Quality of statistics
The Statistical Office of the Republic of Slovenia (SORS) focuses on providing quality and user-friendly national statistics, which means the statistics are presented and accessible in a user-friendly way.
In the past, the quality of national statistics was dealt with mostly in connection with data accuracy in the narrow sense (mainly as coherence between statistical data and exact values). In the last decade the statistical profession has made great progress towards broader understanding of the quality of statistical data. The most developed countries implemented a new approach in national statistical institutes: total quality management at the level of organization. Considerable change in dealing with quality in the European Statistical System was brought by the European Statistics Code of Practice, which was adopted in 2005.
Strategy of Quality in National Statistics
SORS understands the quality policy as a corporate culture with five basic quality pillars (main factors), mutually linked with modern management tools. These pillars are:
- Independent national statistics. The current arrangement of the national statistical system in Slovenia assures high level of professional independence of national statistics (i.e. the national statistical office together with the authorized producers of statistical surveys). This position will be further strengthened, since only professionally and politically independent statistics are trustworthy and thus relevant for the users.
- Data users and data providers. Balancing between users’ requests for statistical data and information on one hand and demands to provide data presented to the data providers on the other is becoming increasingly difficult. It is therefore extremely important to monitor to what extent published statistics meet the expectations and needs of users and to monitor burden caused on data providers due to their obligation to report the data for statistical purposes. Reducing the burden of data providers and assuring confidentiality and protection of the submitted data (which must be used exclusively for statistical purposes) will continue to be fundamental tasks of national statistics.
- Quality of statistical products and related services. In order to provide good quality statistical products and services, the office complies with the standard definition of quality, as well as with the principles of the European Statistics Code of Practice. The dissemination of quality in national statistics made great progress with the documentation of the quality of statistical surveys in the form of standard quality reports.
- Process orientation of national statistics. In the process of preparing statistical data and information in the framework of individual statistical surveys sources used, methodologies, procedures and also costs related to the statistical survey play an important role. With transparent statistical process and clearly documented procedures, better quality of the results and better cost efficiency can be obtained.
- Human resource development. Training of employees in order to increase the level of quality of statistical products and services includes several aspects: methodological knowledge, information know-how and the promotion of good practices exchange. It is important that the employees in the system of national statistics – at SORS and at the statistical units of authorized producers – are aware of the content of the European Statistics Code of Practice and that they work in accordance with it in their everyday professional work.
The main pillars (factors) of quality presented above are defined and thoroughly described in the Medium-term Programme of Statistical Surveys 2008-2012. The strategic directions from the Medium-term Programme of Statistical Surveys are in detail presented in the Total Quality Management Strategy 2006-2008.
The European Statistics Code of Practice
With the adoption of the European Statistics Code of Practice (in 2005) Eurostat and the statistical authorities of the EU Member States have committed themselves to an encompassing approach towards high quality statistics. The Code builds upon a common definition of quality (which was developed and accepted within the European Statistical System) and targets all relevant areas: institutional environment, statistical processes and statistical outputs. The Code is nowadays a recommendation which should be followed as much as possible; Member States and Eurostat should regularly report on implementation to the European Commission. The first report for 2005 was done in the form of a self-assessment questionnaire. As the next step towards the implementation of the Code, Eurostat organised peer reviews to complement the self-assessments. They are considered a vital element for the implementation of the Code in practice given their capacity to encourage the sharing of best practice and to contribute to better transparency of the whole statistical system. The peer review of SORS took place in May 2007. The report includes the description of the situation at that time of the implementation of the Code at SORS and the recommendations for further improvements. The peer review reports from other countries are available on Eurostat's website.
Data Quality Components
SORS adopted the Eurostat’s common quality definition. According to this definition the quality of statistical data is composed of the following six components:
- Relevance. Relevance is the degree to which statistics meet current and potential user needs. It refers to whether all statistics that are needed are produced and the extent to which concepts used (definitions, classifications, etc.) reflect user needs.
- Accuracy. In the general statistical sense this concept denotes the closeness of computations or estimates to the (unknown) exact or true values. Statistical data are namely not equal to the true values because of variability (values vary due to random effects/errors that appear at the implementation of the survey) and bias (values vary due to systematic effects/errors that appear at the implementation of the survey).
- Timeliness and punctuality. Timeliness of publication reflects the length of time between the period when the statistical phenomenon was observed and the release date of data. Punctuality refers to the time lag between the announced date of publication (for example in the release calendar) and the actual release date of data.
- Comparability. Used concepts should be harmonized, so that the obtained data and information are comparable over time, between geographical areas and between domains.
- Coherence. Coherence of statistics is their adequacy to be reliably combined in different ways and for various uses. The problems with coherence can occur when data originate from different sources or from different statistical surveys, where the used concepts, classifications and methodological standards are not harmonized.
- Accessibility and clarity. Accessibility refers to the physical conditions for users to access the statistical data: where and how it is possible to order data, delivery time, how much it costs (clear pricing policy), access to microdata and metadata, availability in various formats. Clarity refers to the environment in which the data are presented: are data accompanied by appropriate metadata, by graphical presentations, by information on their quality and by information about the extent to which additional assistance is provided by the national statistical institute.
Besides the mentioned quality components, SORS also adds a seventh component, costs and burdens. This component measures the cost efficiency of statistical surveys and the burden of reporting units (data providers) when they report the demanded data for statistical purposes.
Reporting About the Quality of Statistical Surveys
Standard quality reports for statistical surveys have been regularly prepared by SORS since 2006 and they cover a broad scope of quality indicators for statistical surveys (in line with the Eurostat’s standard for quality reports). Later on also annual quality reports for statistical surveys were introduced; these are shorter, translated also in English and include only the most important quality indicators about individual statistical surveys and are prepared every year. These quality indicators provide an overview of various quality components and enable comparability between statistical surveys and between countries. Individual quality indicators can be divided into producer and user oriented: the former measure quality from the point of view of producers and the latter from the point of view of users of statistical results. More information about the quality definition and quality indicators is available on Eurostat's Quality homepage.
User Satisfaction Surveys
User satisfaction with statistical data and services is regularly monitored with various user satisfaction surveys. The results of these surveys are an important source of information about the needs of users and problems they face.