Tina Kauh: I am Tina Kauh, a senior program officer in the Research-Evaluation-Learning Unit at the Robert Wood Johnson Foundation, and I am so excited to spend time with some colleagues who are doing some really incredible work around race and ethnicity data. Before I start, I want to say a few words about why this work is so important. As we all know, data collection, analysis, reporting, and dissemination beyond the broad racial and ethnic categories that have been the standard for the last 25 years are critical to advancing health equity. For years, we have consistently seen that the federal minimum standards for collecting and reporting race and ethnicity data place people under broad categories that really miss significant heterogeneity within those groups.
Part of acknowledging and valuing everyone’s unique experiences involves accurately representing them through data. For nearly ten years, the Robert Wood Johnson Foundation has supported research and advocacy initiatives to improve policies and practices, so race and ethnicity can be disaggregated in more meaningful ways.
We are excited that for the first time since 1997, the Office of Management and Budget (OMB) has expanded its minimum race and ethnicity standards to include historically marginalized communities, ensuring that they are visible in data. These changes mean that communities that have been invisible in the past will now be properly represented in the data collected.
Assuming that the changes are implemented well, better data will enable policy makers, researchers, and healthcare providers to pinpoint and address inequities more effectively, and this will help us direct resources and interventions where they are needed most, especially to communities facing the greatest health challenges.
This is a long time coming, and the importance of high-quality implementation is critical as this has broader implications.
High quality implementation means that there are resources at all levels of government and partnerships with community-based organizations that are needed to foster broad adoption. So, with that, I want to introduce my panelists today. I will ask each of you to introduce yourselves, briefly describe the work that you do, and the organization that you are with.
Meeta Anand: My name is Meeta Anand, and I am the senior director of Census and Data Equity at the Leadership Conference on Civil and Human Rights and the Leadership Conference Education Fund. My work in a nutshell is to advocate for policies that ensure inclusive data collection at federal, state and local levels, and to ensure that we have an infrastructure of engagement for communities toward such results.
Maya Berry: I am Maya Berry, the executive director of the Arab American Institute. We are a national civil rights advocacy organization that works on the civic engagement and political empowerment of the roughly 3.7 million Arab Americans. I say “roughly 3.7 million” because we have never had a category to accurately count our community. So, the work that we do regarding data equity is central to the effective representation of our community, and it dates back to our founding in 1985. Our first official partnership with the Census Bureau was in 1990. The first request we made to update the standards and do the work we believe is needed for accurate data collection was back in 1995 before the standards were last revised in 1997.
Ninez Ponce: I am Ninez Ponce, and I am the director of the UCLA Center for Health Policy Research. I am also professor and chair in health policy and management at the Fielding School of Public Health. The center houses the California Health Interview Survey, a project that I feel I grew up with in my career at UCLA. It is the largest state population health survey in the nation. I think I have some bragging rights with it, because we were able to work with multiple community voices, as well as foundations and funders and academics, to create the content and a multicultural technical advisory committee. With all these voices, we have been collecting more granular data since 2001 and continue to share that knowledge with other data collectors.
I am just so happy to be part of Robert Wood Johnson Foundation’s and Tina’s team pushing for data disaggregation and data equity, and now culminating in the formation of sharing this knowledge in a new data equity center that Robert Wood Johnson funded. So, for now, it’s more about the how. There are our new mandates with the OMB. But then “the how” is: How do I think we enter this space? As a data producer I think I have a pragmatic view of how to make this happen. But, as a data advocate, I also have the aspirational view of what it should look like.
Tina Kauh: Welcome to all three of you. We will start with a big picture question: How do you think the updated minimum race and ethnicity standards will benefit data that we have about individuals in our nation, including census data, health and public health data, and beyond?
