To use earnings research appropriately the underlying statistics need to be understood.
Data providers sort earnings into different categories (e.g. Paterson or Unskilled & Semi-skilled). The earnings in those categories are ordered from lowest to highest and then divided into 4 quarters. Values mostly provided are Q1 / 25% / lower quartile, Q2 / 50th % / median, and Q3 / 75th % / upper quartile.

This method is used to exclude outliers that fall below Q1 and above Q3 that will distort the average earnings. This means that there can be claimants in the same job category that earn less than Q1 or more than Q3.
The median is the mid-point of the data set and the most reasonable indication of earnings in this data set. Median should not be confused with average.