Africa Regional Report
The 3rd Edition of the World Wide Web Foundation’s Open Data Barometer (ODB), released in April 2016 and covering 92 countries, ranks nations on three open data criteria:
- Readiness: How prepared are governments for open data initiatives ? What policies are in place?
- Implementation: Are governments putting their commitments into practice ?
- Impact: Is open government data being used in ways that bring practical benefit ?
In this regional report we dig deeper into the Barometer’s results to take a closer look at the performance of the 21 Sub-Saharan African countries featured in the latest edition. The purpose of this regional analysis is to use the rich data to assess the state of play of open data across Africa, evaluating the readiness of African governments to implement open data practice and realise its potential to impact positively on the lives of citizens.
The results of the 2015 ODB show that Africa is lagging behind other regions in the use and impact of open data. No Sub-Saharan African country ranks in the top 40, while there are six counties in the bottom ten. Globally, there is still much work to be done: only 10% of the 1,380 datasets analysed were open. Alarmingly, only two of these datasets were in Africa.
Note on the selection of countries in this report:. This report focuses exclusively on the Sub-Saharan countries in the 3rd Edition of the Open Data Barometer: Benin, Botswana, Burkina Faso, Cameroon, Ethiopia, Ghana, Kenya, Malawi, Mali, Mauritius, Mozambique, Namibia, Nigeria, Rwanda, Senegal, Sierra Leone, South Africa, Tanzania, Uganda, Zambia and Zimbabwe. Any reference to ‘Africa’ in this report will mean the countries of Sub-Saharan Africa listed here. The focus on these countries is for analytical purposes only. The three African countries not included in this report — Egypt, Morocco and Tunisia — are grouped for the purposes of analysis as part of the Middle East and North Africa (MENA) region, as presented on the Data Explorer on the Open Data Barometer website.”
For countries with more limited resources and arguably more pressing demands on those resources, outranking high-income countries in an open data grading may seem a remote possibility. And this may, in turn, bring into question the value of the Barometer ranking to developing countries in Africa and elsewhere. This would, however, miss the true value of the 3rd Edition of the Barometer. For the first time, with three years of data, it is possible to detect trends at the country level. While rankings may be an effective policy lever, the trends revealed by the Barometer data allow each country to reflect on their own performance over time and within the specificities of their context.
But it’s not just about country-level performance. Data has become a development cause embraced by everyone from grassroots activists to the UN Secretary-General. There’s now a clear understanding that we need robust data to drive global goals promoting democracy and development.
Last year, the world agreed the Sustainable Development Goals (SDGs) — 17 global
commitments that set an ambitious agenda to end poverty, fight inequality and tackle climate change
by 2030. Recognising that good data is essential to the success of the SDGs, the Global Partnership for
Sustainable Development Data and the International Open Data Charter were launched as the SDGs
were unveiled. Meanwhile, Africa adopted the African Data Consensus — a roadmap to improving data
standards and availability across the continent. But while much has been made of the need for bigger and better data to power the SDGs, this year’s Barometer follows the lead set by the International Open Data Charter by focusing on how much of this data will be openly available to the public.
“Inequality and poverty are about more than income – they are also about information. Seven years after I first demanded that governments open up their data to all, open data initiatives are now in place in more than half of countries we track. Yet their quality is variable, and benefits are concentrated in rich countries. Now is the time to resource and implement open data throughout the world, through projects such as the International Open Data Charter.”
— Sir Tim Berners-Lee, commenting on the findings of the 3rd Edition of the Open Data Barometer
2 Key findings
In this report, the focus is on the emerging trends apparent in the Barometer data in Africa. While some of the observations may be generalised to the regional level, we encourage each African country to reflect on its own performance in unlocking the potential of open data for the development of the continent. Looking at the trends, we have identified five key findings.
- 1Performance across the continent is relatively poor globally, and in comparison to leading countries in the Global South
|47||South Africa||Upper middle||26.77|
Table 1a: Country rankings for the Open Data Barometer 3rd Edition (Sub-Saharan Africa).
Table 1b: Rankings for upper-middle-income countries including the top-ranked upper-middle-income country in the Global South (in bold)
Table 1c: Rankings for lower-middle-income countries including the top-ranked lower-middle-income country in the Global South (in bold)
Table 1d: Rankings for low-income countries including the top-ranked low-income country in the Global South (in bold)
Table 1a shows that the highest ranked Sub-Saharan African country in the 3rd Edition of the Barometer is Kenya, ranked 42nd out of 92 countries. However, Kenya does not publish a single, fully open dataset — health, education and legislation data are open licensed but fall short of being fully open because the data are not available in bulk. At number 46, Rwanda is the only other African country in the top half of the global rankings table. The country is on the rise (see Figure 2) and challenging Kenya for the top spot in the regional rankings.
