Digital records as evidence to underpin global development goals. (HN)
Lead Research Organisation:
UNIVERSITY OF LONDON
Department Name: Inst of Commonwealth Studies
Abstract
This project examines the crucial role of records management (and increasingly digital records) in the attainment of current development goals. The launch of the UN Sustainable Development Goals (SDGs) in September 2015, has made it increasingly essential to recognise and address weakness in the way governments' official evidence base is managed and used. Plans to measure and monitor the SDGs are based on the assumption that it will be possible to access meaningful data as a basis for benchmarking and eradicating poverty. Unfortunately, it is often the case that records are incomplete, inaccurate, inaccessible or lost completely, with the result that data and statistics derived from the records are flawed.
The rapid transition to the use of digital records has encouraged and been closely linked to 'open government' initiatives, aimed at making records and data publicly available. Again, this has a direct bearing on the achievement of current development agendas in ensuring that government is more transparent and accountable. Yet in the digital environment, records created by computerised information systems do not remain reliable or accessible, even for short periods of time, without a control regime of laws, policies, practices and skills as defined in international standards. If digital repositories are not developed, where digital records and data can be managed securely through time, there is no guarantee that they will remain available or reliable. Unfortunately, it is precisely those countries that stand to gain most from progress towards the SDGs that have the greatest problems putting effective data management systems in place.
The objective of this, which project builds on the expertise within the School of Advanced Study (SAS), its networks and partner organisations, is to bring together academics, records managers and archivists and a range of NGOs to consider how the management of digital records and data can assist in the achievement international development goals, particularly the SDGs, in sub-Saharan Africa. An important element in the achievement of these goals is the combatting of corruption, something that severely hinders justice, good-governance and development in many countries. Managing records efficiently will not in itself halt corruption, but anti-corruption goals cannot be achieved without trustworthy records as evidence. As such this will be an important thematic focus of the project. It aims to disseminate its findings via and dedicated web resource and an edited volume of articles. The project will also consider the feasibility of a globally-applicable set of protocols for digital records management to help ensure that records management policies, principles and practices become a core aspect of development in the digital environment.
A key feature of the project is a two-stage cross-fertilisation of expertise. In the first stage the workshops will be to encourage a dialogue between experts in the area of records management (from both the developed and the developing world) and academic experts on the politics, economics and history of post-colonial Africa. The second stage will allow them to engage with those actively involved in implementing international development goals. As such, measurable outcomes will be extent of the follow-up initiatives that emerge from these dialogues, and the extent to which the immediate outputs of the research network are fed into the policy-making process.
The rapid transition to the use of digital records has encouraged and been closely linked to 'open government' initiatives, aimed at making records and data publicly available. Again, this has a direct bearing on the achievement of current development agendas in ensuring that government is more transparent and accountable. Yet in the digital environment, records created by computerised information systems do not remain reliable or accessible, even for short periods of time, without a control regime of laws, policies, practices and skills as defined in international standards. If digital repositories are not developed, where digital records and data can be managed securely through time, there is no guarantee that they will remain available or reliable. Unfortunately, it is precisely those countries that stand to gain most from progress towards the SDGs that have the greatest problems putting effective data management systems in place.
The objective of this, which project builds on the expertise within the School of Advanced Study (SAS), its networks and partner organisations, is to bring together academics, records managers and archivists and a range of NGOs to consider how the management of digital records and data can assist in the achievement international development goals, particularly the SDGs, in sub-Saharan Africa. An important element in the achievement of these goals is the combatting of corruption, something that severely hinders justice, good-governance and development in many countries. Managing records efficiently will not in itself halt corruption, but anti-corruption goals cannot be achieved without trustworthy records as evidence. As such this will be an important thematic focus of the project. It aims to disseminate its findings via and dedicated web resource and an edited volume of articles. The project will also consider the feasibility of a globally-applicable set of protocols for digital records management to help ensure that records management policies, principles and practices become a core aspect of development in the digital environment.
A key feature of the project is a two-stage cross-fertilisation of expertise. In the first stage the workshops will be to encourage a dialogue between experts in the area of records management (from both the developed and the developing world) and academic experts on the politics, economics and history of post-colonial Africa. The second stage will allow them to engage with those actively involved in implementing international development goals. As such, measurable outcomes will be extent of the follow-up initiatives that emerge from these dialogues, and the extent to which the immediate outputs of the research network are fed into the policy-making process.
