RJ. Kenneth Blackwell`s Remarks to the Urban Institute
Good afternoon. It is a pleasure to be with all of you today. I want to express my appreciation to Janet Norwood for inviting me to participate in the Urban Institute`s First Tuesday issues series. Director Prewitt and I are here to provide an overview of the 2000 decennial census. While we both agree that the 2000 census needs to do a much better job counting people of color, the poor and children – those who were disproportionately undercounted in the 1990 census – we disagree on the best way to do it.
At the risk of sounding divisive, I will start with an issue on which many disagree: statistically adjusting the census count for a second set of numbers. Discussing the census, I tend to think of three friends who went hunting: a Republican, a Democrat, and a statistician. A deer appeared in a clearing, and the Democrat and Republican opened fire. The Democrat missed ten feet to the left, the Republican missed ten feet to the right, and the statistician jumped up and shouted, “We hit it!”
I am skeptical about the Bureau’s plan to adjust the numbers for primarily one reason. Adjustment is not accurate at the block level. Even those who support the plan recognize this problem.
Just last month, Dr. Ken Prewitt, Director of the Census Bureau, told Congress, “[A]t state level, sampling can be effective. But when you get to census block, I think there’s going to be a real division. When you’re only talking about 50 people in a block, and there are six, seven million blocks, there is a legitimate debate by scientists, by statisticians, that [sampling] is not the most accurate.”
Dr. Prewitt’s comments echo those of other experts at the Bureau and across the country. Even the National Academy of Sciences, which broadly endorsed the use of some kind of sampling in 2000, noted that any national statistical adjustment “will be less precise on average” at the local level.
Adjustment gets increasingly less accurate at smaller levels of geography. A survey can give you a pretty good idea of the national or state population. Big cities are a gamble. Small cities are worse. By the time you get down to neighborhoods and blocks, the data is simply unreliable.
Dr. Prewitt and the Bureau say block-level inaccuracy is okay, because adjusted numbers are more accurate at the congressional district level: areas with about 500,000 people. However, I believe blocks and neighborhoods – areas with just a few thousand people – are exactly where we need the most accuracy. The largest undercounts repeatedly occur in small areas – neighborhoods – that are very hard to count. Low-income neighborhoods. Black , Hispanic and Asian neighborhoods. Racially and linguistically isolated neighborhoods, whose residents could benefit most from the funding and representation that come with being counted.
Let me give you an example. Last week at this time, my colleague on the Board, Mark Neuman, and I met with residents and community leaders in Robert Taylor Homes. Robert Taylor is the nation’s largest single public housing project. It covers about 15 blocks of prime real estate in downtown Chicago. That’s just four census tracts – less than half the size of the smallest congressional district. It has about 22 sixteen-story buildings. 99% of the residents are African American. 84% earn less than $10,000 a year.
Robert Taylor is the picture of a hard-to-count neighborhood: low-income residents in a high-crime area who don’t trust government. In 1990, the Bureau counted 8,787 people in Robert Taylor. But the Chicago Housing Authority estimated the population at over 12,000 – 3,500 more than the census found. The adjustment – the statistical adjustment – would have added only 673 people.
Adjustment would have added over 100,000 people to Illinois. But only 673 would have been added where they were needed most – in the fifteen blocks of Robert Taylor Homes.
If we count the people of Illinois, but we don’t count the people of Robert Taylor, we will not alleviate the social injustice of the census. No one denies the census will miss people. It is common sense that, in a nation as large, diverse and mobile as ours, we will miss some people. But it is unconscionable to consistently miss the same people in the same communities – the very communities that depend on the census for the schools and health care and child care that come with being counted. And it is impossible to count those communities without the help of the people who live and work there.
We need people like Reverend Herbert Martin of the Progressive Community Church, who joined us for two focus groups with 24 men and women of Robert Taylor on Monday, and met with us again Tuesday. He is a pillar of the community, known throughout Chicago and in the neighborhood. His church operates a food and clothing pantry, and is building affordable housing in the area. The people of Robert Taylor are his congregation. They know him. They trust him.
We also met with Gwen Long, the principal of Farren Elementary, right across the street from Robert Taylor. She works with parents from Robert Taylor daily, parents who shepherd their children to and from school for safety.
