The Census Project Blog

Entries from November 2009

“Are We There Yet?” Part II: Setting Realistic Goals for Census Response

November 24, 2009 · 2 Comments

by Terri Ann Lowenthal

Last week, I tried to cut through the confusion surrounding census response rates – what they do and don’t tell us about participation and accuracy.  Now let’s dig a little deeper, to help community organizations and elected officials set appropriate goals and realistic expectations during the first phase (mail-out/mail-back) of the 2010 count.

In 2000, the initial mail response rate (based on a universe of all housing units) was 64 percent.  The final response rate, which included late mail returns and telephone and Internet responses, was 67 percent.  Then-Census Director Ken Prewitt challenged state and local governments to boost their response rates by five percent over 1990, and many met that target.

Would a similar competition work in 2010?  Regrettably, the devastating consequences of the recession and foreclosure crisis, and lingering upheaval from Hurricane Katrina and other natural disasters, might make a plus-5 goal unrealistic in many communities.  That’s because high vacancy rates in neighborhoods hit hard by these events will skew mail response calculations.  In 1990, an operational fiasco rattled public confidence early in the process, when millions of questionnaires were deemed “undeliverable” because the Postal Service didn’t recognize addresses where residents had their mail delivered to a post office box, no doubt contributing to a plunge in mail-back rates.  In 2010, civic leaders must anticipate artificially low response rates in distressed communities and focus the public’s attention not on the uphill battle to match or exceed their 2000 benchmark, but on ways participation can bring much-needed government resources and private investment to struggling neighborhoods.

The Census Bureau can devote more energy and resources to coaxing wary residents from behind closed doors if the mail return rate – a truer benchmark of public cooperation because it is based only on occupied housing units – goes up (although we won’t know this figure until the census is history).  The national return rate in 2000 was 78 percent (80 percent for homes that received the short form).  If you’re a glass-half-full kind of person, that means four in five households did their civic duty and mailed back their forms (or phoned-in their answers).  Not surprisingly, though, return rates for harder-to-count households[i] were markedly lower: Hispanic (69%) v. non-Hispanic (79%); Black (64%) v. White (82%); renter (66%) v. owner (85%); younger (for 18-24, 57%) v. older (for 65+, 89%).[ii]

How can we boost the return rate in 2010, especially for less trusting, more fearful, and isolated population groups?  Social science research has shown that people can be persuaded to do the right thing if they see neighbors and friends – people they know and trust — doing it.  Some community-based organizations are planning group events to help people fill out their forms (while respecting the privacy of individual responses); arrangements could be made with local post offices to pick up the completed and sealed questionnaires.

Failure to anticipate the effects of high vacancy rates on initial mail response, stakeholder expectations, and the scope of door-to-door operations, and to explain the problem clearly, could damage public confidence early in the process.  Reasonable goalposts and credible progress reports could go a long way towards keeping enthusiasm and cooperation high next spring.

[i] Response and return rates are based on the characteristics (e.g. race; Hispanic origin; age) of the “householder” – the person listed first on the census form.

 

[ii] All mail return rates taken from National Research Council (2004), “The 2000 Census: Counting Under Adversity,” Panel to Review the 2000 Census.  Constance F. Citro, Daniel L. Cork, and Janet L. Norwood, eds., Committee on National Statistics, Division of Behavioral and Social Sciences and Education.  Washington, DC: The National Academies Press.

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

“Are We There Yet?” – Measuring Census Progress and Accuracy

November 17, 2009 · 2 Comments

by Terri Ann LowenthalTerri Ann Lowenthal

Here’s a brain-teaser: What’s the difference between mail response, final response and mail return rates? And do any of these measures tell us about census accuracy?

How often have you heard a statement like this: “For the 2000 census, only 54% of Queens residents filled out the mail-in survey, well below the national average of 67%” (New York Daily News, 11/10/09). Makes it sound as if half the borough’s population wasn’t counted! (If that were the case, the outcry would have been deafening.)

