Fig. 3From: Reliability in evaluator-based tests: using simulation-constructed models to determine contextually relevant agreement thresholdsThis scatterplot shows the average Krippendorff’s alpha and percent error values for each of the 49 simulated evaluators, with the size and shading of the markers representing random error and systematic error, respectively. An envelope exists which describes the upper-bound of percent error for any given value of Krippendorff’s alpha for this video analysis taskBack to article page