ment. The exam includes some transfer tasks that have been observed to be highly motivating for students (Morland, 1994). For instance, there is a task that assesses whether students can read articles dealing with applications of physics that lie outside the confines of the syllabus. Students know they will be presented with an article they have not seen before on a topic not specified in the syllabus, but that it will be at a level they should be able to understand on the basis of the core work of the syllabus. This task assesses students’ competency in applying their understanding in a new context in the process of learning new material. The only way for students to prepare for this activity is to read a large variety of articles and work systematically to understand them.
Another goal of the U.K. physics curriculum is to develop students’ capacity to carry out experimental investigations on novel problems. Students are presented with a scientific problem that is not included in the routine curriculum materials and must design an experiment, select and appropriately use equipment and procedures to implement the design, collect and analyze data, and interpret the data. Again, the only way students can prepare for this task is by engaging in a variety of such investigations and learning how to take responsibility for their design, implementation, and interpretation. In the United Kingdom, these portions of the physics exam are administered by the student’s own teacher, with national, standardized procedures in place for ensuring and checking fairness and rigor. When this examination was first introduced in the early 1970s, it was uncommon in classrooms to have students read on topics outside the syllabus and design and conduct their own investigations. The physics exam has supported the message, also conveyed by the curriculum, that these activities are essential, and as a result students taking the exam have had the opportunity to engage in such activities in the course of their study (Tebbutt, 1981).
In Chapters 4 and 5, we illustrated some of the kinds of information that could be obtained by reporting large-scale assessment results in relation to developmental progress maps or other types of learning models. Assessment results should describe student performance in terms of different states and levels of competence in the domain. Typical learning pathways should be displayed and made as recognizable as possible to educators, students, and the public.
Large-scale assessments of individual achievement could be improved by focusing on the potential for providing feedback that not only measures but also enhances future learning. Assessments can be designed to say both that this person is unqualified to move on and that this person’s difficulty lies