Rwanda is also the top-ranked low-income country globally, with a score of over double second-placed Nepal (Table 1d). When compared to their peers on the basis of income level, other African countries do not do as well as their Latin American and Asian peers. The top-ranked lower-middle income country outside of Africa is the Philippines. Of the African lower-middle income countries, only Kenya comes close to the Philippines (Table 1c). Mexico is the top-ranked non-African upper-middle income country in the Global South. Mexico scores more than double the highest placed upper-middle income country in Africa, South Africa (Table 1b). This shows that even with limited resources, many countries across the continent could be doing more to adopt and implement open data.
- 2A downward trend is all too common in the overall Barometer scores from 2013 to 2015
Figure 1: Change in overall scores from the 1st to the 3rd edition of the Open Data Barometer for African countries
Figure 1 shows that in nine (50%) of the African countries for which data were available, there was an improvement in their overall Barometer score from the first to the current edition. However, in seven of these countries, the improvement was from a very low base and, with the exception of Nigeria and South Africa, the improvement was marginal. The only two datasets in the 10% of datasets that are open globally come from Nigeria. This accounts for Nigeria’s rise in the rankings. It also suggests that little effort is being made in other African countries to publish data that are fully open.
The other half of the countries saw a decline in their overall Barometer scores from the first to the current edition of the Barometer. In terms of the downward trend in the Barometer scores of these nine countries, the decreases are, generally speaking, pronounced. An analysis of the changes in the median score per region show that in Africa and in other comparative regions of the world, the trend is a general decline in the median scores, particularly in South Asia (Table 2a). In fact, Africa compares favourablye to the other global regions, albeit that there is a decline in the median scores in all four regions.Africa’s performance is similar to other global regions even when one excludes high-income countries to make the comparison more equitable (Table 2b). This confirms the finding in the Global Report that implementation and impact of open data commitments is stalling, and open data availability and capacity remains heavily concentrated in the rich world, where mean scores are increasing. The inclusion of two new indicators could account for the disappointing mean scores in the 3rd Edition of the Barometer and further analysis would have to be done in this regard.
|East Asia & Pacific||36.29||34.38||5%|
|Latin America & Caribbean||31.21||28.73||8%|
Table 2a: % change in the median of the overall Barometer scores (1st to 3rd edition): ALL COUNTRIES
|East Asia & Pacific||21.29||21.16||3%|
|Latin America & Caribbean||28.96||28.73||1%|
Table 2b: % change in the median of the overall Barometer scores (1st to 3rd edition): EXCLUDING HIGH-INCOME COUNTRIES
- 3 Open data initiatives lack long-term commitment and resources, resulting in short-term gains that are unsustainable
The data for those countries for which there was an improvement in their overall Barometer scores from 2013 to 2015, conceal a worrying trend: in all but two countries (Cameroon and Nigeria) the 2015 scores were lower than the 2014 scores. Figure 2 shows that Nigeria is the only country in Africa where the overall ODB score has increased consistently. Only Cameroon has seen a net gain between 2013 and 2015, and an improvement from 2014 to 2015. While there is a net decline in the scores for Kenya and Mauritius between 2013 and 2015, there is a recovery in their overall scores between 2014 and 2015.
Figure 2: Overall Open Data Barometer scores for selected African countries, 2013-2015.
Spikes in the overall Barometer scores as reflected in Figure 2 point to a worrying trend: improvements in the readiness and, implementation of open data in many African countries are prone to be one-off occurrences. There are several possible reasons for this:
- because of the project-focused, short-term benefits of development aid parachuted into a country;
- because the external support provided to African governments is not aligned with the needs expressed by African governments;
- because African governments are not committed to open data for the long-term; and/or
- as Web Foundation research in South Africa and Kenya has shown, in cases where commitment is apparent at the highest levels of government, such commitments are not penetrating the working machinery of governments in Africa.
Greater levels of cooperation and coordination between governments, funders, civil society advocacy groups, and open data users and intermediaries are required to initiate sustainable, long-term open data initiatives that will embed open data practice in government and reveal to all stakeholders to value of open data.
The relatively lower levels of open data implementation shown in Figure 3 appear to confirm the absence of embedded open data practice in African governments.
- 4ODB implementation scores are lower than readiness scores, and the gap is not closing.
Figure 3: Open data readiness and implementation scores for selected African countries, 2013-2015.
