Census 2000 Is Coming!

Article Tools
  • PrintPrinter-Friendly
  • EmailEmail Article
  • ReprintReprints
  • CommentsComments

The 1990 U.S. Census was easy. Public school students were classified as being white, black, Native American, Asian/Pacific Islander, or "other," as far as their race was concerned. ("Hispanic" was not a race, but an ethnic group, so Hispanics also had to pick a race.) Only five races and one ethnic group; unchanged since 1977, when the boss of the Census, the White House Office of Management and Budget, issued Directive 15, outlining the racial/ethnic categories for the Census. Interestingly, in the first paragraph of that directive, one reads, "These categories have no scientific validity." If not scientific, what are they? Answer: one of the biggest public opinion polls ever--if you think you're black, you're black.

Race is of the utmost importance historically and politically, but is scientific nonsense. Earlier censuses had diagrams of various lips, eyes, noses, and hair that were used by the census-taker to decide what race a person was, which was even worse than having people decide for themselves. In Census 2000, all bets are off. For a decade, people of mixed ancestry have been lobbying for a census that would describe everyone accurately. Project RACE, which stands for Reclassify All Children Equally, is one of a number of such organizations. There are a minimum of 3 million schoolchildren who are racially mixed, and probably triple that number. In addition, 35 percent of U.S.-born Hispanics are marrying non-Hispanics, and 45 percent to 55 percent of U.S.-born Asians are marrying non-Asians, following a trend begun by Americans with European backgrounds. Only 15 percent of U.S. citizens of European extraction now represent, say, an Italian married to an Italian, a German to a German, or Swiss to Swiss. Yet in 1900, if a German married an Italian, it was called miscegenation. (See American Demographics) , December, 1997.

As a nation, we "melt" by marrying people of different racial/ethnic/national origins. Yet "race" remains a major political force, even as the physical differences we associate with it are fading away through intermarriage. Today in America, the darkest fourth of the white population is darker than the lightest fourth of the black population.

This Commentary was selected for inclusion in The Last Word: The Best Commentary and Controversy in American Education, published in 2007. Get more information on the book from the publisher.

Responding to these pressures, as well as to court decisions that have upheld the principle that children cannot be forced to choose between their mother's and their father's ancestry, the Federal Interagency Committee's final recommendations for the 2000 Census, issued in October 1997, suggested a major expansion in our definition of race/ethnicity: Black and African-American become two categories, as there are 3 million black Hispanics, mostly Caribbean, who do not consider themselves African; Asian becomes two more--Asian and Native Hawaiian/Pacific Islander (most Chinese do not consider themselves in the same category as Hawaiians); Indian becomes two more--American Indian and Alaskan native; plus white and "other," for eight categories.

However, the 1997 statement allows every American to select as many categories as he or she wishes. The simple checking of one of five boxes in the 1990 Census becomes a matrix of 8 x 8 in the 2000 Census, or 64 boxes. If you allow Hispanic/non-Hispanic for all categories, the matrix becomes 128 boxes. (There is considerable pressure to add categories for the approximately 3 million people from the Middle East--1,000 mosques now exist in the United States--plus the 3 million Americans from the former Soviet Union. Both are as much an "ethnic group" as Hispanics, but this would push the combinations well into the stratosphere, if not outer space.)

William P. O'Hare, the Kids Count coordinator at the Annie E. Casey Foundation, has suggested that this will be a data-collection nightmare for school administrators (no federal money for Title I, special education, school lunch, and so on, unless you include an analysis of your students by "race"), and even more so for policy analysts.

The young golf sensation Tiger Woods, for example, has claimed that he is a "Cablinasian," allowing him to check all four--Caucasian, black, Indian, and Asian--boxes. Fine. Next question: Does Tiger Woods count for four people with his four checks? Does each checked box equal exactly one-quarter of his ethnic heritage, and if so, by what logic? Can Tiger Woods be counted as black for affirmative action purposes, and as white for housing and health surveys? What of the Hispanic child with a Mexican mother and a Cuban father--is that "mixed race"? (The odd answer is that a Hispanic couple, one from Cuba, one from Mexico, will be forced to choose only one nation of origin, while races can check as many as they wish in Census 2000.) How can you add all these boxes to get back to the individual who was split up in the first place? (Tiger Woods' four checks must "add up" to only one person.) If you just count all the boxes checked, the U.S. population could go from 280 million to 600 million. In addition, the black population could leap from 32 million to 50 million, or it could decline to 20 million, depending on how people perceive themselves, and how many cells they check.

