How to start your responsible data journey

07 December 2021

If you’re working with social justice, international development or human rights, there is a high chance that you’re working with data – maybe even highly sensitive data. As practitioners, we have a responsibility to mitigate against potential harm when working with data, and that’s when the concept of responsible data comes into play.

What is responsible data?

Responsible data (RD) is a concept that refers to our collective duty to prioritise and respond to the ethical, legal, social and privacy-related challenges that come from using data in advocacy and social change work. Below, we share a few considerations for those who want to practice responsible data in their projects.

What to keep in mind when working with data

Responsible data should be an overarching theme to be considered throughout (and after!) a project, implemented by all involved, from leadership to staff. Managing data responsibly requires remembering a few key principles – which we touch upon in this post – and adopting processes in line with them.

Your priority should be the rights and dignity of the people whose identities or lives are reflected in the data you hold. Especially if you are working with sensitive data, it’s better to err on the side of prudence and keep in mind the humanitarian principle of doing no harm. Even if you’re not collecting names or seemingly ‘personally identifiable information,’ there are still ways in which your data can reveal the identities and behaviour of individuals, as well as of groups and communities.

Always reflect on the unintended consequences of the data you’re using. Though it’s hard to say how data collected for a project may be shared further or used in the future, we have a responsibility to assess how the data we work with could generate or exacerbate risks for the people reflected in the data. If it isn’t possible to truly assess the potential risks and harms and/or design a plan to mitigate these harms, it’s time to reconsider the project.

Practise data minimisation by only collecting and holding onto the data that is really needed. What is the minimum amount of data necessary to reach your project’s goals? Simply put, just because you can collect and store data, it doesn’t mean you should. Identify what data you really need - and no more- and plan for how you intend to use it.

Make sure you have been given informed consent. Have the people who are reflected in the  data given permission for you to collect and use that data for your work?

Keep your assumptions in check. All data contains a certain amount of bias due to factors like how the data was collected or the way in which the data is structured. To help you spot problems in your data, you can adopt strategies such as asking what perspectives are missing from your data, verifying for sampling bias, looking out for outliers, comparing your data with similar collection efforts from other actors, and getting expert opinions from those with deeper topical or cultural expertise.

Responsible data at different stages of a project

Designing a project with responsible data in mind: Considering responsible data practices from the very beginning can help set the stage for effective adoption of responsible practices through the entire project.

Make sure to conduct a risk assessment that considers:

  • What types of data are you collecting?
  • What potential vulnerabilities and threats can you foresee? 
  • What are possible mitigation strategies?

Also think about the data lifecycle of your project, which can include questions such as: 

  • Where will you store the data? And for how long? 
  • Who are the people responsible for archiving or deleting data at the end of the project? 

In your project design, make sure that the people who are reflected in the data have the most say over how the data is used. To ensure that more power lies with those represented in the data, you need to consider aspects such as inclusion (asking yourself questions like ‘who does the data leave out? And how this might impact them?’), accuracy (is this data as accurately reflective as possible of the reality of the people involved?), and ownership (who owns the actual data and what does that mean for the people involved?).

Collecting data responsibly: After you’ve determined what data you need, you should design a data collection plan to map out what you plan to collect, how you will collect it, and how each type of data relates to your project. You should think about who will provide the data, how to get consent for this specific use, and what tools will be needed to perform data collection. These questions will help you define a detailed data collection plan, brainstorm potential risks and appropriate responses, and refine your understanding of what the minimum amount of data is that you need to collect.

Managing data responsibly: Throughout the project duration, it is important to reflect on how you’re storing the data you’ve collected. Things to include are where the data will be stored, who will manage it, how often you’ll conduct backups, and when you will delete the data.Thinking about access and control is also key: your team needs to carefully assign user access and permissions to minimise the risks identified during the design of the project. There should be regular audits to check who has access to what and remove people who no longer need access (e.g. temporary workers such as contractors and interns).

Sharing data responsibly: Always make sure to weigh the benefits and risks of sharing and publishing data. While sharing data can bring benefits – for example, providing stronger evidence for advocacy, increasing the efficiency of service delivery, or better informing decision-making – it can also lead to unintended consequences. The wider you share, the less control you have. Before sharing your data, consider if and how the data is (or may be in the future) linked to any individuals or groups, and how it might impact an individual or community. 

Closing a project: When closing a project, think through the types of data you have and where it is stored, and identify what you want and need to keep (and why) – this will help you plan for responsibly retaining, archiving or disposing of that data.

Useful resources

Practising responsible data isn’t a simple task, but each step counts. If you want to learn more about how your organisation can incorporate responsible data principles, we encourage you to explore the following resources:

We’d also love to hear about your experience with responsible data work. You’re welcome to reach out to us hello [at] theengineroom.org.

 

You've enjoyed this blog post? Sign up to the IM Resource Portal's newsletter to be alerted when the next one comes out!

 

This publication was funded by the French Development Agency (AFD). Nevertheless, the ideas and opinions presented on this post do not necessarily represent those of the AFD.