The Census as a Tool of American Democracy

Dan Bouk Talks About His New Book, “Democracy’s Data”

By Rachel Hooper

Author Dan Bouk was first lured to the census by the stories and the people that it contained, to the inherent drama of a nation seeking to know all of its people, and the extraordinary difficulty of getting whole complex individuals to fit into a data system. Because the U.S. Census Bureau is obligated to keep all individual records confidential for 72 years, the 1940 Census was the most recent data set to which he had complete and total access. These records serve as the starting point for many people when researching the stories of their families or their neighborhoods.

Dan Bouk teaches U.S. history at Colgate University and resides in New York City. He studied computational mathematics as an undergraduate before earning a PhD in history from Princeton University. His first book, “How Our Days Became Numbered,” explores the life insurance industry’s methods of quantifying people. His newest book, “Democracy’s Data,” tells the stories behind the 1940 Census, the Census Bureau, and the importance of the census as a tool of American democracy.

This interview was edited for length and clarity.

Let’s zoom out for a moment. What does it mean to be counted in the census?

At its most fundamental level, the census is called for in the U.S. Constitution and used to determine how people are represented in government. To be counted in the census means that you can have somebody stand in for you in the U.S. House, and that you and your state will have a say in who is chosen as president by the electoral college. Over time, the census came to be used as a powerful data source for the national government, as well as state and local governments, to plan for their futures. Since the 1940 Census, census data have also increasingly been used to allocate funding, often directly to communities for social goods like roads, schools, health care, and public transportation.

Less often remembered, except when people start to dig into their records for genealogy, is that a census is one of the only sets of records that are well-preserved and capture the stories of everyday people across history — people who deserve to be remembered. The census is a kind of history for people whose stories might otherwise be erased or lost to the past. 

Which people are you referring to when you say that they deserve a place in history?

As a historian, one of the things most startling, and most distressing, is that the people whose papers are kept in university archives, or libraries that are accessible, tend to be political elites, political leaders; they tend to be wealthy men, White men. For everybody else, it’s much harder to find out anything about their past. The census doesn’t discriminate on any of those factors. It’s supposed to include a little bit about everybody.

It is clear that our democracy is facing an existential crisis. Why is a fair and accurate census essential to the healthy functioning of democracy, and what will it take to help people across America understand the role of that unique data?

We need to be able to trust the people who make our facts. We need to be able to trust the counters. It is really important that people can reliably understand the Census Bureau’s practices and trust that the entire process is nonpartisan. We need to commit to the idea that the people and institutions responsible for making the census are committed to accuracy and confidentiality — a confidentiality that will protect people from having their privacy invaded.

Some people have been pushing for the Census Bureau to be given an even more independent position within the U.S. government outside of the executive branch, so as not to be swayed by political forces — a particular danger we saw during the 2020 Census. This is a good reminder for folks that right now, eight years away from the next census, is not the time to rest. Rather, this is the time to get involved and to make sure that the questions, procedures, institutional process, and funding for the census are put in place and ready for the next count in 2030.

Given your historical knowledge of the census, what do you believe are the most concerning gaps that persist in the census data and how it is disaggregated for reporting?

We’ve had evidence of systemic undercounts for more than a century. Some of the first serious investigations for undercounts were done by Black scholars in the 1910s and 1920s. The Census Bureau partially supported some of that research. Yet, because they segregated the bureau, while one wing of the Census Bureau had scholars who did the work, other parts were much more suspicious and denied that this was a systemic problem. But, by mid-century, it became clear that several communities were being systematically undercounted. That’s a persistent and troubling gap that I wish I had a great solution for, but it remains to be given the necessary political attention, process energy, and creativity. The earliest analyses of undercounts identified and estimated the number of African Americans who had gone uncounted. But since then, as in the 2020 Census, evidence indicates the high likelihood that children, American Indians and Alaska Natives, Hispanic people, and other groups have been undercounted too. In general, the census is built to be a census of households, and that means that people who are transient, people living on a low income, and people who live outside of the traditional concept of a nuclear household with one head of the household are harder to capture with current census methodology.

How do we end our nation’s long history of weaponizing the census against certain communities and ensure that it accurately captures our rich diversity? What’s at stake if we get it wrong?

During the efforts to prepare for World War II, existing law held that individual information produced by the census was all to be kept confidential. At the time, a lot of people did not have birth certificates or the documentation to prove their citizenship that was required to get a good job in a factory — a job that many people needed to support their families in the middle of a depression. That meant that thousands of people wrote the Census Bureau asking for certified records of their own census responses, which they could then use as identification materials. The Census Bureau, which is supposed to supply statistical data, ended up becoming a mass producer of individualized data.

Significantly more troubling, the Census Bureau jumped with two feet into the process of trying to prepare for the war. In particular, that meant looking for and providing special tabulations about the location of Japanese Americans, which then became an important part of the infrastructure for the evacuation and mass incarceration of Japanese Americans during World War II. The laws that made some of those individual disclosures of personal data legal during war time were repealed back in 1947, and the Census Bureau officially apologized in 2000. The bureau has done a lot to ensure that direct weaponization of its data is now much more difficult. Still, it’s important to tell these stories and to remember how statistical systems can be misapplied and used to harm individuals or groups, especially in times of war and national crisis. 

The final thing I would say is that the stories themselves can be, and often are, weaponized. They can be amplified as a way to try to convince people who might be in marginalized communities or immigration statuses to be wary of participating in the census. With this longer-term kind of weaponization, whole groups of people can be made less visible and therefore miss out on representation, funding for their communities, or having politicians understand their needs. There’s a real cost when people are dissuaded from participating. 

What did you learn while researching and writing this book that left you feeling hopeful about the story of America or the story that we have yet to write?

There are a lot of great stories of people in the census. The census itself, though not a perfect institution, is actually pretty darn great, especially compared to so many other institutions at a point in our history when we as a society care about data-driven everything and are poised to hand over so much power to automated systems. The Census Bureau is subject to democratic control; it is not a private corporation. It is deeply committed to privacy and the ideal of not harming the people whom it represents. It aims to include everyone and acknowledge the dignity of each individual. Can you imagine that kind of mission statement from most other data-driven institutions? I think it’s an impressive model for what we can expect and demand from other institutions as we evolve toward more data collection and automation.

If you could change one thing about the census today, what would it be? 

Expanding a bit more broadly, the census in Ireland recently included what they called a time capsule. It is an open blank box where people can write whatever they want to about themselves. This is an opportunity for people to define themselves not as the government or a form defines them, but as they want themselves to be remembered a hundred years from now. I would love to see the U.S. census incorporate its own time capsule element. We need the census form to evolve to allow people to see themselves reflected in the choices presented to them.

“Democracy’s Data” is the inaugural selection of The Leadership Conference’s Census Book Club, which features books to deepen members’ understanding of the census, its rich history, and its real-life impacts. Meeta Anand, senior program director of census and data equity at The Leadership Conference, was a reviewer of this book. This interview does not represent an official endorsement of the book.

Rachel Hooper is a senior communications manager at The Leadership Conference on Civil and Human Rights.