Meeta Anand: Looking at census data, in 2020, there was a huge number of people that were reported as having “some other race.” That was really a result of people not seeing themselves on the form and a result of the design of the form and question format. What we hope this new question format and new categories will do, is reflect a decline in “some other race,” and start having a more accurate measure of how people truly see themselves. Why does this matter? Well, what happens is while the decennial census has a category of “some other race” that is not a category found in other surveys or in the rest of the federal government. So, that meant people in that category had to be reassigned to a different racial category for those data to be usable by other arms of the federal government. Thus, those data start being less accurate and less meaningful by not having categories that people can identify with and self-select. Also, allowing people to multi-select so that they can mix and match, to truly reflect who they are, that actually allows for a greater and more accurate understanding of who we are as a nation.
Maya Berry: The new race and ethnicity standards are quite remarkable for all of us that care about data equity. Adding a new minimum reporting category, combining race and ethnicity into one question, and requiring detailed data collection are three significant developments for Arab Americans and other Middle East and North Africa (MENA) populations. It is actually transformative in terms of our ability to gather information about our communities for the first time ever. Historically, we have relied on the American Community Survey (ACS). Prior to that, we relied on the ancestry question on the long form census, which was eliminated after the 2000 Census. So, getting to a place where one, the race and ethnicity standards are updated for the entire country, so that we are all getting more accurate data about our communities, and two, specifically, the addition of a category that will do better in terms of capturing our information will change everything. I would say though, Tina, it is what you led with, which is getting it right. We have the MENA checkbox, now we must implement it correctly.
Ninez Ponce: I think the key change is minimizing the other race category in public health. The census data are our denominator data. When we come up with estimates on serious psychological distress, on exposure to hate crimes, not having access to healthy food, hospitalization readmissions, or other indicators of healthcare access and the wider social safety net, we do not have those data and, we cannot compute a rate on a specific population without the denominator.
In California, the Department of Finance dictates what the major race and ethnicity categories are and how they are collected in the state, and they were from the old 1997 standards. Also, it was a “Hispanic trumps all” rule, so everybody gets rolled up to Hispanic/Latino including indigenous Californians who are Latino. For public health, it is about detecting need and aggregation hides needs. You know this, Tina, we have been working on this together for almost a decade. This “model minority” of the Asian American aggregate category hides the identities under that broad category. For example, Koreans in California had one of the highest uninsured rates, but it was not seen.
Mental health needs vary across different groups. The aggregation of Asian people with Native Hawaiian and Pacific Islander people totally masks the needs of the Native Hawaiian and Pacific Islander population. Their demographics, educational attainment, opportunities to access good healthcare and promote wellbeing, are very different. Aggregation creates this hiding and suppression of understanding the needs of this population. So, in some cases, granular data expose privilege and oppression.
Meeta Anand: I want to follow-up on a point Ninez made, which is often not said as much as it should be or understood. There was a forced binary choice in the two-question format, and people were forced into one of two tables or categories, and everything flowed from there. So, what we see with the new standards is actually more of an ability to mix and match and be flexible. The reason this matters is because you have these forced aggregations or false choices, where people are picking a race because they are an ethnicity, because they do not really know which one to pick, and they are doing as best as they can. You are actually not capturing the lived experiences, not only of the community that people are feeling not seen, but it also changes what is going on in the community they select into. So, it actually creates an inaccurate perspective for the aggregate group as well as for the group that is contained within it that feels they are not really expressing their own lived experiences.
Tina Kauh: Thank you for that. I think we started getting into this a little bit as we talked about aggregated data and the need to disaggregate. But how do we best assure that the revisions to the race and ethnicity data standards contribute to a more accurate and nuanced understanding of demographic trends and inequities in our society?
Maya Berry: I can jump in on the disaggregate point and the disparities that you find within communities. Think about creating a new category, Middle Eastern and North Africa (MENA), which is a broad geographical category that includes Arab Americans, who will be the single largest segment of the MENA populations here. But it includes others, obviously, within that. There are transnational communities like Armenian Americans and Chaldeans and Kurds. There are Iranians and Israelis, for example. MENA is a broad geographical category. If you produce data exclusively on “MENA populations,” I am not sure what that tells us, to be honest. This is where the new Standards’ requirement of detailed data is critical.
