From Open Data Initiatives to Open Knowledge Initiatives!

Aside

ODIsMany organizations recently launched Open Data Initiatives (ODIs) to facilitate the production of a wide-range of socioeconomic data, ensure open access to data and improve the statistical capacity of countries and international partners. In 2008, the World Bank decided to throw open its rich database of more than 8,000 plus time series indicators for free. During the second birthday of the Bank’s Open Data Initiative on April 20th, 2010, President Robert Zoellick stressed that statistics tell the story of people in developing and emerging countries and can play an important part in helping to overcome poverty. The United Nations Economic Commission for Africa (UNECA), under the new leadership of Dr. Carlos Lopes, is undergoing an extensive restructuration of its activities with a strong emphasis on statistics as “the backbone of proper planning for Africa’s future.” Knowledge management and development was one of the key components of the African Development Bank (AfDB) mid-term strategy (2008-2012) and the Bank will continue enhance its knowledge services, a necessary and still underserved complement to its lending activities.  Other institutions like Data.gov and the Open Development Technology Alliance has also developed their ODIs.

The main objective of each ODI is  to improve data quality (timeliness, reliability, accessibility and relevance) and its availability. Data types suitable for an ODI include statistics used by citizens, governments, and the international and donor communities. The types of use for statistics depend on various factors and contexts including government policy measures, information to citizens, democratic debate on issues of public policy, need for information by potential foreign investors and visitors to the country, accountability and fiduciary responsibilities to foreign governments and international institutions for their assistance, and national surveillance requirements.

The balance between these different priorities and the overall demand for statistics has been affected by changes in the aid architecture. Poverty Reduction Strategy Papers (PRSPs) incorporate a set of indicators to monitor and  select outputs and outcomes. All developing countries and their development partners have signed the Millennium Declaration and need to monitor the associated Millennium Development Goals (MDGs) that assess development outcomes. The introduction of budget support and performance-based budgeting has led to a newer demand for more detailed information relating to service delivery and to results from smaller geographical areas. This latter demand more closely follows some of the needs of governments in managing the delivery of services and seems to be leading to stronger demand for national data.

These statistics are usually generated by economic surveys, and regular administrative collections corroborated by occasional sample household surveys. Improving data quality involves the art and science of creating useful surveys to provide to citizens, data about local conditions, headline per­formance indicators, topical, social, political, and economic issues. It also involves collecting high definition local data and fore­most economic and financial statistics for country management (including policy design, pub­lic sector performance, and resource allocation) and nationally representative data on development outcomes and impact (such as those captured by the MDGs) as prioritized by the international community.

Technically, an Open Data Initiative (ODI) is a freely available online service for the creation and dissemination of data for public consumption using Web 2.0 technologies. The idea is simple: you have the data and you provide the service to disseminate it to the public. Users discover and explore data in a rich, interactive, and intuitive application, rather than browse or read large documents of published tables and charts. The end-user can select and visualize any combination of data. It can be exported, printed, linked to, and shared in collaboration environments.

Various technologies are used for the development of the online open data systems. These tools include: a) Microsoft Visual Studio 2012, the development IDE for developers using C#;  b) the Web Technology ASP.NET MVC 4 and WEB API;  c) the Javascript Library for creating Web controls called Kendo UI; d) the famous Javascript library jquery; e) the Javascript Library providing a canvas for creating diagrams called Northwoods GoJS; f) Microsoft SQL Server 2008 for databases storage; g) Microsoft Team Foundation Server 2010 to provide source control for the code, integrates with Visual Studio 2012, and provides a powerful bug and task management system for internal development use as well as testing at a later stage; h) Microsoft Windows Server for hosting the web and windows servers and services (MS SQL and TFS); and i) the Web server  Microsoft IIS. Google AppEngine Cloud, for instance, enables providers of public data to create engaging and rich Web 2.0 experiences.

The issue of weak statisti­cal capacity and the impact this has on development is well documented. In a synthesis report on the first phase of the evaluation of the Paris Declaration, the authors emphasized that reliable statistics are vital for good policy, to measure progress, and to report on development results at local, national and international levels. Yet the findings of the first phase of this evalu­ation highlighted that of all the Paris commitments, progress has been slowest on ‘managing for development results’; and that there is therefore a strong need to strengthen statistical capacities and to use them more ef­fectively for decision-making.

The benefits of such initiatives are the following: a) the potential reduction of the workload on data analysts and researches (cost reduction and time efficiency); b) the provision of data that is complete; c) the provision of a new online data service to the public (data as a service); d) the provision of an engaging and rewarding experience for the public (improving trust relationship); e) the confidence that the public are seeing the right numbers, graphs, and maps, and reaching the correct interpretation and understanding behind those numbers (data integrity); f) the reduction of  the time between data collection and data dissemination to ensure maximum relevancy of the data to the audience (data responsiveness ); and g) the promotion of online experience that can be captured and shared by the public in collaborative environments such as blogs, LinkedIn, Facebook, Google+, or Twitter.

