For each cognitive measure, outliers above or below 3 SD from the mean were excluded.
Measures: LOS outliers were defined as cases with a LOS exceeding the geometric mean plus two standard deviations of all the stays in the same DRG.
For all remaining experimental measures, no outliers were observed three standard deviations above or below the group mean, and therefore no data were excluded on this basis.
In addition, as we are not concerned with the extreme measures and outliers, the Infit mean-square statistics are most useful to our analysis.
Regardless of what you are measuring, the outliers tend to be tiny counties it is much easier for 100 people to be far above or below average than for millions to be.
Distribution of scores obtained for Set A (orange box) and Set B (red box) when applying a model trained on all of the data in Set A (left) or a model trained on all of the data in Set B (right) as a measure of outliers and domain.
The INFIT index is sensitive to unexpected behavior affecting responses to items near the person's measure level (inliers) while the OUTFIT index is sensitive to unexpected behavior by persons on items far from the person's measure level (outliers) [ 18].
Moreover, to measure how outlier-prone the distributions were, the kurtosis was also calculated.
For all electrophysiological measures presented here, outliers were removed based on the 2 SD rule.