For example, in the case of Arab Americans, ACS data tells us we have higher education rates and higher homeownership rates.
*
But if I do not disaggregate that data and look deeper, I am not going to capture the issues and problems that might be encountered by more recent immigrants, specifically refugees from Iraq, Syria, or Somalia. So, the urgency of getting to disaggregate data specific to individual communities within MENA first, and then subpopulations within those communities is obviously there. A lot was said about our efforts to secure the MENA category. People were saying, “You know, we’ve never seen a community not want to ‘be white.’” And I kept trying to explain, “That’s missing the point.” It is missing a lot. Frankly, it was not about not wanting to be white, it was about identifying as an ethnicity that can be of any race. There are white people from MENA. There are Black people from MENA. There are Arab Latinos. It is important to understand that you simply want to capture accurate data, which means that you better ask the question correctly so people can self identify. I think the critical piece is to get the question right, to get the question format right, to get the categories right. I know we are going to spend some time talking about the subpopulations and check boxes and check-ins, but also to make sure that the disaggregate data requirements are real and truly become part of how these agencies, the state, and folks on the local level operate. I mean, we have a researcher with us, and she immediately mentioned that we cannot get an accurate denominator. If we are not collecting the data accurately and making sure that it is presented and tabulated in ways that respect the disaggregation that is required, we will not be able to do what we need for our communities.
Tina Kauh: When we talk about these new standards, we talk a lot about the census, which happens every ten years. But there is another survey that happens annually. Meeta, will you talk a little bit about the importance of these changes for the American Community Survey?
Meeta Anand: Not only every year, but every month a subsample of us in the nation receive the American Community Survey, which is conducted by the Census Bureau, and it gives us more detailed data on all populations. For context, what we commonly think of as the decennial census is ten questions in ten minutes or less.
This survey asks a lot more than that, including the cost of your house, how many bathrooms you have, and your level of education. Why does this matter? Well, when you are reading any article, or you are doing any research in healthcare data or social determinants of health or understanding where educational gaps are, these are the underlying data we are relying upon. So, the American Community Survey is actually going to see the changes to the race and ethnicity question first, before we see it on the decennial census.
And the other piece that makes it so important is within the new standards, there is mandatory data disaggregation, mandatory, detailed data, which is part of what Maya and Ninez have been referring to. Those categories have been established, but they can also default to the categories being used on the American Community Survey. So, there is a “refresh function” that is built into the standards. We can use the American Community Survey to really understand our subpopulations and to help determine how we are going to be collecting that detailed data going forward.
Tina Kauh: All of you have touched on the importance of implementation. I think it is a critical focus now that these changes have been made. We know only with good implementation, will we actually have good, usable, meaningful data.
What additional testing and research do you feel are required for successful implementation of these new standards? I will start with the researcher in the group. Ninez, do you have some thoughts on that?
Ninez Ponce: How do we make it happen? It is by knowledge, awareness and incentives. If we want states and healthcare data ecosystems to adopt this, then there needs to be more forums like this, a translation of” What does this mean?”, and how to do it. There could be incentives for doing this. I always say, data equity is data quality. So, it is not equity or quality. You are actually improving your data systems with better measurement. And, as I said, exposing where there are pain points, where there are assets, because then we will have more efficient resource allocation. You can bridge over time or space, so bridging is from the previous ways we have collected data, which was the two-question Latino/Hispanic, and then the five major racialized categories. If you want to look at trends over time, like uninsured rates over time, for example, that can be difficult even with American Community Survey data. Looking at the percent who have broadband, since we are getting into telehealth, the percent who have plumbing, since that was shown to be a big social determinist factor in infection rates and COVID-19, how do you do it? How do you look at trends over time if you have two different ways of collecting data? In bridging data collected over time, do we reassign only people who identified as White into the new MENA category? That is an assumption, and we need more guidance and elaboration on that. Bridging is also over space. You want to look at an analysis of trends across states. But some states may collect data differently.
