A Sequential IRT Model for Multiple-Choice Items and a Multidimensional Extension.

Area of Science:

• Educational Measurement
• Psychometrics
• Cognitive Psychology

Background:

• Traditional multiple-choice models often assume simultaneous comparison of all options.
• Alternative theories suggest a stepwise evaluation process for response options.
• Existing models may not fully capture the cognitive steps involved in complex item responses.

Purpose of the Study:

• To introduce and evaluate a sequential response model for multiple-choice items (SRM-MC).
• To model distinct proficiencies associated with separate cognitive steps in item solving.
• To compare the proposed SRM-MC against established simultaneous response models.

Main Methods:

• Development of a sequential response model for multiple-choice items (SRM-MC).
• Application of the SRM-MC to a sentence correction test.
• Estimation of model parameters using a fully Bayesian approach.
• Empirical comparison of SRM-MC against competing statistical models.

Main Results:

• Empirical data supported the proposed sequential response model.
• The model demonstrated distinct proficiencies for error recognition and error correction steps.
• The SRM-MC provided a better fit to the data compared to simultaneous models.

Conclusions:

• The sequential evaluation of response options is a viable process for certain multiple-choice tests.
• The SRM-MC effectively models distinct cognitive steps and associated proficiencies.
• This approach enhances the understanding of item response processes in educational and psychological assessments.