January 20, 2012
The Future Starts Now: Designing the 2020 US Census
Photo ©2010 Quinn Dombrowski (Flickr: quinn.anya). Used under a Creative Commons License (BY SA-2.0)
As a lover of history, I always find myself drawing parallels between today’s headlines and consequential events or figures of the past. So when someone asked me why the Census mattered, I immediately thought of Robert Caro’s biography of Lyndon Johnson, and a passage that recounted LBJ’s first years in the United States Congress. A snippet:
“When, in 1937… Johnson became their Congressman, Hill Country farmers were still plowing their fields with mules because they could not afford tractors. Because they had no electricity, they were still doing every chore by hand, while trying to scratch a living from soil from which the fertility had been drained decades before. They were still watching their wives made stooped and old before their time by a life of terrible drudgery, a life that seemed, as one Hill Country woman put it, ‘out of the Middle Ages.’ Four years later, the people of the Hill Country were living in the twentieth century. Lyndon Johnson had brought them there.”
Legislative horse-trading aside, this passage speaks to the dramatic changes that the New Deal brought to families and communities across the country, and helps us grasp the incredible impact that government can make on the lives of citizens through policies, services, and investments. Government doesn’t have a monopoly on social impact, but it certainly has significant market share.
As such, Caro’s account helps us understand why the Census matters. It is not simply a counting exercise; it is also a tool to assess where people are, how they live, and what they need in light of changing times. It takes the pulse of an ever-expanding, ever-changing nation, and provides a diagnosis that directly informs the next decade of national priorities and initiatives – who gets a congressman; who gets a road; who gets an infusion of health services.
As the country grows, the Census becomes even more important. But it also becomes more complex because of an increasingly diverse population, not just in terms of race – although this is a critical factor, with minorities accounting for 98 percent of population growth in large metropolitan areas between 2000 and 2010 – but also in terms of geography, technology uses, and patterns of engagement with government.
As complexity grows, so does cost. The 2010 decennial cost a whopping $13 billion, in large part because counting as many people as possible became far more difficult to do in an efficient and effective way. This challenge has two key drivers: the people that we’re counting and the tools we use to count them. Take this statement from US Census Director Robert Groves in his April 2011 congressional testimony:
“We know of no single method of collecting census data that is optimal for all residents of the U.S. Some residents have told us they do not want people visiting their home… some residents want to use the Internet at any time of the day… and some want to speak by telephone to someone guaranteed to speak their language.”
This statement highlights a challenge faced by many leaders with whom Reboot works, leaders who are striving to deliver critical social services and programs. Like Mr. Groves, they work with populations that do not fit into simple categories nor exhibit predictable behavioral patterns. These leaders also commonly face increasing demands to harness the power of technology to reduce costs and to maintain or improve quality.
Time and again, however, this task is easier said than done. We find that most difficulties arise when such efforts are undertaken with a one-size-fits-all solution (technological or otherwise), developed without a deep understanding of citizen behavior.
These difficulties are not simply logistical flaws or signs of bad intentions. Instead, they are design flaws – failures that flow from the lack of a structured process for sourcing fundamental insights about the people that an organization aims to serve, using such insights to design contextually appropriate tools, and iteratively improving the tools so that deployment is no longer a crapshoot.
Design research is an investigative tool that helps us probe the nuanced needs, behaviors, and constraints of people and complex ecosystems. It borrows heavily from ethnography’s mastery of interviews and observation, and it uses the tools of interaction design and cognitive science – such as personas, citizen journey maps, and mental models – to better understand and serve citizens.
With these tools, we can work to answer some of the key questions facing the Census Bureau in advance of the 2020 decennial. What is the best mix of modes and strategies by demographics and geography? What technology will be feasible for self-enumeration? How can household follow-up be improved or eliminated?
While quantitative research helps clarify the “what” – such as determining which populations are least likely to participate – design research helps us understand the underlying “why” – whether certain groups disengage because of technical barriers to participation, or because of more complex dynamics in their relationship with government.
To get to the heart of such challenges, we embed with communities to understand the citizen experience – as it is, not simply how we perceive it to be. We conduct in-depth interviews, stakeholder consultations, and map a wide array of citizen experiences to determine opportunities for more effective engagement and service delivery methods.
In this sense, design research drives innovation, as it uncovers hidden challenges to be addressed or opportunities to be seized. Its value is directly related to Mr. Groves’ realization that the Census “must innovate if we are to remain useful and relevant to the country.” The innovation challenge, however, is not simply a technological one, though part of any 21st century solution will very well include online and or mobile options where appropriate. The challenge instead is to truly understand where those technology applications can be most effective, and also where such applications will produce unintended consequences. We can’t meet the challenge without meeting the people who will determine success or failure of the enterprise – the citizens of America.
Design helps us to do just that. It can be useful in the tests that the Census Bureau is rolling into the American Community Survey. It will make the iteration process more productive through citizen research pre-survey, user testing during the survey, and citizen feedback post-survey. It will also help create a continuous feedback loop that can naturally flow into the larger design process of the full decennial. Design research is a critical beginning, but a beginning nonetheless, as it is never complete without proceeding to the design and deployment phase. This is the type of progress that Mr. Groves has demanded, and that a citizen-centered design process will enable.
As the Census Bureau moves towards 2020 and strives to radically change the way it has operated over the past 40 years, it has a brilliant opportunity to create the first citizen-centered census in American history – a tall task indeed, but one that design will make far more attainable.
This is the first in a series of posts that will explore the unique challenges and opportunities surrounding the 2020 United States Census. Stay tuned!