To label or not to label.

Welcome to Data Uncollected, a newsletter designed to enable nonprofits to listen, think, reflect, and talk about data we missed and are yet to collect. In this newsletter, we will talk about everything the raw data is capable of – from simple strategies of building equity into research+analytics processes to how we can make a better community through purpose-driven analysis.

This week, there is no other way to start than asking – how are?you? With everything happening in the world,?really, how are you? Overwhelmed? Dejected? Nauseated? Exhausted? Or perhaps switching between anger and fear? I could go on with the words, but you and I know that there aren't many (words) to describe everything the world has seen within the past two weeks alone…

I will speak for myself – lately, it has been tough to show up where I want and need to be. I have found myself circling "why bother" and "does it even matter" enough to confidently say that I haven't found a perfect healing path. But I know that we need to be there for each other– to support through the wounds and broken hearts. Our course of healing may not be straightforward, but it is collective. So, I am collecting my courage, hope, and belief to continue what you and I do best - building this space together.

Centering the desperately needed racial and social justice we seek, let's talk today about the role of labels in social identity data. More specifically, six fundamentals to remember when designing labels-based identity-related questions. These are guidelines or truths for the times when you are wondering whether to label or not. Labels around gender identity, sexual orientation, race/ethnicity, age, nationality, language, immigration status, and other aspects of people's identity.

Before we get into the guidelines, let's remind ourselves of a fundamental truth – we will never get all labels right. This is continuous learning. There may be times when you are afraid of picking the wrong labels and offending your environment. But, regardless of how uncomfortable that may be, you must lean on your community to learn and adapt.

Because these labels are not just choices/options on a data collection tool. These labels are the reason for decisions and actions of access and support for someone. These labels are also a way for someone to share their truth with you. For example, imagine someone who decides to share with their workplace about their ADHD through one of the org-wide surveys but sees no intentionality or authenticity in the survey. They decide otherwise – continuing in their lack of belongingness.

So, our fear of getting labels or not, with consequences right or wrong – must not hold us back. Because the data we collect from those labels (or without the labels) impact our outlook and actions towards justice.

Keeping that in mind, here are six guidelines for the next time you and I are wondering to label/not and what labels to pick. Note that these are "universal" guidelines which means this list applies to race/ethnicity just as much as nationality (or any other aspects of identity).

1. Labels must not lead to "otherness".

How we set up labels in our data collection tools matters. It can make someone feel they belong to a group or far from belongingness. Depending on the audience, reason, and medium for data collection, choose labels with a valid rationale that does not lead to a feeling of "otherness" – both during data collection and when the report/insights are built out of that data.

For example, I have been part of surveys in the past that were meant for a small internal audience (that means identity data is not all unknown beforehand). When those surveys included social-identity questions, it was apparent that my responses had a complete chance to be singled out. So, I felt compelled to "normalize" some of my responses to become part of the "average", while skipping obvious identity-revealing questions.

2. Labels (or not) must prioritize context.

Whether or not you need labels should be driven by the context of the data collection. The complexities and nuances of social-identity data can only be managed enough when we focus?

• Who is funding this data collection?
• Why is this data collection needed?
• What will happen with this collected data?
• Who and how will the analysis be performed on the collected data?
• Who and what sort of next steps/decisions can be made out of it??

All?of that constitutes context. And perhaps more that you and I will uncover together. But let that dictate whether or not you need labels.

3. Labels must center people, not the ease of analysis.

That means whether you ask?

• "Please self-describe your identity" with an open-ended text box or
• create five independent closed-ended questions?

both must center your community (i.e., the people going to respond to the survey) instead of your ease of analysis.

Yes, close-ended questions make analysis much more manageable. Bar charts and themes, check and check. But what that does not offer to your audience is the flexibility to describe as they find most appropriate. It also leaves a big responsibility on the shoulders of the designer to include all inclusive labels; otherwise, they risk "otherness" as described above.

4. Labels represent community, not mirror census.

Census is the most common place where most of us seek guidance on social-identity data. But that's all it should be – guidance. You and I need to remember that there may be situations when the local community demography is changing faster than updates on the census.?

That means we cannot afford to exclude populations from our data collection just because our guiding document still has time to update. My usual recommendation in most spaces around this issue is – talk to your community. Talk to local consultants. Or seek input on your social media page directly from the community. Or perhaps co-design a governance team with the community on such nuanced data pieces that you collect.

Regardless of your direction, labels represent the community, not mirror the census.

5. Labels must be grounded in the language of trust and humanity.

Language is important. Words we choose to speak and write matter. Because they (words) represent our choices and truths. Therefore, it is crucial to ensure that the words we pick for the labels lead with trust and humanity.

We are asking someone to share their most personal information through those labels. Not only does insensitive choice of words lead to loss of trust from your audience but also sparse data that you collect.

Pick words that speak to your intentions of honoring people when asking them to share about their identity.

6. Labels must be able to distinguish between medical-first vs. social-first approaches.

Many social-identity-based research questions we have today are built around the needs of health and medical access. I am not saying that identity questions based on a medical-first approach are wrong. However, you and I need to be careful how we frame those questions, especially when the decision-making from it is for social reasons, like funding and grants.

Evaluate how you want to approach labels in your questions, and from there, backtrack your next steps. If you need a more social-first approach, lean on your community and local experts to guide you in making the shift in the language.

?

*********************************

This is a continuous, evolving list. There is no permanent set of rules or one expert who can guide us through this alone. This is a collective work for you and me around our data collection.

Besides, if not now, when is the time to be more conscious, purpose-driven, and intentional about our data collection, especially the identity data. That identity data (and other critical data points) leads to decisions and actions affecting our belongingness and safety.?

The question…no, the challenge here isn't managing any harmful outcomes of that data we have. It is not doing anything about the data we are yet to collect. It's the responsibility of building awareness about why and what we choose in our data that empowers us a little to claim the justice we find missing.

?

*** So, what do I want from you today (my readers)?

Today, I want you to

• Share how you approach adding labels (or not) in your data collection? How does that impact your analysis and narrative building?

?

*** Here is the continuous prompt for us to keep alive the list of community-centric data principles .