Science magazine

Why Statistics? POPULAR MEDIA AND SCIENCE PUBLICATIONS SOUND THE DRUM: “BIG DATA” WILL DRIVE OUR Marie Davidian is future, from translating genomic information into new therapies, to harnessing the Web to William Neal Reynolds untangle complex social interactions, to detecting infectious disease outbreaks. Statistics is the Professor of Statistics science of learning from data, and of measuring, controlling, and communicating uncertainty; at North Carolina State and it thereby provides the navigation essential for controlling the course of scientifi c and soci- etal advances. This fi eld will become ever more critical as academia, businesses, and govern- ments rely increasingly on data-driven decisions, expanding the demand for statistics expertise. Statistical Association, The melding of science and statistics has often propelled major breakthroughs. Last year’s and executive editor Nobel Prize in Physics was awarded for the discovery of the accelerating expansion of the uni- verse. That discovery was facilitated by sophisticated statistical methods, establishing that the Biometrics. E-mail: fi nding was not an artifact of imprecise measurement or miscalculations. Statistical methods davidian@ncsu.edu.
also allowed the trial demonstrating that zidovudine reduces the risk of HIV transmission from infected pregnant women to their infants to be stopped early, benefi ting countless children. Statistical principles have been the foundation for fi eld trials that have improved agricultural qual- ity and for the randomized clinical trial, the gold standard for compar- ing treatments and the backbone of the drug regulatory system. Statistics often informs policy development. For example, in the United States, billions of dollars are allocated to school districts based on county-specifi c estimates of income and poverty, derived by com- bining data using statistical methods. In evaluating pollutants, statisti- cal modeling isolates true associations with illnesses and deaths. Big Data payoffs can be enormous, but there are many pitfalls. Take the promise of personalized medicine: Achieving this goal will require the integration of vast landscapes of genomic, clinical, and related data from legions of patients. The potential for false discovery looms large. New statistical methods will be needed to address some of these issues. Similar challenges arise Thomas A. Louis is from the haystacks of information on social network, time-use, economic, and other activities that can be mined to benefi t science, business, and society. Close collaboration with statisticians at the Johns Hopkins is the best way to ensure that critical issues are identifi ed and solutions found. A dramatic increase in the number of statisticians is required to fi ll the nation’s needs for expertise in data science. A 2011 report by a private consulting fi rm projected a necessary increase of nearly 200,000 professionals (a 50% increase) by 2018.* Graduates specializing in Statistics Section (U), statistics are equipped with skills that allow them to pursue diverse careers, and there has been a surge in applications for graduate education in these fi elds. But available places are limited; for example, the ratio of qualifi ed applicants to slots in Society, and former Online the Johns Hopkins Biostatistics program exceeds 10 to 1. Resources must sciencemag.org be found to expand the number and size of graduate programs.
No amount of statistical intervention can circumvent fl awed subject- http://scim.ag/ed_6077 matter models or salvage valid conclusions from poorly designed stud- ies, and even sound statistical analysis may fail to yield straightforward answers. The future demands that scientists, policy-makers, and the public be able to interpret increasingly complex information and recognize both the benefi ts and pitfalls of statistical anal- ysis. It is a good sign that the new U.S. Common Core K-12 Mathematics Standards intro- duce statistics as a key component in precollege education, requiring that students be skilled in describing data, developing statistical models, making inferences, and evaluating the conse- quences of decisions. Embedding statistics in science and society will pave the route to a data- informed future, and statisticians must lead this charge.
*www.mckinsey.com/Insights/MGI/Research/Technology_and_Innovation/Big_Data_The_Next_frontier_for_innovation. 6 APRIL 2012 VOL 336 SCIENCE www.sciencemag.org

Source: http://statistics.rice.edu/uploadedFiles/MStat/WhyStatistics.Science.2012.pdf