NIST logo

Statistical Engineering Division

The Statistical Engineering Division (SED), founded in 1946, develops and applies statistical and probabilistic methods and techniques supporting research in measurement science, technology, and the production of standard reference materials.

In particular, SED:
  1. develops and applies best practices for the characterization of measurement uncertainty, in particular to enable the intercomparison of measurements in the context of interlaboratory studies and calibrations;
  2. implements methods and techniques for experimental design, data analysis, statistical modeling and probabilistic inference in computer software;
  3. disseminates such methods and techniques throughout U.S. industry, and the scientific and academic communities at large, by publishing technical and educational materials in print and on-line, by offering training courses and workshops, and by participating in professional conferences.

SED Mission Statement...

The Statistical Engineering Division (SED) of the Information Technology Laboratory (ITL) of the National Institute of Standards and Technology (NIST) conducts fundamental and applied statistical research on problems in metrology and collaborates on research in other Divisions of ITL, in other Laboratories of NIST and with NIST's industrial partners.

The role of SED extends across NIST; SED staff actively collaborate with more than 90% of the scientific Divisions at NIST, both on the Gaithersburg and Boulder campuses, and provide statistical support for some of the administrative offices as well. Basic collaborations include core support to provide a statistical basis for certification for Standard Reference Materials produced at NIST, statistical methodology and documentation for NIST calibration services, and education and training of NIST scientists and engineers in the implementation of appropriate statistical methodology.

As members of multidisciplinary teams, SED staff collaborate more fundamentally to scientific research at NIST to define research objectives, to formulate statistical strategies and develop statistical methods for process characterization and to analyze experimental data. The statistical expertise central to a particular multidisciplinary research project may lie in any of many sub disciplines of statistics (including experimental design, generalized linear models, stochastic models, Bayesian inference, time series analysis, reliability analysis, statistical signal processing, image analysis, spatial statistics, quality control, exploratory data analysis, statistical computation and graphics, etc.). However, the SED objective is always to strengthen the fundamental research design and to implement the most powerful statistical tools for drawing inferences and for estimating uncertainties. Success in these collaborations is largely due to the deep involvement of SED staff with the science itself via their scientist colleagues.

SED also contributes internationally to metrology through the development of statistical methodology and statistical tools for metrologists to use worldwide. Increasing attention to international intercomparisons and international acceptance of standards from national metrology laboratories has increased the prominence of statistics in metrology.

Problems unique to metrology as well as problems unique to NIST that require extraordinarily high precision in their formulation often necessitate innovations in statistical methodology. This fundamental statistical research at SED, whether in probabilistic modeling, in design of experiments, in theory and methodology of inference, in computationally intensive statistical tools or in Bayesian inference and modeling, expands the statistical methodology available to NIST scientists, to US industry and to metrologists worldwide. This research also contributes in a fundamental way to the discipline of statistics.

The educational role of SED within NIST extends to development and presentation of short courses and workshops on topics in statistical methodology. These are designed to equip scientists with sufficient understanding of basic statistical methodology to be competent data analysts for standard experiments and to be astute customers when specialized statistical methodology is needed. Increasing attention is being given to making statistical tools available via both internal NIST and external NIST web pages. A highly successful program of statistical research opportunities gives undergraduate and graduate students the chance to explore the sub discipline of statistical metrology while contributing to SED activities.

The professional staff comprises two Groups of mathematical statisticians with graduate degrees. One of the Groups is located in Gaithersburg, Maryland, and the other is in Boulder, Colorado. Also integral to SED activities are Visiting Faculty from several universities.

SED Publications


  • Will Guthrie and Antonio Possolo receive a 2011 Achievement Award from U.S. Secretary of Energy Steven Chu: In recognition of your contributions to the Flow Rate Technical Group / Nodal Analysis Team's swift and effective response to the Deepwater Horizon oil spill [...] helping to speed the ultimate solution and reduce the environmental cost of the disaster.
  • Antonio Possolo receives a 2010 U.S. Geological Survey Director's Award for exemplary service to the Nation in relation with the Deepwater Horizon oil spill: your answers and insights helped guide important decisions and made a very real and positive difference during the response to this unprecedented oil spill event.
  • The Statistical Engineering Division has a leading role in a newly funded, 2010 Innovations in Measurement Science (IMS) project on Shape Metrology. This is an interdisciplinary project involving other ITL divisions, as well as divisions from EL, MML, and PML.
Information Technology Laboratory (ITL)

Statistical Engineering Division (SED)

Antonio Possolo, Chief

100 Bureau Drive, M/S 8980
Gaithersburg, MD 20899-8980

301-975-2853 Telephone
301-975-3144 Facsimile

If you have any questions regarding this website, or notice any problems or inaccurate information, please contact the webmaster by sending e-mail to: