# Spatial analyses

Each phenomenon that can be described statistically has a spatial-temporal component. This makes it possible to use statistical data for spatial analyses. Spatial analysis is the process of examining the distance, location, position and relationships of features through mathematical algorithms in GIS and mapping software.

All the results of the 2000 and 2011 Population and Housing Censuses (e.g. population composition by sex and age, ethnic nationality, command of languages, dwelling indicators, etc.) have been linked to an address, which corresponds to a certain building. This enables the use of census data for various spatial analyses such as transportation and hinterland, choosing the optimal location, migration and the impact of phenomena and objects. Since 2015, population statistics has been georeferenced as well. Population statistics is based on the census figure of population which has been adjusted with registered changes of population – with data of births, deaths and migration. Under-coverage has also been added to the persons enumerated in the population census.

Building-based statistics can be presented independently from the usual presentation method which is linked to the administrative unit. This enables creating grid maps where the statistical data has been aggregated into uniform territorial units or grids. The grid cell is the smallest territorial unit for which Statistics Estonia publishes statistical data. The territory of Estonia has been divided into 1 km x 1 km grids, and cities with their more densely populated hinterlands into 500 m x 500 m grids. The territory of major cities Tallinn, Tartu, Narva, Pärnu and Kohtla-Järve has been divided into 100 m x 100 m grids.

Controlled rounding method is used upon publishing grid data (except for in the case of population figure by sex and age group). All values are rounded to the nearest number divisible by three. The value 0 is not changed in the rounding process; the values 1 and 2 are randomly rounded to 0 or 3. Upon publishing the population figure, sex and age by grid, the following methods have been applied to protect the confidentiality of the individuals:

- if there are fewer than 4 persons are living in the grid or there are uninhabited buildings in the grid, the value of the grid is <4;
- if there are fewer than 6 persons or fewer than 3 dwellings with fewer than 20 inhabitants in the grid, the values of sex and age groups are not published.