Stịm·mig: Assessing prosodic comprehension in primary school

Abstract

The German speaking countries have tested L1 listening proficiency in large national assessment studies during the past decade. However, testing prosodic comprehension—that is, students' ability to understand prosodically encoded content—has remained a blind spot, primarily because test items focusing on this specific aspect of listening have been lacking. The project stịm·mig aims to fill this gap by developing and evaluating test items that measure students' ability to understand prosodically encoded content in auditory texts. In this article, we explain the basic process of item construction, and we present sample items to illustrate the item design. Thanks to the collaboration of the Institute for Educational Quality Improvement (IQB) in Berlin, we were able to administer and evaluate the items in a large pilot study in 252 third-grade classrooms (N = 4,893 students). The main goal of this large-scale assessment was to evaluate the suitability of the reading and listening items so that they might be used for national, large-scale assessment studies. We tested the effects of the presentation modes (written vs. auditory) of the stimulus texts and test items in a multiple matrix sampling design. Our findings show that prosodic com-prehension is a construct that is empirically distinguishable from both verbal comprehension and reading comprehension. However, more detailed analysis is needed to fully understand the structure of the prosodic comprehension construct.