At the Tableau Foundation, we are on a mission to use facts and analytical reasoning to solve world problems. We provide software, financial support and training to help organizations use data to improve or accelerate outcomes. We build the capacity of others to have a positive impact in every corner of the planet.
Quality data are pivotal to delivering quality education—to getting good outcomes. Good data tell us what works and what doesn’t, which schools are good models and which ones need help, and more.
And delivery of quality education produces opportunity for people, and has even been shown to sow the seeds of peace.
Unfortunately, quality, actionable data are hard to come by in many countries. Addressing this problem is an area in which the Tableau Foundation has much to offer the world.
To get a handle on what kind of support and insight we could lend to this challenge, we joined the Education Data Solutions Roundtable, organized by the Global Partnership for Education (GPE), an organization that supports close to 70 developing countries to ensure that every child receives a quality basic education, prioritizing the poorest, most vulnerable and those living in countries affected by fragility and conflict.
The Tableau Foundation worked with other foundations and companies, developing country partners, multilateral organizations and nonprofits as part of the Roundtable. With facilitation by GPE, the country representatives identified their most pivotal challenges.
We then co-created with those partners an agenda for potential solutions—to help us all focus our efforts and coordinate action. As part of that, we identified some ways in which the Tableau Foundation could help.
In February this year, I was part of a group of roundtable partners who visited Ethiopia to learn what the country already does to collect and analyze education data, and what further help it needs. Two key observations stood out to me.
First, we talked with teachers and principals who were manually gathering data that they were required to provide to the government and their funders. In one case a principal told me he had to provide similar data to six different funders – all manually.
While manual data collection is challenging, the bigger problem is that the data didn’t come back to the teachers or principals in a useful form. They are doing all this work to collect data but are getting almost no value from it.
This problem is not unique to Ethiopia. It is far too common for governments, funders, and others to extract data from communities without the communities getting value from the data.
Drawing on a successful experience addressing the malaria epidemic in Zambia
The Tableau Foundation’s initial grants were in the health sector, including a 2014 partnership with an organization called PATH to help Zambia’s Ministry of Health eliminate malaria by 2021. Our support was in the form of software, financing and training through PATH to build the capacity needed in Zambia’s institutions.
Before we engaged, Zambian authorities were collecting data on malaria at the district level and reporting it to the national level. They organized a meeting every June to discuss data aggregated from all the various districts and make decisions based on those data.
It was a laborious process, often requiring that records be transcribed manually and analyzed through spreadsheets. Real-time analytics were virtually impossible to develop, much less make available to decision makers around the country.
Through our collaborative efforts with PATH and the Ministry, the data were disaggregated all the way down to the community and clinic level allowing for more micro-targeting of resources.
Data standards were set and workflows were automated, breaking the dam on analytical insights and putting reliable, timely information in the hands of decision makers. In fact, the ministry also moved to weekly data collection and analysis, and today holds data meetings every Thursday in some clinics instead of every June.
Now, information about malaria infections is spreading faster than the disease itself! The timely use of data to inform decisions has led the Southern Province of Zambia to report a greater than 90% reduction in malaria deaths and a more than 80% reduction in cases since 2014. Other provinces and countries have adopted this data-driven approach.
Through our work in other countries and sectors, we’re proving that the approach of tracking disaggregated data down to the individual level and empowering people throughout the decision-making chains to use data in a timely manner works across issues, geographies, and political, social and cultural contexts. That is why we believe that this approach could work for the education sector too.
Better data visualization for better analysis
Communities can get value from data collected about them when those data are given back in useful, insightful forms such as data visualizations. Providing visualizations allows people to see and understand their data so they can make data-informed decisions about how to improve their work.
Today, exam data is collected electronically as students fill out forms that are scanned. The exam results are used for analysis at the ministry level but are delivered back to the school in paper form, typically with rows and columns of data, but with no analysis on what they mean.
In Ethiopia, I saw that some schools manually analyzed the data, asking questions such as: are girls and boys performing differently? On which subjects? The answers to those questions are in the data.
Providing principals with simple visual analytics instead of rows and columns of raw data in spreadsheets could help officials make informed decisions on how to improve outcomes.
The desire for analysis
I was encouraged to meet people throughout the education system who were asking what works and doesn’t work for their students. They have the desire to use data and the questions about effective approaches, but they didn’t have the data or tools at hand to answer those questions. That curiosity is an important pre-requisite for finding success in data systems improvement projects.
Alpha Bah, the Head of the ICT unit at the Ministry of Education in The Gambia, who was also a member of the Roundtable, is one of those intellectually curious people. His vision is to get to a point where they can track every child in every school.
We believe his vision is achievable because we have already seen success across the global health, homelessness, and water and sanitation sectors.
Getting to actionable recommendations
After visiting Ethiopia and the Gambia, the Roundtable developed specific and actionable recommendations to address what our developing country partners told us they needed and wanted.
The recommendations have some common themes that can be adapted based on each country’s context and needs. They also support countries building on the data infrastructure and tools they already have, which is always better than creating the whole data system from scratch. Meeting countries where they are while iteratively improving over time is essential to achieving success.
For example, Ethiopia doesn’t yet have clear plan or protocols in place to address operational aspects of education management information systems, an important piece of the puzzle for any country that wants to improve its data system. The Gambia already does.
So, a specific recommendation could be for Ethiopia to develop such protocols, on the basis of best practices put together from other country examples--including from the Gambia.
Based on our work together with our Roundtable colleagues, and based on the diverse experiences of other Tableau Foundation partners, I am convinced now is the right time to make a meaningful change in education data systems and use.
GPE’s leadership is essential because, without it, various actors working to improve data systems in developing countries risk diluting their impact by acting separately.
With only ten years left to achieve the SDGs we think better coordination among actors working to improve data systems is essential for accelerating progress for all.