No label defined (Q4539030)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1763378
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 1763378 |
Statements
4 July 2002
0 references
mental measurement
0 references
item responses
0 references
0.88044614
0 references
0 references
0.8666206
0 references
0.86175716
0 references
0.86175716
0 references
0.8534117
0 references
0 references
The book under review introduces to the theory and practice of mental measurement, based on item responses. A mental `latent' attitude or ability (e.g. transitive reasoning ability), say theta, has to be measured, based on answers given by the subject to questions/items designed by the researcher. In the case of binary, correct/incorrect possible answers, the basic result is that if the probability \(P(\theta;i)\) of correct answer increases with theta for all items \(i\) proposed, then different individuals can be (first-order stochastically) ordered by the total number of correct answers. If moreover different functions \(P(\theta;i)\) and \(P(\theta;j)\) do not cross, also the items can be ordered, on a `difficulty' scale. Nonparametric item response theory only assumes the above monotonicity of the \(P\) functions; parametric models assume specific, usually logistic, functional forms. The advantage of the former approach -- which the book is about -- is that items whose corresponding \(P\) functions do not fit a particular functional form but are nonetheless monotonic can be admitted, leading to a more precise final measurement result. The book includes worked out examples, and in the final part deals with items which call for many-valued (rather than two-valued) possible answers.
0 references