For research purposes, it's clear that Census 2000 won't be compatible with the 1990 results. If you want to compare the growth of the black population from 1990 to 2000, the 1990 figure is easy, but what number should be used for 2000? All the people who checked only black? Should they be given more statistical weight than those who checked black along with other "races"? What happens to the valuable studies of the National Assessment of Educational Progress for various racial/ethnic groups through time? Are minority students going to college in greater numbers in 2000 than in 1990? Are federal and state programs and social services geared to certain racial/ethnic groups (for example, the Indian Health Services) going to the "right" people? All longitudinal studies of student performance through time, like the National Educational Longitudinal Survey and the Schools and Staffing Survey, will be in serious difficulty.

It is true that for most school districts, this new census requirement will not change things much, as 85 percent of America's 3,065 counties are still 90 percent or more white. Indeed, just 10 states--California, Texas, New York, Florida, Illinois, New Jersey, Arizona, New Mexico, Colorado, and Massachusetts--contain 90 percent of the country's Hispanic population, while 43 percent of Asian-Americans live in only three metropolitan areas: San Francisco, Los Angeles, and New York City.

Just 10 states contain 90 percent of the country's Hispanic population.

But even in a county that is 0.5 percent black, 0.2 percent Hispanic, and 0.2 percent Asian (99 percent white), the 64- or 128-cell school survey will have to be filled out. And if the "diverse 1 percent" is 30 students, each of whom checks six (different) boxes out of the 128, the point should be made. Billions and billions of taxpayers' money will be allocated based on these categories, yet no one has suggested that all the Congressional legislation be modified to include these new categories or their combinations.

Most important, the human and fiscal cost of reporting to the federal government on the racial and ethnic diversity of every student in every school of every district of every state in the Union could easily triple, and the results could be much less useful for the educational enterprise, at all levels. (Student financial aid in America's colleges and universities will have to be given out in some very new and complex ways.)

Many others are salivating at this new complexity, especially business marketers who are hoping to discover many new niche markets within the gross totals available to the users of previous censuses. On questions of where to put that new bowling alley, the new ethnic restaurant, what languages to use in the new bank ATM machines, what kind of toothpaste Koreans will buy, and so on, Census 2000 may well become a bonanza. But for America's schools and colleges, the next Census may become a complex, ambiguous, and basically nonuseful tool, which will have to be followed in providing district, state, or even federal information, or face the risk of losing federal money.

If the 2000 U.S. Census is to be as accurate a description of the American people as possible, a "mirror held up to nature," it is likely that the image will be so blurred as to be unrecognizable. And that may turn the country away from a concern solely with racial inequities and toward the ultimate problem: a 20 percent poverty rate among youths in the United States. Being black is no longer a universal handicap, as 20 percent of black households have a higher income than the white average.

Being born poor is a universal handicap, regardless of race or ethnicity. Many people are beginning to think of economic desegregation as a more productive way to a better society, and the confusion produced by the 2000 Census regarding race may only increase this new commitment, particularly regarding the economic inequities between city and suburb.

Vol. 19, Issue 5, Pages 34, 48

Published in Print: September 29, 1999, as Census 2000 Is Coming!
Web Resources
Notice: We recently upgraded our comments. (Learn more here.) If you are logged in as a subscriber or registered user and already have a Display Name on, you can post comments. If you do not already have a Display Name, please create one here.
Ground Rules for Posting
We encourage lively debate, but please be respectful of others. Profanity and personal attacks are prohibited. By commenting, you are agreeing to abide by our user agreement.
All comments are public.

Back to Top Back to Top

Most Popular Stories