PUB - Publikationen an der Universität Bielefeld

Linguistic features as a task-related feature influence the difficulty of mathematical tasks. To reduce this influence (e.g., in testing situations), studies on linguistic simplification focus on modifying linguistic features. These studies show little or no effect on increasing test performance. An open question is whether a quantitative–exploratory approach with texts from a specific domain can be an additional model for reducing the linguistic influence on mathematical tasks. To answer this question, generalized linear mixed models were used to determine the effects of linguistic factors, the requirements of the items, and the effects of linguistic factors when differentiating the requirements of the items, while controlling for further person- and item-related effects. The results show that linguistic factors can have either a negative or positive influence on test performance. The findings indicate that for mathematics assessments and teaching, it might be essential to consider the influence of language factors and task requirements.