Amstat News invited Robert Groves, director of the U.S. Census Bureau, to respond to the following questions so readers could learn more about him and the agency he leads. Look for other statistical agency head interviews in past and forthcoming issues.
What have you enjoyed most about being head of the U.S. Census Bureau?
Some of the most enjoyable experiences have nothing to do with statistics. During the data collection period of the 2010 decennial census, I visited many small and large towns throughout the country. I met with grassroots groups that were helping us ensure that each of their members was counted. Many of the groups were new immigrants. I learned that new immigrant groups are spread throughout the country, not just in big cities on the coasts (as was true in prior immigrant waves). This will mean that the socialization of these groups will be part of smaller cities versus the large urban centers. It was fun to see these new groups seeking to achieve the American dream just as earlier groups did.
What do you see as the biggest challenge(s) for the Census Bureau, and have they changed significantly since you started this position?
The costs of statistical data collection for surveys and censuses are increasing faster than inflation; the demand for more timely and smaller area statistics is increasing; the pace of innovation in technologies adaptable to data collection is increasing; more and more administrative data are digitized and more data are being generated by the Internet each day (some of these may be useful for statistics); and the federal government budget for statistics is likely to be flat or declining in the mid-term.
Some of these are daunting hurdles, but others are exciting potential breakthrough solutions. The future most of us feel we are building is one of multiple sources of data (each subject to missing data and measurement error) being used to support one another (e.g., Internet, telephone, face-to-face surveys, administrative records). The multiple sources must be evaluated with statistical modeling—for real-time switching among modes of data collection—and real-time estimation. More timely estimates will be accomplished by active analysis during data collection and statistical stopping rules for data collection. To do this, we need statisticians with strong design and modeling skills.
Describe your top two or three priorities for the Census Bureau.
Increasing innovation in statistical design, data collection, analysis, and dissemination of statistical information through enhancing the statistical, economic, and sociological talents of the staff and forming partnerships with academic researchers through the National Science Foundation – Census Bureau Research Network. Strengthening the breadth of experience of Census Bureau statisticians through assignment to different areas over their careers. Fostering the development of generalized systems for mixed-mode data collection and real-time management and estimation. Enhancing the utility of web-accessible statistical information from the bureau.
What do you see as the role for the broader statistical community in supporting the Census Bureau?
We need the brightest minds in the United States working to solve the statistical design, data collection, and analysis issues the bureau faces. We have undergraduate internship, doctoral dissertation support; postdoctoral fellowships; and visiting faculty opportunities that ASA members can pursue. We have research data centers throughout the country for research uses of data and a research network of nodes at Carnegie-Mellon, Colorado, Cornell, Duke, Michigan, Missouri, Nebraska, and Northwestern. We actively seek research collaborations with the broader statistical community.
What do you see as the biggest accomplishment (to date) of the agency during your tenure?
All my goals for the U.S. Census Bureau are long-term ones; none of us will know for several years whether they will be achieved. Many will be achieved if the career statisticians at the Census Bureau continue to innovate in their methods.
What I would have liked to have asked Directer Groves is why the U.S. Census Bureau advertises employment opportunities for statisticians, then, whenever I submit an application, I receive an initial acknowledgement, then many months pass by, then I receive an email saying the application cycle has ended and inviting me to submit another application for the next application cycle. If the Bureau needs statisticians, why does it treat them as if they are not needed? If it does not need them, why does it advertise that it needs them?