Quantitative measurement methods must be precise and accurate to justify their clinical use. The accuracy reflects the difference in measurement groups compared to another, often expressed in compliance percentages, standard measurement errors, coefficients of variation or Bland-Altman diagram. We propose a variance component analysis (VCA) to assess the impact of errors due to certain elements of a pet scan (scanner, time, observer, etc.) to express the composite uncertainty of repeated measurements and obtain relevant repeatability coefficients (RCs) presenting a unique relationship with Bland-Altman plots. Here we present this approach to evaluating intra- and inter-observational variations with PET/CT, illustrated by data from two clinical trials. Cohens – can also be used if the same counsellor evaluates the same patients at two times (for example. B to 2 weeks apart) or, in the example above, re-evaluated the same response sheets after 2 weeks. Its limitations are: (i) it does not take into account the magnitude of the differences, so it is unsuitable for ordinal data, (ii) it cannot be used if there are more than two advisors, and (iii) it does not distinguish between agreement for positive and negative results – which can be important in clinical situations (for example. B misdiagnosing a disease or falsely excluding them can have different consequences). The statistic of – can take values from 1 to 1 and is interpreted arbitrarily as follows: 0 – concordance that corresponds to chance; 0.10-0.20 – light approval; 0.21-0.40 – fair agreement; 0.41-0.60 – moderate support; 0.61-0.80 – substantial agreement; 0.81-0.99 – near-perfect chord; and 1.00 – perfect chord. The negative results suggest that the observed agreement is worse than one might expect. An alternative interpretation is that Kappa values below 0.60 indicate a considerable degree of disagreement. Modeling a more complex situation, which takes into account both fixed and random effects, naturally leads to a mixed effects model, as in our study 2. Here we used VCA to provide relevant RCs.
However, the estimate of fixed effects and random components was subject to great uncertainties that were reflected in the latitudes of 95% of 1000 CI. In general, estimating the components of variance requires larger sampling sizes than estimating fixed effects, since the former are the basis of the second moments and the first strokes of theatre, the first moments of random variables . How many observations are sufficient to demonstrate the concordance?. Subsequent extensions of the approach included versions that could deal with “under-credits” and ordinal scales.  These extensions converge with the intra-class correlation family (ICC), which allows us to estimate reliability for each level of measurement, from the notion (kappa) to the ordinal (or ICC) at the interval (ICC or ordinal kappa) and the ratio (ICC).