Amstat News invited Barry D. Nussbaum, chief statistician at the U.S. Environmental Protection Agency 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 chief statistician at EPA?
In many regards, the role of EPA’s chief statistician is to serve as a cheerleader for implementing the use of proper statistical techniques and the proper interpretation of the resultant statistical inference. We have many analysts scattered throughout EPA’s programs, regions, and laboratories. Establishing an EPA statistics users group to try to provide some cohesion and periodically meeting with these analysts has been my most enjoyable endeavor. It has enabled our folks who are busy on their own individual projects to share problems, techniques, approaches, experiences, and, ultimately, successes.
Incidentally, this is a statistics users group, not a statisticians group. I picked that term very carefully to include those who are in associated fields but perform statistical analyses. Initially, I was frequently met with remarks such as “I am not a real statistician.” To me, this is all the more reason to partake in these discussions. I notice I now get fewer and fewer of that type of response and far more acknowledgements and appreciation for opening up this dialog among users of statistics.
What do you see as your biggest challenge(s), and have they changed significantly since you started this position?
The biggest challenge is communication. I am well aware that sounds like a well-worn, threadbare answer taken from any field, any application, or any endeavor involving more than one person! But it is real. EPA faces many pressures affecting its rulemaking and enforcement activities. These include environmental, legal, economic, scientific, health-based, and data concerns. By properly weaving all these concerns together, EPA has been successful. But that means communicating the supportive role of proper statistical analysis to decisionmakers confronted with all these concerns who sometimes feel their degrees of freedom have gone negative. The EPA statistician cannot afford the luxury of confining him or herself to just the statistical area, but must integrate it all and effectively communicate the statistical ramifications. I am happy to report that this enhanced interactive communication has improved since I started in this position. I am more than willing to acknowledge that I am not EPA’s smartest statistician, but I do rank quite high on the effective communicator tail of the distribution.
Describe your top two or three priorities going forward.
It is no secret that EPA is under increased scrutiny in its regulatory activities. This means its data acquisition, information handling, and resultant inferences also will be under the magnifying glass. Simultaneously, budgets to accomplish these efforts are essentially stagnant. This is both an opportunity and a challenge for statistical analyses in EPA. My first priority is to ensure we effectively use all the possible statistical tools and sampling techniques in a comprehensive, cost-effective manner.
Hand-in-hand with this is a second priority to make sure we take our voluminous stock of administrative data and integrate the information from these sources in a proper, useful manner.
The next priority recognizes that more and more of our efforts involve questionnaires and surveys. Frequently, the allure of the ease of implementing current survey software is so compelling that one neglects the basic tenets of fundamental survey design. I have to see that this is not neglected!
What do you see as the role for the broader statistical community in supporting your work?
Answering this is based on serving as the former chair of the ASA’s Statistics and the Environment Section and my current chairmanship of the ASA’s Statistical Partnerships Among Industry, Academe, and Government (SPAIG) committee. I am routinely jealous that at forums such as JSM, attendance in environmental sessions rarely is “standing room only.” Similarly, I have noted that most of the action, particularly involving the annual SPAIG award, seems to be collaborations between industry and academe, with little government involved. I think governmental statistics in general, and environmental statistics in particular, have lots to offer. I would love to garner additional interest in these areas.
I also think the statistical community could do a better job of highlighting successes that have been achieved by effective statistical work. This would help not just EPA, but all those who practice this trade. I believe the trick is to spotlight the accomplishment in the applied area and then show it was based on the statistical work. After all, if Hal Varian, Google’s chief economist, said statistics is the sexy profession of the decade, let’s take advantage of it.
What do you see as the biggest accomplishment (to date) of you and your colleagues during your tenure?
One of my major goals has been to put EPA on the ‘statistical map.’ EPA, of course, is a regulatory agency and justifiably not primarily acknowledged for its statistical work. Yet, we have done some excellent work and I am seeing that recognition. I was absolutely thrilled that, in 2011, EPA statisticians won both the Jeanne Griffith Mentoring Award (won by Jenise Swall) and the Roger Herriot Award for Innovation in Federal Statistics (won by Mike Messner). This is the first time either award went to an EPA recipient. We are on our way!
Somewhat more seriously, I think I have seen a better integration of EPA data analysts with the program offices and labs. We are doing a better job of merging the statistical analyses with all our other concerns. By keeping our eyes on the prize, I believe we are far more effective in supporting environmental outcomes.
In the 1967 movie “The Graduate,” Dustin Hoffman was given the advice, “Plastics.” What is the one word of advice you give out most as EPA chief statistician?
Stratify.
Leave a Reply