U.S. making doubly sure Census isn't overcounted
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
April 30, 2024, 02:46:49 PM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  General Politics
  U.S. General Discussion (Moderators: The Dowager Mod, Chancellor Tanterterg)
  U.S. making doubly sure Census isn't overcounted
« previous next »
Pages: [1]
Author Topic: U.S. making doubly sure Census isn't overcounted  (Read 504 times)
Tender Branson
Mark Warner 08
Atlas Institution
*****
Posts: 58,156
Austria


Political Matrix
E: -6.06, S: -4.84

Show only this user's posts in this thread
« on: July 27, 2009, 07:22:04 AM »

By Haya El Nasser and Paul Overberg, USA TODAY

During the 2000 Census, Kenneth Prewitt's mother proudly informed him that she had performed her civic duty: She filled out her Census form — twice.

That's not what Prewitt, director of the Census Bureau at the time, wanted to hear.

Although most of the Census effort to ensure an accurate head count next year focuses on rooting out people the government can't find, double-counting is just as much a problem.

A review of the 2000 Census revealed that for the first time in history, the count was actually too high. It took two years to nail down the scope of the problem: About 10.2 million people were missed, but almost 11.6 million — such as Prewitt's mother — were counted twice.

"My mother had moved April 1 (the day of the official tally)," Prewitt says. "Obviously, she got a form at the new house they moved into that day and at the old house they moved out of that day. She was so proud to have helped me by filling out her forms."

Counting people twice distorts the demographic and geographic profile of the population. Census counts are used to allocate federal funds, apportion seats among the states in the U.S. House of Representatives and redraw political districts.

Errors have an impact

Double-counting a white woman in Iowa who winters in Florida is as big a problem as failing to count a young black man in North Carolina. If the Census does both, it compounds the error and creates an imbalance that ripples from local to national politics.

That's why the Census must:

• Count everyone — once.

• Count everyone in the right place — where they're living on Census day.

The two groups — the double-counted and the uncounted —are often polar opposites:

• Many of the overcounted are much like Prewitt's mother — older white women who own a home, sometimes two. Others include young people who are in college, the military or prison.

• The undercounted are more often African American and men aged 18 to 49 who don't own homes.

Limiting double-counting

"We have made a lot of changes in 2010 in ways that would reduce the errors," says Daniel Weinberg, assistant director for the decennial census.

The biggest is in the Census form itself. Instructions now specifically warn people not to include people in their households who are living separately even if they plan on returning home. "Otherwise, they may be counted twice," the form warns.

Another change is a question that asks if people in the household sometimes live or stay somewhere else and, if so, where. The choices include prison, college, a seasonal home or a nursing home.

"While Mom and Dad are paying a bundle for tuition, they list the kid on their form," Weinberg says. "The kid lives in a dorm ... at UC-Santa Cruz and the parents live in New York City. He should be (listed) in Santa Cruz."

For the first time next year, forms will be scanned to pick up duplicates in birth dates and names. Follow-up calls will be made to clarify some returns, and field workers will double-check some addresses. Nevertheless, Weinberg says that "the media campaign is heavily weighted toward harder-to-count populations."

That includes immigrants — especially those here illegally — the poor, renters and any population suspicious of government. That group may be larger since 9/11 because of crackdowns on immigrants and covert surveillance. Historically, hard-to-count groups are more likely minorities.

"Both undercounts and overcounts undermine the accuracy of the Census and the ability of policymakers to evaluate community needs fully," Wade Henderson, president and CEO of the Leadership Conference on Civil Rights, says in an e-mail.

The bureau is spending millions of dollars to publicize next year's national tally.

In addition to a media campaign in 28 languages, the Census is partnering with groups fluent in 69 languages. Bilingual staff will be at information centers, and 13 million households will get questionnaires in English and Spanish.

The 2010 push could lead to more overcounting, but it's a chance the government will take because weeding out duplicates is easier than finding people who don't come forward, Weinberg says.

The Census Bureau has workers focusing specifically on double-counting, says Angelo Falcσn, president and co-founder of the National Institute for Latino Policy and a member of a Census advisory committee.

http://www.usatoday.com/news/nation/census/2009-07-26-overcount_N.htm
Logged
Queen Mum Inks.LWC
Inks.LWC
Atlas Superstar
*****
Posts: 35,011
United States


Political Matrix
E: 4.65, S: -2.78

P P

Show only this user's posts in this thread
« Reply #1 on: July 27, 2009, 08:37:19 AM »

That's pretty interesting - I never really thought that that would be a problem, but it makes sense.
Logged
Tender Branson
Mark Warner 08
Atlas Institution
*****
Posts: 58,156
Austria


Political Matrix
E: -6.06, S: -4.84

Show only this user's posts in this thread
« Reply #2 on: July 27, 2009, 08:51:05 AM »

That's pretty interesting - I never really thought that that would be a problem, but it makes sense.

I think the overcount will be smaller than in 2000, mostly because the US population in these troubled times isn't moving to other states in such a rate like in 2000 when the economy was in good shape. Many studies show that the rate of movers is down by a considerable amount since 2000, increasing the likelyhood of more accurate counts. But if it happens and people fill out 2 forms, the weeding out is not really complicated. If you have someone who's name is John Apple, born on July 1, 1970 in California and another John Apple, also born on July 1, 1970 in Missouri, that would be a strange coincidence and a followup will be made. Much will also depend on the outreach of the info-campaign (this time in Spanish and other languages too), but as far as I΄ve read the % of mailed-back questionnaires is on the rise in recent Censuses, probably offsetting the post 9-11 surge of paranoid, government-hating Americans, who'll refuse to fill out the form.
Logged
Pages: [1]  
« previous next »
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.223 seconds with 11 queries.