The Census Project Blog

Entries from October 2009

2010: The Long And The Short Of It

October 27, 2009 · 1 Comment

by Terri Ann LowenthalTerri Ann Lowenthal

When the 2000 census came around, my daughter prayed we’d be one of the lucky 17 percent of homes to receive the “long form” – pages and pages of queries (53 in all!) about our income, education, housing conditions, ethnicity and other key socio-economic characteristics. It wasn’t to be; we got only the streamlined shorter form sent to most households, seeking barebones information. (My bi-racial daughter, hopeful that one of my reporter friends would appreciate her views on filling out the much-maligned longer questionnaire, had to settle for mention in a New York Times article about the first census race question to allow more than one response.)

Next spring, the Census Bureau will send Joelle’s off-campus apartment the same 2010 census form it sends to every other address in the country: 10 questions! 10 minutes! Short is the new long! Except …

Each month next year, roughly 250,000 homes – spread across the country – will get a version of that longer form called the American Community Survey. No flimsy excuse for a statistical exercise here! The ACS, which replaced the traditional decennial long form, clocks in at a hefty 69 (person + housing) questions. It’s an ongoing measurement of important national indicators, down to the local level, that drive vital policy and fiscal decisions by government, nonprofits and businesses large and small.

Three million homes will get the ACS next year, along with the universal 2010 census form. What will they think – that someone is playing a cruel joke on them? The first-time intersection of a census and its ghost from decennials past presents a significant public relations challenge for the Census Bureau. The agency recognizes the looming problem; it is modifying the ACS advance letter and mailing envelope to distinguish the two surveys and to highlight the importance of filling out both.

Still, the agency needs to do more to educate 2010 census partner organizations about the looming confusion. In speaking to hundreds of community-based groups working to promote the census nationwide, I am starting to hear more questions about “that form some people are getting already … the one with lots of questions … what’s that all about?” The 2010 census Web site includes concise, clear information about the ACS and its relationship to the census. But I think a little pro-active education for partners, letting them know that some homes will get both surveys and offering pointers for encouraging response to both, will help avert a collapse in ACS response and significant misunderstanding next year.

Joelle can wish upon a star to be in the ACS sample … except she’ll be studying abroad during spring semester and left out of all counts next year. Some people miss all the fun!

Categories: Census Form · Census Outreach Efforts · Partnerships
Tagged: , , , , , , , , , , ,

Jonathan Swift Would Have Been Proud!

October 20, 2009 · Leave a Comment

by Phil SparksPhil Sparks

When I was the associate director in charge of communications at the U.S. Census Bureau in the late 1990s, I was exasperated about the complex and time-consuming approval process regarding any public statement that came out of the Bureau. Obviously, the data and the various demographic reports issued by the Bureau are important. And, the Bureau must carefully consider how it portrays the data reports so as not to take a public position on the policy implications embedded in most of its releases of information. But, the entire process, involving various layers of the Bureau and then the Department of Commerce, which supervises the Bureau, are positively Lilliputian!

By my count, in the late 1990s I needed 14 different sign-offs in order to release a statement about anything concerning the Bureau or its data! The drill was the same every time. My communications team would write a press release or a draft statement within a few hours. I would review and redraft the material within a few more hours. Then the real process began. The release/statement spent about 5-7 days going through regular channels at the Census Bureau. Then, it was sent to the Department of Commerce for further approvals.

In good times, the whole process took 8-10 working days. Now imagine how this process is going to work given the emerging controversies regarding Census 2010. The Census Bureau approval process is an artifact of a much earlier, more deliberative day. But, it’s facing a 24/7 news cycle.

Unfortunately, we already have a glimmer of what it will be like for the Census Bureau when it faces mounting attacks on Census 2010 from both the left and the right. Two weeks ago Senators Vitter and Bennett introduced their amendment to the census budget (see Terri Ann’s Lowenthal’s posts on 10/6, 10/8 and 10/15), which would require an extra question on the 10-question Census 2010 form. The authors designed the extra question so that undocumented residents could be excluded from the congressional apportionment count. And by the way, who would be the winners if this terrible amendment were enacted? Mostly Southern, Republican-leaning states. They would gain congressional seats. The losers? Mostly, Democratic states like California, Illinois and New York.

