This study aims to explore the quantitative and qualitative learning performance of a class of composition in L2 French (junior students) in Taiwan by applying statistical methods such as SPSS and Excel. The students of a French department following a course of writing during the academic year 2015–2016 are chosen as the research object. The data collected includes students’ scores, class attendance, students’ composition assignments (almost one per week during two consecutive semesters), and several questionnaires. The results of numerical analysis are used to clarify whether our designed teaching methods can improve students’ writing skills. Through discussing the effect of teamwork, the indicators selected to evaluate students’ writing level, and the impact of writing topics, we attempt to figure out a flexible teaching/learning method suitable for different levels of students. The key evaluating indicators contributing to students’ good or poor writing ability are also discussed. Using mechanical error correction methods can notably help teachers identify students’ most common and recurrent mistakes. It also appears that students who are not native speakers prefer their instructor not only to systematically highlight their errors, but also to correct their French. The proposed learning improvement mechanism presented in this study may also be applied to other fields or other languages in future studies.
The European Conference on Language Learning 2017: Official Conference Proceedings