Description:

Keynote Address by Dr. Erwin Tschirner (University of Leipzig)

Valid assessments have positive consequences for all stakeholders. They provide feedback for students, instructors, and departments, and help align goals and curricula. For language assessments to be useful, they need to be task-based and theory-based; i.e., they need to be based on a clear understanding of what students need to be able to do at what level of ability. The latter requires an understanding of how language abilities develop over time. The presentation will start with a
current model of task-based language assessment called Evidence-Centered Design (Mislevy et al., 2002), and will show how it may help to differentiate between useful and not so useful foreign language assessments. A presentation of a large-scale benchmarking study to gauge speaking, listening, and reading levels of college students will follow to provide a basis for setting achievable proficiency goals. Finally, implications of these findings for curricular renewal will be discussed,
outlining how assessments may be used to measure the results of renewal processes and provide ongoing quality control for achieving proficiency goals.