Harmonizing measurements: establishing a common metric via shared items across instruments

Area of Science:

• Psychometrics
• Developmental Psychology
• Data Science

Background:

• Instrument-specific metrics hinder cross-population comparisons.
• A new method is proposed to extract a common metric from different assessment instruments.
• This facilitates transforming original measurements into scores with a standard unit.

Purpose of the Study:

• To develop and validate a method for creating a common metric across dissimilar assessment instruments.
• To enable standardized scoring and population-level comparisons.
• To address the challenge of instrument-specific metrics in research.

Main Methods:

• Utilized existing data from four child development assessment instruments across three child samples.
• Developed "equate clusters" of items measuring similar behaviors differently.
• Employed a probability model and item characteristic curves to identify anchor items and psychometrically equivalent clusters.

Main Results:

• Simulations confirmed the feasibility of generating a common metric from varied instruments and respondent abilities.
• The method demonstrated successful application in early childhood development assessments.
• Equate clusters effectively linked different instruments.

Conclusions:

• Equate clusters offer an intuitive and flexible approach to establishing a common metric.
• This standardization allows for population-level comparisons on a global scale.
• The method transforms measurements to a standardized scale, enhancing data comparability.