Combining Item Purification and Multiple Comparison Adjustment Methods in Detection of Differential Item Functioning.

Area of Science:

• Psychometrics
• Educational Measurement
• Statistical Modeling

Background:

• Differential item functioning (DIF) detection methods often assume some items are DIF-free for item purification.
• Controlling for multiple comparisons is crucial in DIF analysis to prevent false positives.
• The interplay between item purification and multiple comparison adjustment in DIF detection is not fully understood.

Purpose of the Study:

• To investigate the impact of combining item purification and multiple comparison adjustment on DIF detection.
• To propose a novel iterative algorithm that integrates both item purification and multiple comparison adjustment.
• To evaluate the performance of the proposed algorithm using simulation studies and a real data example.

Main Methods:

• Development of an iterative algorithm that simultaneously performs item purification and adjusts for multiple comparisons.
• Simulation study to assess the properties and effectiveness of the proposed algorithm compared to existing methods.
• Application of the algorithm to a real data set to demonstrate its practical utility in DIF detection.

Main Results:

• The combined approach of item purification and multiple comparison adjustment significantly impacts DIF item detection.
• The newly proposed iterative algorithm demonstrates favorable properties in simulation studies.
• The method is effectively demonstrated on a real-world data example, showing its practical applicability.

Conclusions:

• Integrating item purification with multiple comparison adjustment offers a more robust approach to DIF detection.
• The proposed iterative algorithm provides a valuable tool for researchers seeking accurate DIF identification.
• This method enhances the reliability of psychometric analyses by improving the control of Type I errors in DIF detection.