Figure 3 shows the Barometer implementation and readiness scores for the top five African countries in the 2015 rankings. Ghana is included because of the significant drop in its implementation score in 2015 which, in turn, led it to tumbling 24 places in the global rankings.
Three patterns are evident in Figure 3. The first is the consistent gap between readiness and implementation in the case of Kenya, South Africa and Mauritius. For these three countries, readiness and implementation are clearly correlated. The second pattern is a closing gap between readiness and implementation evident in Rwanda and Nigeria. The third pattern is a sharp drop in open data implementation evident in Ghana which shows a narrowing gap due to improved implementation in 2014 followed by a disastrous slump in its implementation score in 2015 (see Box 1).
A deeper analysis of the Barometer scores for Rwanda and Nigeria (see Box 2), indicate deteriorations in their readiness scores due to shifts in the regulative and policy environments. The increase in implementation scores is attributable to successful implementations of projects that boost the scores for these two countries from a low base. Perhaps because these initiatives are project-based rather than programmatic, they are less sensitive to the changing policy environment. Time will tell whether project-based initiatives can make a sustained impact resulting in improved implementation scores in future editions of the Barometer. What is however critical, is to address the issues impacting on open data readiness in African countries. As stated in the Open Data Barometer Global Report, open data advocacy groups need to be more mindful of political and institutional conditions, and should coordinate their efforts more effectively to create enabling environments.
The Case of Ghana:. During the research period, Ghana’s open data catalogue was down or inaccessible for prolonged periods of time, and most of the datasets could not be located elsewhere on government websites. Not being able to access the government datasets has resulted in the low implementation score in 2015 because datasets cannot be evaluated for their openness if they cannot be accessed. It could be that Ghana was preparing for a full revamp of the data portal, in which case the country may resume its position in the top of the rankings of African countries. In fact, at the time of writing, the portal was back online. However, what is troubling is that during the assessment period, no information was provided as to why the data was unavailable and when it was likely to become available again. This severely dents users’ trust in government as a reliable data provider and limits the potential impact of government open data.
The Case of Rwanda:. A drop in Rwanda’s open data readiness score is the result of problems encountered in the implementation of its relatively new Freedom of Information Law as well as poor data management policies (a new indicator in the 3rd Edition of the Barometer). The decline was marginally offset by improvements in other areas such as the country’s open data initiative, engagement with CSOs and government support for innovation. The improvement in its implementation score is the result of particular advances made in relation to map, crime and election data (particularly from the first to the 2nd edition of the Barometer).
The Case of Nigeria: . The fall in Nigeria’s open data readiness score is mostly explained by the poor performance on new policy and data management indicators. The improvement in implementation is attributable to the two fully open datasets on education and health published as the result of a particular project — the government-funded Nigeria MDG Information System (NMIS).
Figure 3 does not include scores for the third component of the overall Barometer score: open data impact. The reason for this is that only seven (33%) African countries returned a score for open data impact (South Africa, Kenya, Rwanda, Malawi, Cameroon, Ghana and Nigeria), and only three (14%) countries scored more than ten out of a possible 100 (South Africa, Kenya and Rwanda).
- 5There is no stand-out performer in Africa.
Africa is the only region without a clear open data champion. In Latin America, Mexico, Brazil and Uruguay all rank in the top 20 globally. In the East Asia and Pacific region, there are seven countries in the top 40 globally, including the Philippines, a country that rose 17 places since the last edition of the Barometer. In previous editions, Ghana and Kenya looked likely to assume this role, but the data show that the performance of these countries is erratic. Pretenders to the vacant throne are mostly in West Africa — Burkina Faso, Senegal and Nigeria — with Rwanda carrying the flag for East Africa.
Since the completion of the Barometer research phase, there have been encouraging developments on the continent. Rwanda and Tanzania have published draft open data policies for public comment, there is strong multi-stakeholder support for open data in Burkina Faso, Tanzania and Côote d’Ivoire, with the latter adopting open data as one of the key pillars of its National Open Government Partnership (OGP) Plan. Sierra Leone now has a functional Open Data Council, tasked with implementing its National Open Data Day. Membership of the OGP in Africa is growing, and institutional support for open data at a continental level has remained steady. Future editions of the Barometer will tell whether the performance of these African countries will continue to improve, and whether any of them will emerge as open data leaders in the region.
Using open data to battle Ebola
In the parts of West Africa affected by the Ebola epidemic, roads, village names and villages are missing on many online maps. OpenStreetMap (OSM), a free, crowdsourced mapping tool provided critical mapping information to Sierra Leone’s National Ebola Response Centre, to the United Nation’s Humanitarian Data Exchange and to the Ebola GeoNode to assist them in co-ordinating the response to the epidemic. This demonstrates how providing missing data in open formats can play an important role in fighting an epidemic as complex as Ebola.
