roundtable discussion: Lust for Statistics

By Published on .

For those with an affinity for numbers, the decade-long wait is finally over.

The oasis is in sight. Demographers and researchers, who have awaited more accurate numbers on the makeup of America for 10 years, finally get to quench their thirst. Amid the changeover from a manufacturing to a knowledge-based national economy, diversity is the theme that underlies the huge national database of people and numbers. Detailed results of Census 2000 will begin to trickle out in the coming months, and social scientists, researchers, and strategists will get first dibs on plumbing the data. In late December, the Census Bureau launched the grand countdown by announcing America's population figures — 52 numbers in all, which included national and state population counts. [See Sidebar]

That was just a tease. This month and next, the bureau will develop analyses that will lead to redistricting and the apportionment of federal funds. In anticipation of all the fun to come, American Demographics canvassed some of the country's leading demographers and market researchers for their first-blush view of Census 2000. Race and ethnicity information is, by far, the most anticipated piece of the census pie. Previously, the bureau limited racial identification to a single race, but Census 2000 allowed Americans to identify themselves as more than one race, leaving the door open to a seemingly endless combination of racial and ethnic identities. Researchers of all types are anxiously awaiting a first peek into America's true colors.

Between June and September, the bureau has promised a plethora of additional statistics: age, sex, relationship status, household type, migration — and more race data. By then, we'll learn the age of those races. How old on average are, say, Hispanics, blacks, whites, and all combinations thereof? The numbers may perhaps be the final wake-up call for marketers. Among the predictions likely to be confirmed: Younger America is more racially diverse, and the oldest age groups have the greatest percentage of whites. For a complete calendar of census release dates, check the bureau's Web site, For a sharper view of how the numbers appear to add up, read on.

Q: What does the data already released tell you?

FREY: Migration to the Sunbelt is no longer flowing exclusively to the traditional Southern and Western states like California, Texas, and Florida. Rather, migratory trends are creating a “new Sunbelt� — states in the Sunbelt region that grew more in the '90s than in the '80s — including Nevada, Colorado, Arizona, Utah, Georgia, and North Carolina. These are states that some people previously characterized as backwater, but now we're seeing that they will continue to be very strong magnets.

FORSTALL: The state totals were, without exception, higher than estimated, notably for the District of Columbia. This suggests that the population of older central cities may turn out to be higher than anticipated. If so, the race and ethnic composition information in comparison to 1990 census numbers will be of special interest. Is it a recent tide of immigration that has caused the unexpected gains? Will we find that older cities like St. Louis, New York, Chicago, and Philadelphia have actually grown significantly — or had much smaller losses than expected? If older cities are actually growing, redistricting will not reduce their political clout as much as has been the case after the last few census counts. Diminishing political clout for these older cities may finally be reversing.

HODGES: Our immediate interest is in the redistricting data. Although limited to total population and population by race and Hispanic ethnicity (as well as an age break at 18), these data will provide us with the first small area counts for Census 2000 geography. It will also provide a new base for estimates of population and population by race and ethnicity. Although there are other Census 2000 products of interest, the “main event� for many business users is Summary File 3 [scheduled for release between June and September, 2002]. The file will provide a wealth of tabulations from the census long form, including income, home value, education, occupation, language, commuting, disability, and many others.

SPAR: The really important information will come out by April 1 — the deadline for producing the Redistricting File, which contains data by race and age, down to the block level. Business is going to start looking at these data on April 2. I doubt that businesses looked too closely at the state data — it's too large of a geographic area for business purposes. Projections of state population were already available and there weren't many surprises in the actual numbers.

Q: What data are you most looking forward to getting your hands on?

FREY: The new data coming out on race is on everyone's front burner. For most of the country, the responses will still be predominantly single race (i.e. black, white, Asian), but in areas like Los Angeles, New York, and South Florida, we will gain important new insight into the result of the melting pot on intermarriage.

