Assessing students’ competencies (either implicitly or explicitly) is a routine task for teachers, with many decisions – for example, concerning grading, ability grouping, and difficulty level of questions or materials – being based on these teacher judgments. Given this, judgment accuracy (often referred to as diagnostic competence) is regarded as a central element of teacher professionalism and is thought to affect students’ learning outcomes. So far, however, empirical evidence concerning teachers’ diagnostic competence is incomplete and partly contradictory. As argued throughout this paper, these mixed results can be attributed, to some extent, to the fact that judgment purposes and judgment demands are not adequately taken into account in current research literature.