Abstract

Abstract Some texts are easy to write, others are a real struggle. This article presents a brief review of how fluency has generally been measured in writing research. In addition to these ‘traditional’ measures, we define a wide range of complementary measures that might be diagnostic of fluency, by taking also more process-related characteristics into account. These complementary measures are derived from keystroke logging data, which were collected from an experiment among 68 students who wrote two descriptive texts, one in their mother tongue and the other in their second language. By using correlation and principal component analyses, we have reduced the set of variables and created a new multidimensional model to better address the complexity of fluency in writing. This model consists of four dimensions: (a) production, (b) process variation, (c) revision, and (d) pause behavior. These four components together create a multidimensional perspective on writing, which enables us to differentiate between fluent and less fluent writers.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call