Dernière mise à jour le :13/10/2016


The mean is the simplest indicator to summarise information provided by a set of statistical data: it is equal to the sum of these data divided by their number. It can therefore be calculated in the knowledge of only these two elements, and without knowledge of the full distribution.

The mean of a distribution is not always the best indicator: the median is often more relevant. But calculating the median requires knowledge of the full distribution, or at least its central part.


A large number of economic values have a lower limit but not an upper one. For example, the hourly wage has a lower limit provided by the SMIC, but certain wages can be very high indeed. These high wages pull the mean upwards, despite the fact they are few in number, and the mean is therefore generally higher than the median.

Also, the uncertainty surrounding the extreme values, and in particular high values, has an impact on the mean but does not affect the median. In this respect, the latter is therefore a more reliable indicator.