Showing posts with label Census sampling vs. headcount. Show all posts
Showing posts with label Census sampling vs. headcount. Show all posts

Friday, May 15, 2009

NY Times: Census nominee shuns sampling as counting method

From the NY Times:

Robert M. Groves, President Obama’s nominee to lead the Census Bureau, told senators at a confirmation hearing on Friday that he would not seek to use statistical sampling in the 2010 census, offering reassurance to Republicans who have long opposed any effort to adjust for inevitable counting errors using mathematical models.

Mr. Groves, in testimony before the Senate Homeland Security and Government Oversight Committee, said that he supported the view of the Secretary of Commerce, Gary Locke, who oversees the Census Bureau, and who said there were no plans to use statistical sampling in the 2010 count.

“I agree fully with Secretary Locke’s testimony that statistical adjustment of the census is eliminated as an option for reapportionment and further that statistical adjustment will not be used for redistricting,” Mr. Groves said in his prepared opening statement. “The 2003 decision of Director Kincannon, consistent with this, assured that no implementation infrastructure for adjustment was put in place for 2010.”

Mr. Groves, 60, is a sociology professor and survey expert at the University of Michigan. He served as associate director of the Census Bureau in the 1990s, when President Clinton battled fiercely with Republican lawmakers in an effort to adopt statistical sampling as a way to correct for counting errors. House Republicans sued Mr. Clinton and won a court victory barring the use of sampling for the reapportionment of Congressional seats or for redistricting.


Many experts believe that sampling would be more accurate than using only an actual count, because millions of people are missed or refuse to be counted. Studies have found those missed are far more likely to be poor and ethnic and racial minority Americans in urban areas. Counting them would likely favor Democrats.

Many Republicans remain concerned that the Obama administration will try to adopt sampling as part of the census count.

Friday, May 8, 2009

Groves discusses sampling, managerial styles during nomination process; pledges to run politics-free Census

From the AP:

Groves pledges to run 2010 census free of politics
By HOPE YEN – 19 hours ago

WASHINGTON (AP) — President Barack Obama's pick to lead the Census Bureau is pledging to stand up to congressional Republicans — and the White House if necessary — and to stick to sound science when leading the high-stakes head count.

If he cannot, he will resign, nominee Robert M. Groves said.

In a 41-page Senate questionnaire obtained Thursday by The Associated Press, Groves defended his push for statistical sampling in the 1990 census to make up for an undercount of millions of mostly minorities, who tend to vote for Democrats — a move that was then decried by the Republican commerce secretary as political tampering. Groves is a veteran survey researcher at the University of Michigan.

While Groves said the use of sampling in 2010 was unlikely given the little time remaining, he would not say whether he would support other measures such as a government halt to immigration raids. "I will work with all agencies of government to assure the best census this country can achieve," Groves wrote when asked if he would seek to scale back enforcement.

On matters of science, Groves was unequivocal. "The White House can have no role," he said. "If the director is perceived to be a pawn of one or another political ideological perspective, the credibility of the statistical system is threatened."

Groves said if he encounters undue partisan interference from the White House or elsewhere that he cannot resist, "I will resign and work outside the system to stop the abuse."

The Senate Homeland Security committee is scheduled to hold Groves' confirmation hearing on Tuesday.

In his questionnaire, Groves also:

_Cast the Census Bureau as woefully outdated, saying it lacks scientific talent due to a recent and upcoming wave of retirements.

_Acknowledged he lacked extensive management experience to run the bureau's sprawling operations but said he was up to the task.

Separately, the Commerce Department said it had hired Kenneth Prewitt, who headed the 2000 census count, as a part-time paid consultant to Secretary Gary Locke on 2010 census issues. He will be part of a team of experts who will present recommendations later this year to the census director.

Prewitt, now a Columbia University professor, served last year as an adviser to Republican Commerce Secretary Carlos Gutierrez. At the time, Prewitt privately urged the Bush administration to push for a change in federal laws to allow for broader use of statistical sampling, or "accept the likelihood of a distributionally inaccurate census that is more undemocratic."

"I appreciate that ... sampling will be viewed as an unrealistic option," Prewitt wrote in the memo of March 17, 2008, provided to the AP, noting the likely political resistance from Republicans who oppose census sampling methods that tend to favor Democrats.

"However, there are times when crisis conditions produce political courage," he said.

Gutierrez never followed up on that recommendation for broader use of sampling.

Groves, 60, a former census associate director from 1990-92, said he joined several officials in recommending the 1990 census be statistically adjusted because the "adjusted estimates were of higher quality than the unadjusted."

The recommendation came amid a fierce political dispute that prompted White House staff under President George H.W. Bush to call advisers to the bureau and express opposition. The Census Bureau was then overruled by Republican Commerce Secretary Robert Mosbacher, who called the proposed adjustment "political tampering."

The Supreme Court later ruled in 1999 that the wording of the federal law barred broad uses of statistical sampling to apportion House seats. Justices, however, indicated that adjustments could be made to the population count when redrawing congressional boundaries and distributing federal money.

In his questionnaire, Groves indicated that he believed adjusted counts could help improve the census but said he had no plans to push sampling in 2010 or beyond.

"I believe the Supreme Court ruling stands as the guidance on this issue," he said.

