Miloš Švaňa

ai, python, software engineering, decision-making and more

Would I do a PhD again?

In May 2024 I successfully defended my dissertation and ended a 5-year-long adventure that was doing a PhD. Knowing what I know now, would I do it again?

Answering this question is anything but straightforward. So instead of doing it directly, I want to talk about different aspects of doing a PhD including traveling, teaching, money, writing paper, and dealing with bureaucracy. I hope this reflection helps those of you who are currently thinking about starting a PhD, but are unsure if it’s the right path for you.

Be aware, though, that no all PhD journeys are the same. Factors like field of research, location, or choice of supervisor matter a lot. Your previous experiences, goals, and beliefs influence your perception, too.

Traveling and Meeting People

Doing a PhD allowed me to travel and meet interesting people. I visited many countries, including Switzerland, Finland, or the Netherlands. I met fellow PhD students, industry professionals, and quirky professors from every corner of the world.

My language skills improved, I learned how things work at different places, and had fun with the locals. Academic conferences were a beast of its own. You might be surprised, but people in academia know how to party.

Traveling also helped me see my own country from a different perspective. In Czechia, we idolize countries like Germany, Switzerland, or Norway, but the actual day-to-day differences are not that big. And I dare to say that many things actually work better in Czechia; I have yet to visit a city with better public transport than Prague or Brno.

But traveling also has its cons. It’s not that you CAN travel a PhD student, you HAVE TO. Spending time abroad is an official requirement. If you have a family or a job, this can be a deal-breaker.

Teaching

All full-time PhD students at my university must teach. Our responsibilities are usually limited to computer lab exercises. I mostly taught students how to use statistical and mathematical software across courses on multiple-criteria decision-making, econometrics, or linear programming. And I enjoyed it. Sharing my knowledge with others felt awesome.

My greatest teaching achievement was a short “Introduction to programming in Python” course I designed from scratch myself. I enjoyed the freedom — I could teach what I deemed important and use methods I found most effective. It was gratifying to see that the course was popular among both students and faculty.

Then again, there were some negative aspects to the whole experience. Although I liked teaching, it was very exhausting. I came home and I had to rest for a few hours. My mental energy was depleted.

I also encountered many disinterested students. Although I understood that some students will be bored or try to cheat no matter what I do, I still felt like I failed.

I sometimes had to do things I didn’t agree with. For example, I disagreed with the idea that a course on optimization methods should be mandatory for business management students. I thought that their time could be used more efficiently. Still, I had to teach the course.

The part I hated the most was tests and grading. It’s no secret that students dislike tests. But most teachers, including me, don’t enjoy them either. Grading is boring, repetitive, and unproductive.

Overall, though, I see teaching in a positive light. It sharpened my presentation and communication skills, which are useful in any knowledge-based profession. And even more importantly, it made me feel like I had a positive impact on the world.

Money

Many people considering a PhD feel uncertain about their income. In my experience, a PhD student can earn to cover their basic expenses, but they won’t be buying their own apartment anytime soon. And the outcome depends strongly on both your own effort and luck.

Our university paid me a basic stipend. It started at 9,000 CZK (350 EUR) and increased up to 17,000 CZK (670 EUR) as I progressed in my studies. Given that I preferred to have my own place, this wasn’t quite enough.

Fortunately, I was able to make extra money by participating in various research and education projects. The two most important sources of additional funding were a grant from the regional government, and a student grant competition organized by our university every year. Adding them to the basic stipend, my average monthly income was well above 30,000 CZK (1150 EUR). This was enough to cover all my basic and some not-so-basic needs.

While earning extra money was certainly a benefit, I believe this system has several drawbacks. Writing grant proposals and navigating the bureaucracy takes a lot of time. And success is never guaranteed. I could have spent countless hours on a grant proposal only to have it rejected.

Besides doing a PhD, I was also worked at a regular company. I had to sacrifice some of my free time, but I was happy to pay this price. The most obvious advantage was the additional income, which allowed me to actually save some money.

But there was also a less tangible benefit. Academia often seems disconnected from the real world; there is a gap between what’s happening in the industry and what people teach and research at universities. I was quite surprised that only a few researchers used programming languages like Python or R. By working in industry while doing a PhD I was able to bridge this gap, stay informed about what’s happening “out there”, and transfer knowledge between these two worlds.

Writing Papers

Research skills are actually very useful in practice. My industry jobs always revolved around data science, data engineering, and machine learning. Reading papers, designing experiments and statistical testing always were a crucial part of my work. Writing a paper is a great way of fostering these skills.

But what discourages me is the business of academic publishing. The way things usually work is that a scientist submits an article to a journal, the editor sends the article to other scientists for review (which are not always objective) and then accepts or rejects your contribution.

