29.11.2018

Following Webinar 1 Capacity Development in Data and Statistics: A waste of money? a few participant questions are addressed below.

Q: Is there a strategy to introduce politics to the importance of capacity development for statistics?

A (Lisa Denney): There are a few ‘movements’ or communities of practice that seek to do this. The thinking and working politically community of practice, doing development differently community, and the problem-driven, iterative and adaptive approach are all good examples. Adaptive programming (for instance, USAID’s Collaborate, Learn, Adapt approach) is another example. These are not about capacity development specifically, but about how development assistance can better engage with politics, and the political drivers of dysfunction.

A (Shaida Badiee): While there is not one universal strategy to involve politics into the importance of capacity development in statistics, there was hope that including the statistical capacity-related SDG indicators (17.18.2, 17.18.3, 17.19.1) in the SDG framework would create a political buy-in for statistical capacity building and increase the resources allocated to it. There are also initiatives at the international, regional, and national level to involve politics in statistical capacity development. At the international level, the United Nations Statistics Divisions has placed efforts to bridge technical and political communities together through the United Nations World Data Forum. This conference marks an opportunity to raise the political profile of data and statistics by involving different government departments or ministries and policy-makers. This often happens at the regional level as seen at the African Open Data Conference specifically and events held by UN Economic Commissions more broadly. At the national level, there have been successes through partnerships between data producers and users. This means involving end-users (usually the government policy-makers) in data conversations early on in the process. Showing policy-makers what they can accomplish with data creates the incentives to not only use the data for policy-making but to build support (political and financial) and demand for statistical outputs from the national statistical office. Here is a good example of this playing out in practice.

Q: Is it possible or useful to conduct an impact evaluation study to extract concrete/actionable learnings from capacity development interventions at a more aggregate level? (Hearkening to Lisa's initial point on lack of enough learning taking place in the field overall due to a lack of innovation and actors employment of rather homogenous models across sectors etc.)

A (Lisa Denney): So many studies of capacity development and its limitations have been done (see EDCPM’s excellent work on this, for instance) that I am not sure more reports and lessons learning exercises are the answer! I think the question to ask is why, despite all the studies and lessons learned that show our tried and tested models produce limited results, do we keep doing the same thing? I think this speaks to the political economy of the wider aid industry:

  • The push for tangible, easily quantifiable outputs that we can put in log frames and results frameworks;

  • A fundamental belief that more knowledge or skills will change behaviour when we know that our theories of change need to be more nuanced than that.

Without addressing those things, we are likely to keep doing the same thing. In the SLRC synthesis report, we argued that we need a re-politicisation of capacity development. We now think of CD as very much a technical issue – but it’s not. It’s about a process of social change that we hope will produce development outcomes from which some people will win and others stand to lose (at least in the short term). I think we need to re-engage that with politics and put it front and centre a bit more.

Q: Do the panellists think that the SDGs provide a good hook/momentum for capacity development for statistics?

A (Shaida Badiee): Yes, the Sustainable Development Goals (SDGs) provide a good hook for capacity development for statistics. From the beginning, the SDGs have been closely tied to the United Nations Data Revolution and the 2030 Agenda is underpinned by the need for high-quality data. There was also a lot that was learned from the Millennium Development Goals, the predecessor to the SDGs. Moving into the next (far more ambitious) global agenda, the SDGs, it was well understood that without the right international support to increase statistical capacity, monitoring the SDGs would be almost impossible. Not only are data needed to measure and track progress towards each of the individual targets, the national capacity for statistics is found within indicators: Indicator 17.18.2: Number of countries that have a national statistical legislation that complies with the Fundamental Principles of Official Statistics. Indicator 17.18.3: Number of countries with a national statistical plan that is fully funded and under implementation, by source of funding. Indicator 17.19.1: Dollar value of all resources made available to strengthen statistical capacity in developing countries. There is a consensus that the SDGs place a tremendous demand on the international and national statistical systems. Alongside this consensus is the understanding that not all countries have the capacity to collect, produce, and publish the data necessary to meet those demands. Besides traditional methods of data collection such as census, household surveys, or civil registration systems, the SDGs also put pressure on statistical systems to produce new data using new sources, implement new standards, and develop new methodologies to provide missing information for critical indicators. Strengthening capacity for statistics is critical to close the gap and produce the necessary inputs for sustainable development. There is a saying that you may hear echoing within the walls of conferences and panel sessions: Leave no national statistical office behind. It means that as the leave no one behind agenda promises to ensure all individuals – no matter what race, ethnicity, age, sex, disability status – benefit from the gains of sustainable development; this must also be the case for all national statistical offices – no matter the country, capacity level, or income level. If the development community is serious about achieving the SDGs and understanding our progress towards them, we must also get serious about capacity development for statistics and the SDGs are one entry point to do this.

Q: Barbara's Swiss/Albanian story is wonderful but how could you apply the lessons with National Statistics? Could you have a programme that said we are going to improve statistics but we can't say which statistics we will be able to improve until one or one and a half years into the programme?

A (Barbara Baredes): The main question before designing a programme or intervention is “what do we need statistics for?” The answer that we all have in mind is “to inform decision-making” (for citizens, researchers, the media, private companies, the government, international organizations and the public at large). Evaluations of statistical capacity development interventions-such as World Bank’s 2017 Data for Development: An Evaluation of World Bank Support for Data and Statistical Capacity and UN’s 2016 Evaluation of the contribution of the United Nations development system to strengthening national capacities for statistical analysis- acknowledge that one of the main shortcomings of current programmes is the lack of engagement with users (especially from the government) to promote effective use of data for decision-making. We are currently focusing on producing statistics, but we are less concerned with understanding how users can access them and what purposes they will serve (e.g. research, monitoring policies).

Indeed, waiting for a year to find out what data is needed by national or municipal stakeholders and how they plan to use it seems too time consuming, and includes no concrete results. However, as we propose in the Capacity Development 4.0 Framework (and in line with the recommendations of such evaluations), relationships between stakeholders of the National (or Municipal) Statistical Systems are capabilities in themselves. Following this train of thought, the full year of consultation counts as part of the project because it contributes to developing statistical capacity –as much as providing technical support or training statisticians do.

Q: It is interesting to hear from all panelists a focus on leadership, political willingness, and political/contextual analysis as key indicators of success for capacity development for statistics. What advice do you have for those working on statistics of populations groups/topics of high political sensitivity? For example, we work on statistics of refugees and IDPs as part of an Expert Group which is why I am asking this question.

A (Barbara Baredes): Indeed, there are topics that are more politically sensitive than others. Migration is one of them, since there has been an upsurge of anti-immigration (and refugee) groups that threaten the stability of governments. Even more, this has been the main focus of several (successful) political campaigns in the past few years. In this sense, it is important to understand the particularities of the context (and the subject) in which we are intervening before proposing any intervention. First of all, analysing how various actors in the country would use the data on refugees or IDPs, in order to outline a strategy to outweigh identified negative effects. Second, leveraging on the work of other actors, such as that of the International Organization on Migration–together with UNSD- for the 2010 Census Round. These organizations recommended to identify individuals’ country of birth, citizenship and year of arrival. As a result, 87% of 149 countries that sent data to UNSD collected at least one question on the matter.Finally, working together with the national government to build consensus on what needs to be done, and focusing on small wins to build up from there.