PREWITT: I'm obviously excited about the redistricting data, which is the first detailed data from Census 2000 with race, ethnicity, and some age data all the way down to the block group level. It's a very early snapshot of the composition of the American population and where it's living. The entire country will be watching the patterns of the race data. All early indications are that the overall levels of multiracial identification will not be that high, but we expect to see some areas of high concentration.

WITECK: We're interested in uncovering new, unprecedented data on same-sex, unmarried couple households. We will analyze geographic distribution, age, income, and behavioral characteristics for same-sex households that have never before been captured effectively by U.S. census efforts. We are especially interested in comparing the migration of those in same-sex households (who are generally understood to move to highly urbanized areas) to that of the aggregate population, which, as we found from the earliest census release, is largely moving to the South and to the West.

DARGA: One important issue to watch is how successful the efforts were to confirm vacancy status. In 1990, for example, 31 percent of Michigan's municipalities had vacancy rates over 25 percent, and 14 percent had vacancy rates over 50 percent. This is largely due to seasonal migration by people with more than one home. In such communities, it is critical to distinguish vacant housing from other non-responding housing units. Even a modest rate of error in determining vacancy status can have a serious effect on population figures for communities with high vacancy rates.

Q: What will be THE big changes from the 1990 census to Census 2000?

PREWITT: We know that there have been high immigrant flows [into the U.S.] over the past decade — we're now over 10 percent foreign born — and soon we're going to see those flows in more geographic detail. For example, in 1996, Columbus, Ohio had about 50 people from Somalia. That group today is expected to number more than 15,000. We might find fascinating concentrations of our newer residents that did not exist in 1990.

Q: What implications will Census 2000 have for business?

FREY: Because the first wave of Boomers are now about to move into their early retirement years, we'll get our first detailed look at which people are poised to move to retirement communities and which ones will be satisfied where they are. Marketers will also have to pay attention to the vast socioeconomic divide of the elderly, which will only widen as the Boomers age. The yuppie elderly will be a new market for all sorts of products. There will also be a lot of Boomers who are just scraping by, and because it's the Boomer generation, there will be a lot of them.

HODGES: Census 2000 is the first to identify persons of multiple race, and provides unprecedented detail by reporting totals for specific multi-race combinations. The large number of categories poses a challenge, as does the additional work required to identify 1990 to 2000 census trends, but the additional detail will help some businesses refine their targeting of population segments.

O'HARE: One can't help but wonder if the new racial categories might erode some of the power of the traditional minority/majority categories. It may add to the re-examination of many public policy and private sector programs built around the black/white dichotomy that dominated public thinking on racial issues in the 1950s and 1960s.

Q: What will be the “unsung stories� of this census?

PREWITT: The obligations between the census and the American people is a significant story. Census 2000 was the people's census, with numerous civic and community groups stepping forward to make this a good census. They did their job; now it's [the bureau's] job to get the data out there for the people to use. I'd like to see the media cover the community's use of the data once it's delivered back to them.

FREY: The black movement back to the South may go underreported. Where are middle class blacks moving to, and how are they different? We'll be able to tell where exactly they're moving to, which will give us a better idea if and how they're becoming integrated in the community.

the panel

Kenneth Darga State Demographer, Michigan

Richard Forstall Consultant, (Retired Staff Member, Census Population Division), Washington, D.C.

William Frey Demographer, Milken Institute, Santa Monica, California

Ken Hodges Director of Demography, Claritas, Inc., San Diego, California

William O'Hare Demographer, Annie E. Casey Foundation, Baltimore, Maryland

Kenneth Prewitt Dean, Graduate Facility, New School University, New York City (Former Director, U.S. Census Bureau)

Ed Spar Executive Director, Council of Professional Associations on Federal Statistics (COPAFS), Alexandria, Virginia

Bob Witeck Partner, Witeck-Combs Communications, Washington, D.C.

Most Popular
In this article: