Defending my dissertation

Well, I did it! I wrote my dissertation. I’m Dr. Murray now. A year of pre-k. 12 years of grade school. 5 years for my undergraduate degree. 5 more for this PhD. 23 years of my life have been spent in school! It’s a strange, scary, and exhilarating feeling to have finally reached the zenith of this long experience, and to stare into the distance ahead, wondering what’s next. The next steps, however, will have to wait. First, I want to catch my breath and look back at this latest achievement.

I titled my dissertation Embracing Complexity in the Science of Science. Its a “stapled” dissertation format, merging 4 studies that I wrote while in grad school. The narrative thread I chose to connect each study was to look at science through the perspective of Complexity Science, considering how treating science as a complex system might inform us about its inner workings. You can read it here, if you want (its quite long). If you’re more of a visual person, you might also appreciate this presentation, which I think did a pretty good job of summarizing it.

The first of the four studies investigates the extent of gender and national bias in peer review, observing the role of homophilic preferences in not only shaping review decisions, but structuring the composition of teams of referees. The second study leverages public and private data to explore the factors driving student-teacher evaluations, an important but often overlooked metric in a scientist’s career; we find evidence of demographic bias, namely that women and non-White faculty are at a disadvantage in these ratings, putting them at a disadvantage in hiring, promotion, and more. The third study attempts to quantify the extent of disagreement across science, revealing that it is most common in the social sciences and humanities, and lowest in physics and maths; yet there is also a great deal of heterogeneity across sub-fields, resulting from their the physical realities of their research topic. Finally, the fourth study explores the examines the global landscape of scientific mobility, creating a dense representation of mobility that illustrates the importance of different kinds of proximities, such as geography, language, culture, shared history, and prestige, in structuring the mobility decisions of individual scientists.

Together, each of these studies constitutes a significant contribution to the Science of Science. Yet when viewed collectively through the perspective of complex systems, they show insights about the organization and behavior of science. For example, homophilic preferences, demographic bias, physical realities, and proximity are all examples of bottom-up forces that govern the actions and interactions of individuals, allowing them to self-organize into complex social arrangements such as teams, communities, and disciplines. Concepts like feedback, too, explain how selection mechanisms, like peer review (eLife), metric evaluations (student-teacher evaluations), peer judgements (disagreement), and hiring (mobility) naturally produce and perpetuate inequalities and hierarchies in science. Yet feedback also allows for radical change across science. The chaotic nature of science sometimes triggers feedback loops that amplify the effects of certain events, such as an individual’s success or a breakthrough discovery, to such an extreme that they remake wide swathes of science over a few short years.

By viewing science as a complex system, it shifts the attention of the Science of Science away from grand theories of science and universal laws, and instead towards the everyday actions and behaviors of individual scientists. By simply shifting our perspective, we can reveal something about the fundamental workings of science, and maybe even spot a path towards improving it, making it more open, equitable, and effective.

There are of course tons of details left in my dissertation. Writing it took a long time, many revisions, and a lot of reading! Although I have wonderful mentors who praised my work, I remain intimately familiar with all of its shortcomings. Yet finally, I am beginning to look back and appreciate the work I have produced. It will never be perfect, but it was good enough to get me to where I want to go. And I’ve learned a lot from it too. I hope to take what I learned here and do even better research moving forward, because the dissertation shouldn’t be a barrier to your career, but a stepping stone.

Avatar
Dakota Murray
PhD in Informatics

I study the social dimensions of science