Meet a Participant: Kristine Gierz
Mathematics student rethinks data distributions to improve
Kristine Gierz is pursuing a doctoral degree in statistics at Old Dominion University. However, her interest in statistics grew from a passion for the environment and sustainability.
Gierz completed a bachelor’s degree in biology, but decided to veer from the biology path after a three-month environmental conservation internship in Fiji.
“I saw the gap between research and data analysis, and was inspired to pursue graduate studies in the data analysis field,” said Gierz, who focuses on biostatistics. Gierz looked for an opportunity to test her knowledge and gain experience outside of the academic world.
That opportunity came when Gierz applied for the National Science Foundation’s (NSF) Mathematical Sciences Graduate Internship (MSGI) Program.
The NSF MSGI program offers research opportunities for mathematical sciences doctoral students to participate in internships at national laboratories, industries and other facilities. NSF MSGI seeks to provide hands-on experience for the use of mathematics in a nonacademic setting.
Gierz began a research project in data analysis at the National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland.
In almost any type of research, a hypothesis is made, and data are collected and analyzed in pursuit of more knowledge and information. Researchers use statistics to tell if the results of data analysis are meaningful. Statistical error and uncertainty quantification provides a measurable, universal way to ensure that conclusions drawn from research are valid.
However, in order to statistically assess and summarize data, including generating uncertainty estimates, certain assumptions are placed on the data in question. Most methods of data analysis require some type of assumption about the distribution, or the organizational breakdown, of the data. However, data sets don’t always fit into predefined boxes, and analyzing data using those typical assumptions can lead to significantly misleading results.
In an effort to explore a different way to evaluate data, Gierz began a project to redefine how researchers create their statistical uncertainty estimates. The goal was to create a method for uncertainty analysis that made as few assumptions as possible. Instead of using predefined distribution shapes, Gierz attempted to allow the data to form their own reasonable range of potential distributions.
Gierz collaborated with three mentors during her internship: primary mentor Hariharan Iyer, Ph.D., Steven Lund, Ph.D., and William Guthrie, Ph.D., all of whom research in the Statistical Engineering Division at NIST.
During her 10 weeks at NIST, Gierz provided first steps in developing an alternate method of investigating the effect of assumptions regarding data distributions. Her contribution included the creation of a web application that allows users to upload their own data, generate distributions and conduct preliminary analysis. The project is ongoing; if it can be further developed, it could create a modern alternative to analyzing data without relying on a single data distribution model.
Gierz appreciates her time spent at the internship not only for her project and the technical skills she learned, but also for the professional experiences she acquired. Gierz had the opportunity to interact with professionals in her field and gain an understanding of what researching in applied mathematics is like. “This opportunity has been an amazing experience that will boost my professional connections for future collaborations. Overall, I had a wonderful experience that was challenging, but rewarding,” Gierz said.
Gierz returned to Old Dominion University to complete her doctoral degree. Upon graduation, she hopes to begin a career in data analysis and research. Throughout her education, she has never forgotten about her love and passion for the environment, and how it has shaped her path.
The NSF MSGI Program is funded by NSF and administered through the U.S. Department of Energy’s (DOE) Oak Ridge Institute for Science and Education (ORISE). ORISE is managed for DOE by ORAU.