Abstract
Relationships between a mathematical measurement model and its real-world applications are discussed. A distinction is made between large data matrices commonly found in educational measurement and smaller matrices found in attitude and personality measurement. Nonparametric methods are evaluated for estimating item response functions and (unconditional and conditional) inter-item covariances.
| Original language | English |
|---|---|
| Pages (from-to) | 295-299 |
| Number of pages | 5 |
| Journal | Applied Psychological Measurement |
| Volume | 25 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - Sept-2001 |
Keywords
- inter-item covariance estimation
- item response function estimation
- Mokken model
- monotonicity of item response junctions
- nonparametric item response theory
- ordinal person measurement
- POLYTOMOUS IRT MODELS
- LATENT CLASS ANALYSIS
- UNIDIMENSIONALITY
- RELIABILITY
- SCORES
- TRAIT