In this experimental study, two response formats (oral vs. written responses) are examined within a computer-based test environment. Both varieties assess German-as-a-Second-Language (GSL) competency of (pre-service) teachers, using video-based items. The study investi-gates the influence of the response format on correctness and comprehensiveness. Moreover, the study examines if the measured effects main-tain when GSL-relevant learning opportunities are controlled or if interaction effects between learning opportunities and the response mode appear. The random sample (N = 90) consists of students of educational studies (pre-service teachers, B. A. / M. Ed.). The results of this study show that the response mode has no statistically significant effect on the correctness but on the comprehensiveness of the responses. Oral re-sponses were statistically significant more comprehensive than written responses. Test persons who had more GSL-relevant learning opportuni-ties were more likely to answer correctly, regardless of the response mode. There are no interaction effects. The implication for the test construc-tion therefore is that oral response formats do not lead to unfair test results while testing students, compared to a written format.