LEARNER CORPUS RESEARCH IN SLA FRAMEWORK: A SYSTEMATIC REVIEW OF EXPERIMENTAL DESIGNS

Abstract

Learner Corpus Research (LCR) has become increasingly relevant to Second Language Acquisition (SLA), yet its systematic integration within experimental research designs remains relatively limited. This systematic review investigates how LCR has been incorporated into experimentally oriented SLA studies published between 2015 and March 2024. Guided by the PRISMA 2020 framework, ten studies were identified from the Scopus database that explicitly combined learner corpus data with experimental or quasi-experimental methods. The review examines these studies in terms of their targeted SLA constructs, types of learner data, methodological orientations, and analytical procedures. The findings show a strong preference for a quantitative approach and performance-based constructs, with particular emphasis on the complexity–accuracy–fluency (CAF) framework. The reviewed studies also show a clear preference for employing Natural Language Processing tools, particularly Coh-Metrix and syntactic complexity analysers such as SynLex. Across the reviewed studies, learner corpora consistently serve three key functions: accounting for experimentally observed effects, validating instructional outcomes through learners’ extended performance, and facilitating triangulation across multiple data sources. Despite its clear methodological and theoretical value, the integration of LCR into experimental SLA research is still uneven.