In fact, 54 percent was a “final response rate.” Here’s the Cliffs Notes version of that and other census yardsticks:

  1. Mail response rates: Percent of all housing units– occupied, vacant, and potentially non-existent — on the master address list that return a census form by mail. Calculated in real time during the mail-out/mail-back operation (mid-March through April) and posted on the 2010 census Web site daily for all governmental units.
  2. Final response rates: Broader than the initial measure, final response rates also include late mail returns, Be Counted forms and telephone responses. (Census 2000 final response rates included Internet response, an option not available in 2010.) Essentially, represents the percent of all housing units on the master address list that respond in some fashion before enumerators start knocking on doors. Reported in late summer 2010.
  3. Mail return rates: A truer measure of public cooperation, the percent of occupied housing units that respond to the census by mail or telephone. Reported in 2011; available by race and Hispanic origin (of the householder).

Let’s stop for a minute. Notice that response/return benchmarks reflect housing units, not people, counted, and that they only tell us about the first phase of the census. There’s still a long way to go, given that census takers have to visit roughly a third of all addresses. So how do we know how accurate the population count was?

Enter the coverage measurement survey (called Census Coverage Measurement for 2010). Highly trained field workers visit a large representative sample of homes after census door-knocking ends, to ask who lived there on Census Day. Responses are matched to the information gathered in the census, to gauge misses and double counts. The results are then projected for similar households (looking at geography, race and ethnicity, gender, age) to generate estimates of undercount (omissions) and overcount (duplicates and other erroneous counts). We’ll see these coverage estimates – for the nation, regions and states – in 2012. By then, we’ll be well on our way to planning the next decennial count.

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

Why Are Young Children Missed So Often in the Census?

November 10, 2009 · Leave a Comment

by Dr. William O’Hare
Senior Fellow, The Annie E. Casey Foundation

Even people who follow the census closely are often surprised to learn that young children are the age group that is missed most often in the census. Data from the Census Bureau’s Demographic Analysis indicate that there was a net undercount of 4 percent for children under age 5 in the 2000 Census… this amounts to about 750,000 young children. This is a higher percentage than young adult males (although young minority males have a higher net undercount rate). The chart below shows the net undercount rates for children year-by-year from 0 to age 17.

Despite that fact that young children have been undercounted for decades in the census, there is little understanding of why. Potential explanations vary from their living in hard-to-count neighborhoods to the fact that the census form only has space for complete demographic information on six household members.

Here are some key findings regarding the undercount of children.

  • Young children are missed at a higher rate than most other groups.
  • Minority children are missed most often.
  • Children are over-represented in hard-to-count neighborhoods.
  • The undercount of children results in reduced funding for needy families.

There are a number of trends which suggest getting an accurate count of young children will be more difficult tin 2010 than it was in 2000:

  • Minority children have higher undercount rates, and the share of children age 0 to 4 who are from a Hispanic or racial minority population rose from 41 percent in 2000 to 47 percent in 2008.
  • More children live in families with one or more undocumented immigrants. Nearly half (47 percent) of unauthorized-immigrant households are couples with children. The number of children (of all ages) with at least one unauthorized immigrant parent increased from 3.9 million in 2003 to 5.5 million in 2008.
  • The housing crisis will cause more families to double up in one housing unit or in other temporary and unusual housing situations. It is likely that at least 2 million children will be impacted by the housing crisis.
Undercount of Children: Percent Net v. Age. A minus sign indicates a net overcount.

Undercount of Children: Percent Net v. Age. A minus sign indicates a net overcount.

What Can Be Done

How can we reach households with young children on a systematic basis? There are several avenues that could be used by the Census Bureau, other federal agencies and child advocates to reach parents of young children with a message about the importance of including children in the census. Some of these are outlined below.

  • Materials promoting participation in the census could be sent to everyone enrolled in the Women, Infants, and Children Program (WIC), which provides nutritional food for 8.2 million pregnant women and infants each month.
  • Materials promoting participation in the census could be sent home with the nearly one million children enrolled in Head Start (and early Head Start) programs.
  • Materials promoting participation in the census could be sent to every family enrolled in childcare facilities.
  • Groups could work with the American Hospital Association to give a written notice (in appropriate language) to all women who have a birth in January, February or March of 2010, that it is important to make sure they record their new child in the census. One would expect about one million births during these three months.
  • Material promoting the census could be sent out to the offices of the 60,000 members of the American Association of Pediatricians, where parents of young children visit often.

Without a concerted effort, it is likely that young children will again have a higher undercount rate than any other age group.

