An outlier is an observation that appears to be unusual or not well described relative to a simple characterization of a dataset. A discomfiting possibility is that these data come from a different population than the one intended to be studied.

An outlier is an observation that appears to be unusual or not well described relative to a simple characterization of a dataset. A discomfiting possibility is that these data come from a different population than the one intended to be studied.

However, outliers are not necessarily bad or wrong, nor do they necessarily need to be removed from data for further analysis of that data set. However, outliers (of which there can be more than one in any set of data) indicate that some data at least appear to differ from the bulk of the data set, suggesting they should be individually examined and understood. Also, some statistical procedures are sensitive to outliers: this means that removal of one or more outliers could substantially change the conclusions of those procedures.