For the Census Bureau, and its supervisor the Department of Commerce, a response in the form of a talking points memo issued by the Secretary of Commerce was circulated in record time — It took only five days! Meanwhile, the 24/7 news cycle had been grinding on without a response from the Obama administration for days and days.

Now, the fate of the Vitter/Bennett amendment is up in the air. It could pass. The administration lost valuable time and political capital by not responding to this ridiculous proposal much sooner.

But, the larger question is how will the Bureau, and its supervisor the Department of Commerce, respond next year when the going really gets tough? With more than a million census takers and billions of taxpayer dollars on the line to fund Census 2010, problems are sure to arise. The Census Bureau can’t be responding a week late to crisis after crisis!

Categories: Census Outreach Efforts · Politics & Census 2010
Tagged: , , , , , , , , , , , , ,

Census Guilt Trip

October 15, 2009 · Leave a Comment

by Terri Ann LowenthalTerri Ann Lowenthal

I’m feeling guilty today. In my last weekly blog post (Tues., 10/6/09), not anticipating the firestorm to come, I challenged the prudence of a Latino census boycott, the motives behind legislation to add a citizenship question to the census, and the serious consequences of both for census fairness and accuracy.

I specifically wondered how the Census Bureau would determine the number of non-citizens or undocumented residents in each state, in order to exclude them from the congressional apportionment base. I wryly offered a sample question: Check here if you are in the country illegally!

Note to self: Not everyone appreciates sarcasm. A day later, Senators David Vitter (R-LA) and Robert Bennett (R-UT) proposed an amendment that has stalled completion of the Commerce Department’s 2010 funding bill. The sponsors blithely suggested that, only months before the decennial count starts, the Census Bureau could add an “errata sheet” to ask 300+ million people their citizenship and immigration status without disrupting the entire process. Just like that! And to think the Census Bureau spent the entire decade constructing the 2010 census, when they can just print and mail on a dime!

Okay, I’m being churlish. But really, folks, does anyone honestly believe the Census Bureau can include another question at the 11th hour and still start the enumeration on time, as planned? When all instructional, promotional and training materials; all data capture and processing equipment; AN ENTIRE $7 BILLION WORTH OF PREPARATION centers around a ten-question census that doesn’t ask about citizenship and legal status? And once the start is delayed, all other dominoes fall, because the census is pegged to April 1, 2010. Counting college students in their dorms? Not if it’s June 1, or August 1, or however long it would take to get the process back on track. Homeless people seeking shelter from the late March chill in group facilities? Not in the summer. And according to the Census Bureau, Americans are most likely to move during the summer, creating less than optimal conditions for taking an address-based count.

But I digress. Because on the face of it, the proposal is unquestionably unworkable. And Sen. Vitter has now put forward a puzzling alternative: Just add a question on citizenship, which is currently asked in the ongoing American Community Survey. Yes it is, but it’s not a part of the TEN-QUESTION 2010 CENSUS THAT DOESN’T ASK ABOUT CITIZENSHIP. So we’re back to derailing the start of the count (see consequences above). It’s a puzzling position, though, because if the sponsors’ intent is to exclude only undocumented residents from the state population totals used for apportionment, how does a question only on citizenship help accomplish that goal?

I rest my case (for now).

Categories: Census Form · Politics & Census 2010
Tagged: , , , , , , , , , , , , ,

My Worries About Census 2010

October 13, 2009 · 1 Comment

by Martha Farnsworth Riche
The author is a former director of the U.S. Census Bureau.

How people are feeling about government always plays a part in the success of the census—even George Washington noted it when he reported the 1790 census results. If the media is correct in reporting that Americans are more than unusually unhappy with government, fewer people might cooperate with the 2010 census. This will drive up costs and lower the quality of the results, since the Census Bureau is obliged to count everyone.

