Summed-score linking using item response theory: application to depression measurement.

Area of Science:

• Psychological Measurement
• Psychometrics
• Mental Health Assessment

Background:

• The Center for Epidemiologic Studies Depression Scale (CES-D) is a widely used measure for assessing depression.
• Accurate and consistent measurement of depression risk is crucial for patient outcomes research.
• Linking different versions of assessment scales is essential for longitudinal studies and comparing data across different time points or populations.

Purpose of the Study:

• To present and demonstrate an item response theory (IRT) approach for test linking using summed scores.
• To calibrate a modified 23-item CES-D scale to the standard 20-item CES-D scale.
• To establish a method for translating scores between the two CES-D versions.

Main Methods:

• Utilized an item response theory (IRT) framework, specifically F. Samejima's graded IRT model.
• Simultaneously calibrated responses from 1,120 participants to both the original and modified CES-D versions.
• Linked the two scales using derived summed-score-to-IRT-score translation tables.

Main Results:

• The study successfully linked a modified 23-item CES-D to the standard 20-item CES-D using an IRT summed-score approach.
• A cut score of 16 on the standard CES-D was found to correspond most closely to a summed score of 20 on the modified version.
• The derived translation tables provided a practical method for score conversion.

Conclusions:

• The IRT summed-score approach is a straightforward, valid, and practical method for test linking.
• This methodology can be applied in various research settings requiring the comparison of scores from different but related assessment instruments.
• The findings facilitate more accurate depression risk assessment and outcome tracking in research.