Advancing the application of Rasch models to the level of tests: estimating person parameters independently of all test parameters. The project advances recent theoretical work on Rasch models which makes it possible to compare estimates of person locations on a scale independently of all properties of the tests to which they have responded. The project aims to circumvent the current methods of test equating where various distribution assumptions are implied.
Controlling empirical factors to measure educational achievement in invariant units. Over a million students take part in National and State assessments which are used to compare schools and to determine tertiary entrance. The aim of the project is to study and control the factors that affect the unit in which achievement is measured so that such comparisons and decisions are based on more valid and precise information.