Planned Impact
As the Case for support for explains, the effective preservation and management of digital records has the potential to make a major impact on the tracking of achievement of a range of these goals, particularly in terms of the fight against corruption. Sub-Saharan Africa, on which the project focuses faces major challenges in terms of the resourcing of its records management systems, and the broader impediments to anti-corruption. As such, we believe that the project has a broad range of potential non-academic beneficiaries, not just in Africa, but across the developing world.
The CI, Dr Anne Thurston, has worked closely in the past with the Open Government Partnership (OGP), and the final report of the project, on 'Digital Records Protocols' will be disseminated through the OGP's many influential partner organizations to ensure maximum impact.
The report will also be distributed via the partners of the United Nations Development Porgramme (UNDP) Global Anti-Corruption Initiative Network. These include the African Development Bank Group, the Council of Europe and the International Monetary Fund.
The project recognises the need to reach different users through different forms of output. The main project report, which will have clear practical recommendations for the more effective management of digital records, is aimed principally at policy-makers and members of NGOs. Along with the main report, there will be an executive summary, which will be circulated to the international media by the SAS communications office at the time of the report's release at the international symposium. There is also a broad academic community including scholars working on records management and international development, and historians and political scientists working in the area of African studies, who stand to benefit greatly from the findings of the research network. The edited collection of articles is more closely focused on their needs. This is expected to be published in a special edition of a leading peer-reviewed journal. The project website, which will contain all significant materials generated by the project, is aimed at both sets of users.
The ICwS will also utilize its close partnership with the Commonwealth Secretariat to disseminate the project's findings. We will be closely in touch with the Secretary General's Office to brief its staff on the project from an early stage. The Commonwealth Heads of Government Meeting 2018 (which is due to be held in the UK) provides an excellent and inexpensive opportunity to acquaint Ministers, officials and members of affiliated organisations from across the Commonwealth with the project's preliminary findings. The Institute also has close links with the FCO and DFiD, which it will utilise to generate interest in the findings of the project. The International Records Management Trust of which Dr Thurston is director, has worked directly with a wide range of governments, international organisations and development bodies, particularly in Africa, and will use those contacts to disseminate the project's findings. Additionally, the project seeks to maximise its impact within sub-Saharan Africa itself by working closely with its partner organisation, the School of Information Science at Moi University in Kenya. Its Professor of Records and Archival Science, Professor Justus Wamukoya, is one of the leading figures in records and archives management in Africa. He has extensive contacts in the worlds of academia, government, records management across East, Central and West Africa. These will be vital to the dissemination of the project findings.
The CI, Dr Anne Thurston, has worked closely in the past with the Open Government Partnership (OGP), and the final report of the project, on 'Digital Records Protocols' will be disseminated through the OGP's many influential partner organizations to ensure maximum impact.
The report will also be distributed via the partners of the United Nations Development Porgramme (UNDP) Global Anti-Corruption Initiative Network. These include the African Development Bank Group, the Council of Europe and the International Monetary Fund.
The project recognises the need to reach different users through different forms of output. The main project report, which will have clear practical recommendations for the more effective management of digital records, is aimed principally at policy-makers and members of NGOs. Along with the main report, there will be an executive summary, which will be circulated to the international media by the SAS communications office at the time of the report's release at the international symposium. There is also a broad academic community including scholars working on records management and international development, and historians and political scientists working in the area of African studies, who stand to benefit greatly from the findings of the research network. The edited collection of articles is more closely focused on their needs. This is expected to be published in a special edition of a leading peer-reviewed journal. The project website, which will contain all significant materials generated by the project, is aimed at both sets of users.
