An overview of the family of rasch measurement models
暂无分享,去创建一个
The family of Rasch measurement models is a way to make sense of the world. Experience is continuous. But the moment we notice experience, it becomes discrete. We sense the fragrance of flowers. The sensation is continuous. But when we distinguish between flowers—with and without fragrance; strong from weak fragrance, fragrance we like, don’t mind, or dislike, then our observations become discrete. As we notice and remember particulars, we begin the counting that can become measurement. Counting is never accidental. It is always underpinned by the intention of replication. But replication is never exact. Its approximation depends on the situation, how much we care and what we are going to do with the count. A vacationer may count seashells according to size, shape or color. But an Aboriginal would count them according to whether or not their contents were edible. Any idea that all seashells are sufficiently identical to be counted is based on a fiction that each shell makes an equal contribution to an intention—which for practical purposes we keep constant. INTRODUCTION TO RASCH MEASUREMENT
[1] D. Andrich. A rating formulation for ordered response categories , 1978 .
[2] D. Andrich. A General Form of Rasch's Extended Logistic Model for Partial Credit Scoring , 1988 .
[3] J. Linacre,et al. Many-facet Rasch measurement , 1994 .
[4] J. Linacre. Detecting multidimensionality: which residual data-type works best? , 1998, Journal of outcome measurement.
[5] D. Andrich. Rating Scale Analysis , 1999 .