Merging statistics and geospatial information, 2020 projects - Malta

Integration of geospatial information and statistics – implementation of a 1 km x 1 km grid; 2020 project; final report December 2022

This article forms part of Eurostat’s statistical report on the Integration of statistical and geospatial information.

Full article

Problem

There is a need for data presented according to a 1 km x 1 km grid, as opposed to data for localities.

Objectives

The main objective of this project was to improve access to geospatial data and the development and integration of a geocoding infrastructure for production processes based on address registers for 1 km grid datasets.

Method

A spatial layer called realistic building defined street layer (ReBuDS) was created by combining two other layers:

a street layer showing the road network,
a utility spatial layer showing the location of smart meters used for water and electricity consumption from which it can be interpreted where people live / buildings are located; buffers were used to match the location of the utility meters to nearby streets.

The street layer was clipped based on the buffered utility layer resulting in a polyline representation of the location of buildings. The polylines were transformed to points through the creation of vertices and a centroid for each clipped line (see Figure 1). These points have the same attributes as the street layer. The street layer was then standardised according to the common database (CDB) – the official source used by various government entities – of which street names and street codes were used to match the spatial layer with the street-based addresses.

Figure 1: Transformation of a polyline to points

All of the statistical databases to be analysed were standardised by implementing the names and codes of the CDB. This was a lengthy process mainly due to the large amounts of data received and the lack of standardisation. Various methods were applied to geocode records for addresses that could not be located and therefore lacked a street code.

A spatial analysis based on a 1 km x 1 km grid was performed for the newly geocoded databases.

Education – enrolments: the focus is on the location of students’ residence and the location of the school they attend, with an analysis of the distance between these as well as by the type of institution (state, church or private).
Education – graduates: the focus is on the location of each graduate’s residence.
ICT usage by enterprises: the focus is on the number and share of enterprises offering web sales.
Job vacancy survey: the focus is on the average quarterly number of occupied posts and average number of vacancies which arose during the quarter.
Labour force survey: the focus is on the population of persons aged 15 years or more, with a distinction between employed, unemployed, and economically inactive persons.
Social benefits: the System for the Administration of Social Benefits contains micro data for beneficiaries of the social security pension scheme and for contributors; the analysis is focused on the count of persons receiving benefits and the value of the benefits.
Social income and living conditions survey: the focus is on the average disposable annual household income as well as the share of persons living in dwellings that
- were too dark / did not have enough light,
- had noise from neighbours or from the street,
- were in an area with crime, violence or vandalism.
Tourism: the focus is on the number of collective accommodation establishments and the number of nights spent in these establishments; the amount of publishable data was limited by confidentiality issues.
Business register: the focus is on the number of enterprises and the number of local units.
Structural business survey: the focus is on turnover, production value, value added, the gross operating surplus, purchases of goods and services, personnel costs, wages and salaries, social security costs, investment, the number of persons employed, the number of employees and the number of enterprises.
Population register: the focus is on a count of the population.

Figure 2: The 1 km² grid showing concentrations of local units

Results

Through the standardisation of street names according to the CDB and the assignment of street codes, individual databases could be geocoded via an in-house built point spatial layer (realistic building defined street layer (ReBuDS)) which shows the location of buildings in each street. Using this, databases with an address as one of its fields were geocoded and data analysed.

Efforts are being made to geocode other databases and to update those that were already geocoded.