Abstract
AbstractThe concept of measurement in which the magnitude of a property is quantified in a common unit relative to a specified origin is a deep abstraction. This chapter shows the application of measurement in a social science context where the motivation is transparency and equity rather than the advancement of scientific laws. However, to achieve these, the realization of measurement needs to be no less rigorous than it is in the advancement of scientific laws. Rasch measurement theory provides the basis for such rigor. The context in this chapter is competitive selection into universities in Western Australia based on a summary performance on a series of instruments which assess achievement in a range of discipline areas. Such selection tends to determine life opportunities; therefore to ensure consistency and fairness, performances on different instruments need to be transformed into measurements which are in the same, explicit unit relative to a specified origin. Because the illustrative context is complex, it is considered that the Rasch measurement theory applied in this chapter could be applied to a range of social contexts where assessments on different instruments need to be transformed to measurements in a common unit referenced to a common origin and where the focus is on making decisions at the person level.
Publisher
Springer International Publishing
Reference47 articles.
1. K. Alder, The Measure of All Things: The Seven-Year Odyssey and Hidden Error That Transformed the World (Free Press, 2002)
2. E.B. Andersen, Sufficient statistics and latent trait models. Psychometrika 42, 69–81 (1977)
3. D. Andrich, A rating formulation for ordered response categories. Psychometrika 43(4), 561–574 (1978)
4. D. Andrich, A structure of index and causal variables. Rasch Measur. Trans. 28(3), 1475–1477 (2014)
5. D. Andrich, The problem with the step metaphor for polytomous models for ordinal assessments. Educ. Meas. Issues Pract. 34(2), 8–14 (2015)