Promoting the Common Metadata Framework
Strategy Paper Approved by the METIS Steering Group
1. The Steering Group on Statistical Metadata (METIS) has been developing the Common Metadata Framework (CMF) for several years. There are now tangible outputs for all four parts of the framework. Having invested in developing these outputs, maximum return on investment will only be achieved if these outputs are known, understood, valued and actively used by their target audiences (which vary to some degree between parts). Therefore this paper sets out the strategy for promoting and publicizing the CMF.
2. The CMF currently comprises four parts, which are, to some extent, targeted at slightly different audiences. The METIS Steering Group has recently agreed a maintenance strategy for the CMF with different periodicity and updating arrangements for each part. For these reasons, after considering some general points, this paper also considers specific promotional activities for each part.
II. General points on promoting the CMF
3. Most material related to the CMF is available in English only. The exceptions are Part A, which is also available in Russian, and the Generic Statistical Business Process Model within Part C, which is available in French and (partly) in Russian. More language versions would help reach a wider audience, but would significantly increase the maintenance costs. Parts B and D of the CMF are specifically intended to be living documents, existing only in electronic format, and regularly updated. Formal translation for these parts is therefore not practical.
4. Possible solutions include inviting countries to share translations of material from the CMF, and posting these on the METIS Wiki. This has already been done with the GSBPM, where a French translation prepared by INSEE is now available alongside the English version. This approach is most suitable for material that is expected to remain unchanged for several years. For material that changes more frequently, one option is to provide links to automatic translation tools, such as Google Translate or Babelfish. Although the quality of automatic translations is not normally sufficient for published documents, it is usually good enough to get a basic understanding of the content of a piece of text. This quality is also gradually improving over time.
5. The CMF is primarily intended to be published in electronic format via the METIS Wiki. However, Part A was thought to be relatively stable over time, so following the last update a limited number of printed copies were produced in English and Russian (see also the discussion under Part A below). The main drawback of electronic resources is the lack of something physical to hand out at meetings and use for advertising purposes. For this reason, the UNECE secretariat has created a brochure to publicize the CMF (see Annex).
6. To promote an electronic resource, it is important to have as many links as possible from other sites, and to ensure that the web site is given the highest possible rating by Internet search engines. Whilst the CMF can easily be found when it is searched for by name, it is not so easy to find using search terms such as “statistical metadata”. Some research on improving “Google-friendliness” could help.
7. Linked to the issues of format and language is the issue of accessibility. The information in the CMF should ideally be available to potential users whenever, wherever and however they need it. Free public access via the Internet helps greatly with this. However the web service needs to be reliable (UNECE is currently working to improve this), and the layout of the information needs to be clear and logical.
8. However, accessibility should not be considered just as a passive concept. Just because people can access something, doesn’t mean that they will. Part of accessibility is making sure that potential users are aware that there is a resource they can access. In other words the METIS Steering Group should actively disseminate CMF documents via e-mail and other relevant means, to user groups such as:
the heads of methodology departments in statistical organisations
the heads of information technology departments in statistical organisations
the participants of METIS Work Sessions and Workshops
other international statistical bodies.
Clarity of message
9. Even if users access material such as the CMF, this does not necessarily mean that they can understand it. Statistical methodology materials are often criticized as being written by a small group of experts for a small group of experts. We need to be careful that the CMF does not fall into this trap. It should be clearly understandable to all those working with official statistics, either as producers or users.
10. As authors are often not the most objective judges of the clarity of their outputs, independent feedback should be sought. This could take the form of a review by a trained technical editor, and/or comments from users. Where possible, feedback should not just be limited to clarity, it could also cover the general concepts of quality and usefulness of the CMF.
Promoting the CMF in relevant international meetings
11. To reach potential users outside the statistical metadata community, the CMF should be promoted whenever relevant in other international meetings and forums, particularly those that report to the Conference of European Statisticians (e.g. Management of Statistical Information Systems, and Work Sessions on Statistical Data Editing and Statistical Data Confidentiality). Promoting discussions of CMF outputs and ideas with these groups could increase the potential for future collaboration.
III. Promoting Part A – Statistical Metadata in a Corporate Context
12. The main aim of Part A is metadata advocacy, explaining the importance of a good metadata management system for the efficient production of statistics. It is targeted at senior managers and subject-matter statisticians, as well as those responsible for making business cases to create or enhance statistical metadata systems. It was first published on the Internet in 2006, but was updated and enhanced by the Steering Group on Statistical Metadata to reflect feedback from various users. The resulting version was published in electronic and paper formats, in both English and Russian, at the end of 2009. The UNECE secretariat still has a stock of these publications, which are distributed at meetings, and can be mailed free of charge to anyone who is interested.
13. However the METIS Steering Group has discussed that to really get the attention of senior managers, perhaps a shorter version of Part A is needed, in the form of a brochure containing an executive summary. A sub-group has been formed to prepare this brochure.
IV. Promoting Part B – Metadata Concepts, Standards, Models and Registries
14. Part B exists as an electronic resource providing information about different metadata standards. The target audience for Part B includes those responsible for developing and implementing statistical metadata systems. Some of these people attend METIS meetings, so should be aware of Part B. The challenge is how to reach the others.
15. Possible solutions include:
targeted mailings to organisations not represented in METIS meetings
encouraging METIS participants to pass information to their colleagues
giving links to the resource descriptions in Part B whenever metadata standards are referenced in papers and other documents
V. Promoting Part C – Metadata and the Statistical Business Process
16. Part C is currently the least developed part of the CMF. It will explore the role of metadata throughout the statistical production process. The main output so far is the Generic Statistical Production Process Model (GSBPM), which was developed to give a standard framework to describe the production process.
17. In many ways the GSBPM has been the most successful output of the work on the CMF, particularly in terms of its recognition and use in statistical organisations. However, the Task Force currently working on developing materials relating to the GSBPM has identified a number of communication issues. One issue is that there is often a perception that the model is very linear and survey focussed. The Task Force is looking at alternative presentations of the model to counter these misconceptions.
18. Other tools to help promote the GSBPM have been proposed, including posters and mouse mats. However the main obstacle is financial and a sponsor would probably be needed.
VI. Promoting Part D – Implementation
19. Part D comprises a set of case studies from national and international statistical organizations, describing their experiences in developing and implementing statistical metadata systems. These case studies follow a standard template and focus on lessons learned. They provide a valuable resource to help users to benefit from the experiences of others. Australia provided a paper for the 2010 METIS Work Session evaluating the uses of the case studies, with generally positive conclusions. This idea could be developed further to provide a promotional tool.
20. In common with Part B, the target audience for Part D is those responsible for developing and implementing statistical metadata systems. Those that attend METIS meetings should be aware of Part D, but the challenge is how to reach the others. The possible solutions are the same as those identified for Part B.
Annex – CMF Brochure
Back cover Front cover