Correlation does not imply agreement: A cautionary tale for researchers and reviewers. Issue 4 (6th July 2021)

• Record Type: Journal Article
• Authors: Edwards, Christopher; Allen, Heather; Chamunyonga, Crispen
• Is Part Of: Sonography. Volume 8:Issue 4(2021)
• Page Start: 185
• Page End: 190
• Publication Date: 2021-07-06
• Subjects: agreement -- biostatistics -- Bland–Altman plot -- correlation -- intra‐class correlation; Diagnostic ultrasonic imaging -- Periodicals; 616.07543
• DOI: 10.1002/sono.12276
• Languages: English
• ISSNs: 2202-8323

Abstract

Abstract: Sonography researchers are often confronted with questions concerning the relationship between measurements acquired in clinical practice. This educational commentary provides clarity on the definition of the terms correlation and agreement. We discuss the statistical tests used to assess correlation and outline some common pitfalls authors fall into when reporting these. We provide examples of the inaccurate use of these tests in the sonography literature and recommend better alternatives for assessing agreement. This review will benefit authors embarking on studies comparing measurements and reviewers considering manuscripts to ensure the methodology used is sufficient to justify the claims. The authors also provide plots created in the R statistical software package, and the supplementary data file supplied can be used and adapted for those wishing to perform similar statistical tests.