Data is everywhere and more data are continuously generated. Without data management, the data lose much of their value. Data management is necessary to make sure that data could be quickly found and effectively used.
Data management allows:
- to quickly find the right information from data;
- to provide services and make procedural and other decisions on the basis of data;
- to offer data as analyses and statistics for making management decisions.
Why should an organisation have well-organised data management?
Well-organised data management is one of the cornerstones of success when organisations are making necessary rapid changes. A great benefit is also savings in time and money. A third gain becomes apparent when employees change. If there is well-organised data management, new employees can quickly start to use data.
Where to start with data management?
To start, determine who is responsible for data management in the organisation and prepare an activity plan. The first task is to get an overview of datasets and identify which data volumes should be managed, then describe, i.e. catalogue, these.
What are the more general goals of data management?
A more general goal of data management is to help to organise the use and protection of data. Data use creates benefits, which is why it is important to enable the reuse of data. Equally important is monitoring the reuse of data, in order to make sure that the data are protected. If reuse is possible, one can start to reduce duplication in data collection and to apply the once-only principle, i.e. asking for data only once.
What is the role of Statistics Estonia in data governance?
Pursuant to the Public Information Act and the Official Statistics Act, it is a duty of Statistics Estonia to coordinate data governance. The activities of Statistics Estonia do not duplicate other public sector duties in the development of information society, its services and the state information system, protection of personal data and guarantee of cyber security, instead they concentrate on the following data management requirements:
- to have an up-to-date and content overview of databases and datasets that are used in analyses and statistics;
- to standardise data descriptions so that datasets, incl. open data, could be quickly found and the data would be described once with quality, which would allow to understand the necessary data and support their adoption;
- to monitor data quality and improve it so that users would be quickly convinced that the data are correct, complete and timely.
Data management is connected with the use of classifications. Statistics Estonia manages the system of classifications and monitors that the same classifications are used in databases and information systems. In addition, Statistics Estonia coordinates the adoption of international classifications and standardising lists of consistent objects used in information systems and creation of classifications of these.
How does Statistics Estonia coordinate data governance?
For the coordination of data governance there is an action plan and a cooperation network. The data governance action plan lists the duties and responsible bodies for years 2021 and 2022.
Statistics Estonia focuses on data in databases and analytical datasets, not on media files or text corpora, digital culture or heritage. As the sources for statistics are expanding, areas and data types without data are added, such as spatial data and data on information society services.
What software applications to use for data management?
For an overview of data, data catalogue applications are used, which include a list of managed datasets and their descriptions. Statistics Estonia together with the Information System Authority are developing a data description tool (RIHAKE) for organisations. It allows describing the organisation’s datasets and transmit the description in a machine-readable format to the administration system of the state information system (RIHA). Data descriptions can also be managed as data models. A data catalogue and data descriptions can also be prepared and the data quality can be monitored also with very common applications, such as Excel or Jira.