Models in collaborative writing among CLIL learners of English in primary school: Linguistic outcomes and motivation matters. (November 2022)

• Record Type: Journal Article
• Authors: Villarreal, Izaskun; Lázaro-Ibarrola, Amparo
• Is Part Of: System. Volume 110(2022)
• Publication Date: 2022-11
• Subjects: Models -- CAF -- YLs -- WCF -- Motivation -- Collaborative writing -- Complexity -- Accuracy -- Fluency -- Thermometers; Language and languages -- Study and teaching -- Periodicals; Langage et langues -- Étude et enseignement -- Périodiques; Electronic journals; 407
• DOI: 10.1016/j.system.2022.102922
• Languages: English
• ISSNs: 0346-251X

Abstract

Abstract: Model texts in the context of collaborative writing with young learners (YLs) have been shown to promote noticing and incorporation of features in students' subsequent drafts. However, few studies have analysed the quality of the drafts and even fewer have considered task motivation, despite its great impact on task completion. To fill in these gaps, our study measured draft quality and task motivation (with thermometers gathering students' ratings and motives) with Spanish YLs of English in primary school. Thirteen pairs of students (aged 10–11) wrote the same composition twice (pre-test and post-test) divided into a control group (CG), which received no feedback, and a model group (MG), which worked with model texts. Draft quality analyses revealed non-significant and medium-sized effects in syntactic complexity (clauses to T-units, dependent clauses to clauses and T-units, r = 0.44) and fluency (number of clauses, r = 0.32) in the MG. Meanwhile, a medium magnitude effect was detected in accuracy gains for the CG (L1 structures, L1 structures to clauses and lexical errors to words). Regarding task motivation, students' ratings were high, mainly justified by peer work and significantly higher after-task and across tasks in both groups, although the MG evidenced a significant and medium-sized ( r = 0.53) drop in the post-test. Our findings tentatively support the potential of models to generate upgraded texts but also suggest a possible motivational cost worth further investigation. More generally, our study enhances the motivational value of collaboration among YLs.