The Internal Validity of Discrete Choice Experiment Data: A Testing Tool for Quantitative Assessments

Area of Science:

• Health Economics
• Health Services Research
• Biostatistics

Background:

• Discrete Choice Experiments (DCEs) are widely used in health research to understand preferences.
• Ensuring the internal validity of DCE data is crucial for reliable results.
• Existing tools for validity testing may not be readily accessible or user-friendly.

Purpose of the Study:

• To develop and deploy a user-friendly tool for assessing the internal validity of DCE data.
• To collect and analyze summary test results from health researchers.
• To demonstrate the tool's utility across diverse health research applications.

Main Methods:

• A previously developed Gauss program was translated and adapted into MATLAB.
• The MATLAB tool was compiled and deployed for researcher use.
• Sixty-seven authors of published DCE studies were invited to use the tool and submit test summaries.

Main Results:

• Twenty-one researchers from 10 countries submitted results from 55 DCE datasets.
• Attribute dominance was the most frequently applied test, while stability showed the highest failure rate.
• Internal validity checks were feasible for most datasets, though rarely reported.

Conclusions:

• The developed tool facilitates the evaluation of internal validity in DCE data.
• Improved reporting and design of DCE studies can be achieved through accessible validity testing tools.
• Wider adoption of such tools can enhance the quality and interpretation of stated preference data in health research.