In 2000, a census-taker came to my door to ask me about people down the block who hadn’t returned their form, or answered the door. That cost the taxpayer extra money, and the information I gave wasn’t helpful. If other neighbors weren’t more helpful, eventually the census computer “imputed” the characteristics of another household in our neighborhood. In short, the Census Bureau will come up with data for every housing unit on its address list, but it can cost more and describe less than if the people who live there participate willingly.

When you add such circumstances to the difficulties already present in counting some neighborhoods or communities, census success becomes less likely. And data for addressing a wide range of concerns—not just voting, the raison d’être of the census, but also jobs, health, education, and so on—become less reliable for small population groups and communities.

There is one tried and true way to address these concerns: with effective partnerships. My mantra in talking to people about the 2000 census plan was, “We don’t want to do it for you; we don’t want to do it to you; we want to do it with you.” The Bureau delivered on this promise by setting up over 90,000 partnerships across the nation.

This wasn’t easy, and it caused a lot of angst at the Bureau. There’s good reason government employees generally avoid involving the public with their work. Not every one observes appropriate boundaries or standards of behavior, and transgressions come back to harass the agency involved. But the Census Bureau recognizes that community groups have local knowledge and credibility that no Federal agency can provide.

If the signals of greater-than-usual public unease are correct, the 2010 census is going to require even more from partners. The Census Bureau will do a good job of taking the census. But at the margin, the ultimate success of the census will depend on partners not just telling their communities that it’s important to answer the census, and to help take the census, but also actively countering those who advocate non-participation for their own purposes, no matter how well intentioned.

Categories: Complete Count · Partnerships
Tagged: , , , , , ,

Let’s NOT Take A Census!

October 8, 2009 · 2 Comments

by Terri Ann LowenthalTerri Ann Lowenthal

Want to throw away the $7+ billion Congress has already spent to research, design, plan and prepare for the 2010 census? Want to postpone the constitutionally required decennial count for the first time in our nation’s history, or put off reapportioning the House of Representatives for only the second time (post-1920 census was the first)?

The Vitter-Bennett amendment to the FY2010 Commerce, Justice, and Science Appropriations bill (H.R. 2847), waiting for a vote in the U.S. Senate today, is the answer to your prayers. Sen. David Vitter (R-LA) and Sen. Robert Bennett (R-UT) wisely proposed adding a question on citizenship and legal status to the 2010 census, in order to facilitate the exclusion of undocumented residents (or all non-citizens … not clear which) from the state population totals used for congressional apportionment.

I won’t rehash the exclusion proposal’s legal infirmities or probable effect on census accuracy (see October 6 blog post), except to reiterate that both Republican and Democratic administrations have concluded unequivocally that the Constitution requires a count of all persons residing in the country. I’ll try to contain my frustration at the fact that CONGRESS requires the Census Bureau to submit the census topics and questions three and two years, respectively, before Census Day, so it can decide if it wants to add or drop anything. (Weren’t these Senators serving in their posts at the time?)

Let’s just look at why the amendment is the height of irresponsible and misinformed legislating: If enacted, the 2010 census would come to a halt. It would be delayed for who knows how long, and with that, apportionment and subsequent redistricting wouldn’t happen either. Why?

Sen. Bennett suggested that the Census Bureau could simply print an addendum. If it were that easy to add a question to the form, Congress wouldn’t have allocated $7+ billion over the past decade to plan the census. Questions must be tested, forms designed carefully in order to elicit the intended information. The Census Bureau would need to re-do all of its informational and promotional materials (10 questions, 10 minutes … no wait, 11 questions!); all of its assistance guides (59 languages!); all of its instructional materials for census staff. It would need to redesign all data-processing software. It would need to reprint 180 million new forms, a process which already took most of a year and most of the nation’s printing capacity. The list goes on and on.

Perhaps the biggest irony in all of this? If apportionment were postponed, UTAH would not get the fourth congressional seat it has coveted since losing it post-2000 census (despite filing two federal lawsuits and piggybacking on the D.C. statehood bill). Surely Sen. Bennett would not want to deprive his state of an additional House member one day more than necessary!

