When is it justifiable to exclude 'outlier' data points from statistical analyses?
Data analyzers inspecting tables or figures might decide to exclude from statistical analyses unusual data points sometimes called 'outlier' data points. Statistical patterns and conclusions might differ between analyses including versus excluding outliers.
The exact underlying mechanisms that create outlier data points are often unknown. People might always find arguments to exclude or keep data in analyses. How important is familiarity with model species or model systems in the justification of data point selection, or the definition of statistical rules in general?