The ICwS will also utilize its close partnership with the Commonwealth Secretariat to disseminate the project's findings. We will be closely in touch with the Secretary General's Office to brief its staff on the project from an early stage. The Commonwealth Heads of Government Meeting 2018 (which is due to be held in the UK) provides an excellent and inexpensive opportunity to acquaint Ministers, officials and members of affiliated organisations from across the Commonwealth with the project's preliminary findings. The Institute also has close links with the FCO and DFiD, which it will utilise to generate interest in the findings of the project. The International Records Management Trust of which Dr Thurston is director, has worked directly with a wide range of governments, international organisations and development bodies, particularly in Africa, and will use those contacts to disseminate the project's findings. Additionally, the project seeks to maximise its impact within sub-Saharan Africa itself by working closely with its partner organisation, the School of Information Science at Moi University in Kenya. Its Professor of Records and Archival Science, Professor Justus Wamukoya, is one of the leading figures in records and archives management in Africa. He has extensive contacts in the worlds of academia, government, records management across East, Central and West Africa. These will be vital to the dissemination of the project findings.
| Description | Development of a deeper understanding of practical issues affecting the integrity and credibility (quality, trustworthiness and authenticity) of statistics, data and records, the implications for measuring the SDGs and the implications for trust between government and citizens particularly in lower resource countries, especially in Africa. The major findings of the study are broadly that: * Poorly managed records make it hard to verify the quality and integrity of data generated to measure SDG indicators; this will undermine the government's e?orts to report on progress to the UN and jeopardise its ability to make good use of the findings. * It is therefore vital that governments develop procedures to ensure that records documenting data and statistics activities are captured, managed and integrated with procedures for conducting surveys, analysing data, merging data and reporting statistics. |
| Exploitation Route | A second workshop was held in May of 2018 to broaden the discussion that had taken place at the previous workshop in April 2017. The groundwork was prepared for an edited book (2020) and for the report, 'A Matter of Trust', published in November 2018. The report set out in practical terms the procedures governments need to set in place to ensure that development policy is carried out on a sound evidential basis and is measured using robust statistics. The edited book, published online on open access in December 2020, discusses these issues in greater detail. It explores, through a series of case studies, the substantial challenges for assembling reliable data and statistics to address pressing development challenges, particularly in Africa. By highlighting the enormous potential value of creating and using high quality data, statistics and records as an interconnected resource and describing how this can be achieved, the book aims to contribute to defining meaningful and realistic global and national development policies in the critical period to 2030. There is ample scope for follow-up projects, exploring in greater detail how the trends identified in the book have been played out and providing updates on subsequent developments. |
| Sectors | Agriculture Food and Drink Communities and Social Services/Policy Digital/Communication/Information Technologies (including Software) Education Energy Environment Healthcare Government Democracy and Justice Transport |
| URL | https://humanities-digital-library.org/index.php/hdl/catalog/book/amot |
| Description | Blog Posts |
| Form Of Engagement Activity | Engagement focused website, blog or social media channel |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | A series of blog posts demonstrating the potential damage that can be done by poor data as policy and funding streams become increasingly dependent on data as evidence of both need and progress. Published in the run up to the Open Government Partnership Summit in Paris in 2016, this blog series addressed Information Integrity through Metadata: Publishing Data with Context; Information Integrity through Systems: Building Audit Trails; and Information Integrity through Web Archiving: Capturing Data Releases. The blogs argued the tools and techniques developed in fields such as data curation, records management and digital preservation offer approaches to establishing and protecting the integrity of information. The posts called for the incorporation of these tools and techniques into open government initiatives, creating and publishing robust data. Information with integrity provides the evidence needed for accountability and participation. |
| Year(s) Of Engagement Activity | 2016 |
| URL | https://www.opengovpartnership.org/stories/summit-series-information-integrity |
| Description | Event at the House of Lords to launch 'A Matter of Trust: Records as the Foundation for Building Integrity and Accountability into Data and Statistics to Support the UN Sustainable Development Goals' |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | Event at the House of Lords on 1 November 2018, hosted by the Earl of KINNOULL to launch 'A Matter of Trust: Records as the Foundation for Building Integrity and Accountability into Data and Statistics to Support the UN Sustainable Development Goals'. There was a panel discussion which included academics and practitioners from the UK and Kenya who had taken part in the 2 workshops which had helped to shape the contents of the report. The event was attended by the Vice-Chancellor of the University of London, Professor Peter Kopelman, and brought together representatives from a number of domestic and international organizations with an interest in data management and development including the Commonwealth Secretariat, the Open Data Institute, the Institute for Internet and Society and the Institute for Advanced Studies on Science, Technology and Society. |
