Bayesian Methods for Measuring Agreement in Repeated Measurement Studies
Author Information
Author(s): Philip J Schluter
Primary Institution: Monash University
Hypothesis
Can Bayesian methods improve the assessment of agreement in repeated measurement method comparison studies?
Conclusion
The proposed Bayesian models allow for full parameter uncertainty and handle unbalanced or missing data effectively.
Supporting Evidence
- The study presents two examples illustrating the advantages of Bayesian methods in measuring agreement.
- The models can handle unbalanced data and provide credible intervals for estimates.
- Bayesian methods allow for the incorporation of prior information and can be generalized to complex study designs.
Takeaway
This study shows how to use special math methods to check if different ways of measuring something agree with each other, even when some data is missing.
Methodology
The study uses two multivariate hierarchical Bayesian models to analyze repeated measurement data.
Potential Biases
Potential biases may arise from the selection of subjects and the assumptions of normality in the data.
Limitations
The assumption of normality may not always hold, and the models may not be suitable for small sample sizes with high autocorrelation.
Participant Demographics
The study involved 85 subjects for the blood pressure measurements and 9 preschool children for the step count measurements.
Statistical Information
Confidence Interval
95% credible regions provided for estimates.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website