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

“Just Say No!”: Undermining Public Confidence in the Census

November 3, 2009 · 2 Comments

by Terri Ann LowenthalTerri Ann Lowenthal

Dear fellow Americans:

When the nation’s decennial population count starts next spring, be very wary.  That questionnaire in your mailbox?  Government bureaucrats are trying to steal your data DNA.  Be judicious with your information: The number of people in your home — that’s all you must tell them.  (And don’t let Big Brother fool you with that sneaky reference to federal law.)

The Constitution does not say it’s okay to ask anything else.  Well … maybe we could add one more teensy weensy little question: Are you an authentic American citizen?  Okay, two questions (two minutes?): Person tally and citizenship!  But if our fourth-and-long bomb to remake the census fails to scare away enough people who shouldn’t be here anyway, a few of our Hispanic friends have helpfully called on Latinos to sit out the count.  (What? … The same Constitution doesn’t say anything about counting “citizens” or “lawful inhabitants,” only “persons?”  Jeez, that quill-and-ink on parchment paper is hard to decipher… )

Anyway, if you still can’t bring yourself to mail back the form, then you’re likely to get a knock on the door.  Man the barricades!  The Census Bureau is hiring more than a million temporary enumerators, and there are likely to be criminals among them.  Census officials say they are working with the FBI to improve the fingerprinting process.  But until the Census director signs a blood oath promising to weed out all criminal elements, do not unlock the deadbolt.

Okay, so maybe there’s no history of former felons getting their hands on your personal information or threatening your safety during field operations.  Don’t let down your guard: A threat still looms in the form of … labor unions (the AFL-CIO! SEIU!).  They are partnering with the Census Bureau and will take over door-to-door canvassing (as we understand the process), in a naked attempt to grab power for liberals.  (In case you’ve forgotten, you can thank us for ensuring that no one associated with the group ACORN will find you.)

And lest you think your children are safe … did you know the Census Bureau is co-opting their fragile minds in support of big government by teaching how census data helps us make wise decisions?  Better keep the little ones at home when the lesson plan turns to census taking.

Now, let’s be frank: You can save this country a lot of money if you cooperate from the start.  For every one percent of households that mail back completed forms, the cost of the census drops by $90 million.  And let me tell you, this census has a whopper of a price tag!  In fact, we are now suspicious that the Obama Administration … er, we mean the Census Bureau … has wasted billions of your tax dollars on this sorry excuse for a civic exercise.  This is the very same incompetent bureaucracy that has resisted spending millions more dollars to hunt down bona fide fellow citizens who have left our shores and are scattered around the globe doing important (albeit private) work.  Americans should be outraged at the misplaced priorities!

So, fellow countrymen (and women), now that we have awoken from our nine-year slumber, we must be extra vigilant.  We must alert real Americans to the danger lurking six months from now.  We must spend whatever it takes to undo a decade of planning and preparation because (can you believe this!) the Census Bureau has just decided that non-citizens count as much as the rest of us.  (In case you missed it, though, we have pledged our full support and cooperation to help the Census Bureau ensure a successful 2010 census.)

But if you’re not cowering in the cellar yet; if you aren’t ready to toss that “short” (ha!) form indignantly in the trash can; if you haven’t locked your race, ethnicity, and birth date in a safe deposit box… if we haven’t shaken your confidence in the process to the core…  could you please put down that you live in Utah?

Our 45th state has been suffering for a decade from congressional district-envy.  (In fact, Utah is so envious of states with greater political representation that it chose the California seagull as its state bird!  I kid you not.)  We may just have scared so many people away from the count, they’ll need all the help they can get to pick up that fourth House seat demographers have been telling us for nine years they’re certain to get after 2010.  Or, we could just spook the rest of the country into hiding in the attic next spring.

Yours in civic solidarity,

The “Count Me If You Dare” Coalition

#     #     #

Postscript:  If you haven’t yet joined the Coalition of the Unwilling, it’s time to redouble your efforts to highlight the positive aspects of the 2010 count.  No, the census process is not perfect.  Yes, the Census Bureau has made mistakes.  But this hard-working agency is doing its best to pull off a monumental task only months from now, and the constant alarm bells – many exaggerated or misinformed — do little more than play to the worst biases of an already skeptical public.  The naysayers scare the populace at their own political peril.  The rest of us will keep working to encourage pride in the simple civic act of being counted.

Categories: Census Form · General · Politics & Census 2010 · Uncategorized