Posted by | May 22, 2012 11:42 | Filed under: Top Stories

by Stuart Shapiro

Last week the House of Representatives eliminated the American Community Survey, an extension of the Census that is used by government and businesses alike.  There is so much wrong with this decision that it would take pages to document it (or you could click through).  I want to focus on one reason given for the elimination given by Rep. Daniel Webster (R-Fl).

“We’re spending $70 per person to fill this out. That’s just not cost effective,” he continued, “especially since in the end this is not a scientific survey. It’s a random survey.”

One of the first things you learn in statistics is that randomness is the gold standard for a survey (I have no idea what he means by a “scientific survey”.  Randomness is the way you assure that your sample is representative of the population you are examining.  I often rant about how poor American understanding of very basic statistics is and the cost of this poor understanding.  Apparently the confusion includes some of our highest officheolders and the cost is awful policy decisions.

By: Stuart Shapiro

Stuart is a professor and the Director of the Public Policy
program at the Bloustein School of Planning and Public Policy at Rutgers
University. He teaches economics and cost-benefit analysis and studies
regulation in the United States at both the federal and state levels.
Prior to coming to Rutgers, Stuart worked for five years at the Office
of Management and Budget in Washington under Presidents Clinton and
George W. Bush.