The Politics of Statistics

Opinion

Published: Monday, March 9, 2009 - 11:47

For the record, I’m a registered Independent voter. With that said… Recently, President Obama’s nominee for commerce secretary, Sen. Judd Gregg, unexpectedly withdrew from the position. Was it because of skeletons in his closet: unreported nanny taxes, inappropriate personal lifestyle, questionable business practices? No, the reason evidently had to with a disagreement over who would take control of the Census Bureau—the Commerce Department or the Obama Administration. At the center of the debate, however, was how the census would be conducted. Would it use the traditional actual count method or statistical sampling techniques?

Before explaining the differences between both approaches, I need to share some background information on the census. Every 10 years, our nation conducts the ritual of counting the population. The purpose of the census is more than just reporting it in high-school textbooks. The census determines how many seats each state gets in the House of Representatives and helps to determine where the district lines are drawn within each state. Literally, billions of dollars are at stake because these population-driven financing formulas can determine where federal spending is allocated.

Republicans and Democrats have been at loggerheads for years over how to conduct the census. The traditional count method consisting of mail-back surveys and knocking on doors is cumbersome, expensive, and time consuming. Moreover, almost everyone agrees it fails to count millions of Americans, particularly minorities, immigrants, the poor, and the homeless. Democrats have argued that the solution is to use statistical sampling models to extrapolate figures for the uncounted. Republicans have responded by saying statistical sampling is unreliable and that the Constitution mandates an actual count. 

Where is the truth? As a certified Six Sigma Master Black Belt and certified quality engineer, I use statistics on a daily basis to analyze data to make rational business decisions in a timely and cost-effective manner. Statistics is a branch of mathematics that allows us to make inferences or predictions based on a sample. Statistical models and formulas were developed over the centuries by preeminent mathematicians and statisticians such as Blaise Pascal, Thomas Bayes, Frederick Gauss, Simeon Poisson, William Gosset, Sir Ronald Fisher, and Dr. Edwards Deming. In many industries such as pharmaceutical, chemical, and aerospace, statistics is the accepted course to estimate the probability of events occurring, acceptance/rejection decisions, failure rates, differences among populations, confidence intervals, and sample size estimates. Ever wonder how professional pollsters such as Newsweek or Gallup survey just a little more than 1,000 adult voters to be able to forecast election outcomes with 95-percent confidence and only +/- 3 percent margin of error? It’s through binomial statistical sampling techniques. 

Statistical sampling is accurate and timely, and it’s economical. In light of our ballooning federal deficit and given our country’s need for reliable census data conducted in a timely and cost-effective manner, we should feel comfortable using statistical sampling. If it’s good enough for industry, why not the federal government.

About The Author

Peter J. Sherman

Peter J. Sherman is a managing partner of Riverwood Associates, a lean Six Sigma certification training and consulting firm based in Atlanta. Sherman brings more than 20 years experience in designing and implementing process improvement programs. Sherman is a certified lean Six Sigma Master Black Belt, an ASQ-certified quality engineer, and an APICS-certified supply chain professional. He holds a master’s degree in engineering from MIT and an MBA from Georgia State University.