Signal-to-noise ratio (SNR) as a measure of reproducibility: Design, estimation, and application

Naser Elkum, M. M. Shoukri

Research output: Contribution to journalArticle

3 Citations (Scopus)


This paper proposes the use of signal-to-noise ratio (SNR) as another index of a measurement's reproducibility. We derive its maximum likelihood estimation and discuss confidence interval construction within the framework of the one-way random effect model. We investigate the validity of the approximate normal confidence interval by Monte-Carlo simulations. The paper also derives the optimal allocation for the number of subject and the number of repeated measurements needed to minimize the variance of the maximum likelihood estimator of the SNR. We discuss efficiency in estimation and cost considerations for the optimal allocation of the sample resources. The approach is illustrated on two examples: one from MRI data and the other on the WHO immunization coverage data.

Original languageEnglish
Pages (from-to)119-133
Number of pages15
JournalHealth Services and Outcomes Research Methodology
Issue number3
Publication statusPublished - 2008
Externally publishedYes



  • Delta method
  • Likelihood inference
  • Monte-Carlo simulations
  • Random effect model

ASJC Scopus subject areas

  • Health Policy
  • Public Health, Environmental and Occupational Health

Cite this