Count Me In
Census 2000: There's Power in Numbers
By Mike Clark-Madison, Fri., Feb. 11, 2000
To people who still have some faith in "civics," it may seem weird and sad to have come to this, marketing the census -- promotion, media events, the inevitable Capital Metro bus wraps. But counting everybody is hard work, especially in cities like Austin that are on the demographic move. And a decade is a long time; unlike voting, which happens in Austin seemingly every few weeks, the census only happens once in a generation, and the past 10 years have been tumultuous ones. So this constitutional obligation, once familiar to and anticipated by everyone, now has to make its case.
In Austin, it should be an easy case to make; if you don't respond out of duty, you should at least act on curiosity. Anyone trying to make or follow public policy in Central Texas over the past three years has struggled with 1990 numbers that by this point are meaningless. How much have we really grown? Where did the people come from? What are they doing here? Where are they living now, and who or what did they displace? How big is the gap between the rich and poor sides of the digital divide? How much does it cost to live here? Can we afford it? All the questions we toss about, over the bar stool and onto the council dais, can be answered, at least in part, by an accurate census in 2000.
Any of you reading this could end up being missed by the census for any of a number of reasons. But being in the statistics business, the Census Bureau and its local partners are focusing on, in official parlance, "difficult to enumerate" populations, or DTEs. "People don't respond for a variety of reasons," says city of Austin demographer Ryan Robinson, "but most of them are cultural." He offers a scenario: five men, Mexican nationals, living together in an apartment off North Loop, working in the construction trades, saving money, and expecting to return to Mexico. "And then you get one questionnaire, probably in English, from the U.S. government. What is your motivation to respond?"
The Bureau and its local partners identify which census tracts -- the basically neighborhood-sized units most often used for census-data analysis -- have the most DTEs, using formulas that combine income, ethnicity and language, and home ownership, along with the tract's 1990 mail-back response rate. In Austin, the designated DTE tracts are where you'd expect them to be -- east, southeast, around campus, and so on (see map at right).
"These are areas where we need to really get the community involved," says Leslie Lawson, the Census Bureau's Austin-area "partnership specialist" in charge of coordinating local complete-count efforts. "We need to get the message out, and it may be a neighbor-to-neighbor message -- your pastor telling the congregation, or the lady who takes care of your kids telling you that this form is going to come."
To that end, Lawson (and her peers in the nation's 520 local census offices) are being helped by local "complete count committees," consisting of civic leaders or, in Austin's case, civic activists. (Travis County, and surrounding cities like Lago Vista, have their own committees.) Austin's committee is chaired by Cherrywood Neighborhood Association president, Mueller Neighborhoods Coalition leader, and all-around man-on-the-scene Jim Walker. "If we don't get people to fulfill their civic responsibility to be counted, I'm not sure what the point of the rest of our organization and activism is," Walker says. "If you complain there isn't enough money for what you want, if you don't participate in the census, you're part of the problem. So I sought this one out."
Along with Walker and the equally ubiquitous Will Bozeman (president of the Austin Neighborhoods Council), the nine-member committee includes reps from, and liaisons to, Austin's minority communities, including Aida Douglas, a Capital Metro marketing specialist working to reach Hispanic populations. "We're mostly looking at grassroots efforts," she says. "We're working with agencies like El Buen Samaritano, Caritas, Meals on Wheels, and we also want to get some businesses involved. And we want to have forums in areas where Hispanics are predominant, addressing reasonable fears that they have -- like whether this information is confidential."
It is -- federal law prevents the Census Bureau from reporting individual information to anyone, including other government agencies (such as the Immigration and Naturalization Service), and census workers who leak private data face five years in prison and a $5,000 fine. Nor will the Bureau report data that might itself compromise people's privacy -- for example, if a census tract or (smaller) block group contains only one African-American family.
"It's pretty amazing that in the 200 years of the census, we've never once had a breach of data security," says Lawson. "The government takes it very seriously, because this is the cornerstone of democracy. When you lose the trust of people in the reliability of our representative form of government -- which is in turn ensured by the census -- that's bigger than all of us."
But it's hard to get that message out, the complete-count advocates say. "I think in general, people [in the Hispanic community], especially people who are newly arrived, have a big distrust of government, and they tend not to respond, which is why the undercount is so big," says Douglas. "But if we can explain that this benefits the community where they reside, that can have an impact. Their children are in school, they're using health clinics and other services, and the more people respond, the more money will be available for them and their children."
