Outlier
A data point that lies far from the rest of the distribution, which may reflect an error, a rare case, or a genuinely extreme value.
An outlier is an observation markedly distant from the other values in a dataset. Outliers are commonly flagged by inspecting plots or by rules based on standard deviations or interquartile range. They matter because many statistics, including the mean, variance, and regression estimates, are sensitive to extreme values, so a single unusual case can distort results. The correct response depends on the cause: correct it if it is a recording error, but if it is a genuine extreme case it carries real information and deserves careful treatment rather than reflexive removal.
In a thesis, never delete outliers silently. State how you identified them, what you decided, and why, and consider reporting analyses with and without them to show whether conclusions hold. Robust alternatives, such as the median or non-parametric tests, are also defensible options. Transparent handling of outliers signals analytical maturity to examiners.
Writing the thesis this term belongs to?
CiteDash takes a thesis from research question to a compiled document, with AI that cites only real papers and verifies every claim against its source.