Daniela Witten is an Associate Professor of Statistics and Biostatistics at University of Washington. Daniela develops statistical machine learning methods for the analysis of large and complex data sets arising from genomics, neuroscience, and other fields. Daniela has co-authored a textbook, “Introduction to Statistical Learning” which provides an accessible introduction to the field of statistical machine learning, and is currently used to teach statistical learning worldwide. Daniela is passionate about the applicability and necessity of statistics to better understand a wide range of scientific problems. As a woman in data science, Daniela loves the opportunity to be a role model to the next generation of women data scientists.
What is your path in getting to Data Science?
Daniela began her undergraduate studies at Stanford with aspirations to study foreign languages, although eventually ended up double majoring in mathematics and biology. She knew that she wanted to pursue a PhD but felt unsure about what subject she felt passionate enough to pursue for the rest of her life. “I also knew that I wanted to develop a skill set that would allow me to answer a broad range of scientific questions over the course of my career.” Daniela decided to pursue a co-terminal degree at Stanford, wherein she could earn a Masters degree alongside her undergraduate studies. She describes scrolling through the catalog of possible programs, getting to the `S’ section, and stumbling upon ‘Statistics’. She pursued her MS in statistics at Stanford and liked it so much that she stayed to complete her PhD there, as well. When recalling that she initially intended to study foreign languages, Daniela jokes, “in a sense maybe I stuck with the original plan, since to many people, statistics is a foreign language.”
What part of your job makes you most excited?
Daniela loves that, as a professor at UW, she has had the opportunity to work with many smart students. She describes how fun it is to work with such talented students, to get them excited about her research and new questions, to teach them what she knows, and to learn from them in the process. Daniela loves her career and cannot imagine doing anything else. “Being a statistician is a great career choice for a natural skeptic who is also extremely argumentative”, Daniela jokes, and she feels fortunate to have a career that allows her to make fundamental contributions to a broad range of scientific problems.
What are your hobbies?
Daniela loves spending time with her family: her husband and two children, ages 2 and 3. When time allows, she also loves getting together with friends, running, eating her way through Seattle’s amazing restaurant scene, traveling, wakeboarding, hiking, skiing, biking, and enjoying the natural beauty of the Pacific Northwest.
Can you pinpoint one moment or person that was instrumental in your academic decisions?
Daniela credits much of her success as a graduate student, and in her career since then, to the mentorship and support of her PhD advisor, Rob Tibshirani. Dr. Tibshirani is a Professor at Stanford in Statistics and Biomedical Data Science, and with Daniela, Gareth James, and Trevor Hastie, wrote the aforementioned textbook, “Introduction to Statistical Learning.” “He was, and continues to be, and incredible research mentor and friend.”
What advice can you share with aspiring data scientists?
Daniela urges future data scientists to cultivate a strong background in statistics. “Statisticians are the `original’ data scientists!” A solid understanding of statistics is necessary in order to be able to appropriately model data and to differentiate between signal and noise. Daniela recommends taking formal statistics courses at any level, with the encouragement to not get frustrated when the content is difficult. “As with many worthwhile endeavors, it takes a while for the work to pay off.”
What advice can you share with your colleagues?
Daniela worries about that the general public does not understand the importance of statistics. This lack of understanding is a problem in our current era of huge and high-stakes data, when statistics and statisticians are needed more than ever to understand and communicate what this data means. Daniela believes, however, that this public understanding of statistics and its importance need not be overly technical or difficult. Whereas the theoretical details of statistical ideas can become quite complicated, the overall ideas are often simple. Daniela believes that, when we put in the effort to be clear and effective communicators, the vast majority of statistical concepts can be explained to the public. Daniela urges her colleagues to not assume that someone needs to know linear algebra, measure theory, or multivariable calculus in order to have a solid conceptual understanding of statistical ideas. She believes that the spread of statistical ideas to the public will not be achieved by the memorization of dry facts in an introductory statistics class. Instead, she encourages her colleagues to find creative, clear, and compelling ways to explain both basic and advanced statistical ideas to people without an advanced mathematical or statistical background. “We have a responsibility to clearly communicate what we do, and why it is important, so that we can have a seat at the table.”