Different, or simply more difficult? The role of reduced time in the construct of computerised C-Tests in English and German
Anna Timukova, g.a.s.t. (Society for Academic Study Preparation and Test Development),Germany
Franziska Möller, g.a.s.t. (Society for Academic Study Preparation and Test Development), Germany
Anastasia Drackert, g.a.s.t. (Society for Academic Study Preparation and Test Development), Ruhr University of Bochum, Germany
https://doi.org/10.58379/USDT5701
Volume 15, Issue 1, 2026
Abstract: Speeded C-Tests are a time-constrained variation of the traditional C-Test, requiring test takers to complete partial word gaps in increasingly complex short texts within about two minutes per text, compared to the standard five. While the original C-Test is a well-established measure of global language proficiency, the speeded version – promising greater reliability and cost-effectiveness – remains underexplored. Reducing time typically increases test difficulty due to the speed-ability trade-off but may also shift the construct by altering the skills assessed. The study examined university students’ performance on computerised C-Tests in L2 English (N = 237) and German (N = 191) under two timing conditions. Time constraints significantly lowered scores in both languages, although the effect was small. Less proficient learners were more affected, but the difference in point loss was not statistically significant. An interaction between time and text difficulty was observed in German, but not in English. The speeded C-Test showed slightly stronger correlations with oral proficiency and marginally better predictive power in both samples, particularly among more advanced learners. In sum, the speeded C-Test is more difficult and appears to emphasise automatised, oral-like language skills, with the construct being sensitive to proficiency and (in German) text difficulty.
Keywords: global language proficiency, placement test, speeded testing, computerised language testing