The standardisation of data is necessary for more meaningful thematic map presentation (e.g. costs per inhabitants instead
of the total of absolute costs). The resulting common denominator enables a comparison between different types of collected
data.

Moreover, we standardise data if we wish to show the relation between our collected data and another dataset. Thus, we
standardise data in order to make our data comparable to others, to show the ratio, and enable a better analysis of our data.

In the next paragraph, we will introduce different standardisation approaches for numerical data.

The most common method of standardisation approaches is the simple ratio standardisation. For ratio standardisation,
we divide an area-based numerical dataset by another area-based numerical dataset. As the numerator and the denominator both
consist of the same measurement units, the result is a proportion. This proportion can also be expressed as a percentage.
An example may be the division of the water run-off of a considered catchment area [mm/m2] by the measured precipitation
of the same catchment area [mm/m2]. The resulting ratio between run-off and precipitation is the "run-off coefficient" of
the catchment area [%].

When standardising data to indicate the density, we divide a non area-based variable by an area. For example, If we want
to calculate the soil erosion per area [kg/m2], we have to divide the measured eroded soil weight [kg] by the area size [m2].

Another way of standardising data is to compute the ratio of two non area-base variables. Accordingly, the resulting
units are always rates. Example: We may calculate the damage costs per person after a natural hazard [USD / person]. Another
example of rate standardisation is the calculation of the available hospital beds per inhabitant [hospital bed/inhabitant].

Sometimes we standardise data by dividing an area size by a non area-based variable. Example: When we divide the total size of a village's estates [m2] by the number of land owners [persons], the emerging result is the average size of estate per land owner [m2/landowner], which is a so-called area-based rate (Slocum 1999).

It is sometimes necessary to standardise data before we can compare it to other datasets. However, we do not need to
standardise our data, if we compare our collected data with data that refers to one and the same basic measurement settings
(e.g. same location, same measurement method, etc.).

Example: If we measure the precipitation at a specific place x over one year each day, we can compare these measurements without
standardising them. We can even use an invented unit for comparison, as long as the measurement system and its units remain
the same during the considered year. However, as soon as we want to compare our measured precipitation data with the precipitation
data of another place, we need to standardise the data in order to obtain the same units (a common denominator).