Refocusing writing style education? Relationships between stylistic lapses and the quality of Dutch secondary school students’ argumentative texts

Jimmy van Rijt, Brenda van den Broek, Huub van den Bergh

Research output: Contribution to journalArticleScientificpeer-review

Abstract

In Dutch L1 classrooms, style in non-fictional genres is typically taught by means of normative exercises in which students are tasked to identify stylistic lapses. Not much is known about the effectiveness of such exercises when teaching style. Unknown factors include what kinds of stylistic shortcomings are found in Dutch students’ writing, and how the occurrence of certain stylistic lapses relates to writing quality. The current study empirically explores these scarcely investigated issues. Teachers rated 125 argumentative texts written by tenth-grade pre-university students by means of comparative judgement. Additionally, these texts were manually analyzed to investigate the occurrence of stylistic lapses, taking into account stylistic lapses that are common in text books and other types of
style related errors. Multilevel regression analyses revealed that none of the stylistic lapses that students are typically educated in negatively influence text quality as evaluated by teachers. Two other style related lapses negatively predicted text quality: the use of detached phrases and mistakes in question mark use. A final model including those two predictors explained 11.1% of the variance in text quality. The article discusses the implications of these findings for non-fictional style education, suggesting that it might need to be refocused.
Original languageEnglish
Pages (from-to)299-330
JournalJournal of Writing Research
Early online date19 Nov 2021
DOIs
Publication statusPublished - 2022
Externally publishedYes

Fingerprint

Dive into the research topics of 'Refocusing writing style education? Relationships between stylistic lapses and the quality of Dutch secondary school students’ argumentative texts'. Together they form a unique fingerprint.

Cite this