A two-level logistic model for DIF detection in CAT improves control of spurious findings over single-level models in simulations, especially for short tests with low exposure rates.
Table B1 Statistics of examinees and provisional ability estimates (for part of the items)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.AP 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Dependencies in Item-Adaptive CAT Data and Differential Item Functioning Detection: A Multilevel Framework
A two-level logistic model for DIF detection in CAT improves control of spurious findings over single-level models in simulations, especially for short tests with low exposure rates.