It Takes Eight Minutes
The Bureau's goal is to get most of us to respond to the census by mail -- forms will be arriving at our households in March, and would ideally be back to the Bureau by April 1, the official "snapshot date" for census data. (Children born after April 1, for example, would not be included in the 2000 census, and if you respond late, you're supposed to answer the questions as you would have on April 1.) But even in the best of circumstances, the Bureau only expects to get a 65% mail-back response rate -- even though the form, they want you to know, takes only eight minutes or so to fill out, unless you're one of the one-in-six households that gets the long form. (More on that later.)
That's where the door-to-door census taker comes in -- to follow up with people who don't or can't respond, for whatever reason, via the mail. This is quite a logistical effort; even in a midsize city such as Austin, it takes 1,500 full-time enumerators to count that missing third of the populace. (Since many census jobs are part-time, the bureau will hire more than 1,500 people.) It takes 50 people, for example, simply to do "update and leave" -- hand-delivering forms to people who don't have mailboxes. All the more reason to mail in your form, says Lawson: "It costs the taxpayers $25 million to count 1% of the U.S. population" that doesn't return its forms, "and I can think of much better uses of my tax dollars. Plus, if you return the form, a census worker won't knock on your door during dinner."
And what if you don't have a door? Counting the homeless involves a three-night blitz that sends census workers to soup kitchens, shelters, and known encampments, using a special short census form. (Many of the questions asked on the standard forms are about housing.) "Hopefully, these enumerators will be hired from the service-provider community," says Lawson. "We want people who are known by the folks being counted, who know the area, and who will be trusted and welcomed in. That's our goal."
If that describes you, or if you're looking for short-term work, the Bureau is now hiring enumerators (the jobs run from late March through June, or as long as it takes to count everyone). It's also soliciting volunteers to work in the assistance centers -- which will be all over the place, including at every HEB and all Austin Public Library locations -- where people can go to get help filling out their forms. "We especially need bilingual volunteers, to explain how to answer and why the census is so important," says Lawson. "It's a very short-term opportunity to make a big difference in the community." (Call 477-7183x113, if you want to volunteer; the census job line is 477-1470.)
Bilingual, of course, typically means Spanish, the language used by four out of five non-English-only Austin households. The form itself is available in Spanish, Korean, Vietnamese, Chinese, and Tagalog, but to get a non-English form you have to respond to a letter that will come before the census (which is, natch, written in English). If you or yours speak a less common language, you can get a guide at the assistance centers, or talk to an operator on the Census Bureau telephone hotline, in more than 40 different languages.
Some Will Get Long Form
With all of this effort to combat undercounting -- much of it being tried for the first time in 2000 -- it's tempting to wonder how valuable the census actually is as a data source. Perish the thought, says Ryan Robinson. "We tend to slam the census and concentrate on the undercount, but it's the best census in the world," he says, "even compared to the former Communist countries with their central planning. It's incredibly rich data for which there's no substitute."
It would be even better data, Robinson and other demographers feel, if the Census Bureau were allowed to do what every other statistical operation does and use sampling techniques to collect hard-to-count numbers. Indeed, the Bureau already uses sampling -- the aforementioned long form, which only one in six of you will get, is where the bureau collects much of what Robinson calls "the juicy stuff," things of current import to Austin like how long it takes you to get to work, or how much of your income goes to housing costs.
But as you no doubt have heard, the Census Bureau's effort to use sampling to count DTEs in 2000 was firmly stymied by congressional Republicans and has floundered in the courts. The GOP "would not hear of scientific sampling, even though it would have improved the count and was endorsed by Republican experts [from] the Reagan and Bush administrations," says U.S. Rep. Lloyd Doggett. "Rejecting the best science is hardly new for this Congress; it's just a matter of politics. Republicans like Tom DeLay and Dick Armey feared a census that was too accurate; to them, that just meant more Democrats and more people eligible for federal aid to education and child care."
It's ironic, if not entirely accidental, that the people to whom much of that federal aid is targeted are the people who most often don't get counted. But federal formula grant funding -- that which is dependent on census data for allocation -- covers a wide range of programs and, according to the General Accounting Office, nearly $167 billion of the federal budget (see "Federal Money at Stake," above, for a complete list). Not all of these programs are used only by the poor and disenfranchised.
"It's just a question of money," Doggett continues. "The undercount in 1990 has cost Travis County about $40 million and the state of Texas almost $1 billion. Another undercount means that federal resources that our community deserves will be directed elsewhere. We could lose millions that could help us address traffic congestion, public schools, health care, and workforce development."
At the local level, you can find that money in various places, including Austin ISD, the city/county health and human services department, and the city's Neighborhood Housing and Community Development office, which is almost entirely supported by federal funds. Those funds "aren't just apportioned on population, but on formulas that include lots of statistics, all of which come from the census," says NHCD director and city housing officer Paul Hilgers. "The way they estimate living conditions is all found in the census.
