Carnegie Mellon University
Skip navigation and jump directly to page content

Aug. 26: Save The Census: Carnegie Mellon's Stephen E. Fienberg Warns That Demographic Data Provided Is Too Valuable To Lose

Contact: Shilo Raube / 412-268-6094 / sraube@andrew.cmu.edu

Save The Census: Carnegie Mellon’s Stephen E. Fienberg Warns
That Demographic Data Provided Is Too Valuable To Lose


Stephen FienbergPITTSBURGH—While concerns about privacy, cost and response rates have placed scrutiny on census taking around the world, Carnegie Mellon University’s Stephen E. Fienberg and Columbia University’s Kenneth Prewitt argue that reliable and detailed demographic data, used in a variety of ways — from allocating legislation representation and deciding where hospitals are built to distributing government funding and tracking and identifying sources of problems like poverty — are too valuable to lose or not collect as accurately as possible.
     
Both the United Kingdom and Canadian governments are drastically changing their census methods while the U.S. Census Bureau was recently forced to execute an aggressive and controversial communications strategy to confront issues during the 2010 census process. Now the bureau faces calls from members of Congress who propose curtailing the American Community Survey, which has replaced the mandatory long-form census.
     
In their opinion paper, published by Nature, the professors view moves such as the Canadian government changing a mandatory long-form questionnaire sent to 20 percent of the population to a voluntary survey sent to 30 percent, and the U.K.’s consideration of only using records from post-office address lists, drivers license records and health registers as devastating.
      
Fienberg and Prewitt wrote, “Census data provide the gold standard against which all other studies on such issues can be corrected and judged. National statistic agencies must learn to make use of administrative records and digital data, but much work is needed to ensure that this approach will produce data that are equivalent or superior to those produced by current methods.”
     
Censuses in the present questionnaire, mail-out form are not working, especially like they did in the past. The cost of mandatory surveys keeps rising while response rates to any types of surveys in general are decreasing. Privacy concerns also have become a hot topic, which causes even more difficulties to ensure high response rates.
     
Countries including Germany, Sweden, the Netherlands and Denmark have already made the switch from traditional censuses to using data from administrative records. However, those methods are often incomplete, inaccurate and out of date.
     
“Taking short-cuts and using unreliable data sources leaves countries without vital information that they need,” said Fienberg, the Maurice Falk University Professor of Statistics and Social Science and leading U.S. census expert. “Yes, censuses in their current form face challenges, but sound, proven statistical methods are pertinent to providing governments at every level with the best population information possible. It can be done — the 2010 U.S. census faced problems but, with a lot of hard work, was able to match 2000’s response rates and come in $1.6 billion under budget. That being said, the U.S., like many other countries, still has a long way to go to make the process better.”
     
Technology offers ways to count people in ways never before possible; digital footprints are left every time a person swipes a bank card, surfs the Internet, boards an airplane or uses a GPS. “Privacy issues obviously need to be worked out using these sources, and ultimately governments will need to determine if these sources can provide a cheaper, quicker stream of data about a nation’s population,” Fienberg said.
     
Fienberg and Prewitt realize the immense challenge but contend that “scientists need to appreciate that a vital scientific resource is on the line.”

###

Pictured above is U.S. census expert Stephen Fienberg, the Maurice Falk University Professor of Statistics and Social Science at Carnegie Mellon.