Keine Zeit für den C-Test?
Eine empirische Untersuchung zum Einfluss einer Geschwindigkeitskomponente auf das Konstrukt des C-Tests
Publishing year: 2019
The C-test has come to be known as an objective, reliable and valid means of measuring general language proficiency. Many studies support its interrelation with several language sub-skills: reading comprehension, listening comprehension, speaking, writing, vocabulary, grammar. By contrast, the S-C-test – a substantially speeded up version of this testing format – has not yet been investigated in depth. The purpose of the study at hand is to help fill this gap.
This thesis deals with the question concerning the influence the drastic reduction of time allotted to complete the test has on the C-test.
Underlying is an assumption made by Grotjahn et al. (2010) hypothesizing that adding a speed factor to the C-test, thus simulating real time language use, might lead to higher correlations with listening comprehension and speaking tasks.
After a theoretical discussion of the origin, construction, variants and applications of C-tests, studies dealing with correlational analyses or using a speeded C-test will be examined.
In the second part of this thesis, C-tests and speeded C-tests will be compared, with data gathered from adult learners of German in a university context. As will be seen, S-C-tests reach a sufficient reliability. In fact, correlational analyses show that S-C-tests are more closely related to listening comprehension than commonly used C-tests. As far as speaking is concerned, both test versions yield similar results. Correlations with both language skills are considerably stronger for advanced learners than for less advanced ones.
To sum up, the S-C-test appears to be a reliable test instrument that is appropriate for measuring foreign language competence in low stakes situations, especially for advanced learners.