Application of Bayesian Decision Theory in Detecting Test Fraud

Area of Science:

• Psychometrics
• Statistical Decision Theory

Background:

• Test fraud poses a significant challenge in educational and professional assessments.
• Existing methods for detecting test fraud may lack statistical rigor or simplicity.

Purpose of the Study:

• To propose a new, statistically grounded approach for test fraud detection.
• To develop a straightforward decision rule for identifying fraudulent examinees.

Main Methods:

• Utilizing Bayesian decision theory principles.
• Developing a decision rule based on computing posterior probabilities of test fraud.
• Applying the method to a dataset with confirmed instances of test fraud.

Main Results:

• The Bayesian approach yields a practical decision rule for test fraud detection.
• The method was successfully applied to a real dataset, demonstrating its applicability.

Conclusions:

• Bayesian decision theory offers a robust framework for identifying test fraud.
• The proposed method provides an effective and computationally simple tool for assessment integrity.