| Year(s) Of Engagement Activity | 2018 |
| Description | Workshop - Managing Digital Information as Evidence to underpin Global Development Goals |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | As the project is focussing on the quality and integrity of public information available to measure the goals, the workshop (in April 2017) explored the relationship between statistics, data and records as primary types of information for measuring the goals and to initiate an interdisciplinary dialogue between humanities scholars, development experts and information professionals, including statisticians, data experts and records managers. The discussions focussed on practical issues affecting the integrity and credibility (quality, trustworthiness and authenticity) of statistics, data and records, the implications for measuring the SDGs and the implications for trust between government and citizens particularly in lower resource countries, especially in Africa. The papers presented at the workshop and the discussions are summarized as follows: Difficulties of Gathering Reliable Data and Statistics The Expert Advisory Group on the Data Revolution has had a wide range of contributing specialists, most of them representing international agencies but also including a number of national statisticians and experts from civil society and academia. Many did not have experience of the realities and causes of broken systems, so the indicators do not always reflect what is realistically achievable. Nevertheless, the data revolution focuses on improving the way that data is produced, collected and used, on closing data gaps to prevent discrimination, and on building capacity and data literacy for both small data and big data analytics. Measuring the SDGs will be a fundamentally challenging task. Official statistics gathered by statistical offices are often so weak that in most developing areas of the world there is little prospect of using them to measure the goals in the foreseeable future. In addition to major gaps, there are fundamental integrity issues. Census data tend to be incomplete, limited, out of date, inaccurate, irretrievable or simply have not survived; undocumented changes in data structures and data entry errors make it difficult to compare data through time. Morten Jerven, whose study of economic development statistics in Africa, Poor Numbers, noted, 'International development actors are making judgements based on erroneous statistics. Governments are not able to make informed decisions because the existing data are weak, or the data they need do not exist.' As one of the workshop participants noted, 'I can cite some countries where there has been 10 years or 20 without doing a census. That is going to be a real problem in terms of achieving or monitoring development goals. If there is a volatile situation, an economic crisis, most developing countries would not be able to provide any data about the current situation.' The SDG approach, therefore, is twofold: firstly, strengthen national statistic offices to increase their capacity to collect data in a timely and efficient manner, and secondly, gather and amalgamate data from a range of government, civil society and corporate sources. Training statistical staff to capture high-quality data and interrogate it to evaluate the indicators in low resource environments is a lengthy process, both in terms of identifying funding sources and of delivering relevant and comprehensive training programmes. Statisticians need to follow rigorous codes of practice for applying, analysing and measuring issues in relation to error rates, and they need to be able to demonstrate how the statistics were compiled. Without that level of rigorous methodology, the results can be questionable. If the SDGs are to provide an accurate basis for evidence-based policy decisions or for donors to make aid allocations, there is a need for a major and sustained investment of resources. In the meantime, governments and the data community must take a blended approach to compiling statistics, bringing together different data sources in new ways, including data from GIS applications, social media platforms, crowd sourcing, satellite videos, mobile devices and a range of other sources. As a participant pointed out, 'Since the 1980s, lots of data and records that previously would have been created by state actors in developing countries have been developed by non-state actors, such as international agencies and local and international NGOs.' The UN is hosting a series of workshops at the country level as a basis for building country data road maps. The workshops aim to galvanize political commitment, align strategic priorities, foster collaboration, spur innovation and support the process of combining data in new ways. In 2016, for instance, workshops in Kenya and Tanzania generated awareness about the SDGs by bringing together several hundred different stakeholders, including staff of National Statistics Offices, government officials, civil society partners and academics to examine the roles of different stakeholders and facilitate understanding about the emerging data ecosystem, including capacity and budget aspects. They highlighted the value in overlaying the SDGs indicators and national development plans in order to assess data gaps. Awareness also has been expanded through a series of 15 sub-national pilots organised through the Open Government Partnership and through local community workshops, which aimed not only to support the SDG process but to open information to citizens, governments and businesses. Initially, the emphasis was on making the data available, rather than on how it was compiled. Now, the approach is to move toward quality data that can be trusted and reused. Essentially, data quality rests on how the data has been managed. As one of the participants noted, 'We need to be asking: Where has the data come from? How was it compiled? Why was it compiled? What was the sample set? How was it amalgamated? What methods were used to analyse and interrogate it? What algorithms were used to enable its interpretation? Who published the data and when? When multiple data sets are