If confirmed by the Senate, Groves will take over at a critical time. Census officials acknowledge that tens of millions of residents in dense urban areas — about 14 percent of the U.S. population — are at high risk of being missed because of language problems and an economic crisis that has displaced homeowners.

The government is devoting up to $250 million of the $1 billion in stimulus money for outreach, particularly for traditionally hard-to-count minorities.

But Hispanics, blacks and other groups are warning that traditional census outreach will not be enough, citing in particular rising anti-immigration sentiment after the Sept. 11, 2001, terror attacks. A nationwide group of Latino ministers has called for a boycott of the census unless Obama moves forward on his pledge to pass immigration reform.

Obama last month announced the selection of Groves, who has spent decades researching ways to improve survey response rates and designing surveys for agencies from the U.S. Bureau of Justice Statistics to the EPA and National Institutes of Health.

Thursday, April 2, 2009

Breaking news: Obama nominates Groves as Census Director

The AP has an informative piece regarding Obama's pick to head the Census Bureau, which oversees the upcoming decennial headcount of America. Obama's choice - Robert Groves - is a professor who advocates the controversial "sampling" method of counting individuals, as opposed to actual enumeration. Read the story below or click here for more information.

Obama to nominate sampling expert to head census
By HOPE YEN –

WASHINGTON (AP) — President Barack Obama has chosen Robert M. Groves to be the next census director, turning to a professor who has clashed with Republicans over the use of statistical sampling to lead the high-stakes head count.

The White House will announce the selection of Groves on Thursday, a Commerce Department official told The Associated Press. The official demanded anonymity because the individual was not authorized to speak before the announcement.

Groves is a former Census Bureau associate director of statistical design, who served from 1990-92. He has spent decades researching ways to improve survey response rates. If confirmed by the Senate, he will take the helm less than a year before the decennial count, which has been beset by partisan bickering and will be used to apportion House seats and allocate billions in federal dollars.

House Republicans quickly expressed dismay over the selection of Groves, saying Obama's choice raised serious questions about his political intentions.

"This is an incredibly troubling selection that contradicts the administration's assurances that the census process would not be used to advance an ulterior political agenda," said Rep. Darrell Issa, R-Calif., the top Republican on the House Oversight and Government Reform Committee. "Mr. Groves will have every opportunity to address these concerns during the confirmation process."

When he was the bureau's associate director, Groves recommended that the 1990 census be statistically adjusted to make up for an undercount of roughly 5 million people, many of them minorities in dense urban areas who tend to vote for Democrats.

But in a fierce political dispute that prompted White House staff to call advisers to the bureau and express opposition, the Census Bureau was overruled by Republican Commerce Secretary Robert Mosbacher, who called the proposed statistical adjustment "political tampering."

The Supreme Court later ruled in 1999 that the use of statistical sampling cannot be used to apportion House seats, but indicated that adjustments could be made to the population count when redrawing congressional boundaries.

Commerce Secretary Gary Locke has made clear that sampling will not be used for apportionment. He stated during his confirmation hearing that there are no plans to use sampling for redistricting, while indicating that sampling could be used to measure census accuracy or collect a wider range of demographic data.

Census experts have said it would be difficult at this point to make plans for sampling in the 2010 census for congressional redistricting purposes since it is only a year away. It is more likely that Groves could have an impact on statistical adjustment and other decisions as part of long-term planning for census surveys after 2010.

Groves, 60, a professor at the University of Michigan, would take over at a critical time. Census officials acknowledge that tens of millions of residents in dense urban areas — about 14 percent of the U.S. population — are at high risk of being missed due to language problems and a deepening economic crisis due to the financial meltdown that has displaced homeowners.

The government is devoting up to $250 million of the $1 billion in stimulus money for outreach, particularly for traditionally hard-to-count minorities.

But Hispanics, blacks and other groups are warning that traditional census outreach will not be enough, citing in particular rising anti-immigration sentiment after the Sept. 11, 2001 terror attacks.

Arturo Vargas, executive director of the National Association of Latino Elected and Appointed Officials, praised Groves as a well-regarded academic, calling the question of statistical adjustment in the 2010 census a "non-issue" because there are no plans for it.

Rep. William Lacy Clay, D-Mo., who chairs a House subcommittee on the census, said Groves will be a strong and effective manager for the bureau. "I look forward to working closely with him to reduce the undercount of minorities," said Clay, speaking also on behalf of the Congressional Black Caucus.

Republicans have been crying foul after the White House earlier this year indicated that it would take greater control over the census to address minority group concerns about Obama's initial nomination of GOP Sen. Judd Gregg as Commerce secretary.

Gregg later withdrew his nomination, partly citing disagreements over the handling of the census. The White House has since made clear that Locke will make the final decisions regarding the 2010 head count.

Democrats and Republicans for years have disagreed on whether the census should be based on a strict head count or cross-checked against a "statistical adjustment" to include hard-to-track people, particularly minorities, who might have been missed.

Meanwhile, the cost of the 2010 census is estimated to be $15 billion, the most expensive ever, and experts have long said the Census Bureau must do more to reduce a persistent undercount among minorities, as well as to modernize what is basically a paper mail-out operation that has been in place for decades.

Wednesday, February 4, 2009

The Census - sampling vs. headcount

U.S. News and World Report blogger Michael Barone discusses the usage of sampling versus actual headcounts to redistrict legislative districts, as well as gives a brief overview of the Census's political history. It's an interesting piece with an interesting perspective.