History teachers (N=22) and their 11th-grade students (N=451) rated the importanceof four types of questions (verbatim, paraphrase, inference and skill items) by indicating the degree to which these questions were to be expected in an upcoming teacher-made test about a particular textbook chapter. In addition, teachers classified their actual tests according to these four question types. Importance ratings of individual students showed a low correspondence with those of their teacher and no correspondence with the actual test questions (as classified by their teacher). However, teachers' importance ratings also showed a considerable number of discrepancies with the actual test questions. These findings suggest that in study-test situations, more clarity is needed about processing demands.