Wanted! Statisticians in the Federal Statistical System

Tommy Wright is chief of the U.S. Census Bureau’s Center for Statistical Research and Methodology. He earned his master’s and PhD degrees in statistics from The Ohio State University, his master’s in mathematics from the University of Tennessee, and his bachelor’s in mathematics from Knoxville College.

Roderick Little is the Richard D. Remington Collegiate Professor in the biostatistics department at the University of Michigan and associate director for research and methodology and chief scientist at the U.S. Census Bureau. He earned his bachelor’s in mathematics at Cambridge University and his master’s and PhD in statistics and operational research at Imperial College, London.

How many “principal statistical agencies” are there in the U.S. government? There are a lucky 13, actually, and they all hire statisticians!

Students with undergraduate or graduate (MS or PhD) degrees in statistics or closely related areas are not always aware of job opportunities in this sector (federal job openings can be found in Amstat News or on job sites such as USAJOBS.gov). U.S. citizenship is a requirement for many, but not all, federal statistics jobs; check individual postings for requirements.

Our own experience concerns the U.S. Census Bureau, the largest of the federal statistical agencies, though we emphasize that opportunities exist at other statistical agencies, too. We’d like to expose some myths about statistics in federal agencies, present some pros and cons of the federal statistical system as a potential source for rewarding job experiences and careers, and comment about current federal statistical topics that make the government sector a vibrant place to work.


• Government statistics jobs involve tedious number-crunching. In fact, problems in statistical agencies are complex and intellectually challenging. For one recent example of an interesting modeling task, see the Census Challenge competition for the Decennial Census. Even better, enter the competition and win a valuable prize!

• Federal agencies only hire sampling statisticians. Actually, there is a distinct trend toward increased use of modeling in statistical agencies, and people with strong statistical modeling skills, as well as statistical computing skills, are in high demand.

• Federal statistical jobs are on the decline, under the pressure of current federal budget deficits. It is true that federal budgets are tight, but there continue to be openings for qualified statisticians. Moreover, new challenges arising from measuring an increasingly complex and dynamic society and recent and continuing waves of retirements are two factors that ensure opportunities for individuals with strong statistical skills.

• Federal salaries are not competitive. Salaries for mathematical statisticians at the Census Bureau are not always at the level of industry and top academic jobs, but they are competitive and higher than you might think. In addition to offering recruitment bonuses to new graduates and new hires, we can usually match a competing written salary offer.


• You get to work on important problems that have an impact on society. High-quality information is vital for government and industry. For example, the Decennial Census is a cornerstone of our democracy, the basis for allocation of representatives to Congress, redistricting, and allocation of large sums of money for federal programs. Statistics based on federal statistical agency data such as the monthly unemployment rate, consumer price index, and crime and health statistics are at the center of the political debate and closely watched.

• You work in an interdisciplinary environment with smart colleagues, where your expertise is valued and important. There are also offer opportunities to interact with researchers at universities and other research institutions.

• In general, work hours are less onerous than in some other environments. There are times when you may need to work extra hours to complete a project within a deadline, but you’ll also have some free time.

• Job security is generally better than in industry and academia, after a relatively short probationary period. Additional current benefits include flexible work schedules, telework, mentoring programs, campus-like environment, fitness center, mass transit subsidy and access to Metro, thrift savings plan, subsidized health and life insurance, and free parking.

• Most federal jobs are in the Washington, DC, area, a vibrant city with diverse cultural, educational, sporting, and entertainment options.

• There is an extensive job market for federal statisticians, and hence many opportunities for advancement for strong performers.

Federal Statistical System
Bureau of Economic Analysis (Commerce Department)
Bureau of Justice Statistics (Justice Department)
Bureau of Labor Statistics (Labor Department)
Bureau of Transportation Statistics (Department of Transportation)
U.S. Census Bureau (Commerce Department)
Economic Research Service (Department of Agriculture)
Energy Information Administration (Department of Energy)
National Agricultural Statistics Service (Department of Agriculture)
National Center for Education Statistics (Department of Education)
National Center for Health Statistics (Department of Health and Human Services)
National Center for Science and Engineering Statistics (National Science Foundation)
Office of Research, Evaluation, and Statistics (Social Security Administration)
Statistics of Income Division (Treasury Department)


We may be biased, but the only con that comes to our minds is … ????

Statistical agencies face challenging problems in meeting their missions in the modern world. The mission of government statistical agencies is to produce relevant, timely, and credible statistics about key social and economic phenomena. Statistical agencies face increased demand for data products, and, as we noted earlier, the questions asked by our society are becoming increasingly complex and hard to measure.

On the other hand, individuals and organizations are less willing to respond to requests for information, voluntary or not. Sample surveys and censuses have increasing costs and are increasingly challenging to mount. Combining information from a variety of data sources is attractive in principle, but difficult in practice. Disseminating information for small geographic areas or small subpopulations is subject to dangers from disclosure of confidential information from respondents. In short, the standard statistical approach of taking a random sample of the target population and weighting the results up to the population has limitations and additional statistical tools are needed.

The United States just conducted a census of people and housing, and the evidence suggests it was successful. However, the traditional design of the census is increasingly being questioned, because of difficulties in finding and interacting with mobile populations, nontraditional families, rapidly changing dwelling structures, increased use of the Internet over “snail mail,” the increasing expense of personal interviews, and the difficulty of including people who are hard to reach or reluctant to participate. New technology offers new data collection tools, and administrative records offer some promise, but how should they be used? How can the quality of census results be measured?

The world of survey methods also is changing. New technology offers new data-collection tools, but deploying them in an optimal way is tricky. The traditional survey should be increasingly seen as one of an array of data sources, including administrative records and other information gleaned from cyberspace. Tying this information together to yield cost-effective and reliable estimates is not simple and requires modern statistical analysis tools.

Meeting these challenges requires continuous innovation, which requires scientific methods, including controlled comparisons of alternative approaches. In short, it’s an exciting time to be a statistician in federal statistics, and people entering the field have the opportunity to advance this crucial statistical endeavor and make a difference.