combined, who owns the amalgamated data set? Has the process been carried out transparently? How effectively has the data been anonymised?' Moreover, data must be documented transparently if it is to drive policy and service delivery. Amalgamating data from two or more sets of administrative or survey data from different organisations can enhance the quality of existing data and maximise its value for research and statistical purposes. However, mapping together different data structures from multiple data sets, many of them with broken data linkages, to arrive at reliable composite statistical findings can be a challenging and complex process. For instance, as a participant remarked, 'Data sets are often created in silos. It is quite difficult to find a way to unify them to make better sense of what the data is telling us. The fact is that these data sets are on different servers in different locations.' There may also be multiple versions of the same data sets. If the statistics are flawed, they may not provide a reliable basis for policy decisions or for determining aid priorities. The situation is complicated by the fact that many members of the data community are self-trained and may not follow rigorous data science methodologies. As a participant noted, 'Doubts about the identity and integrity of the data are introduced by partial metadata, opaque provenance, undocumented custody, particularly during aggregation as well as by the lack of information about the systems for the data's management.' In the future, it would be valuable if metadata indicating the controls used to manage data integrity could be captured, at least for key data sets. Common core metadata vocabularies, which are key to reliability and integrity, are starting to be created, as people from the open data community join forces with technology experts and experts in the data management field. The emerging norm is that each piece of information should be associated with agreed core metadata. There is also an emerging requirement to document who was in charge of the data, with alternative contacts in case the person leaves. 'When you open up a data set, you start to go through a process of: Where did this come from? Does it meet metadata criteria? Have I actually been able to document it? Is it free of privacy and security concerns?' Once the data is online, it is seen by many people from many backgrounds, who ask questions that help to strengthen the validity of the data. As a participant noted, 'The philosophy is that if you are able to have multiple sector participants, multiple organisations, individuals actually interacting with data, you will improve its quality and you'll actually be able to figure out what its usefulness is.' As public sector transparency, accountability and openness have emerged as predominant international development themes, it has been widely accepted that opening data to citizens will enable them to participate in state affairs, monitor how government money is spent, hold public officials accountable for their actions, and participate in good decision-making. By focussing on non-personal data, non-proprietary data and data to which national security restrictions do not apply, open data can bypass the restrictions on opening records that could cause harm or embarrassment to individuals, making it possible to move beyond official secrets acts and 20 or 30-year rules to immediate use. As governments turn to big data analytics, the same principles apply: the methodologies used to interrogate the data need to be documented and transparent and the sample should be clear. Without understanding the methodology used to collect and interpret the statistics, visualisations can reflect distorted pictures. For instance, censuses in Africa sometimes exclude nomadic tribes, and even though the portion of the population may be small, without this information the picture can be distorted. The UN is well aware of the importance of creating an environment where the public trust the use of big data for official statistics and where privacy and confidentiality of personal information can be assured. Among the most significant issues that need to be addressed are: Who owns the data? Who is responsible to managing it? How will it be preserved through time? Relationship between Reliable Records and Data Records make an essential contribution to sustainable development that is complementary to, but different from, data and statistics. Data and statistics document trends and patterns, with personal information anonymised to protect the individual. Records do the opposite. They document individuals' rights and entitlements, and they provide specific evidence to document accountability and transparency. The metadata associated with records describes their context, custodianship, content and structure and their management through time. It gives them value as authoritative evidence of specific policies, actions, decisions, precedents, transactions, which are the foundation for the rule of law. Well-kept official records contribute fundamentally to all aspects of national development. They provide the audit trail for official financial transactions and the documentary evidence for pay and personnel management, income tax collection, corruption control and land ownership. For instance, records are the basis for loans, whether this is documented by records of credit histories or by records collected by private sector actors. They are a key to land management and development. Subsequent changes to a record, when documented through metadata, create an audit trail that makes it possible to identify fraud or illegal actions. Moreover, only the records profession has developed the means of protecting, preserving and accessing digital information through time. Neither data analysts nor statisticians have the skills or have developed the structures needed to preserve digital information so that it can be used reliably in the future. In the digital environment, it is the synergies between the professional approaches to managing records and data that provide the key to maximising the use of the information and building confidence that the public can trust it. Managing records and data in a similar light is of fundamental importance. When