Response to Sabour, "reproducibility of computed tomography to evaluate ankle and hindfoot fusions; statistical issue to avoid misinterpretation".


Dear Editor: We appreciate the review of statistical limitations of the κ test as elaborated by Dr Sabour. In his examples using imaginary data with 2 observers, he demonstrates that when the total numbers of concordant and discordant observations are the same, the κ test can produce different results with varying proportions of concordant observations in… (More)
DOI: 10.1177/1071100714567089


  • Presentations referencing similar topics