that presents a learning challenge to French L1 learners using a parallel concordancer, focusing on teachers’ and students’ perceptions on their training, the time needed, the completion of the tasks and the monitoring.
Three intact Cegep (college) ESL classes, taught by two teachers, used a French-English parallel concordancer, Tradooit, over a 6-week period. The teachers were first trained, and then created and taught three laboratory tasks emphasizing L1/L2 differences between French and English. During those labs, students completed a discovery task, a proofreading task and an investigation task to correct L1 interference errors. Student performance was measured through accuracy scores on these tasks, and student and teacher perceptions were recorded using observation notes, teacher logs, a post-study questionnaire administered to students, and semi-structured interviews. The results suggest that DDL can be successfully implemented and that students and teachers are willing to use DDL in the future. However, both students and teachers would have liked more guidance for selecting keywords and finding grammar patterns. Students were able to complete the labs within the time allotted, but with varying degrees of accuracy; some grammar features such as the tense-aspect system seemed to be more challenging. Teachers could also have received more feedback on the labs. Implications include better training on pattern hunting and wording grammar patterns, and a better use of feedback.
Divisions: | Concordia University > Faculty of Arts and Science > Education |
---|---|
Item Type: | Thesis (Masters) |
Authors: | Dufour, Dave |
Institution: | Concordia University |
Degree Name: | M.A. |
Program: | Applied Linguistics |
Date: | August 2017 |
Thesis Supervisor(s): | Collins, Laura |
Keywords: | Data-driven learning, DDL, grammar, parallel concordancer, language interference |
ID Code: | 982802 |
Deposited By: | DAVE DUFOUR |
Deposited On: | 09 Nov 2017 16:12 |
Last Modified: | 18 Jan 2018 17:55 |
Repository Staff Only: item control page