Categories: Census Form · Politics & Census 2010
Tagged: , , , , , , , , , , , , ,

Undocumented Immigrants & the U.S. Census

October 6, 2009 · 3 Comments

by Terri Ann LowenthalTerri Ann Lowenthal

“State GOP hopes to block census from counting illegal immigrants” (New Orleans Times-Picayune, 8/29/09)

“Immigration activists urge census boycott” (Boston Globe, 9/8/09)

What do these headlines have in common? Not hard to figure out, I’m afraid. They describe two independent campaigns – one to exclude non-citizens from the census counts used for apportionment; the other a call for immigrants to boycott the census until Congress enacts comprehensive immigration reform — that are likely to produce the same result: An undercount of immigrants and Latinos in the 2010 census.

A number of lawmakers and other opinion leaders are shocked – SHOCKED! – to “discover” that everyone, regardless of legal status, is counted in the census. In fact, this is a “perennial decennial” issue – one of several concerns that surface every ten years. Long enough, I suppose, for people to forget that both Democratic and Republican Solicitors General have concluded that the U.S. Constitution envisions congressional apportionment based on a count of all people living in the country (with exceptions carved out expressly … more on that in a moment).

No matter – just last month, the latest bill to propose excluding non-citizens from the census surfaced in Congress. Similar proposals would leave out undocumented residents only. Constitutional questions aside, how would we accomplish this policy? Put a question on the census form, according to the new legislation. Maybe something like this?

Mark one. This person is –

  • A U.S. citizen.
  • A legal permanent resident of the U.S.
  • Residing in the U.S. illegally (trust us; we won’t tell anyone).

Congress is not going to stop the 2010 census in its tracks to add a question to the 180 million forms rolling off the presses (after years of research and testing). So why introduce a bill now? With the Census Bureau heavily promoting a lean-and-mean enumeration, hoping to ease fears among immigrants, any talk of including a question on legal status may well cause confusion and dampen participation in these communities.

Enter the boycott effort, which advocates a broad rejection of the census not just by undocumented immigrants but by all Hispanics, unless Congress moves more quickly on immigration reform. The brainchild of a Latino evangelical clergyman, the campaign’s premise – legalize us before you count us – plays to disappointment with the long-term failure of political leaders to address the status and treatment of the undocumented.

I understand the frustration … but not the strategy. I don’t see a good outcome for anyone with a stake in the battle to advance a sensible immigration policy. Congress is not easily swayed to action by threats, and the Administration has already put off immigration reform another year, making the boycott a useless tactic. More alarming, if large numbers of Latinos avoid next year’s count, they will have little leverage in the redistricting process, potentially leaving them with fewer lawmakers who support their cause and represent their interests. A Boston-based boycott leader suggested that an incomplete census would “tak[e] away money and power from the politicians.” No … an undercount would take money and power directly from Latinos and their communities.

Arturo Vargas, executive director of the National Association of Latino Elected and Appointed Officials Educational Fund, has called the boycott campaign “immoral.” I think I understand what he means.

Article I, section 2, of the Constitution, apportioned seats in Congress based on “the whole Number of free Persons” (including indentured servants but excluding Indians not taxed) and three-fifths of all slaves, whose numbers would have boosted the political influence of southern states. We tried to right that wrong with the Fourteenth Amendment; sitting out the 2010 census would be a 234-year step backwards for those seeking an equal voice in our democracy.

Regrettably, it boils down to this: Boycott leaders, if successful, would accomplish for the pro-exclusion advocates what those lawmakers cannot now achieve legislatively — remove large numbers of Latinos from the state population totals used for apportionment. The specter of advocates for comprehensive immigration reform directly aiding those who largely oppose greater rights for immigrants, is hard to comprehend.

It’s time to stop using the census as a vehicle for pressing all sorts of unrelated political causes. Let’s have debates over immigration policy and reform; go ahead and squabble over labor unions and low-income advocacy groups — but stop playing all of these games in the census sandbox!

Categories: Complete Count
Tagged: , , , , , , , , , , ,