Abstract:
Timed impromptu writing is widely used in the assessment of writing ability, but its validity has been questioned because of its "timed" feature. This study addressed this issue and investigated the effect of a timed testing condition on the quantity and quality of writing. The study involved 23 L2 learners sitting two essay writing tests: an untimed test followed by a timed one. The timed condition was operationalised as 70% of the time participants had spent on the untimed writing, so that the time allocated was an equal proportion of the time every learner needed. The essays were measured in terms of length, accuracy and complexity and were also analytically scored by a trained rater. The study found that the timed condition had a significant effect on the length and content quality, but did not impact on accuracy nor complexity of writing. Results suggest that time conditions might not be as much a threat to test validity as originally perceived. The study also portrays a more complex picture of the trade-off effect than that which has been described by current theories, and recommends further exploration on this issue.