The bridging suggestions are trying to understand and look at patterns that are apples to apples comparable. This is especially important in multiracial categories, because prior to 1997, everybody had to choose just one race. In 1997, the advance was being able to check more than one race. So, at that time, OMB put forth bridging guidance from the new multiple race identities to the old way of restricting identities to a single race, and suggested ways of tabulating multiracial individuals. There were several ways: the rarest group, the biggest group, the group that had the most need in a particular health outcome. There were different suggestions on how to bridge people who identified with more than one race and the data sets that subjected everyone to a single race category. I know, I was pouring over the bridging documents. I think it was very good on the Latino piece, but not on the MENA. Our team is looking for guidance on the wording for soliciting a response of the granular categories. So, for example, if you check that you are Asian, then we have a follow-up response: “You said you are Asian. What is your detailed ethnicity?” That is what we have used. There is apparently no guidance on the follow-up questions to get at granularity. Do you use the term ethnic subgroup, ethnicity, subgroup, subpopulation, origins, heritage, ancestry? From the conversations I have been part of, we decided to not use heritage or ancestry. But it is a mixed bag after that. Do you say origins? Do you say ethnicity? So, we looked at the American Community Survey. It basically says, “Check the box.” It does not solicit identity, which is neutral in some ways,in lieu of using the terminology of ethnicity or subgroup or subpopulation. But I would love more guidance.
Tina Kauh: Thank you for bringing that up. I think the issue of collecting the subgroup data is so important. We have made a big deal about these new minimum standards. But, ultimately, you need better subgroup data in order to be able to disaggregate within those broad categories.
Ninez Ponce: That is the whole point of this.
Tina Kauh: I think that is why the required collection of subgroups is such an incredible change. But so much must be done to make sure that works.
Maya Berry: I think the way Ninez laid out her questions ought to inform the research and testing agenda that the Census Bureau can put in place. There are two particularly important items. The first is an observation. I think Ninez is quite accurate when she said that bridging data was presented thoughtfully with regards to Latino populations, but less so for MENA. Regrettably, I would say MENA testing and research–a more nuanced understanding of populations from MENA–has not been what it should be and that dates back to the 2015 national content test
†
. There has been great work done on testing the combined race and ethnicity question and determining how it can better solicit accurate data about populations. While MENA was part of that, it was not given the full attention it warranted. When we saw the release of these new standards, it was a manifestation of that. And now when we talk about bridging data and how we get there, you are going to continue to see that. Our colleagues at the Census Bureau have identified this as an issue in terms of accurate data collection for a long time. They understand there is an undercount. They are committed to improving it but it is clear that MENA subject area experts must be added to the bureau beyond the great race and ethnicity team that exists there.
Second, it is also clear as we talk about the bridging issue that it actually presents us with an opportunity. MENA is a brand-new minimum reporting category. We are going to need to update the way in which we collect these data, report them, and tabulate them because we have a brand-new category, not just new standards. Getting that process right is critical, while at the same time, we have standards that tell us,” You need to collect more detailed data.” There is a requirement on additional disaggregate data. So, I think we need to lean into that piece. You must add a new category. There is really no reason not to also make sure that, as you add the new category, you are adding the subpopulation checkboxes and the write-in options. There is frankly no logical reason not to. The education point is clear across the board. It is important when we are doing the outreach to our own communities, to have them fill out the forms completely. But the imperative of detailed data collection must also be clear to stakeholders on the federal, state, and local levels. If city administrators understand those data will improve the work that they are able to do for their own constituents, then I think you get buy-in on a much broader level. We will see a lot less resistance about this idea of, “Oh, that’s a level of detail I’m not ready to do.” Well, you need to add a new category, so you also need to accept this new level of data. It is, after all, a requirement.
Meeta Anand: We have all been talking about the new data disaggregation requirement. But one thing we have not said about it is that, though mandatory, federal agencies are allowed to ask for an exception. What we need to do now is ensure that what exceptions would be allowed is very narrowly tailored and very clear, and that people do not say. “Oh, it’s a burden to collect more data,” and have that be just enough of an excuse to not collect the detailed data. I think Maya articulated the counter narrative very well, which is, “You’re going to need to change something, so you might as well get this right and do the detailed data disaggregation.”
