Robust describe how far an inferential procedure, i.e., estimation or significance testing, gives stable results across different conditions such as different distributional assumptions, presence of outliers, and measurement errors.