Measuring and Assessing Typing Skills in Writing Research

authored by: Luuk Van Waes, Mariëlle Leijten, Jens Roeser, Thierry Olive, Joachim Grabowski
Abstract: In keyboard writing, typing skills are considered an important prerequisite of proficient text production. We describe the design, implementation, and application of a standardized copy-typing task in order to measure and assess individual typing fluency. A test-retest analysis indicates the instrument’s reliability. While the task has been developed across eleven different languages and the related keyboard layouts, we here refer to a corpus of Dutch copy tasks (N = 1682). Analyses show that copying speed non-linearly varies with age. Bayesian analyses reveal differences in the typing performance and the underlying distributions of inter-key intervals between the different task components (e.g., lexical vs. non-lexical materials; high-frequent vs. lowfrequent bigrams). Based on these findings it is strongly recommended to include copy-task measures in the analysis of keystroke logging data in writing studies. This supports a better comparability and interpretability of keystroke data from more complex or communicatively-embedded writing tasks across individuals. Further potential applications of the copy task for writing research are explained and discussed.
Organisation(s): Institute of Psychology
External Organisation(s): University of Antwerp (UAntwerpen)
Nottingham Trent University
Universite de Poitiers
Type: Article
Journal: Journal of Writing Research
Volume: 13
Pages: 107-153
No. of pages: 47
ISSN: 2030-1006
Publication date: 2021
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Education, Language and Linguistics, Linguistics and Language, Literature and Literary Theory
Electronic version(s): https://doi.org/10.17239/JOWR-2021.13.01.04 (Access: Open)

BibTeX

@article{708992632bcc47b0abdd0e8e1227577e,
title = "Measuring and Assessing Typing Skills in Writing Research",
abstract = "In keyboard writing, typing skills are considered an important prerequisite of proficient text production. We describe the design, implementation, and application of a standardized copy-typing task in order to measure and assess individual typing fluency. A test-retest analysis indicates the instrument{\textquoteright}s reliability. While the task has been developed across eleven different languages and the related keyboard layouts, we here refer to a corpus of Dutch copy tasks (N = 1682). Analyses show that copying speed non-linearly varies with age. Bayesian analyses reveal differences in the typing performance and the underlying distributions of inter-key intervals between the different task components (e.g., lexical vs. non-lexical materials; high-frequent vs. lowfrequent bigrams). Based on these findings it is strongly recommended to include copy-task measures in the analysis of keystroke logging data in writing studies. This supports a better comparability and interpretability of keystroke data from more complex or communicatively-embedded writing tasks across individuals. Further potential applications of the copy task for writing research are explained and discussed.",
keywords = "copy task, motor skills, transcription processes, typing skills, writing fluency, writing processes",
author = "Waes, {Luuk Van} and Mari{\"e}lle Leijten and Jens Roeser and Thierry Olive and Joachim Grabowski",
note = "Funding Information: For the technical development of the Inputlog copy task, we would like to thank Tom Pauwaert, Eric Van Horenbeeck and Sebastian Fierens. For collecting the Dutch corpus of copy tasks, we would like to thank all the researchers that contributed to this corpus. Especially thanks to Nina Vandermeulen (NWO-LIFT project), Lise Paesen and Catherine Meulemans (FWO/BOF project) and master students of Multilingual Professional Communication (University of Antwerp) for making their copy task data available for this study. We also thank Thomas Quinlan for proofreading an earlier version of this manuscript. Within the context of the European literacy network (ELN) COST Network (IS1401) and EarlyWritePro (Developing methods for understanding early writing through analysis of process disfluencies) the copy task has been developed in different languages. We thank Lise Fontaine (Cardiff University, UK), Mark Torrance (Nottingham University, UK), Thierry Olive (CNRS at University of Poitiers, France), Esther Breuer (University of Cologne, Germany), Olga Witczak (Adam Mickiewicz University, Poland), Teresa Limpo (University of Porto, Portugal), Anna Sala (University of Barcelona, Spain), ?sa Wengelin (University of Gothenburg, Sweden), Victoria Johansson (University of Lund, Sweden), Gulay Tiryakioglu (University of Lyon 2, France), Vibeke R?nneberg (University of Stavanger, Norway), Anne S?tersdal Myklestad (Western Norway University of Applied Sciences, Norway) and Alessandra Rossetti (University of Antwerp, Belgium). ",
year = "2021",
doi = "10.17239/JOWR-2021.13.01.04",
language = "English",
volume = "13",
pages = "107--153",
journal = "Journal of Writing Research",
issn = "2030-1006",
publisher = "University of Antwerp",
number = "1",
}