Tina Kauh: I could not agree more about the concerns around the exemption from collecting more detailed subgroup data. Because I think that is when you run the risk of systemic racism really rearing its ugly head in this process. Because who gets to decide if something is a burden and why?
Maya Berry: Yes, as there is the added concern of only some agencies choosing to collect more detailed data for entirely different reasons–and not all good. For example, I represent a community where we talk about having been rendered completely invisible in the data by being placed exclusively within the white racial category for years. Yet we are highly visible and targeted by some law enforcement agencies. Some agencies of our government were able to find Arab Americans for profiling purposes very easily. I do not mean to be glib about it. You are rendered invisible at Health and Human Services with regards to health research. However, the Department of Homeland Security, or, say, an NYPD profiling program that was put in place post-9/11 could easily identify those communities. We need all agencies to get it right across the board, without potential motives in place that are less about data equity and more about profiling. I do worry different agencies may prioritize detailed data collection for different reasons. It is critical disaggregate data is prioritized across the whole of government because selective agency use risks causing harm to communities and will negatively impact our community get-out-the-count efforts.
Meeta Anand: I think we need to remember the mantra that data, like democracy, is not a status. It is a practice. We need to think of how we look at our data the same way, and about how we are evolving as a society and how we are capturing that information. The new standards make a commitment to looking at revising this, at least refreshing the standards every ten years, which is great, since we know it has been 27 years since the last one. But I would build into that the ability for refreshing data we collect through the American Community Survey. I think we all need to commit to this notion of pressuring agencies to constantly examine if we are successfully collecting data on our most relevant communities, on the communities most impacted. Engaging with the impacted communities in the process is key. So, seeking to do this regularly is step one but it’s important to ensure that we are engaging with communities as we continue to refine and develop and implement standards.
Ninez Ponce: I want to echo that the testing piece is not going to be as effective if it is done in a vacuum and not with communities. You need to have diverse communities, not to think, “Well, there’s this one group that is the spokesperson for one community”. Try to look for different perspectives, particularly geographic, because racialized minoritized identities differ in space as well. So, making sure it is not just all organizations in the Washington, DC area that they consult with, that they ask national organizations that represent other voices in other places. So, I think that part is important. And Maya, thank you for raising that the choice of collection, for what purpose, exposes the differential use of data as power, will really help us understand what is going on in communities.
Tina Kauh: I really appreciate the two principles that you lifted up around recognizing how people in our society are seeing themselves is evolving over time as well as the importance of community engagement. I think those principles extend beyond just race and ethnicity. It extends to the way that we have defined different demographic characteristics like disability status or sexual orientation and gender identity. So, what we have learned from this process will extend to other critical issues about data equity.
With that, I want to ask questions related to implementation. We have talked about issues at the federal level. But, the reality is that much of this will be implemented at the state and local levels. Are there any states that are implementing these higher standards already, successfully?
Ninez Ponce: The data equity center is actually going to have a state data summit in the summer. We were looking at who holds the authority of making race and ethnicity data collection in a state. Is there a chief data officer? a state data equity officer? Are they the same person? In fact, we were thinking of asking a question, “Are you aware of statistical policy directive changes?”, just to get a level understanding of how different states understand this. In California, I am not sure we are even there yet. I mean, our survey collects granular data for the Asian population, the Native Hawaiian and Pacific Islander population. We ask about tribal affiliation for the American Indian/Alaska native population. We also ask if the tribe is state or federally recognized because that comes with eligibility to access to some services. We started collecting a write-in similar to the 2020 census for the white category and for the black category. We use the write-in categories, and then we tabulated to a new MENA category as well. So, we did do that, but it was a workaround.
