How true is WYTIWYG? (What You Test Is What You Get) This tool brings together evidence of the effect of modes of assessment on teacher approaches in the classroom, and on student approaches to learning. In addition to reports on practice, two research studies are outlined. One is from a large-scale comparison of assessments and practices in two Australian states. The other is a study of the approaches to learning of students in higher education when they are assessed by essays, or by multiple choice tests.

The resemblance between the tasks in a system's high-stakes assessment and the focus of teaching and learning in most classrooms is not enough to show causal influence, either way. But the available evidence on introducing changes in such assessment suggests that the influence can be strong. In particular:

• When curriculum-embedded components are introduced into high-stakes assessment, they happen � teachers and students spend at least the expected time, working on the new component.
• When changes are made in external high-stakes assessment, introducing new task-types, this can have profound effects. The first research study reported, from Australia, showed that the introduction of non-routine problem solving into the Grade 11-12 assessment resulted in such work appearing in classrooms from Grade 8 upwards.

The evidence thus underlines the importance of the close alignment of any system's high-stakes assessment with its curriculum standards and learning goals.

The report also discusses the tendency of innovations to degrade in quality over time, the reasons for this and how it may be minimized. Download the extended report for a review of the evidence of the influence of high-stakes assessments on the pattern of classroom practice in schools � and thus of the likely effects of choices of assessment by system leadership on the implemented curriculum in their schools.