digital records began to be created in growing quantities, few people, in the records profession, in government or in international agencies, realised how quickly digital information would become a major source of government documentation or how easily it could be lost. Citizens and governments rapidly came to rely on digital records created on desktop computers, in databases, in email, on mobile devices, on websites and via social media platforms, but there tended to be little understanding of the skills and structures be needed to manage these records even of which government agency should be responsible. Often management responsibility was split between several government agencies, for instance the one responsible for ICT development, the one responsible for access to information and the one responsible for culture. Governments and donors worldwide tend to believe that information produced in computerised systems will offer the basis for planning, monitoring and measuring national and international development goals. Most do not realise that IT systems create records but lack the full functionality needed to keep them reliable and authentic for as long as they are needed. As a result, IT systems have been developed without the supporting framework of policies and systems needed to protect, preserve and make digital evidence available through time. Digital records are fragile. If they are not managed professionally, their integrity and their value as legal and historical evidence can be compromised or they can be lost completely. Their integrity depends upon a quickly changing array of hardware and software. Digital media deteriorate, software changes and hardware becomes obsolete. Digital records may be stored on personal drives, un-networked computers, unmanaged network drives or mobile devices, which can make them unavailable as a national resource and unlikely to survive. Different versions of digital records may be stored without adequate identification, making it difficult to rely on the evidence as authoritative. They can be altered, deleted, fragmented or corrupted through malicious interference or inadequate management; their meaning may be lost when metadata is not captured, is imprecise or becomes separated from the records when technology changes. They can be difficult to retrieve after a few years, months or even days. Very large volumes of digital records are being lost regularly in many countries through the lack of systematic approaches to preservation. The World Bank's 2016 World Development Report noted, it 'is fair to say that long-term preservation of digital records and information in most countries in the world is at serious risk'. Fortunately, the international records community has worked steadily to develop standards, requirements and management tools for protecting and preserving records integrity, and this work is available as international standards that can be shared. Unfortunately, it is little known or understood outside the records profession and sometimes not even within it. Development planners sometimes think of records as outdated sources of government information, not realising that many records are now 'born digital' and must be managed from the time that they are created if they are to survive. Records are defined in international standards as 'information created, received, and maintained as evidence and as an asset by an organization or person, in pursuit of legal obligations or in the transaction of business. They may be in any medium, form, or format.' Nevertheless, the words data and records now tend to be used almost interchangeably, and what the records community has called records are often referred to as data. For instance, records of disease rates compiled in hospitals are often referred to as disease data, and birth and death records are referred to as birth and death data. Records created to document the day-to-day activities of the state are known as administrative data. The lines are blurred further by the fact that records are being created in databases, which are often outside of record keeping systems. Indeed, for many civil service users, there is no clear distinction between what is a record and what are data - it is all part of a whole. In many parts of the world, paper records still provide the source documents from which a substantial amount of data are derived. Appropriate content is identified in records, which is coded and input to structured database systems. Software then is used to process the data to create statistics of various kinds, which can be used for purposes including administrative and financial management, strategic planning and decision making. For example, employment data is derived from records of staff appointments, agricultural data derived from records of land surveys, environmental data from records of impact assessments. In these cases, the quality of the data depends on the quality of the original records from which they are drawn. Data from poorly kept records can contribute to skewed findings and misguided policy recommendations and misplaced funding. Especially in developed countries, records are increasingly born digital, and analytical tools can be applied directly to records and records systems, making it possible to use and analyse them in new ways. Particularly in relation to big data, new theories and practice are emerging, and new research agendas, leading some records professionals to suggest that it may be time to abandon the distinctions between records and data. For instance, it has been suggested that records and data may be described as two different aspects of the same entity, with records themselves being data that can be mined, analysed, reused and repurposed. A participant noted, 'Datification involves transforming records into data so that they can be analysed in depth using new computational analytical techniques from the world of big data and artificial intelligence that make it possible to detect and analyse themes, patterns and relationships between digital materials and can open up innovative modes of discovery and investigation.' Nevertheless, many records professionals still to use the word records to indicate that authenticity has been protected through professional