WHAT THEY LEARNED: Rebecca Miller ’17

WHAT THEY LEARNED: Rebecca Miller ’17

People often approach statistics as if they present fixed and objective truths, but as Rebecca Miller ’17 discovered, that isn’t always necessarily the case. In her thesis, “The Breslow-Day Homogeneity Test and Its Applications to Apsley v. Boeing,” the mathematics major and statistics minor explored a 2012 age-discrimination class-action court case via its statistical data to try and understand how different analyses of the same data can result in wildly disparate conclusions.

“My thesis ultimately argues that the court ruled incorrectly when they said there had not been age discrimination,” said Miller. “This means that there is reason to feel that the people suing should have been entitled to damages. Obviously, it is too late to do anything about that. I think more broadly, though, it means that courts need to be really thoughtful when they look at statistical data, and rely heavily on unbiased data analysts, who can look at the methods of data analysis used by plaintiffs and defendants to look at whether they are valid and convincing.”

Miller, who entered Haverford thinking she’d become a biomedical engineer, is now following a career path that allows her to continue her passion for using data to tackle real-world problems. Her job as a marketing data analyst at Crossix, a consumer-based healthcare analytics firm in New York City, is a direct result of her thesis project.

“Doing so much data analysis in my thesis really helped me confirm that this was I wanted to do post-graduation,” she said. “I also think it really made me aware of the crucial role data analysis has on this world, and thus how important it is to do it right.”

 

What is your biggest takeaway from your thesis?

I think the biggest thing I learned was how to handle such a massive project. The thesis is an incredible undertaking, and at times felt really daunting. Having finished it, though, I now feel really confident in starting other similarly large projects. The other thing I really found while working on this paper is how different analyses of the same data can come to totally different conclusions. I was working with data from an age-discrimination case. Both the plaintiffs and the defendants did analyses of this data, and came up with totally different arguments for what this data said. Both of these were valid analyses to run, and so part of my job through this thesis was to think about which was more valid. It made it really clear to me that statistics does not necessarily give objectives truths, and that decisions made in how people analyze data really matters.

How did your thesis advisor help you develop your topic, conduct your research, and/or interpret your results?

I knew going into senior year that I wanted to do an applied statistics thesis, but I didn’t really know where to go from there. [My advisor Professor] Weiwen Miao’s background is in law and statistics, so she had ideas about legal cases that had questions that I could turn into a thesis. My meetings with her kept me accountable and ensured I was making consistent progress through the year. She also was there to help when I encountered road-blocks, which ranged from having trouble understanding theory to problems with codes.  In addition, her confidence in me really helped my confidence through this whole project, and made me feel like I could actually tackle this project.

 

What They Learned” is a blog series exploring the thesis work of recent graduates.

Submit a Comment

Your email address will not be published. Required fields are marked *