In research, an outlier is a data point that falls outside of the expected range of variability for the thing being measured. True outliers are typically the result of errors in data collection or experimental methodology, or, in the case of survey data, errors or mistakes by participants.
Because outliers can skew the results of analyses, they are generally identified and excluded from analyses.
The most common method researchers use to identify outliers is by looking for measures that fall outside of three standard deviations from the mean. This standard is a good place to start when looking for outliers. The subject matter expertise of researchers is also important, however. Subject matter expertise may lead researchers to either include or exclude data based on criteria other than whether the measure falls outside of three standard deviations.
