In our research project on diagnostic competencies of pre-service teacher students, item pools serve as the main interface between the case vignettes, the competence structure model and the measurement model. Therefore, this article provides details on the respective theoretical and empirical demands with regard to the generation of items. Special emphasis is laid on aspects concerning the open task format and the validity of the test construct. Besides presenting the item pools of the initial and final case, groups of items will be identified that build different content-based clusters.
