Note that if one of the tests is a reference standard or “gold standard”, the distortion is based on the difference between the result of the new test and the “actual value” of the measured size, and therefore on a precision measurement.10 In these cases, it can be said that the CCC measures both accuracy and consistency. However, if neither test is a gold standard, it is not appropriate to say that CCC also provides an accuracy measure. It is often interesting to know whether measurements made by two (sometimes more than two) different observers or by two different techniques give similar results. This is called concordance or concordance or reproducibility between measurements. Such an analysis considers the pairs of measurements, either categorical or both numerically, each pair having been made on an individual (or a pathology slide or an X-ray). Methods of assessing the concordance between observers according to the nature of the variables measured and the number of observers These two problems – knowledge of objectives and consideration of theory – are the main keys to a successful analysis of compliance data. Below are some other more specific questions regarding the choice of appropriate methods for a given study. Statistical methods of conformity assessment vary according to the nature of the variables studied and the number of observers between whom the concordance should be assessed. These are summarized in Table 2 and are explained below. For a 2×2 table, the percentages of concordances by category are the percentage of positive concordance and the percentage of negative concordance, and the symmetry test is reduced to the McNemar test . Kappa is not weighted and thus treats all different categories in the same way. A full review of compliance statistics and symmetry testing is available at [5, Chapters 8 and 11].
If test B is a gold standard, diagnostic accuracy is measured by (simplified on 2×2 tables): PPV = N22/N+2, NPV = N11/N+1, sensitivity = N22/N2+ and specificity = N11/N1+. Another method of visually assessing the conformity of two tests is to establish a scatter plot of the results of the first test against the results of the second test. If the two tests have a good match, we should expect the points to fall on or near the 45° line (i.e. y = x); Deviations from this line would indicate a mismatch.. . .