What is the new potential of datadriven decision-making in the public sector?
H.E. Justin Ho Guo Shun
(Special to the BINN)
The problem: While a lot of data is collected for various reasons, there is little data that are relevant to address key policy issues.
Why it matters: Policymakers need data to generate actionable intelligence for decision-making.
The solution: We will need to recognise the complexity of policy issues, embrace the mindset of innovation and develop a new capacity to explore and test various possibilities of data for policymaking.
Early this year, United Nations Development Programme (UNDP) Maldives invested in data science in support of the Government of Maldives to explore the potential of data and innovation and address some of the complex development challenges the country is confronted with. UNDP in the Maldives aims to generate actionable intelligence informed by data to support decision-making and create a learning journey to develop a new set of competencies in public administration and civil service in harnessing innovation for sustainable policy options in development.
Against this backdrop, UNDP has partnered with the Ministry of Economic Development to lay the foundation of the Economic Research Centre to support the full exploitation of rich data set at the disposal of the Ministry and derive from its strong and evidence-based policy recommendations while at the same time embedding robust data governance.
After a period of exploration, UNDP is ready to share the first set of reflections on the new potential of data-driven decision-making in the Maldives’ public sector.
The paradox of data
The country is not exempt from the known paradox of data. The Maldives, like other countries, has too much but also too little “usable” data–too much data being collected for various reasons and not used properly, meanwhile, too little data that are relevant to make it to the policymakers confronted by deep routed development issues.
Sometimes, more data doesn’t necessarily translate into more insights because uncertainty is mutable and immeasurable by nature. As Edward Price pointed out, “if we simultaneously have more information about, and far less understanding of, the modern world, our ignorance is born not of fewer, but greater facts.”
Why does the paradox of data happen? It largely lies in the discrepancy between the nature of the complexity inherent in the development challenges we intend to make sense of and the capacity of relevance conveyed by data we can hold on to.
Since there is a network between each element in our society and environment, the direct aftermath of one problem can further generate ripple effects across multiple areas and amplify the impact through this chain reaction.
Going beyond data
Development challenges like climate change, youth unemployment, and social inequality are affected by various intertwining factors like awareness and behaviours in society, demands in the market, and the exploitation of natural resources.
Since there is a network between each element in our society and environment, the direct aftermath of one problem can further generate ripple effects across multiple areas and amplify the impact through this chain reaction. That’s why the complexity of the development challenge is profound. And being able to unpack this complexity is key to developing effective solutions. With the advance of digital and technological development, we have seen more data captured and utilised to help us better identify and understand problems. But often, either the data needed is not available, or the complexity is not intuitively measurable.
So, what did we do in the Maldives about it? What does it mean for data-driven decision-making in the public sector, especially in the context of using it for addressing complex development challenges?
Here are some lessons learned from our early exploration and thoughts on the new potential of data-driven decision-making in the public sector:
Start from the problem to get a direction: It is well understood that data is valuable in terms of providing insightful and actionable intelligence for decision-makers. But what kind of data are we referring to here? How do we know what data is needed? To answer these questions, we need to start from the original development problem we aim to address, ask questions and dig into why certain things happen, and accordingly, create linkages between them and develop a roadmap to guide data exploration towards a better understanding of the problem and the identification of policy priorities. Therefore, rather than going straight to the data from the beginning, we had a series of in-depth conversations on the pain points in unemployment issues from the symptoms to the root causes and generated a mind map of key policy research questions to guide the data exploration.
Widen the scope of data to enrich insights: Acknowledging that data provides a lens for us to peek at the development problems, we could potentially go one step further to situate ourselves in the data to empathise with the stories behind the numbers, ask questions that were not reflected in the datasets, and generate a deeper understanding of the problems. This requires us to delve into people’s aspects–their experience, needs, concerns, desire, beliefs and behaviours and enable a participatory process to enrich the data flow. Therefore, the meaning of data should not be confined to the quantitative realm but can extend to qualitative information that most often has been missed. Realising the importance of making use of qualitative data, we also referred to data sources from interviews and focus group discussions which helped to explain the possible reasons behind the patterns disclosed by the quantitative data analysis. And we are keen to conduct stakeholder consultation and community engagement to gain more perspectives from the people.
Make the connection to hit decision-making points: Eventually, the question of what we can do with this data will come up from the decision-makers. Deriving from the development problem space, the loop needs to connect back to the decision-making process to realise the value of this data exploration journey. This means insights obtained from data analysis need to be translated into policy intelligence that can help decision-makers to take action. To overcome the limitation of the traditional approach that starts from and solely relies on data, the insights generated from data should be combined with other strategic intelligence such as future development, risk management, and capability assessment to ensure relevance for decision-making.
Cultivate the condition to amplify the value of data-driven decision-making: Despite the value of data-driven decision-making being well-known to many government agencies, most of the time, organisational silos hinder the data flow and pose an opportunity cost for missing the value of getting a more holistic picture from intergovernmental data sharing. With the understanding of the concerns over data privacy and protection, robust data governance supported by a strong legal framework, data sharing modality and technology, and governance mechanisms will need to be put in place. Therefore, moving forward, we will continue our endeavour to strengthen the portfolio of data governance and explore more options for mainstreaming open data. Last but not least, the value of data-driven policymaking shall be demonstrated and disseminated across the organisation to incentivise teams beyond the data analytics unit to embrace the new mindset and cultivate the culture of data and innovation.
Once good development professionals were weighted by the mileage of direct experience in multiple contexts, today development is all about how capable you are to ingest and be able to trace connection often with a large set of information–developing new sensorial experience with data is certainly in the deck of a new generations of professionals, it will matter to keep intact the ancestral pleasure of personal discovery.