CALF Analysis of a Learner Corpus of English as a Second Language Students: Research Design Perspectives

Abstract

This study explores the application of CALF (Complexity, Accuracy, Lexical Complexity, and Fluency) measures in analysing the L2 writing. It replicates and expands on previous CALF research, addressing methodological and design considerations using the Swedish Learner English Corpus (SLEC). Using computational linguistics tools such as the L2 Syntactic Complexity Analyzer (L2SCA) and Coh-Metrix, the research evaluates texts from Swedish high school students across multiple dimensions. Through comparison with previous studies, it highlights genre-related differences and contextualises findings within pedagogical and linguistic theory frameworks. Despite limitations such as small sample size and motivational variances, the results contribute valuable insights into the interplay between CALF measures, L2 proficiency, and writing quality. Future research directions include refining learner corpus design and exploring the relationship between CALF measures and genre-specific writing quality.