Meeta Anand: Massachusetts passed a law last year, with a data equity provision included in the state budget. It is exemplary in many respects, because a lot of times when you see data disaggregation at the state level, it is either for education or help, both of which are great. But it is better if you can get it across agencies so that you can really look at many different variables that are affecting people’s lives. Massachusetts has it for all state agencies. What they passed does not have data disaggregation for MENA, does not even have a MENA category. It does not have data disaggregation for Hispanic or Latino, but built into the legislation is a requirement to conform to any new updates provided by OM B’s guidance. So, Massachusetts was very forward thinking in its legislation. I am not a Massachusetts legislation or legal expert, but I was reviewing the bill in anticipation of this exact question and they do need to provide a report yearly. They have to have public meetings yearly, and they do have to conform. I am not sure about the conforming time, but I would say they are probably at the forefront of being ready to implement the standards because they are implementing their new law as the new standards come out.
Oregon is an interesting case where they have a lot of categories and they have done a lot of disaggregation. It is mostly at the healthcare provider level, but Oregon has done a really fantastic job in thinking through “What do we want? What detail categories do we want to see?” I would say they and Connecticut are really at the forefront in terms of thinking through disaggregation, just not necessarily thinking through it across all state agencies. So, I think that the next step is to say, “Look, the federal government is doing it across agencies. This is important at all levels. So, make sure that if you’re doing the data disaggregation anyway, which you are, you should really look to make sure you’re collecting that disaggregated data in a meaningful way, so that your data sets can speak to each other and really paint a portrait of your population.”
Ninez Ponce: California was a leader in data disaggregation with AB. 1726, or the AHEAD Act,
‡
which promulgated state agencies to collect granular data on the Asian, Native Hawaiian and Pacific Islander population. However, it took a while for it to be implemented, and it had no dollars. There was no appropriation for it. That was the criticism. So, you can say, “This is the law of the State.” But if there are no dollars to address that data collection burden, then it is not going to be implemented. Since then, there have been bills on disaggregation of the Latino category. I think Oregon, through the Oregon Health Authority, has been at the forefront of collecting granular data, and on intersectionality, collecting data on disability status, sexual orientation and gender identity.
Meeta Anand: I wanted to give a shout out to the great state of New York; when it implemented its data disaggregation policy law, it did have it as a funded mandate.
**
So, to Ninez’s point, New York did do that part right, which was ensure that there was funding put in place to make sure that it happened.
Ninez Ponce: That is how it must be.
Maya Berry: Absolutely critical across the board. The point I would make is on the MENA category in particular, on Arab Americans, the state with the largest population size is California. The state with the largest number of Arab Americans as a percentage of the state’s population is Michigan. Both of those states do not have a MENA category where we can capture that data. This point is critical to identifying health disparities and, as a result, improving access to healthcare. When we have looked at the states that have added MENA, the healthcare issue has really driven that in some ways. The states that added MENA prior to the federal standards being updated, Illinois, Nevada, and New Jersey, are the only three that have done so across all agencies.
I do not know that there has been enough time for us to look at how they have done it to see the bridging data piece, the way it was implemented on the local level. But it is important for us to reflect on the fact that the states with the largest and most significant concentration of those communities have not added a MENA category yet. This is why the federal standards are so incredibly important and transformative for us.
Tina Kauh: Are there any opportunities to work with the private sector on this issue? Do you have thoughts on how the healthcare industry, for instance, could either support or help advance some of this work?
Ninez Ponce: The private sector healthcare industry, of course, is motivated by delivering excellent care, but also profits right? So, revenue generation accreditation. If the accrediting bodies are saying you cannot be accredited unless you comply at the minimum with OMB’s Statistical Policy Directive (SPD) 15, then that is one way to incentivize change. In the conversations I have had among healthcare finance and policy folks, there is more and more concern about equity because there are bonuses with equity dollars built in. For example, in the Medicare advantage system, if you have four or five stars, you get a bonus. There is an equity rating that will help you bump up your star rating. So, then you can have a bonus versus a penalty. The private sector responds to maximizing revenue. Payers like the Centers for Medicare and Medicaid Services (CMS) also now have more of an equity agenda.
