Task repetition and L2 written performance: A meta-analysis

Abstract

This meta-analysis investigates the effects of task repetition (TR) on the syntactic complexity, accuracy, lexical complexity, and fluency (CALF) of second language (L2) written production. The study included 31 studies, with 65 unique samples. Results from three-level meta-analytic models revealed that task repetition had a positive effect on all CALF constructs, with significant effect sizes for accuracy (d = 1.19), lexical complexity (d = 0.75), fluency (d = 0.50), and syntactic complexity (d = 0.37). These findings suggest that TR is particularly beneficial for accuracy and lexical complexity but less effective for fluency and syntactic complexity. Variance analysis indicated that differences between primary samples accounted for the largest portion of total variance in all CALF constructs. Further moderator analyses revealed that factors such as spacing interval, number of repetitions, and feedback moderated the effects: Longer spacing intervals were associated with smaller effects, while more repetitions generally resulted in larger effects. Compared to TR without feedback, TR with feedback showed a statistically significant advantage for accuracy, a non-significant advantage for syntactic complexity and fluency, and a disadvantage for lexical complexity. This study provides valuable insights into optimizing TR within task-based language teaching (TBLT) to enhance L2 written production.