Simulations were conducted to investigate factors that influence the performance of the logistic regression method in assessing differential item functioning (DIF) for within-multidimensional data. Several factors were manipulated when non-uniform DIF exists. It was found that the matching variable significantly influenced the accurately detection rates under the bidirectional DIF pattern.
|Publication status||Published - Apr 2015|
|Event||National Council on Measurement in Education 2015 Annual Meeting - InterContinental Hotel, Chicago, United States|
Duration: 15 Apr 2015 → 19 Apr 2015
|Conference||National Council on Measurement in Education 2015 Annual Meeting|
|Abbreviated title||NCME 2015|
|Period||15/04/15 → 19/04/15|