Confessions of a DEI data junkie
I am going to start with what may seem like a brag and not a confession.
In pursuing our mission to “get you the information you need to do good,” Candid receives and moves vast quantities of digital information in the social sector on more than 1.8 million active US-based nonprofits. As part of that work and over the past 7 years, we’ve been actively and systematically collecting and sharing demographic data – information about race & ethnicity, gender identity, sexual orientation, and dis/ability status. There are now over 22,000 organizations actively sharing demographic information via Candid’s profiles.
The point of these facts isn’t to brag; rather, it’s to show that we have some hard-won experience to share. In fact, I am going to do the opposite of brag. In this blog, I’m going to give concrete examples of a few of our own mistakes and surprises in collecting and sharing demographic data.
Why confess these at all?
We want to share the mistakes we’ve made – so you don’t have to repeat them. We want to share them so you can learn from our experience in collecting this information systematically over the past seven years. We want to share, because we need your help. We ask that your organization advocate for a Gold Seal of Transparency and demographic data via Candid.
There are sensitive and important questions our sector struggles to understand: Who leads and works at nonprofits? Who is served by nonprofit organizations? Which organizations are supported by philanthropy (or not)?
By asking your grantees, members or peers to participate on Candid, you are working toward answers to these important questions. You are also doing so in a way that is respectful of the time and effort organizations need to go through to provide these answers. Instead of hundreds of custom, one-off surveys between foundations and grantees or associations and members– we can work together toward a collective understanding of diversity in the social sector. That’s the real promise of Candid and of “having information you need to do good.”
Confession: We knew asking about identity would be hard – but we underestimated how hard it would be.
Identity is not simple. But to keep the burden of data collection low and to analyze trends easily, the questions about identity and methods for data collection and reporting need to be. It’s a constant balancing act.
On the GuideStar nonprofit profile, organizations can report on four dimensions of diversity: race & ethnicity, gender identity, sexual orientation, and dis/ability status. Our “How to Collect and Share DEI Data” guide is a resource for the field and has been written with nonprofits who want to start collecting their organization’s data in mind.
The questions and choices of answers have been carefully considered with our expert partners – CHANGE Philanthropy, Equity in the Center, and RespectAbility. We also re-visit the questions based on feedback and by analyzing the data contributed by organizations.
An analysis of data shared by more than 22,000 organizations provides a clear example of how challenging questions about identity can be. We provide an open-response option that invites people to “Please specify” their race and ethnicity. And to date people have specified over 100 ways of describing their identity beyond the structured choices we’ve provided. Some with the intent to educate about their primary identity– by sharing “Middle Eastern” and “Jewish.” Some with the intent to protest the question (e.g. by answering “Human”).
So what can you learn?
- Adopt standards where possible. The Census questions are a kind of standard, as are Candid’s questions shared in the How to Collect and Share DEI Data guide.
- Consult with experts and evolve the standard. We are open to feedback on Candid’s Diversity, Equity, and Inclusion (DEI) questions and we receive comments through our support channel.
- Keep the balance. Respect for identity vs. the burden of collection, depth of data vs. utilization.
Confession: We collected data we couldn’t interpret.
When GuideStar (now Candid) launched the DEI data collection, the balancing act was even tougher, as many organizations did not see the need for these questions on the profile. In our attempts to make the data collection simple, we compromised on how the data was collected to the point where it became difficult to make sense of it.
For example, we did not validate that the number of staff or board members identified within a demographic category totaled to the overall board and staff count. In the course of the data collection, we received some feedback from people trying to use the data. Feedback like:
“Basically … we need further guidance to clean up the data we received. For example, in some cases, the number of staff who checked each category box do not match the total number of staff the organization has.”
We’ve corrected this, of course, in the newest iteration of the data collection launched in August 2019. We also made sure that each demographic question had explicit options to “Decline to state” and for “Unknown” responses (to account for non-responders to surveys of board and staff). These were significant changes that have improved the quality of the collected data tremendously.
So what can you learn?
- Collect data you can use – by actually trying to use it.
- Include best practice response options like “Decline to state” (for those who decline to identify themselves explicitly) and “Unknown” (to account for non-responders to surveys of board and staff).
- Learn from the responses you get.
Confession: We assumed nonprofits had the data to share.
In the early days of DEI data collection, we noticed that adoption among nonprofits was rather slow. There are many hesitations in collecting and sharing this data but one we identified early on was that some nonprofits did not have this data to share in the first place. Or perhaps they did have the data but in different categories and formats.
In response, we created the How to Collect and Share DEI Data guide to take the guess work out of collecting DEI data. The guide includes introductory text, questions and definitions.
So what can you learn?
- Create resources and offer grant support to help nonprofits collect the data.
- Build DEI into your application process.
- Don’t let perfect be the enemy of the good – acknowledge challenges with missing data and move forward with the analysis.
Confession: The variety of reactions and feedback surprised us.
If your organization advocates for the collection of DEI data – be prepared for a variety of feedback. We’ve received many comments over the years both positive and negative.
Some of the feedback will be negative:
“We hire employees based on experience, skills and heart, not to fulfill a social-political agenda.”
“I’m offended at the profile questions. I am not going to ask employees what their sexual orientation is nor am I going to be made to feel guilty because my staff is all white (because our community is white).”
Some might be heart-breaking, for example when the questions you ask might not do justice to someone’s identity:
“I just wanted to let you know there’s a few issues with terminology on the demographics piece for orgs to fill out. As someone [who is] non-binary, I’m not cis but I also don’t identify as trans.”
What can you learn?
- Respond to all and let them know how much you appreciate their feedback.
- Collect constructive input and consider for the next iteration of revisions.
We started recognizing organizations for sharing demographic data about their leader with a Gold Seal of Transparency in October 2020. Since then, we’ve seen tremendous growth in the number of organizations doing so via Candid’s profiles – from just over 6,000 in October 2020 to now over 22,000 organizations and counting. But there is more we can do together.
With more than 1.8 million active nonprofits and almost 200,000 with at least one full-time employee in the US alone – there is a lot of room to grow participation and we need your help. We ask that your organization advocate with your peers, members or grantees for a Gold Seal of Transparency and demographic data via Candid.
Organizations like the Jessie Ball duPont Fund, the Charles Stewart Mott Foundation, and the David and Lucile Packard Foundation have already done so. In their message to grantees, the Mott Foundation said why: “to promote a just, equitable and sustainable society compels us to strive to do the best we can to promote diversity, equity and inclusion in our grantmaking.”
With the unprecedented level of resources and attention on diversity, equity, and inclusion in our own sector – we are poised to make real and lasting progress – but only if we are willing to work together and hold ourselves accountable to the data.