Developing a Computerized Adaptive Test to Assess Stress in Chinese College Students

Area of Science:

• Psychometrics
• Psychological assessment
• Health psychology

Background:

• Stress is a prevalent issue impacting health and disease prevention.
• Accurate stress measurement is crucial for effective intervention.
• Existing methods may lack efficiency and precision.

Purpose of the Study:

• To develop and validate a Computerized Adaptive Test for Stress (CAT-S) based on item response theory (IRT).
• To create an efficient, accurate, and intelligent tool for stress measurement.
• To evaluate the psychometric properties and validity of the CAT-S.

Main Methods:

• Developed a 93-item stress item bank based on eight scales and psychometric evaluation.
• Simulated adaptive testing using the Expected a Posteriori (EAP) method for score estimation.
• Utilized greatest item information for item selection and various stopping rules for test termination.
• Assessed criterion validity using the Perceived Stress Scale (PSS) and Receiver Operating Curve (ROC) analysis.

Main Results:

• The final stress item bank demonstrated good psychometric quality.
• CAT-S scores showed high correlation with the item bank and PSS scores (r > 0.5).
• Area Under the Curve (AUC) exceeded 0.7 across all stopping rules, indicating good diagnostic accuracy.
• The CAT-S achieved high precision with a minimal number of items.

Conclusions:

• The CAT-S is a psychometrically sound instrument for stress measurement.
• It offers advantages of speed, accuracy, and efficiency compared to traditional methods.
• The CAT-S enables dynamic and intelligent assessment of stress.