Transcription: A Key Capability Underlying Effective Writing

• In Event: 54.030 - Translating Effective Research on Reading and Writing Into Practice at the Elementary Grades

Abstract

Transcription, which draws on spelling and handwriting (or typing) to externalize language in the form of written text, is a defining and necessary condition of writing. In behavioral observations of writers producing texts, it was noted that transcription does not proceed continuously from start to finish. Instead, it is well established that text production is built up through a succession of language bursts interspersed by long pauses (usually longer than 2 s). Recently, Alves and Limpo (2015) showed that, in Grades 2-7, the number of words a writer can put in a burst is influenced by handwriting fluency and spelling accuracy, and that longer bursts are associated to higher writing quality and fluency. Additionally, Alves et al. (2016) found that promoting handwriting fluency resulted not only in better texts but also in longer bursts during story writing. The current study aims to extend these results, by examining the putative mediating role of burst length and short pauses duration (between 30 ms and 2 s) in the transcription-writing link. Though there is compelling evidence showing that transcription constraints children’s writing (Limpo & Alves, 2013), little research has focused on the underlying mechanisms.
In here, we asked 174 second graders (Mage = 7.3, SD = 0.4; 93 girls) to perform the alphabet task during 60 s and to write a story with smartpens running the HandSpy logging software during 10 minutes. Using path analyses, we tested the direct and indirect effects of handwriting and spelling, via burst length and short pauses duration, on writing quality (Model 1) and writing fluency (Model 2). Both models fitted the data extremely well, with Model 1 explaining 46% of the variability in writing quality and Model 2 explaining 80% of the variability in writing fluency. Results for Model 1 showed that handwriting and spelling contributed to writing quality both directly and indirectly via burst length. Results for Model 2 showed that handwriting and spelling did not contribute directly to writing fluency, but only indirectly via burst length (a small indirect effect of handwriting through short pauses duration was also found). Importantly, for both models, the indirect effects of handwriting via bursts were stronger than that of spelling. Overall, this study suggested that the relationship between transcription and writing is mediated by written language bursts or said differently, by writers’ ability to sustain transcription and composition concurrently. These results support transcription as a foundational capability in writing and suggest that, in beginning writing, it influences text production by constraining the efficiency at which children can get their message onto the page.

Author

• Teresa Limpo, University of Porto