The Rasch model, additive conjoint measurement, and new models of probabilistic measurement theory.

摘要: Abstract This research describes some of the similarities and differences between additive conjoint measurement (a type fundamental measurement) Rasch model. It seems that there are many two frameworks, however, their nontrivial. For instance, while specifies scales using a data-free, non-numerical axiomatic frame reference, model numerical reference is, by definition, data dependent. In order to circumvent difficulties can be realistically imposed this dependence, formalizes new non-parametric item response models. These models probabilistic theory in sense they explicitly integrate ideas with statistical order-restricted inference Markov Chain Monte Carlo. The specifications these rather flexible, as represent any one several used psychometrics, such Mokken's (1971) monotone homogeneity model, Scheiblechner's (1995) isotonic ordinal or (1960) proposed applied analyze both real simulated sets.