When I talk about healthcare delivery systems, it is hospitals, electronic health record vendors, health insurance companies, federally-qualified health centers and nursing homes. With COVID, the nursing home race and ethnicity data was terrible. So again, how do you get the private sector to adhere to more race data collection fidelity with the SPD 15? Again, I think the way to do it is through incentives. They are losing dollars on the table. Also, you cannot have high quality delivery of care or effective health systems if you do not have the data equity piece. The other piece is AI, artificial intelligence. The heartening news is that industries beyond healthcare are talking about income inequality, racial justice, and the way to have that resilience as a society. If you can get venture capitalists or CEOs to understand the importance of collecting more granular race and ethnicity data, that is about efficient investment, a way of generating beyond revenue and profits that we call public goods. That is important. But the private sector is also just in love with AI. And the thing is, with AI, unless you have good starting data, it is just going to create wrong insights, possibly harmful insights.
Tina Kauh: I think that is a great example of another way in which systemic racism can impact the quality of these data and how the data are interpreted and applied.
I want to conclude by asking how should government, either state or federal agencies, engage with communities to ensure that the new guidelines for race and ethnicity data standards are understood and accepted?
Maya Berry: There is an interagency group on the implementation of the race and ethnicity standards. If it is exclusively principals working within the government talking to each other, it will not be successful. Stakeholder communities obviously need to be an integral part of that. Research and testing are absolutely critical. One of the things we did not touch on, but I think must be stated, is while the standards were incredibly important and transformative, they were issued in a bit of an unusual way. The provided prescriptive dictated the subgroup population checkboxes, and write-in examples, and that had not been done before.
From our perspective, with regards to the MENA category, it is done in a way where we are worried about what that data collection could look like with those existing boxes, because we do not think that they represent the broad geographical diversity or the broad racial diversity of our communities. So, there is a need to continue to do that type of stakeholder engagement. And obviously, as we get closer and closer to the decennial, the importance of the Census Bureau really prioritizing more testing and more community input is critical. One of the key things that you can hear from our conversation is that the entire enterprise is about collecting better data to improve people’s lives. To get that right, there are two main hurdles. The first is that we are going to do the question wording correctly, we are going to have the right categories. We are going to do the research and testing necessary to get to that. And then the second part is actually on us as community advocates. It is to put that into the field to do the “Yalla, Count MENA In!” get-out-the-count campaign. And you know the question you’re asking about the private sector and others? I was reminded of the work we did to inform our 2020 campaign. We conducted one national poll and several focus groups across the country. When we asked the question about who are the most trusted messengers, the second most credible voice identified for Arab Americans were healthcare providers; the first were family members. This highlights the importance of getting the implementation of these changes right. It does not stop with the federal agencies creating this form, or a state entity adopting this form. Hospitals are now including it, or universities are including it, or corporations are adopting it; all those things are happening. It really does need to be the whole government, the whole of society, understanding the importance of data equity and the modernized race and ethnicity standards are a way to get there.
Meeta Anand: I think Maya captured what we need to engage with the Interagency Committee on Race and Ethnicity statistical standards. The Interagency Technical Working Group that worked on the revision of the standards held listening sessions very deliberately. They did community engagement. We need to see that done again. What questions should we be asking? What is missing? What does implementation look like? Those open forums need to be held once again.
We also need to see engagement by the Census Bureau in message testing. No one has the reach that the Census Bureau has. We need the Census Bureau to work with communities to understand what message testing needs to be done and how to explain the new question to communities. There will be some level of confusion; that is not a bad thing. There is confusion with the old way, too. People need to be brought up to speed, and we need the assistance of the federal government, which has more resources to be able to really do that broad-based testing and rolling out to communities. Then, partnering with community advocates and healthcare professionals and researchers, to see what is going to resonate with people and how we can really engage our communities in this endeavor.
Ninez Ponce: I think it is about accountability and democratization of data. They should build the tools of democratizing this information because you are doing all this work and content testing. You need to make sure there is accountability around which agencies are complying so that it is not just Homeland Security. Then, creating more tools that are public-use-facing and freely available for communities to use and then resources for technical assistance, so that communities are aware of the information and know how to use it.