The value of men and women like Reverend Martin and Principal Long cannot be overstated. If a census taker walks into Robert Taylor alone, most doors will be closed. But if a census taker walks in with Reverend Martin, doors will open. People will offer to help.
We need that help. We need men like Tyrone Galtney and Levi Nawls – unofficial resident ambassadors who also joined us on Tuesday.
Tyrone and Levi don’t have the official title like Reverend Martin and Principal long. But, inside Robert Taylor, they carry just as much weight. They have lived there for years, even though there may be no record of it. They have more knowledge, credibility, and access than anyone from outside. They are the undercounted – the census missed 1-out-of-12 Black men in 1990. The undercount among Black men in public housing was even higher.
Tyrone was very frank. Without my help, he said, “you’ll never find me. I stay in 4130, 1322, 5700” and about four other apartments. Levi put his finger on it when he told a reporter, “There’s a misperception that Black men aren’t interested in the census and how it affects us as residents … I want to help get [residents] the correct information.”
He wasn’t the only one. The question asked most often last Tuesday was, “What can I do?” How can I help? How can I make sure my people, my family, my neighbors, get counted next year?
I have fought the war against social injustice for 22 years of public life, and 15 years before that. For the first nine years of my life, I grew up in a neighborhood like Robert Taylor. I can assure you, the battle of the undercount is won in the trenches. In the churches. In the schools. In the community centers and in the Robert Taylor Homes. Our generals are not the Ken Blackwells and the Ken Prewitts. Our generals are the Reverend Martins, the Gwen Longs and the Tyrone and Levi’s. Our generals live in the undercounted community. They work in the community. They serve in the community. They do not ask, “What can I get?” They ask, “What can I do?”
If we are to win, if we are to be victorious in this largest peacetime mobilization, we must equip our generals. We must ask ourselves, “What can I do?” How can I enlist Reverend Martin? How can I supply Gwen Long? How can I reinforce Tyrone Galtney and Levi Nawl? Here, from Washington, what can I do to help our people on the front line?
Be intellectually honest before you answer with “statistical adjustment.” Adjustment is not accurate on the front line. Adjustment won’t count Reverend Martin’s church, or Gwen Long’s school. It won’t find Levi and his friends. Adjustment won’t put people into Robert Taylor and neighborhoods like it. We can’t win this battle from the air. We need ground troops.
Our ground troops are the trusted men and women of intelligence and experience in undercounted neighborhoods. Men and women trusted by their community.
They need to be deployed strategically. Of the 60,000 census tracts across the country, the Bureau has identified about 2,600 that will probably be very hard to count. Only 2,600, out of more than 60,000. If the Bureau can improve the count in those neighborhoods, it can reduce the differential undercount.
I told you I grew up in housing like Robert Taylor. So did General Colin Powell. Recently, he and I had a conversation about the human information network in our neighborhoods growing up. We had more aunts than our mothers and fathers had sisters. At any given time, our aunties might be on the stoop or at the window, from one end of the neighborhood to the other. Nothing under the sun was hidden from our aunties. If you got into trouble on one end, the news was at the other faster than you could run. Kids today have the Internet. But in my neighborhood, and General Powell’s, we had the Auntie-Net.
As I have grown older, I have lived in other neighborhoods. But I always return to where I grew up and places like it. I am often reminded how people outside the neighborhood – especially here in Washington – mistake low income for low IQ. In other circumstances, my aunties – and men like Tyrone and Levi – would have been doctors, lawyers, and professionals. They would be respected the way the National Academy of Sciences is respected, or the experts at the Bureau are respected.
They deserve our respect. They are the true experts on undercounted communities. Respect them. Listen to them. Hire them. Give them the funding and support they need to count people, to build a platform for people to speak up and speak out and break down the barriers to opportunity.
Until we support the true experts on undercounted neighborhoods, we will continue the social injustice of the census differential undercount. We will continue to drastically undercount places like Robert Taylor Homes. We will miss children and people of color, and continue to deny them a fair share of American representation and opportunity. Until we have a plan to count blocks and neighborhoods as well as we count cities and states, we still have work to do.