A Two-Parameter Logistic Extension Model: An Efficient Variant of the Three-Parameter Logistic Model

Area of Science:

• Psychometrics
• Educational Measurement
• Item Response Theory

Background:

• The three-parameter logistic (3PL) model is widely used but relies on a fixed guessing parameter.
• Quantifying guessing behavior accurately is crucial for effective educational assessment.
• Existing models may not fully capture the nuances of examinee response strategies.

Purpose of the Study:

• To introduce and evaluate the two-parameter logistic extension (2PLE) model, a novel variant of the 3PL model.
• To compare the performance of the 2PLE model against the traditional 3PL model and other guessing models.
• To demonstrate the practical application of the 2PLE model using real-world data.

Main Methods:

• Developed the 2PLE model, which integrates item features based on examinee ability.
• Analyzed model properties and compared them with existing guessing models.
• Conducted a simulation study to assess the 2PLE model's performance against the 3PL model under various scenarios.
• Applied the 2PLE model to two real data sets for empirical validation.

Main Results:

• The 2PLE model demonstrated superior performance compared to the 3PL model in simulations.
• The model's probability of correct response increases with examinee ability.
• At extreme low ability levels, the 2PLE model approximates the 3PL model with a chance-level guessing probability.
• Empirical application showed the 2PLE model's effectiveness in comparison to existing models.

Conclusions:

• The 2PLE model offers a more refined approach to modeling guessing behavior in psychometric assessments.
• This enhanced model provides a more accurate estimation of item and examinee characteristics.
• The 2PLE model represents a significant advancement in item response theory, particularly for educational and psychological testing.