What do we call a data point that is far from others, potentially distorting results?

A data point that is far from others and can significantly distort results is referred to as an outlier. Outliers can occur due to variability in the data or may indicate measurement error or a novel phenomenon. Their presence can skew statistical analyses, such as means or correlations, leading to misleading conclusions.

Understanding outliers is critical in data science and statistics because they can affect the performance of models and lead to improper forecasts. Proper handling of outliers—such as determining whether to remove, modify, or keep them—can significantly improve the accuracy and reliability of data analysis.

In contrast, the other terms relate to different concepts in data analysis. A subset refers to a portion of data selected from a larger dataset, while a cluster denotes a group of data points that are similar to one another. The range is a measure of the spread of a dataset, specifically the difference between the maximum and minimum values. None of these terms describe the phenomenon of individual data points being significantly different from the rest in a way that could distort outcomes.