A methodology for longitudinal research on EFL written production: Capturing writing multidimensionality by combining qualitative and quantitative procedures

Abstract

[eng] This article critically examines the most challenging aspects of the methodology used in a broader research endeavour aimed at measuring the longitudinal impact of Content Language Integrated Learning (CLIL) on the compositions in English as a Foreign Language (EFL) produced by secondary education learners. A pretest/posttest design with repeated measures was used, which appears to be the only way to control for differences in task complexity (Larsen-Freeman 2006) despite claims of carry-over effects (Abbuhl et al. 2014). The major methodological concern was to describe writing in all its complexity and multidimensionality (Alanen et al. 2010). The decisions regarding the communicative written tasks administered, the building of the corpus, and especially the mixed methods for evaluating the written production obtained were made accordingly. All things considered, a combination of quantitative and qualitative methods to measure and characterize EFL writing is advocated in order to capture all its richness.