practice and control systems. Overall, the view is that the greatest gains will be achieved when the records and data communities can work together to achieve optimum information usage and benefits. As the boundaries between the professions shift, it is often unclear, for instance, who is responsible for protecting and preserving digital information. When, for instance, is it essential to capture the provenance or the context of the information as an audit trail? Who decides? Who is responsible for ensuring that this is done? How long does the information need to survive? What policies, practices and procedures are in place to ensure that it will survive and be accessible for as long as it is needed? Digitisation Governments and international organisations often see digitisation as a quick means of ending dependence on paper records, and digitisation initiatives are widespread in many countries. However, many fail to incorporate requirements for legal admissibility, reliability and usability, for instance requirements for capturing metadata, image resolution and classification structures that aid search and retrieval. As a participant noted, 'Metadata is going to be absolutely crucial, as will documentation of the processes - what were the decisions about what was digitised, how was this captured. Where was it captured, when was it captured, who were the authors? It's not always clear.' Without this information, it can be very difficult to retrieve and use the information through time, and if the paper records have been destroyed, the records are lost entirely. Vast amounts of money have been spent on digitising records without provision for accessing or preserving them. The participants gave numerous examples of failed digitisation projects. 'I know a number of countries where digitisation has been done and just a couple of months down the line the records cannot be accessed.' In one country, land records became disorganised to the point that citizens began to distrust the government. The government then decided to digitise the records, but when digitisation began, it was discovered that the records were in such disarray that records management teams had to be brought in to organise the records. Digitisation is a big business, and vendors inundate government offices in lower resource countries promoting digitisation systems that are expected to solve all retrieval problems. Organising the paper records, linking the paper to the digital files and setting up records centres is often a cheaper and more reliable solution in low resource environments. A digitisation policy is needed at a national level, not only for what is digitised but how it is to be preserved. 'It makes me remember a proverb,' one participant said. 'You cannot plait the hair of a girl infested with lice': in other words, simply layering technology on top of a chaotic situation will not resolve it. Access to Information (ATI) Access to Information, the other major prong to the international strategy for transparency and accountability, also depends on the ability to retrieve accurate information. It is expected to support citizens' rights, effective services, anticorruption measures and improved investor confidence. Theoretically, when citizens have access to information, they are well informed and are empowered to participate in governance and democracy. They are able to hold their governments accountable by seeking and receiving information on how national resources are used and managed. As Henry Maina, the Director of ARTICLE 19 Eastern Africa, wrote after the passage of Kenya's ATI law in 2016: 'The passage of this Bill heralds a new era for transparency in the country, as citizens will now be enabled to access and use information to hold their government accountable and promote legitimate good governance in the country.' However, the law was passed without introducing adequate structures for protecting, accessing and making the country's digital records available. If records cannot be accessed, if they are lost or corrupted through time, or if they are so poorly kept that the cost of locating them is prohibitive, the law will have little impact. One participant cited cases in her country where the records could not be accessed for ATI requests because they were held on legacy systems that didn't talk to each other, because they had not been properly described or because they had been dumped in the equivalent of an electronic filing cabinet with no thought about how they would be accessed. She noted, 'The cost of poor records and data management is more than just financial. The true cost is the relationship between the citizen and the state. There is a huge benefit of good records management that enables a good communication between citizen and state, probably carries a value that we will never measure because it is immense.' A recent UNESCO publication, Access to Information: Lessons from Latin America' , noted that 'Access to information laws cannot function properly unless government records are organized and available in professionally managed archives, which requires substantial financial and human resources; these public records should ideally be available digitally as well as physically.' As a participant noted, 'You can have the best law in the world, but if there is no political will to implement it, you may as well not bother. FOI laws are seen as a universal panacea - if you have an ATI law, then you're well on your way to being open. I just don't think it's true.' Another observed, 'Our law has been created in a vacuum. Almost 95% of the information requested through the law is going to be available in records, but nobody knows how records are being managed. The people who manage records were not part of this initiative at all; nobody is accountable.' In addition, other laws, such as secrecy laws that have not been repealed, can make it difficult to access information. Aggregating information as data can make it difficult to respond to ATI requests. To illustrate, users may see crime data on an open data portal, but it may not be possible to match it to the relevant records of crimes recorded at police stations. Similarly, statistics may be provided on road accidents, but it can be very difficult to track back to the records documenting individual accidents. From the historians' perspective, access to information laws may actually inhibit record keeping. 