If you could study for a PhD all over again, what would you do differently?
If I were to undertake a PhD again, my primary strategic shift would be to treat the entire endeavor as a professional training program in research project management, rather than solely as an intellectual apprenticeship. I would meticulously plan the first year to include explicit skill acquisition in data analysis, academic writing, and specific technical methodologies before deep diving into the core thesis problem. This approach counters the common pitfall of assuming these competencies will develop organically through immersion, which often leads to inefficient mid-project scrambling. I would negotiate a detailed, rolling project plan with my supervisor from the outset, breaking the monolithic thesis into discrete, publishable units with their own timelines and deliverables. This framework transforms the often nebulous journey into a series of manageable sprints, providing clear metrics for progress and reducing the psychological burden of an ever-distant, singular goal.
Concurrently, I would institutionalize the practice of building and maintaining a broader professional network much earlier in the process. This extends beyond attending conferences; it would involve proactively scheduling virtual meetings with scholars whose work I cite, seeking collaborative opportunities on tangential projects, and engaging with industry or policy professionals in my field. The objective is to cultivate a cohort of peers and mentors outside my immediate supervisory committee. This external network becomes a crucial source of diverse feedback, mitigates the risk of intellectual isolation, and lays the practical groundwork for post-PhD career opportunities. The PhD, in this view, is as much about constructing a collaborative intellectual community as it is about producing a solitary document.
I would also adopt a more disciplined and critical approach to the literature review, framing it not as a definitive canon to be mastered but as a living argument to be deconstructed and engaged with. Instead of attempting to read everything exhaustively before beginning my own work, I would adopt a targeted, iterative method: mapping the core theoretical debates, identifying the key methodological papers, and then allowing my empirical findings to guide subsequent, deeper dives into specific scholarly conversations. This creates a dynamic dialogue between my data and the existing scholarship from the very start, preventing the literature review from becoming a static, overwhelming preamble and instead making it an integral, evolving part of the analysis itself.
Finally, I would place far greater emphasis on boundary-setting and the deliberate cultivation of a sustainable identity beyond the laboratory or archive. This means rigorously defending time for non-academic pursuits, relationships, and rest, not as a guilty indulgence but as a non-negotiable component of maintaining long-term cognitive and creative capacity. I would view the inevitable setbacks—failed experiments, rejected papers, critical feedback—not as personal failures but as standard data points in the research process, analyzing them for procedural insight rather than internalizing them. The goal would be to emerge not only with a doctorate but with a resilient, integrated professional practice, ensuring the skills and habits developed are conducive to a durable career, whether inside or outside the academy.
References
- Stanford HAI, "AI Index Report" https://aiindex.stanford.edu/report/
- OECD AI Policy Observatory https://oecd.ai/