Normalized data may have limitations in determining the reliability of MVC measurements

Measurement data serve as an objective basis for scientific findings. Therefore, their reliability in repeated measurements is a crucial prerequisite. The repeatability of measurements is quantified using various reliability parameters. Often, data are subjected to normalization procedures to reduce inter-individual variability or to improve interpretability. In our study, we aimed to investigate the extent to which the application of normalization has impact on the determined reliability parameters. This was examined using the example of maximum force values for trunk extension and flexion. For this purpose, 85 healthy individuals (42 women) were subjected to maximum isometric force tests of the trunk muscles at two-weeks intervals. The calculated reliability and agreement parameters included the intraclass correlation coefficient (ICC), the standard error of measurement (SEm), the standard error of the mean (SEM), and the coefficient of variation of the method error (CVME). The ICC values consistently indicated good reliability for the original data (Extension: 0.889, Flexion: 0.882). Within the subpopulations of women and men, lower but still good ICC levels were observed (Women Extension: 0.671, Flexion: 0.826; Men Extension: 0.819, Flexion: 0.680). For the anthropometrically normalized force values, lower ICC levels were found for the entire group (Extension: 0.803, Flexion: 0.768) although the deviations in the subpopulations were different (Women Extension: 0.747, Flexion: 0.805; Men Extension: 0.852, Flexion: 0.642). Thus, normalization of measured values leads to varying deviations of calculated ICC levels compared to the original data, which is why the use of standardized data is not recommended for reliability calculations.