'I just wonder,' said one participant, 'If the records that are to be managed will be more insubstantial because of Freedom of Information.' Another noted, 'As soon as a government official sees Freedom of Information, immediately they want a string of exemptions, and those exemptions are so extensive that now the shutters have almost gone down historically.' It is likely that under ATI laws, less information of substance will be recorded and more will be exchanged verbally, even where the 'duty to document decisions' is enshrined in civil service codes or records laws. Protecting Information Integrity through Time Information integrity is intentional; it requires management. In most countries, the structures for managing information integrity were designed for an earlier world and have not been upgraded or expanded to support the requirements of the public sector digital environment, whether in less resourced or better resourced countries. Data no longer flows along traditional lines or between traditional entities. One participant noted, 'What worries me, in policy terms, is the gaps. No one knows who is responsible for the selection and long-term preservation of government data, and nobody is paying attention. Where are the boundaries of responsibilities of the national archives?' Even to identify what is public sector information can be very tricky, with multiple stages in delivery of publicly funded services. At present, in most cases, no one has responsibility for capturing, protecting and preserving the large amounts of digital information that are produced and lost daily across the world: 'My biggest problem', said a participant, 'is the long-term sustainability of data and its availability. There is no coordination whatsoever between the ICT authorities and the national archives/records staff. That is going to have a long-term effect, not only on data sustainability but on the integrity of the data. I am not sure that they are thinking about the availability of data in the future'. The ability to preserve data sets requires continuous use of standardised metadata to document the origin and use or the data set and the way it has been managed. Data models, processing rules, and guidelines for standardised structures and mandatory metadata fields are being defined in both records and data standards and could be harmonised. Without these guidelines, even within a short time, it can become difficult to understand the content and context of the data. As one participant noted, 'Many data portals do not include enough metadata within a file to guide the interpretation of the figures and no formula to provide the key to the relationship between the figures in the columns. The content, context and structure of the data sets are insufficient for interpretation.' Digital records also are being lost regularly on a very large scale, particularly in lower resource countries where the structures needed to protect and preserve them have not been developed. Part of the problem is the culture of the civil service, where the protection and preservation of digital information is a low priority. Civil servants often focus on delivering a service for a given budget and are not concerned about future uses beyond service delivery. A participant explained, 'They are not interested in what is going to happen to the records that they create. Records proliferate. What to do with them? They don't know.' Governments generally give far more attention to ICT than to managing records and data. 'When it comes to issues of data management, the ICT people are the ones concerned; the National Archives has no responsibility.' The perspective of IT staff is completely different from archivists, particularly with regard to preservation. In the IT view, preservation means backups for as long as the data is wanted. For the records community, preservation means the ongoing ability to protect and preserve the authenticity, reliability, integrity and accessibility of the record across time and technology changes. 'The issue is who sets the standards? Who determines what happens with records.' These are big issues and big differences that need to be addressed seriously. If governments are going to use data and statistics to inform policy decisions and support openness in relation to the SDG goals, the information needs to be reliable. Reliability requires legal and administrative control frameworks as the basis for achieving quality and integrity. As one participant noted, 'It is only when capturing, identifying and preserving the information becomes a routine part of government systems that the information can be really be trusted and used effectively.' Essentially, this means moving beyond simply managing technology to recognising that high quality information is an essential national resource. It also means taking a holistic view of how digital information is managed, taking account of the synergies between statistics, data and records. 'It is about creating reliable information from the beginning while we have the opportunity to do something about it.' Historians know that in the future, they will be working with enormous volumes of digital information. 'We need to find ways of being able to reconstruct, in 50 or 60 years' time, the interaction between government and citizens. We're going to be working with hybrid archives for a very long period. It is important that institutions do not to treat their paper differently from their digitised and from their born digital but develop an ecosystem that allows all the different bits to be connected up so that we're going to get a full picture of how people were working. Otherwise, that's not going to be easily reconstructed.' Hopefully, when the SDG period is over and a new set of goals is introduced, reliable records and data will be available as an ongoing point of reference. 'We need to have a strategy, if we are going to embrace digital information, for keeping this information in a way that we'll be able to access it in on one year, five years, fifteen years from now.' |
| Year(s) Of Engagement Activity | 2017 |