To be analyzed with the GLM, signals must satisfy the assumption of normal distribution [32]; however, the actual responses of regional cerebral blood flow (rCBF) are not necessarily normally distributed.
However, unlike fMRI and PET, various adjustments of experimental design analysis are required in NIRS studies, because of the following issues: Issue 1 To be analyzed with the GLM, signals must satisfy the assumption of normal distribution [32]; however, the actual responses of regional cerebral blood flow (rCBF) are not necessarily normally distributed.
Under the assumption of normal distribution, each eigenvalue can be described by its expectation and variance.
The assumption of normal distribution of tested variables was assessed with the Shapiro Wilk test.
The assumption of normal distribution variables was checked visually by Q–Q plots.
The assumption of normal distribution of all data was investigated by the Shapiro-Wilk method.
The method presented relies on the assumption of normal distribution of this dataset.
The data contradicts the assumption of normal distribution and therefore was analyzed with related-samples Wilcoxon signed ranked test.
Logarithmic transformation of the measured dose was required in order to respect the model's assumption of normal distribution of residuals.
Another limitation of our study was the assumption of normal distribution for our data for statistical tests of significance.
In contrast, the statistical and graphical tools of exploratory data are not based on the assumption of normal distribution of data.