Using budget data to increase child grants
Civil society organisations in South Africa analysed the national budget and used the data to campaign for increased support for the country’s poorest children. They found extra funding in the budget and lobbied for the age range to be increased. The number of children receiving support rose fivefold from 1.9m to 11.1m in 13 years. Civil society efforts have been aided by the fact that South Africa is one of the most transparent countries in the world when it comes to budgets.
Using open data to reduce corruption
Evidence from the Vihiga Municipal Council indicated substantial abuse of local funds in Kenya. This led to civil society successfully intervening by increasing the accountability and management of public funds through citizen monitoring. The Centre for Law and Research International, in cooperation with a local NGO, Community Engagement for Peace and Development, and with financial support from the Partnership for Transparency Fund, have had a demonstrable impact, curbing corruption by 20% and increasing the execution rate of projects by 15%.
Cheaper medicines for consumers
Code4SA took a little known dataset from the Department of Health website and turned it into an easy-to-use format which allowed patients to make sure that they aren’t being overcharged by their pharmacies. Patients are getting cheaper medicines because of this tool. It had also unexpected outcomes. It was expected that middle-to-upper class patients with better online access would benefit from the tool. As it turned out, doctors use the tool to save money for their patients. As one doctor reported: “It very quickly became a must-use website for me. I work in a mixed-income neighbourhood and being able to figure out what works for my patients’ budgets is extremely helpful – there’s is no point in prescribing medicine that the patient cannot afford to buy.”
Open data should not be the preserve of developed nations, nor should it be a luxury for developing countries. Donors, civil society and governments need to work together to close the data divide, and make sure government data is open data. Open data’s legacy must not be a ghost town of outdated portals, one-off hackathons, dusty bootcamps and abandoned apps. To see real impact, governments must embrace open data as a cultural shift and engender a culture of open by default for the long term. This will require more than donor-dependent open data initiatives or signing up to global and regional initiatives without providing resources and departmental-level incentive schemes to back up these high-level commitments.
The African Data Consensus is a step in the right direction. But not a single African country has adopted the International Open Data Charter yet. The Charter provides an opportunity for African governments to formalise their commitments and to participate in a global initiative that focuses exclusively on open data.
The 3rd Edition of the Open Data Barometer shows that there are some African countries supporting their commitments at the level of implementation. Rwanda and Nigeria stand out in this regard, and it is possible that Ghana, South Africa and Mauritius will reverse the declines in their 2015 scores to resume an upward trend in open data provision and use. There is plenty of room for newer entrants to the open government data space such as Burkina Faso, Senegal and Benin to boost their scores. Future editions of the Open Data Barometer will be critical to confirm progress and trends in the provision and use of open data on the African continent.
“We need empirically-based, data-informed decision-making in Africa in order to improve governance and productivity.”
— Kofi Annan, speaking at the African Higher Education Summit 2015
Below are four recommendations to increase the readiness, implementation and, ultimately, the impact of open data for development. These recommendations mirror those in the Open Data Barometer Global Report and are supported by the analysis and conclusions shared in this report.
- 1Get behind the International Open Data Charter.
Not a single African country has signed the Charter, and there is no clear open data champion at the national level in Africa. While the Africa Data Consensus is acknowledged as a pioneering commitment at the continental level, the Charter will lock in political momentum around data needs at a global level by establishing clear and solid guidance on policy, and helping governments to develop long-term plans that set out timelines, resources and responsibilities for implementation. African countries signing the Charter will also be signalling their intent to take the lead in open data implementation in Africa.
- 2Expand and deepen open data practice.
Our analysis of African countries has shown spikes in performance rather than sustained gains. To see real impact, governments need to embrace open data as a long-term cultural shift in governance with sufficient resources and staff, ensuring that infrastructure, laws and policies are strong enough for long-term open data implementation and results.
- 3Finish the job – make sure the government data published is truly open.Only two of the 10% of open datasets in the world come from Africa. This doesn’t mean that African governments aren’t publishing data at all, but they need to follow through by licensing, releasing in bulk and making data machine-readable for the data to be fully open. Only by doing so will the full value of the data be realised.
- 4Harmonise open data, privacy and freedom of information efforts. In several countries, there has been a decline in the readiness scores of African countries; in part due to changing regulatory and policy environments. Open data advocates should work more closely with transparency, privacy and right to information activists to achieve better mutual understanding and coordination of efforts.