Abstract

The current study explored the longitudinal development of lexical cohesion for speakers of English as a foreign language (EFL) at the group level by employing Latent Class Growth Modelling (LCGM). A convenience sample of 276 university students from Southwest China were recruited and invited to write four argumentative essays over four months. Their essays formed a learner corpus, which was utilized to construct unconditional latent class growth models to explore the latent classes of EFL learners' development trajectories of lexical cohesion in their writing. Besides, conditional LCGM including logistic regression analysis was employed to investigate the effect of English proficiency on the classifications. The unconditional models demonstrated that there were different latent classes of development trajectories for local and global cohesion indices, which supports the heterogeneity of their development trajectories. Those models also revealed that two categories were the optimal ones for latent classes of development trajectories for selected local and global lexical cohesion indices. Additionally, logistic regression analyses showed that English proficiency affected latent classes of development trajectories of only one global cohesion index. The implications for EFL/L2 writing instruction and language development research are also discussed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call