[Cohen's kappa or McNemar's test? A comparison of binary repeated measurements]

This article intends to illustrate the combination of McNemar's significance test and Cohen's kappa coefficient in the comparison of repeated binary measurements. Both methods are standard statistical tools of major relevance for the evaluation and comparison of clinical imaging methods and thus have an impact on the corresponding publications. The interpretation of results obtainable with these methods will be illustrated to facilitate their use based on recent statistical software. Examples will further outline limitations and possible pitfalls in their application to clinical data.