Exploring task type effects on L2 accuracy profiles: narrative vs. email writing by B1 Spanish EFL learners

Accuracy has been extensively analyzed in Learner Corpus Research. A significant body of work has explored how accuracy correlates with proficiency (e.g., Díez-Bedmar 2011, 2015, 2018; Gráf Hawkins Thewissen 2015). However, the impact of task type on L2 accuracy profiles has garnered relatively less attention. Exceptions include studies by Alexopoulou et al. (2017), Lan (2015), and Lyashevskaya et al. (2022), which highlight the importance of task design. This limited focus has restricted the identification of criterial features (Hawkins & Filipović 2012) which may differentiate error patterns across task types at the same proficiency level. This study aims to address this gap by analyzing the effect of task type on L2 accuracy profiles. The data comes from the FineDesc Learner Corpus, comprising written production from 100 Spanish learners of English at the CEFR B1 level. Each participant completed two tasks: writing an email and a narrative text, resulting in 100 emails and 100 narratives. The learners’ errors were manually annotated by two experienced annotators using the Louvain Error Tagging Manual. The annotated data were analyzed using non-parametric statistical tests to identify significant differences, while effect sizes were calculated for non-significant results (Plonsky 2015). The findings reveal that task type significantly influences L2 accuracy profiles. Narrative texts elicited a greater variety and frequency of errors compared to emails, with the verb phrase domain being particularly affected. This suggests that narratives impose greater linguistic demands, requiring learners to employ more complex structures. Additionally, task-specific challenges in error analysis were observed, as annotator agreement varied across task types. These variations underscore the importance of refining annotation practices to account for task-related influences. The results are interpreted through the lens of Task-Based Language Teaching (TBLT) (e.g., Skehan & Foster 2012), which emphasizes the relationship between task design and linguistic performance. The findings align with TBLT principles, suggesting that task complexity and communicative demands influence learners’ linguistic output and error profiles. This study makes several contributions to the field. First, it highlights the critical role of task type in shaping L2 accuracy profiles. By identifying criterial features that characterize accuracy within specific tasks, the research enhances our understanding of how task design impacts learner output. Second, the study provides methodological insights, emphasizing the need for task-sensitive approaches in error tagging and analysis. Finally, it offers practical implications for language teaching and assessment. For educators, the findings suggest that tasks should be designed to elicit specific linguistic features, enabling targeted feedback and instruction. Our results also highlight the importance of using task-sensitive rubrics in the evaluation of learner language. In conclusion, this study advances Learner Corpus Research by identifying the task type effect on the accuracy profiles of the same 100 B1 Spanish learners of English in two different task types, narrative writing and email writing, in the FineDesc Learner Corpus. The task effect on error analysis is also discussed. References Alexopoulou, T., Michel, M., Murakami, A., & Meurers, D. (2017). Task effects on linguistic complexity and accuracy: A large-scale learner corpus analysis employing natural language processing techniques. Language Learning, 67(1), 180–208. Díez-Bedmar, M. B. (2011). Spanish pre-university students' use of English: CEA results from the University Entrance Examination. International Journal of English Studies, 11(2), 141–158. 20 Díez-Bedmar, M. B. (2015). Article use and criterial features in Spanish EFL writing: A pilot study from CEFR Benjamins. A2 to B2 levels. In M. Callies & S. Götz (Eds.), Learner corpora in language testing and assessment (pp. 163–190). John Benjamins. Díez-Bedmar, M. B. (2018). Fine-tuning descriptors for CEFR B1 level: insights from learner corpora. ELT Journal, 72(2), 199–209. Gráf, T., & Huang, L. (2022). Persistent errors in spoken English among Taiwanese and Czech learners at CEFR B2 and C1. In A. Leǹko-Szymanska & S. Götz (Eds.), Complexity, accuracy and fluency in Learner Corpus Research (pp. 137–158). John Benjamins. Hawkins, J. A., & Filipović, L. (2012). Criterial features in L2 English: Specifying the reference levels of the Common European Framework. Cambridge University Press. Lan, N.T. (2015). The Effect of Task Type on Accuracy and Complexity in IELTS Academic Writing. VNU Journal of Science, 31(1), 45–63. Lyashevskaya, O., Vinogradova, O., & Scherbakova, A. (2022). Accuracy, syntactic complexity and task type at play in examination writing. In A. Leńko-Szymańska & S. Götz (Eds.), Complexity, accuracy and fluency in Learner Corpus Research (pp. 241– 272). John Benjamins. Plonsky, L. (2015). Statistical power, p values, descriptive statistics, and effect sizes: A “Back-to-Basics” Approach to Advancing Quantitative Methods in L2 Research. In L. Plonsky (Ed.), Advancing quantitative methods in second language research (pp. 23–45). Routledge. Thewissen, J. (2015). Accuracy across proficiency levels: A learner corpus approach. Presses universitaires de Louvain.