"Part of me thinks the census is one of the most serious things you do as a citizen," Hilgers continues. "It is your constitutional obligation, a civic function, and I can get pretty maudlin about how important it is to take part in the process. But practically speaking, we're talking about money for the people in Austin." And probably a lot more money, because as you may have heard, our city has grown quite a bit since 1990. Back then, as you may remember from the city limit signs, our population was 465,622 souls, with the MSA ("metropolitan statistical area") clocking in at 781,572. Since then, metro Austin -- Travis, Williamson, Hays, Bastrop, and Caldwell counties -- has added 400,000 or so people, with about 150,000 of them within the Austin city limits.
Those raw growth numbers, large as they are, come as no real surprise, because the Census Bureau and others regularly estimate population at intervals during the decade. You may recall front-page stories in the daily about the metro area topping one million people, or about Williamson County being the second-fastest-growing in America, or about the city surpassing Seattle in size; those were based on such estimates.
"And all of those estimates will vaporize when the 2000 data hits," says Robinson. "Right or wrong, accurate or not, actual decennial census data is the oracle" for making policy and funding decisions. One of the first of those decisions, in 2001, will be the redrawing of congressional and legislative districts, and the potential drawing of Austin's single-member council districts, if they're approved by voters in November.
Besides, most of the variables counted by the census are not officially estimated (at least by the bureau) between years ending in zero. So Austinites frankly have no idea who the people are who live here, since data on variables such as age, ethnicity, income, employment, and housing are all 10 years old. "It's finally going to be a look at the new Austin -- were we right or wrong? Is the area changing where we thought it was? You don't really know, just as we were surprised back in the early 1990s when we got that data."
For example, Robinson continues, "we saw depopulation in the urban core in 1990, which we didn't expect, but we think the opposite will be true in 2000. We were surprised in 1990 to see ethnic changes -- that not just Southeast but South Central Austin, especially south of Ben White, had become predominantly Hispanic. We'll see more of that, as well as a significant number of Asian households in North Central Austin that weren't there in 1990."
Another variable that will be closely watched is income -- in 1990, the median household income in Austin was just under $28,000, and that number will doubtlessly increase, even though that tide of money is clearly not floating all boats. "One of the things I can't wait to see is how much more money there is in Austin now than there was in 1990, and how much bigger the homes are for the people who have that money," says Hilgers. "But my assumption is that income, both here and nationally, has gone up dramatically, while the level of poverty has not gone down and may have expanded. Which is an indication of how much we need programs like ours, to help people who have been left behind. The census validates the need for public involvement in human opportunities."
And where do we find those needs -- or no longer find them -- in the new Austin? We espouse as our creed that Austin is a city of neighborhoods, and we're acutely aware that, in the last decade, some of those neighborhoods have gotten better, others worse. The maps on page 26 show the relative position of Austin's census tracts (there were 141 in the county in 1990; tracts do not respect municipal boundaries) on variables such as income, ethnicity, and home ownership -- in other words, which neighborhoods were in 1990 richer, or whiter, or less transient than others.
Expect the same maps to look quite different when the 2000 data rolls in. "You're going to see new areas with new needs," says Robinson, "like the south central spine, particularly south of Ben White -- north of Ben White, the problems have been effectively priced out." Conversely, you'll see neighborhoods that used to be solidly average, like Crestview, join the hip 'n' chic urban core, which is itself a lot different than it was in 1990, when slacker renters still dominated. And the 2000 data will be far better equipped than, say, tax appraisals to tell us whether East Austin has actually gentrified.
But an even bigger story will be the transformation of the metro area outside of Austin. "I think you'll definitely see the regionalization of the metro area," says Hilgers, and that will also likely include the diversification -- in ethnicity and income -- of the suburbs. "You'll start to see a more urban mix in places like Pflugerville, Round Rock, and rural Travis County," says Robinson.
Now, even with the complete-count efforts and the high expectations, it's impossible to count everyone and capture everything in a decennial census. "Some of these trends in Austin are really recent, and while the census is a snapshot, it does reflect 10 years of massing," says Robinson. "So if something has just started to change dramatically, it won't look quite as dramatic in a 10-year time frame. If you could do a census in 2003, you could really see it."
But an accurate 2000 census will go a long way toward giving Austin the information it needs to decide what it wants to do when it grows up into a real urban region. "It has so many practical implications," says Jim Walker. "If we don't have an accurate sense of who is where, we'll be bound with that for 10 years. And there is power in being able to get service support when you have an accurate count. There is power, literally, in numbers."
Got something to say on the subject? Send a letter to the editor.