Tools and Techniques to Evidence with Data
Introduction: Effective use of data for health justice involves being creative in how you approach data as a tool and language. While there is no absolute best practice, it is a practice that we each can develop and learn from each other.
Why this lesson is important: With the analogy of data being a language, people have different purposes for using data for their advocacy. Data should be a tool to accentuate the goals, context, and discourse that you already hope to achieve. The tips outlined in this lesson are not the only approaches but reflect our knowledge and experiences in practice as well as those within our community. You can then determine which will complement the work you are currently trying to achieve.
Story: The first step is to clarify whether you're trying to create understanding or inform decisions. Trying to do both at once risks overextending yourself and doing neither effectively.
If awareness about how your community’s health is affected is lacking, your first step might be advocacy to paint a picture of community health dysregulation. Informing decisions usually involves larger authorities where you can start by using existing data to generate new questions.
It’s also essential to understand the language of your intended audience, especially if you're aiming to influence decision-making. When analyzing data, consider causation (e.g. “the smoke causes me to become sick”), correlation (e.g. “I get more light headed when construction activities increase near my house”), and hotspots (e.g. children walking a red route to school also live in multigenerational homes with preexisting illnesses). Those factors may not be directly related but together increase susceptibility.
Set principles and ethics around data use. Labeling something or making it countable can expose it to limits—saying “some” pollution is allowed is different from saying pollution isn’t acceptable. The difference between clean air, cleaner air, and Air, is that by the time you get to Air, you're not trying to give a caveat of saying it should be air enough—you're saying it should actively have properties that heal.
Ethically, think about working backward: establish the need before collecting data. Consider verifying or “ground-truthing” your findings before sharing them publicly. Ethics also evolve as your experiences change. What you were comfortable sharing a year ago may not feel appropriate now—like names, addresses, or personal identifiers—and that’s okay.
When it comes to communicating and visualizing data, treat it as a language. Tailor it to your audience. A policymaker might want long-term trends, whereas a local councillor needs data that reflects daily life. We’ve used an exercise that identifies three types of audience:
Influencers: e.g. councillors, GPs, faith community leaders, educators
Decision-makers: e.g. local authority or regional staff
Mobilisers: e.g. local advocates, students, young people
Understand their roles, responsibilities, motivators, and spheres of influence. Consider how they process and share information. They may care about the overall topic, but in order to do their job, they may not actually have a metric that's tied to what you care about. For example, someone might prefer documents to videos if they need to pass materials to managers. Think about how your data can support their work and decision-making.
Use key questions to guide your strategy: What motivates them? What’s in their control? What story do you want to tell with your data? What other datasets complement yours?
Quantitative data is good for tracking frequency or duration, but be cautious with averages—they may obscure those on the margins. If you're advocating for people most affected, average figures won’t capture their experience. Susceptibility may be difficult to quantify, which is where qualitative data becomes powerful.
Qualitative data lets communities identify important factors—relationships with the environment, physical or emotional health, priorities, or mental load. These are insights best gathered in conversation or participatory settings.
We previously published a framework, Microenvironmental Data for Health Justice, which outlines three stages in data for health justice that also apply generally to science and research:
1. Context – Understand factors around the context in which data is being collected such as terminology, race, air pollution activity and experiences, and commercial activity that influence how the data would be framed.
2. Collection – Identify gaps or conversations that need more support to determine specific questions and methods, such as going to the community centre and asking people who enter and use the centre a specific question every week.
3. Communication – Don’t let analysis sit unused or with visibility only to researchers and practitioners in institutions. It might be the case that the people who picked it up and made the article have a different spin than the people who actually produced the research.
If you're producing data as a community, you may be worried that a politician finds your work who does not have the ability to communicate exactly what the phenomena is using the data that you've provided. You might say, ‘I'm good at setting the scene for research but I want cultural workers to really unearth what is the context in which we're collecting data so that we're collecting ethically.’
Building a repository of data that's accessible to communities can help equip these communities for climate resilience. Data strategies like the ones proposed can help communities understand local and proposed health infrastructures—materials, climate risks, construction changes. They can use data to evidence needs, engage demographics, and support campaigns with consensual pictures of the current environment.
Visualisation is a key tool. You shouldn't overcomplicate visualising information if it's not going to help you clarify or get the point across. Align your visuals with justice goals, not just to showcase technical achievement. In this process, don’t let quantitative data erase qualitative data and lived experience. Don’t be afraid to mix formats: use charts, lists, images, or videos to appeal to different audiences.
We use both our own research tools and public resources. One of our key resources is Right to Know, which helps users explore environmental hazards in their area by entering their postcode. It’s based on a snapshot in time, so remember it’s not real-time, but it’s a strong point of reference.
Another resource is the Environmental Data for Health Justice Working Board, which includes interviews, terminology, reports, and strategies. You can use it to build your vocabulary, understand key players, and position your work—whether you're an academic, journalist, or advocate.
Public tools like Find Open Data (data.gov.uk) and the Office of National Statistics (ONS) provide datasets on population, economy, and society. These are the kind of sources that you can use just to start painting the picture before you get to the specific data collection that you want to do. You might even use these sources to make your case from data that already exists when applying for funding to do the research you want to do.
Hopefully, these strategies and tools help build your confidence using data as a language for health justice and advocacy.
Learning Points:
Decide if you're trying to create understanding or inform a decision in your more immediate plans. You can then determine the role data can play in achieving that goal.
The Macro Environmental Data for Health Justice report gives an additional framework on where to focus your efforts between: Setting the context for data, supporting data collection, or communicating the data to the right audience. You can partner with others to perform different parts of this framework.
Use already available data to generate new questions.
Understand the "language" of the intended audience. Communicating and visualising data is dependent on who you are speaking to.
Ethics are not static but Set some principles and ethics about what should be used as data
Never let quantitative data erase qualitative data or lived experience. Don’t be afraid to experiment mixing types of data visualisations under the same campaign.
The Office for National Statistics (ONS) can be a great place to start in secondary data.