Publishers usually hide the article behind a paywall, or ask the authors to pay a substantial fee to make the article publicly available. In both cases, the entire profit, or most of it, goes the publisher. The three other parties responsible for the creation of the article — the editor, the authors, and the reviewers — get absolutely nothing.

The entire system feels outdated. Journals and publishers made sense in the past. But now we have this thing called the internet. There must be better ways of disseminating scientific knowledge.

Everything is also sluggishly slow. In fields like machine learning, knowledge deprecates quickly. We are talking months, maybe even weeks. In contrast, getting your article published can take even a year.

On the other hand, I see a lot of value in academic conferences. Yes, you still have to pay registration fees, sometimes hundreds of Euros, but at least you get something in return: you meet people from your field from all over the world and get to talk to them face-to-face.

Bureaucracy

Everything I’ve described so far involves a lot of bureaucracy. One professor at my university once said that not all bureaucracy is bad, and when done correctly, bureaucracy is actually helpful. However, I think we are very far from this ideal.

First, there are procedures that seem unnecessary. For example, when I got some bonus stipend for my work at the student grant competition project, I had to sign a document saying that I am fine with getting paid. The funniest part is that I was actually leading the project and had requested the stipend myself. Another example is submitting a request for a stipend increase after passing a course. Everybody knew I passed the course, it was recorded in the system. Still, it wasn’t enough.

The second issue is confusion. If you ask three people, you get four different answers. And you have to use a different information system for everything. Want to update information about your research project? Use system A. Want to do something related to your studies? Use this system B with a completely different interface. Want to go to a conference? Submit several physical documents to a person whose office is at the opposite end of the campus.

The third problem is the inefficiency of many processes. It has a lot to do to with how my university handled digitization. Instead of simplifying, digitization often means that we have to do things twice. We have to fill forms in an information system and then print and sign a physical version. Two examples come to mind: submitting a proposal for the student grant competition and submitting a request for my official thesis defense.

To me, all these issues are symptoms of a deeper problem. Many people blame the legal requirements posed on us by the government. These requirements play a role, but I think it’s only a partial explanation. I believe there is at least one additional factor — lack of trust. Department heads don’t trust their employees. Departments don’t trust other departments. Academic workers don’t trust administrators and vice versa. People don’t communicate as much as they should. It’s no wonder then that some requests have to be approved by 3 or more people.

So would I do it again?

I see value in the skills I’ve gained while doing a PhD and the experiences I’ve lived through. The list includes doing rigorous research, teaching, presenting at conferences, visiting interesting places, or meeting new people. All of this helped me grow as a person. However, I also experienced a lot of frustration from doing things in which I saw little value, having to deal with inefficient bureaucracy, or encountering the business of academic publishing.

I would still consider doing a PhD. However, it wouldn’t be an automatic yes. To maximize the benefits and minimize the drawbacks, I would carefully evaluate universities and professors just as they evaluate students. My goal would be to gather as much information as possible and approach the decision with a rational and analytical mindset.

I would ask about the amount of bureaucracy, the efficiency of different processes, the freedom I have in making my own decisions, the reliability of other people at a given department, cooperation with other departments, or opportunities to make extra money. I would also analyze how people behave and talk. Do they look like someone with whom I could enjoy a glass of beer? Do people in managerial positions actually know how to manage? Do they insist on using academic titles? Are they able to answer my questions clearly?

No matter how much data you collect, you will face uncertainty. A career guide from 80000 hours offers a fitting approach to such problems: act like a scientist and perform an experiment. Do a PhD for a while, say one year, and if you don’t like it, just quit. In this sense, I now see doing a PhD as a job. It is completely okay to quit and look for employment elsewhere if the university fails to provide good conditions or if you discover you don’t like doing research. The world is full of opportunities, and you are not doomed if you make one bad turn.

The similarities between doing a PhD and a regular job are actually quite extensive. I dare to argue that many companies offer some of the same benefits, including opportunities to travel, to attend conferences, to run interesting experiments, or to present your work in writing. The key difference is that in a regular company your efforts will likely be oriented towards a specific problem the company needs you to solve.

So, my last piece of advice is to compare not only different universities and supervisors, but to also consider opportunities outside academia, especially if you are interested in practical fields such as computer science, AI, or marketing. And there is always the option of doing a PhD and having a regular job at the same time. Many universities, including mine, let people study “externaly”, which usually means that your duties are reduced to bare minimum — co-authoring a few papers and writing your dissertation.

Doing a PhD isn’t necessarily a waste of time. It can be a lot of fun and the experience can help you grow as a person. But you also have to be ready to face many obstacles, some of which seem completely unnecessary.

Comments

Leave a Reply

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