Methods Quality guidelines

Quality Manual of the Statistical Offices of the Federation and the Länder

The quality of statistical data has always been of great importance in official statistics. The statistical offices are recognised as the leading providers of high-quality statistical information about Germany. The aim is to continue to guarantee and expand the quality level achieved. For this reason, the Statistical Offices of the Federation and the Länder have developed and approved a quality manual.

The quality manual describes the framework for ensuring data quality in official German statistics. On the one hand, it is intended to inform users of statistical data on the quality assurance of statistical results and, on the other hand, to serve as a guideline for employees of the Statistical Offices of the Federation and the Länder and other offices in Germany that produce official statistics.

Chapter 5 of the handbook, the most important chapter for statistical production, formulates several hundred "Quality Guidelines for the Statistical Production Process", which guide German official statistics in their daily work. The quality guidelines describe concrete procedures for all phases of the implementation of official statistics (e.g. data collection, preparation, dissemination) which serve to ensure the high quality of the statistical processes and products and whose implementation is binding for all statistics. For many statistics, the quality guidelines are not only measures that have already been systematically and comprehensively implemented, but also partly ambitious development targets.

European Statistics Code of Practice

Ensuring high data quality is one of the main functions of the European Statistical System (ESS). Therefore, in May 2005, the European Commission released the European Statistics Code of Practice in a recommendation which has since been updated in 2011 and 2017. In a total of 16 principles, the Code of Practice defines the quality assurance guidelines regarding the institutional framework of statistics production, the statistical processes and the statistical products. This is to ensure that official statistics are produced according to recognised scientific methods and free from external influence in all Member States. The EU statistical institutes have committed themselves to the Code of Practice as a tool for self-regulation. Compliance with the Code of Practice is ensured by the statistical institutes through self-assessments and Peer Review. As regards compliance with the Code of Practice at Eurostat and in the ESS as a whole, an advisory body (European Statistical Governance Advisory Board) consisting of high-level statistical experts every year submits a report to the European Parliament and to the Council of the European Union.

Also, in accordance with Article 11 (4) of the revised EU Regulation No 223/2009 on European statistics, Germany is obliged to report on the implementation of the Code of Practice. This was first done in 2017 when a progress report on implementing the Statistics Code of Practice was sent to Eurostat.

Quality declaration of the European Statistical System

The heads of the National Statistical Institutes in the ESS have established common quality principles for all essential aspects of statistical work. The Quality Declaration, signed by all the authorities of the Member States of the European Statistical System, underscores and confirms the commitment to quality.

Quality requirements of the Regulation on European statistics

The Regulation of the European Parliament and of the Council on European statistics is the legal basis for developing, producing and disseminating European statistics. To take account of the importance of high quality, separate Articles are dedicated to the Code of Practice and to quality assurance. Parallel to the European Statistics Code of Practice, Article 12 defines the following binding quality criteria applying to the development, production and dissemination of European statistics:

  • relevance,
  • accuracy,
  • timeliness and punctuality,
  • accessibility and clarity,
  • comparability and coherence.

The Commission (Eurostat) may oblige the Member States to submit reports on the quality of individual sets of statistics. Specific quality requirements such as minimum standards and targets may additionally be defined here for any statistics.

UN-Fundamental Principles of Official Statistics

On 29 January 2014, the United Nations General Assembly adopted a resolution concerning the 10 Fundamental Principles of Official Statistics, which have applied worldwide since 1994. The 20th anniversary of adopting these principles was used as an occasion to revise the preamble in order to take into account the new global developments of the past 20 years and to enhance the implementation of the principles.

The Fundamental Principles govern the production and dissemination of official statistics throughout the world. They constitute a global frame of reference for uniform professional and scientific standards, among other things, regarding objectivity, transparency, confidentiality, the prevention of unauthorised use, national co-ordination and international co-operation in official statistics.
In the coming years, the implementation efforts will be strengthened by political initiatives, training both the management and the staff involved in official statistics, compiling best practices and providing technical support.

Germany has actively supported the resolution in the United Nations General Assembly. Together with statistical offices worldwide, recommendations are currently being developed for an improved implementation of the Fundamental Principles at a global level.

Error policy

Despite the many quality assurance measures we employ, errors in published statistical products cannot be completely avoided.

The whole procedure of dealing with publication errors in the Federal Statistical Office is now described in a guideline "How to deal with publication errors". It ensures that errors are always treated, corrected and documented identically corresponding to a defined error classification. It also specifies the communication of the error with respect to the users of statistical data.

Revision policy

A revision is a modification of already published results as new or improved data become available and are incorporated into the calculation or when methodological and conceptual changes are made.

The General Revisions Policy of the Statistical offices of the Federation and the Länder describes the revision procedures which apply to all statistical domains in a transparent and comprehensible manner so as to increase the trust in official statistics and further enhance the usability of statistics.

The general revisions policy is supplemented by the revision calendar of the Federal Statistical Office. The revision calendar provides an overview of which sets of statistics are subject to revision and describes the respective revisions cycle by means of a standardised structure.