Looking beyond the answer: The code-breaking world of mathematics assessment

For many years high-stakes testing has been a contentious issue in the education field, and despite findings for and against the use of such tests, their prevalence remains. In an Australian context, the launch of the first National Assessment Program for Literacy and Numeracy (NAPLAN) in May 2008 intensified debate in this area. Since that time there has been a heightened awareness of using such tests to measure students mathematical ability and as a tool to evaluate teachers and schools performance. Even greater controversy was created when the Australian Government announced that schools results would be publicly reported upon and compared to others of similar demographics. With increased accountability comes great responsibility that the tests being used are achieving their intended rationale, in this case accurately measuring a childs level of numeracy. It was for this reason that one of the purposes of this thesis was to investigate the impact of current test item design on students mathematical performance.