Are We Losing The Battle For Data Literacy?March 9, 2020
Data literacy is a crucial skill to help people to navigate a world ruled by data, and riddled with misinformation. There’s no time to waste in sharing these skills far and wide.
As an open data practitioner I find that I’m frequently asking myself “aren’t we fighting a losing battle against misinformation?”. Can open data advocates really keep up in a world where the wealthy and powerful can buy enough skilled workers to distort reality, and spread false information to interfere in election campaigns?
In order to really unlock the benefits of (open) data we need to make sure people can actually use it. The big question is: can we train the population to understand how to analyse data, so they’re able to take more control over their future, and see through misinformation?
For this we need to do a detour from the data itself and rethink literacy. According to many dictionaries, literacy is the ability to read and write. However, in today’s information age, the ability to ‘simply’ read or write is not enough. We need to also think about numeracy, the ability to understand and work with numbers. From finances to science, numeracy plays a big role in innovation. Taken together, literacy and numeracy are, in my opinion, the foundation to data literacy.
However, the way we teach and learn numeracy is flawed in plenty of contexts globally. It’s usually based on the old mechanisms of US university admissions, designed so that only the elite can go to universities and acquire knowledge and social capital. In fact, in most universities, high school math can be covered in a couple of lessons (This fascinating Freakonomics article explains more).
We need to teach numeracy in a way that gives the opportunity for everyone, including marginalised communities, to understand the world around us. We need to teach useful numeracy, and one of the ways to do this is through data literacy.
One way of explaining data literacy is to use the definition created by Bhargava and D’Ignazio — “The ability to read, work with, analyse, and argue with data”. This translates to many skills, from statistics, to computing and visualising.
In today’s world everyone needs to have some level of data literacy, but just how feasible is it to train the world’s population in these skills? As the State of Open Data suggests, data literacy trainings have been carried out across the world, and yet they’ve only reached thousands of people. Are these trainees enough to be the agents of change?
The answer is no. We need to tackle data literacy in many ways, and to take data literacy into the global mainstream. I’m hopeful though, and I believe that we can achieve this at a few different levels.
We need to teach our next generation to use data, so why can’t we put open data and data rights at the core of the curriculum? In Uruguay these topics are part of the school curriculum — children learn about how to be safe online, and how to make freedom of information requests. By teaching these subjects, their parents are also able to learn new skills as they help their children with their homework.
In academia, bodies like The Centre for Continuing Education (CCE) at Birzeit University in Palestine have developed open courses (available on Github) to teach students basic and advanced data skills. Those courses can be taken and be adapted in other universities as well.
Teaching the future generations the skills to engage meaningfully with open data not only unlocks job future opportunities, but also allows them to better understand the world around them, and to become the next agents of change.
Governments produce many (open) datasets. The process of publishing open data — especially in government — should not be just “release and forget”, or designed to meet arbitrary targets (as Soila Kenya described in her recent piece for Open & Shut).
In a recent piece for Open & Shut, Soila Kenya discussed how governments need to start doing more than just publishing and forgetting. (Source)
The process should instead be valued as a learning process that allows organisations to understand what data they actually have, how their data is being structured, what errors are present in the data and, most importantly, to understand what questions the data is able to answer. The process can be even more successful if the publisher asks for feedback and changes their data collection practices as a result.
However, even in 2020, many government employees are still unfamiliar with core data skills. One of the ways they could familiarise themselves with these skills is using the ‘data pipeline’ methodology — a framework which guides people how to understand the ways that data is being produced and used.
The pipeline helps to build data strategies in organisations and communities, which can help them to control the data, or at least to make informed decisions about how and when data needs to be shared. This can be very powerful since it allows data producers and users to understand when they might lose control or power over the data, or how they could produce better datasets.
This is one version of the Data Pipeline, as explained by School of Data.
…in civil society
Many data literacy initiatives have been undertaken by civil society, but organisations still frequently lack either the resources and understanding of how to support data literacy.
In professional contexts, not everyone needs to be a data analyst, but I believe that data literacy should be part of organisational culture, and become a skill that is entwined with every role. This would allow people not only to understand what can be achieved with the data they have available to them, but also to recognise where it could potentially cause harm.
Data can be a great tool for advocacy, and learning how to combine it in those efforts can help to tell stories and help to inspire more supporters for noble causes.
Initiatives like the Citizen Evidence Lab at Amnesty International are undertaking pioneering work to create toolkits and guides to teach civil society organisations how to crowdsource and use data to campaign for human rights. By harnessing thousands of volunteers to support their digital efforts, they are also able to distribute the knowledge and abilities of the project.
…In the future
So, are we fighting a losing battle? Probably not. In the past, most of the world’s population did not know how to read, write or do simple calculus — those skills were still for the few.
The twentieth century brought us public schools and education for all. Our technology is developing at an unprecedented pace with machine learning, and producing vast volumes of data that are expanding exponentially, and so our education system has a lot of catching up to do.
But we are lucky that there are so many open-spirited organisations out there sharing modules and programs to increase data literacy globally, and to equip people with the skills they need to navigate data in this changing world.
I’ve gathered some of my favourite resources below — all of them are free to use. If you’ve used any of these, please write back about your experience, and feel free to remix any of these resources, adapt them to your context, and share them back to the community.
- DataBasic.io is a website with easy-to-use web-based tools that help to demonstrate the basic concepts of working with data;
- The Data Playbook, by the International Federation of the Red Cross and Red Crescent, contains examples, best practices, how-to guides, session plans, training materials, matrices, scenarios, and resources about data literacy in organisations;
- School of Data, a network to promote data literacy, offers free online modules on working with data;
- The Responsible Data Resource list is a great tool to learn about ethical and socially responsible data practices;
- Data Feminism is a book which introduces the theme of gender and data but also discusses the question of data and power.
- Data Maturity Tool is a tool that allows you to self assess how what is the level of data culture your organisation is in and what do you need to do in order to get better.