President Barack Obama is tapping Robert M. Groves, a University of Michigan professor who has pushed the use of statistical sampling, to be the next census director.
A Commerce Department official who demanded anonymity said the White House will make the announcement later Thursday.
Groves is an expert in survey methodology and statistics who served as an associate director of the Census Bureau from 1990 to 1992. He and others recommended that the 1990 census be statistically adjusted to make up for an undercount, only to be overruled by then Republican Commerce Secretary Robert Mosbacher, who called it political tampering.
It is something of an understatement to describe Robert Groves as "an expert in survey methodology." He is one of our nations' most respected survey methodologists and arguably the leading authority on the subject of non-response in surveys. He has served as the president of the American Association for Public Opinion Research (AAPOR), and won many of its awards including the career award for exceptionally distinguished achievement. Interests disclosed: I had the good fortune to study under Groves in classes I took at the University of Maryland's Joint Program in Survey Methodology (JPSM), a program that Groves helped found.
As the first three three paragraphs of the AP make clear, some see nothing but "political tampering" in any reference to "statistical sampling" regarding the census. For those tempted to label Groves as the pawn of partisans in the White House or the Democratic party, I have a warning: The notion of Bob Groves yielding to partisanship is laughable. As in rolling on the floor laughing out loud laughable. Groves is well known and universally respected among survey researchers and Census Bureau professionals alike. He is an ideal choice for this appointment.
I conducted the interview with Groves below, on the topic of non-response, at last year's AAPOR conference. The Bob Groves in this interview is the scientist and professional his students and peers know well:
House Republicans quickly expressed dismay Thursday over the
selection of Groves, saying Obama's choice raised serious questions
about an "ulterior political agenda."
"The fight to protect the
accuracy and independence of the 2010 census has just begun," said Rep.
Patrick McHenry of North Carolina, the top Republican on a House
subcommittee overseeing the census. "President Obama has made clear
that he intends to employ the political manipulation of census data for
Also, the Washington Post has more background on Groves and the sampling controversy, including this:
Groves served as the bureau's associate director from 1990 to 1992 and currently is director of the university's Survey Research Center.
He has researched why people participate in statistical surveys, worked
to develop surveys with lower non-response errors and studied how data
is collected for surveys.
A congressional aide familiar with Census matters said Groves has
"bulletproof scientific credentials" and is "really highly regarded by
his peers as a low-key, determined guy who's been really focused on
reducing error in survey research for his whole career."
Follow Mark Blumenthal on Twitter: www.twitter.com/MysteryPollster