Figure 1.
Schematic representation of Guttman/Rasch data structure.
Representation of the raw data (top) and after sorting of the columns (health states) and the rows (patients) in order to arrive at the hierarchical Guttman/Rasch data structures (the check mark indicates that this health state is preferred over the next health state, the cross mark indicates a misfit) (from: [33]).
Figure 2.
Data collection designs and response processes in measurement models.
Schematic representation of the different data collection designs in combination with the specific response process of these designs and the appropriate measurement models for these four combinations (combination of discrete choice model and Rasch model, block bounded by thick line is multi-attribute preference response model).
Figure 3.
Judgmental tasks used in measurement methods.
Schematic representation of the judgmental task for three health states by: A = conventional monadic measurement (SG, TTO) by a sample of the general population; B = conventional discrete choice task (paired comparison) by a sample of the general population; C = multi-attribute preference response model for individual patients (3 patients in this example, each assessing 2 nearby located health states).
Figure 4.
Example of a response task under the multi-attribute preference response (MAPR) model (based on EQ-5D description).