Why Joined-up Data Standards?

Data has immense potential to help drive poverty eradication and development, but right now it is incredibly difficult to join up the data we have on money, people and results, because it is published in different formats and to different standards. This stops it from being turned into useful information for decision-making and accountability. To solve this, we need to enable existing and future standards to join up. Development Initiatives and Publish What You Fund are working together to make this happen by focusing on technical solutions and political will. Efforts to join up data will help equip decision-makers and those holding them to account with vital information for driving sustainable development. To find out more about our work, see About

SEMIC Conference – reflections on the potentially transferable lessons

3 Share

SEMIC Conference – reflections on the potentially transferable lessons

“If we don’t have interoperable solutions we could not do our work, we could not invent the future” said Markku Markkula, President of the European Committee of the Regions, in his ...

“If we don’t have interoperable solutions we could not do our work, we could not invent the future” said Markku Markkula, President of the European Committee of the Regions, in his closing speech at the Semantic Interoperability Conference (SEMIC 2017) in Malta a few weeks ago.

Mr Markkula emphasised how important it is to ensure investment in data governance, to rethink information retrieval both within and beyond organisations, to share information, to collaborate and co-create and to have in place sound knowledge management systems. His words aimed to summarise a day filled with solutions to the problems in systems in developed, technically advanced Europe. My work mostly revolves around data and systems in developing countries, but I couldn’t help but wonder whether some of the solutions discussed at the conference could be used by the governments, public administration, statistical offices and anyone else publishing and sharing data in developing countries. Here I unpack some of the solutions that may resonate beyond the confines of SEMIC.

Data governance

The main topic at the conference was the lessons learnt and challenges faced when implementing a data governance strategy at the national and institutional level. Data governance in simple terms refers to processes for accessing and sharing data and resolving conflict when the processes don’t work. The concept has existed since private organisations embraced the notion of data as an asset. If data is an asset it needs to be accessible and shareable, with proper data governance in place.

Data governance is now the bread and butter of European public agencies, and this is reflected in the Luxembourg principles on open data portals where information is seen as an asset and must be accurate, integral, protected, accessible and up to date, and should be interoperable by default.

In practice, sound principles and data governance can lead to some truly spectacular products that put citizens at the heart and that harmonise major public and private systems. X-Road – Estonia’s large data platform is a good example of this.

Creating the right culture

Peter Van Landegem of the European Commission explained how its biggest problem is that staff do not always have all the information they need for their work. He mentioned information overload, new working methods and increasingly complex files, siloed data in the parts of the Commission, problems with the quality and robustness of data and information, and a lack of ethos of collaborative working. Sounds familiar?

The Commission wants to ‘walk the talk’ and be a well-performing, innovative knowledge-based institution committed to evidence-based policy making. It acknowledges that to achieve this, the key aspects are: sharing skills and knowledge, finding and reusing good quality information and anticipating policy-making needs. Van Landegem emphasised the need for complete, reliable, relevant and accessible data/information/knowledge, and to foster a culture of cooperation, performance and innovation. As such, the Commission has developed 12 strategic objectives to achieve this:

  1. 1
    Ensure information responsibility
  2. 2
    Establish corporate governance over information
  3. 3
    Ensure system interoperability
  4. 4
    Connect people
  5. 5
    Map knowledge
  6. 6
    Enable collaborative working methods
  7. 7
    Develop data scientist methods
  8. 8
    Exploit the potential of (big) data
  9. 9
    Put in place the tools and infrastructure for data analytics
  10. 10
    Become knowledge workers
  11. 11
    Become learning organisation
  12. 12
    Become and information aware organisation

More often than not these principles go hand in hand with the highly technical solutions used to apply them. But the story behind data governance at the European Centre for Disease Prevention and Control (ECDC) might be an inspiring one in its simplicity. When faced with the problem of creating a curated catalogue of internal and external data for the user (an information asset catalogue), they started off with an assessment of their needs and focus (Figure 1) and ended up with a simple excel spreadsheet that presents a great first attempt at making their data easy to understand and retrieve by anyone who would like to make sense of it.

Figure 1: ECDC Information Asset Catalogue

Addressing new demands

Eurostat spoke about its new policy around data dissemination. This is motivated by the European Commission corporate strategy that focuses on maximising the use of data for evidence-based policy-making. Titus Purdea explored issues connected to this policy. These ranged from the demand to include on Eurostat an ever-broader range of data outside official data sources, to dealing with increased demand for ad-hoc statistics and for releasing more data towards partners and the research community. These seem to be the same demands that many national statistics offices are facing. In response Eurostat is adapting its IT architecture to deal with these issues.

Figure 2: Eurostat’s potential IT architecture

Shared concerns and lessons learnt

Development Initiatives has recently published an excellent discussion paper written by Bernard Sabiti, Data for development in Africa. Ensuring commitments made at the High-level meeting in Kenya are met. Bernard spells out the steps that need to be made by governments, donors and other actors to ensure the vision for development data in Africa becomes reality. He speaks of the need for collaboration between private companies, governments and other stakeholders – the main problem highlighted in each of the presentations at SEMIC. National statistical offices need to embrace change by dealing with non-official statistics , mirrors the steps Eurostat is taking to incorporate data from other sources into its own systems. And they need to connect to community data collection and engagement systems, so reinventing the role of civil society and other non-state actors. Finally, building capacity and walking the talk by ‘putting money where the mouth is’ resonated deeply at SEMIC.

Don’t get me wrong; the problems faced by developing countries are very different from those of EU countries. But the need to invest in sound and sustainable data systems, reusing solutions and principles are shared concerns in both contexts. In that case – can they learn from each other and exchange solutions? Especially if it aids in creating data systems that are published, disseminated, maintained and governed in a sustainable way?

There is much to learn across different contexts, and I look forward to continuing to help identify useful approaches, solutions and principles that can be reused for sustainable progress of the developing countries.

 

http://juds.joinedupdata.org/wp-content/themes/juds.org_theme/media/img/default-avatar-68.png
Beata is the Data Scientist working on the technical side of Joined-up Data Standards from Development Initiatives’ Bristol office. Beata’s work focuses on mapping and linking data standards. She also develops and maintains the thesaurus. Beata finished her PhD in metabolic modelling of microbial systems at the University of Bath in 2015. She holds a MSc in Bioinformatics from Imperial College London and a BSc in Biotechnology from Cardiff University.