The internal correlation coefficient was calculated to assess the concordance between three doctors in the assessment of anxiety in 20 people. There was an absolute mismatch between the three physicians who used the two-way random effects models and the “Single Rater” unit, kappa = 0.2, p = 0.056. Compared to the reliability of the interrater, the ICC selection process of test and intrarater reliability is easier. The only question that must be asked is whether the actual application is based on a single measure or on the average of several measures. With respect to the choice of “model”, Shrout and Fleiss19 propose that the two-way mixed effects model be appropriate for testing the reliability of intrarater with multiple scores from the same assessor, as it is not wise to generalize an assessor`s scores to a larger population of assessors. Similarly, the two-way mixed effects model should also be used in test reliability studies, as repeated measures cannot be considered randomised samples.2 In addition, for test and intrarater reliability studies, the absolute definition of compliance should always be chosen, as the measurements would not be meaningful if there were no match between repeated measures. A clinical researcher has developed a new ultrasound-based method to quantify coliotic deconformity. Before using the new method for his routine clinical practice, he conducted a reliability study to assess the reliability of the tests and repeat tests. He recruited 35 scoliosis patients with a variety of deformities in a pediatric hospital and used his new method to measure their coliotic deformity.
Measurements were repeated 3 times for each patient.