ODIs’ purposes and methods fall on a continuum. Some have far-reaching generalizable effects. Others are conducted to meet specific needs. Some ODIs (sometimes called data or knowledge discovery or data mining) are based on a process of analyzing data from different perspectives and summarizing it into useful information. An ODI can also include a set of advanced models and technologies that transform raw data into meaningful and useful information for business purposes. These business intelligence techniques can handle large amounts of information to help identify and develop new opportunities. Finally, ODIs involve knowledge management techniques with the ability to create and manage a corporate culture and structure that encourage and facilitate the creation, appropriate use, and sharing of tacit, explicit, and embedded knowledge to improve organizational performance and effectiveness.

Obviously, the above open data initiatives go beyond the simple manipulation of statistical data. They are part of a large spectrum of activities called Open Knowledge Initiatives (OKIs).  An Open knowledge initiative is a process that encompasses nine activity classes: five primary activity classes (acquisition, selection, generation, assimilation, and emission) and four secondary activity classes (measurement, control, coordination, and leadership). Knowledge is not data or information. Data is simply raw facts without context. Information is data that comes with context. Knowledge is information that is contextual and relevant of event, as well as actionable by something like human or agent. Knowledge is information in action.

With increasing recognition that organizations operate in hypercompetitive, high-velocity, or rapidly changing environments, the question of how they could manage knowledge merits an explanatory framework that reflects (a) the context where OKIs takes place, (b) processes that show how to conduct OKIs, and (c) consequences of such initiative.

The recent “amazing mea culpa” of the top economist of IMF, Dr. Olivier Blanchard, that the Fund blew its forecasts for Greece and other European economies because it did not fully understand how government austerity efforts would undermine economic growth is one of the reasons to consider OKIs. Dr. Blanchard agreed that he could not actually determine what multipliers economists at the country level were using in their forecasts. This number was implicit in his forecasting models, a background assumption rather than a variable that needed to be fine-tuned based on contextual circumstances and peculiarities.

Beyond this simple mathematical error (using a uniform multiplier of 0.5 when in fact the circumstances of the European economy made the multiplier as much as 1.5), there were recurrent problems in official Greek economic data on public finances, whose reliability has been impaired by inappropriate accounting methods, the application of poor statistical methods and deliberate misreporting. IMF failed to detect these issues. A robust OKI system that collects, collates, and analyzes countries data on deficits and debts could have helped IMF to better understand economic, fiscal, and financial policies that should be proposed to member states. One of the implications for the future of Blanchard’s mea culpa is how IMF should improve its surveillance methodologies by including OKIs for a better preparation of their Public Information Notices (PINs).

Statistics (data) are used not only by citizens or governments but also by the international and donor communities, among whom demand has been particularly strong in recent years. The creation of a global society with possibilities of knowledge sharing is among the contributions of the IT revolution and globalization. In the knowledge society, the value-creating strategies and long-term viability of organizations depend on sustaining open knowledge initiatives.  At their meeting in February 2013, heads of the multilateral development banks, the IMF, and the United Nations (U.N.) agreed to strengthen collaboration between their institutions on issues related to data and statistical capacity building in their member countries. A Memorandum of understanding was signed that sets out principles and modes of collaboration.

Open knowledge initiatives are growing areas of work, particularly with the increasing demand for better knowledge to monitor development efforts. A global strategy for OKIs is vital for planning, implementation and monitoring of country management initiatives and development outcomes and impacts. Among other things it identifies capacity needs, organizational change proc­esses, priorities, and resources requirements.

In terms of Millennium Development Goals (MDGs), a worldwide strategy for OKIs should address the assessment of the blueprint agreed to by all the world’s countries and leading development institutions with the intention of halving extreme poverty rates to halting the spread of HIV/AIDS and providing universal primary education, all by the target date of 2015.

OKIs should address coordination prob­lems and facilitate the understanding of countries challenges with appropriate methods and tools. Extensive open knowledge initiative coordination and collaborations should be carried out among the key international organizations such as the World Bank, IMF, OECD, AfDB, the U.N. system, etc. For example, the various institutional OKIs need to be flexible to facilitate their integration with national databases when available through service oriented architecture and web services.

==========================================================

The author’s involvement in Open Data/Knowledge Initiatives:

Dr. Nkoyock is a scholar and a professional. He is a Professor of Management, Leadership, Mathematics, Statistics and Computer Science. He earned a doctorate degree in Management in Organizational Leadership with a Specialization in Information Systems and Technology, a Master’s degree in Software Engineering and a Bachelor degree in Mathematics. Nkoyock’s PhD quantitative research project used Structural Equation Modeling (SEM) with the multivariate analytical software LISREL 8.8 and the statistical software SPSS 16.0. Currently, he is involved in a research project, called LmiData, an Open Data/Knowledge Initiative that covers the  full life-cycle of online surveys, data analysis, automatic data interpretation, and dissemination. As a professional,  Dr. Nkoyock  is currently a project manager for the development of the Public Procurement Review Software (goPRS, available at goprs.unodc.org), a suite of software packages comprising of goPRS Enterprise, goPRS Web, goPRS Intelligence, goPRS Learn, and goPRS eGP. goPRS is a prevention corruption tool for public procurement regulatory authorities.