A data element that is significantly different from other elements in the data collection, or an element that appears to imply a pattern that is inconsistent with the bulk of the data evidence. An outlier should always be given careful consideration.

Mathematically, an outlier is determined by the following steps:

  1. The interquartile range (IQR) can be calculated by subtracting Q1 (first quartile) from Q3 (third quartile), that is, IQR = Q3 – Q1.
  2. Take the IQR and multiply by 1.5.
  3. Any data element will be labeled an outlier when it is either smaller than Q1 - 1.5x(IQR) or larger than Q3 + 1.5x(IQR).

Related Term: box and whisker plot

English | Español