The pandemic has stimulated remarkable public interest in data. For example, John Burn-Murdoch’s COVID-19 graphs for the Financial Times have been widely replicated and are instantly recognizable as a standard for presenting viral case counts. There has also been a surge in volunteer-led initiatives, from coalitions that aggregate data such as the COVID-19 Canada Open Data Working Group to data-driven activism such as Keep Your Rent’s eviction tracker.

This has coincided with a growing awareness that our biases are reflected in the data we choose to collect and report. In Canada, this couldn’t be more acutely represented by our failure to consistently measure race-based inequities across a range of indicators of health and wellbeing. We know that this is a blind spot because there is a growing movement advocating for it and because other countries are well ahead of us. Perhaps there is no better example than the United States where there has been a stronger tradition of including race and racism in civil discourse. From the outset of the pandemic, US states have collected race-based data as standard practice for case count tracking and this has continued with their vaccination tracking.

While we wait to see if Canadian governments and institutions will respond with urgency on this issue, the non-profit sector can independently take steps to remediate its own data gaps. This short guide summarizes some key considerations for organizations looking to integrate race-based data into their work.

Plan before you collect

It may seem logical to open with a discussion about how to collect race-based data, but before looking at the mechanics, it is imperative that organizations determine how they intend to use the data first. This is not just important to ensure that you collect the right data, but that you also have a plan to put it to use. In other words, your data collection practices shouldn’t resemble that impulse purchase that you thought might come in handy some time in the future but has since been gathering dust. Non-profits should continually ask themselves: why do we need this data? and how will we use it?

Opportunities for using race-based data

So, what are some ways that you can use race-based data in your organization? In the media, we increasingly see it employed to dissect systemic inequities in society. Non-profits that work at the system level may have an interest in emulating these types of stories, though most locally focussed non-profits will find that they do not collect data at a scale that permits such broad conclusions. Furthermore, this type of storytelling tends to focus on the deficits that racialized groups face which obscures a parallel story of resilience and the assets that people bring to their communities. Non-profits should carefully consider which narrative they wish to perpetuate.

Most non-profits will find race-based data to be especially useful for internal needs. For non-profits that deliver community services, race is an important lens for evaluating if a program is reaching its intended target community. For example, program-level race-based data can be compared with local census data to measure the disproportionality between groups accessing a service and those living in the community. Similarly, comparing the racial mix of your staff and board with the people that you serve can raise important questions about representation. While data can tell you how far away you are from parity or help you measure pay equity, the objective here may not be to achieve direct alignment. Ultimately there are many factors to consider when evaluating staff and board representativeness, including the symbolism of certain ranks and roles within an organization as well as your equity and inclusion priorities.

Data collection practices

Turning to the practical details of collecting race-based data, this is a good opportunity for organizations to review whether they have appropriate policies and procedures in place around consent and the collection and storage of personal information. The Ontario Anti-Racism Directorate provides a comprehensive guide to collecting, managing, and using race-based data. We recommend that organizations first review their workflows to determine how and when to collect. This could include direction for staff about the language to use when asking about race, how to explain the purpose for collecting this information, and how to use a self-reporting methodology.

In most cases, we recommend that non-profits use questions that have been previously vetted. Unfortunately, there aren't many Canadian resources available and language and practices are continuing to evolve. Rather than wait for a standard to materialize, organizations would be better served to start collecting now and be prepared to adjust later as needed. Health-funded non-profits will be interested in a recently proposed standard by the Canadian Institute for Health Information that has been tested in primary care settings by the Upstream Lab and which is similar to a standard outlined by the Ontario Anti-Racism Directorate. Other non-profits may be interested in the draft data collection template put forward by Colour of Poverty-Colour of Change, one of the early advocates for collecting disaggregated data in the sector. This is by no means an exhaustive list. For organizations that determine a need for additional detail, we strongly recommend that they develop their own questions with interoperability in mind and with the assistance of a survey designer with knowledge about their communities of interest.

Other considerations

The guide above is far from comprehensive and some will recognize it as fairly conventional in its approach. Race is a sensitive and nuanced topic and it would be naive for organizations to approach this work without recognizing that issues of power, privilege, and colonialism also manifest themselves in the way we collect data. While this can seem daunting, it offers non-profits an opportunity for critical inquiry and reflection. As with other parts of the data journey, improving how we use race-based data is best seen as a work-in-progress both at an organizational level and at the sector level. We look forward to continuing to learn more ourselves and sharing again in this space.