There Is No One Way to Do a PhD

Fawzia Kara-Isitt, Jessica Yauney, Bethany Ahluwalia

Fawzia Kara-Isitt, Jessica Yauney, Bethany Ahluwalia

What does a PhD journey look like? If you ask most people, they would explain to you that completing a PhD involves  identifying a research question, finding a supervisor, conducting research, writing the dissertation, defending it, and moving on.

However, the reality of a PhD journey can be far more complicated and look rather different.

Fawzia, Bethany, and Jessica are all ACM-W members and sat down to chat. They are computer science researchers at different points in their doctoral journeys, working at different universities and on very different research problems. 

“How did we get here?”

Fawzia is a PhD student who is working on her second round of corrections for her dissertation (almost done), but her path to this point has been far from linear. She studied computer science in Pakistan when the field itself was still relatively new, then received a scholarship for a master’s degree in the United States, specializing in AI and software engineering. At the end of her master’s degree, a professor arranged an opportunity for her to pursue a fully funded PhD. She turned it down. Not because she did not want it, but because at that stage of her life, family and cultural expectations pulled her in another direction. Years followed that included working in industry, teaching, moving between countries, and starting a family. Then, suddenly, her PhD dream felt like unfinished business, and she started applying for it again, two decades after her Master’s.  After applying to more than a dozen universities, she received a funded studentship at Brunel University and began her doctoral work in 2019. Her initial interest was in online harm: systems were increasingly able to identify toxicity and hate speech. However, she was more interested in what she viewed as the real question: What could technology actually do to mitigate it?

Bethany’s journey has a very different rhythm. She enrolled in computer science at the University of Southampton. Because she had been away from formal education for some time, she began with a foundation year that ran through COVID. Then something interesting happened. She kept reaching the point where she was supposed to finish and decided that she was not ready to. During her undergraduate degree, she transferred into an integrated master’s degree. During her master’s, one of her favorite lecturers advertised a PhD degree combining AI and qualitative research. It was the only PhD she applied for. She was accepted. Her work now explores human-AI collaboration in qualitative analysis, with an increasing focus on reflexivity. She develops tools that help qualitative researchers examine their assumptions, decisions, biases, and positionality as they analyze data.

Jessica’s route started in a classroom rather than a research lab. She became a high-school teacher in Los Angeles. She enjoyed teaching and expected it to remain central to her career. Her return to a graduate school was something of a “COVID adventure.” The master’s program there gave her an opportunity to explore computing education research. Conversations with advisors led her toward researchers and programs she would not otherwise have considered. She ultimately entered a doctoral program at Stanford University, where her work now sits at the intersection of computer science education, teaching, and participatory research.

The strange problem of becoming an “expert”

Jessica: One thing that seems strange about doing a PhD is that we are supposed to be becoming independent researchers, but so much of our work is constantly being checked by someone else. What does that do to your confidence?

Fawzia: Sometimes the confidence—because we have to go through so many channels of having everything checked, whether it’s our syntax, whether it’s our grammar, whether it’s the formality of the language, whether it’s a paper, and the revisions and the dissertations—somewhere along the line, you kind of lose your self-confidence a bit. The worst thing is this whole exactness of everything, and the preciseness… It’s my language, this is how I want to say it, but no, it starts to feel like it’s becoming the corrector’s language instead.. But, if the person who’s so precise and strict will eventually accept it, then [that] means anyone will accept and appreciate it. So if there are problems, I’m glad that a keen eye has picked it up, but it took me quite a while to get to that mindset.

Jessica: Yeah sometimes it feels really difficult to please everyone. I like quantitative, and I like qualitative methods, so I did some of both, and nobody can agree on how much of each method we should be doing, because some people really like one method, and some people really like the other method. So eventually, it’ll be done.

Remembering why the research matters

Jessica: With all of those requirements, revisions, and milestones, is it easy to lose track of why you wanted to do research in the first place?

Fawzia: We have an idea, a hypothesis, something that will help prove something, and when we’re there trying to prove that something, our deeper intent is just to make the world just a little bit better somehow with that innovative idea. But when you have so many little boxes to check, and I’s to dot, and T’s to cross, sometimes the motivation of what we’re doing in the first place gets lost in that.

Bethany: I’ve always been kind of interested in data and using it for good, as well as, like, the responsible use of AI. We’re now focusing really on using AI to enhance reflexivity rather than to perform analysis. We are building tools that notice, for example,  when a researcher makes a change to how they have coded  their data and prompt them to document why, promoting awareness and reflection.

Change is the only constant in life

Jessica: All three of us started with an idea of what our research might be, but those ideas have changed. How much are you supposed to let a PhD change as you go?

Fawzia: Technology has changed dramatically while I’ve been working on the PhD. There are tools to find toxicity, but what do we do about it? And is there a way to curtail and control it dynamically? I initially approached the problem through sentiment analysis and mathematical modeling. Years later, large language models created an entirely new way to approach the mitigation part of the problem. By the time I finished my PhD… LLMs was the new thing. As we all know, since we’re all in computer science, there is nothing written in stone. What works today may not work tomorrow and seem completely irrelevant and obsolete.

Bethany: I am still using the title that my supervisor advertised the PhD with because I hate coming up with names, so I’m just leaving it until the very end. That name still refers to “next-generation qualitative research” but our scope is now much narrower.

Jessica: Uncertainty is built into my methodology. My dissertation asks young people to participate in deciding what questions are worth investigating, so I cannot determine the research question in advance because doing so would undermine the point of the approach.

Three points on a journey—not three finished stories

Jessica: So after all of this, what do you want to do next?

Fawzia: Definitely more research than teaching. I have had some teaching job offers, but I’ve turned them down, because I don’t want to do pure teaching. Now I don’t see myself doing anything else but research, with some teaching just because I enjoy mentoring young people.

Bethany: I don’t know what I want to do in the future… I’d love to stay in research. I basically have no plans of leaving Southampton as long as they’ll keep me.

Jessica: I was gonna say all’s well that ends well because I’m enjoying what I’m doing now, but I’m, like, two-fifths of the way done with this, on the optimistic view, so still have a while to go.

If you enjoyed this conversation about PhDs, we would be happy to return to some of the questions that surfaced here: 

What are the hardest parts of doctoral study? How does feedback shape confidence and research identity? What makes a supportive supervisor or research community? And what happens when the finish line of the PhD begins to come into view? 

If you have any questions or topics, or would like to be a part of one of these conversations, reach out to Jessica at: acmw-communications@acm.org


  • Facebook
  • Twitter
  • LinkedIn
  • Email
  • Print