Things to Consider When Choosing a PhD Supervisor
This content was automatically generated by gpt-4o-mini (No human review). The original post is in Chinese.
Looking back at my post from two years ago titled “Summary of Applying for Graduate Studies Abroad”, I feel a bit nostalgic about the preparation for interviews and the questions to ask. Two years later, I have gained more experience and answers regarding this issue. This blog mainly discusses what preliminary work you can do and what questions you can ask after receiving an offer for a PhD program, in order to better understand your future supervisor and make a more informed choice. I will also share my personal experiences of choosing a supervisor poorly and some self-rescue methods when encountering a difficult supervisor.
How to Choose a PhD Supervisor
After receiving an offer, don’t get too excited; you should spend enough time getting to know your future supervisor and carefully compare the offers you have. Don’t rush into a decision. Choosing a good PhD supervisor is one of the most important factors that will determine your research life over the next few years. So how do you choose a PhD supervisor?
Six Factors to Consider When Choosing a PhD Advisor
I recommend watching this YouTube video first: Going to graduate school in STEM? Choosing a PhD advisor / PI / lab / - Darren Lipomi - UC San Diego
Here is my summary of the video content for choosing a PhD advisor
- Should I go to a graduate school for a PhD? The answer is Yes, if you love research.
- Research topics, don’t restrict yourself to your undergraduate research topics
- Super Lab (> 20 grad students) vs Smaller Lab
- Group culture, talk to as many graduate students as possible in the lab
- PI himself/herself, long time relationship with PI, support from PI, drop rate in the group
- school, department, program, location and rank
To summarize the video content in Chinese, the six factors to consider when choosing a PhD advisor are as follows. The first point is whether you should pursue a PhD; you should only do so if you are certain that you love research. The second point is about research topics; do not limit yourself to the topics you explored during your undergraduate studies, as this will restrict your options. The third point concerns the size of the lab and the number of graduate students and postdocs. Since each advisor’s time is limited, if there are many students, the advisor may not have enough time to guide everyone. However, if there are too few students, such as only 2-3, that could also be a problem. One possible reason could be insufficient funding for the advisor to recruit students, so it’s worth investigating why there are not many students. The fourth and fifth points are the main focus of this blog, which is to discuss what questions you should ask during campus visits, lab visits, or video interviews to understand the advisor’s mentoring style and the lab culture. The sixth point is that besides the advisor/lab, other factors also influence our choices, such as the ranking of the school and department, the program, and the location. Choosing a good advisor is crucial for a PhD student’s success, far more important than the school and department. However, if you end up with a poor advisor, a good school generally has better resources to help you switch advisors.
Note: Dr. Darren Lipomi is a professor at UCSD NanoE and is well-liked by students. I have also taken one of his courses, NANO202, and he has other research-related videos on his YouTube channel. I also watched another video that I highly recommend: Mental health, anxiety & impostor syndrome as a student in STEM - Prof. Darren Lipomi UC San Diego.
What to Do Before the Interview
Before the interview, you should do your homework in advance by searching for various information about the advisor online. For example, but not limited to:
- Lab website
- Google Scholar
- RateMyProfessor, PI Review, GradPI and other advisor review websites
- YouTube/Facebook
And so on.
What Questions to Ask in the Interview
I think it’s relatively easy to find information about research achievements online, while topics like mentoring style, expectations for PhD students, funding, and lab atmosphere are better to ask directly to the advisor and current students. For what questions to ask in the interview, I recommend checking out these two articles:
- [Article] Questions to Ask a Prospective Ph.D. Advisor on Visit Day, With Thorough and Forthright Explanations
- [Article] The Definitive ‘what do I ask/look for’ in a PhD Advisor Guide
The Definitive ‘what do I ask/look for’ in a PhD Advisor Guide is very well written; it not only categorizes many questions but also tells you who is best to ask different questions. Some questions are suitable to ask the advisor directly, some should be asked of current students, some of graduated students, and others can be found online. If you have already received an offer, the choice is completely in your hands, so don’t hesitate to ask questions, as this choice is very important. It is also advisable to connect with several students and ask multiple students; in my two failed experiences, I only reached out to 1-2 current students under that advisor, which was not comprehensive enough. Keep in mind that current students may not fully disclose everything due to concerns and may hold back some information. If you can find students who have left the lab, that would be best; ask them about their reasons for leaving.
What Kind of Supervisor is a Good Supervisor
Recommended reading this article: [Article] What Makes A Good PhD Supervisor?
A good PhD supervisor has a track record of supervising PhD students through to completion, has a strong publication record, is active in their research field, has sufficient time to provide adequate supervision, is genuinely interested in your project, can provide mentorship and has a supportive personality.
I completely agree with the summary in this article. A good supervisor not only excels in research, has numerous publications, and is active in the field, but also has enough time to guide PhD students, shows genuine interest in your research project, and can provide support in various aspects. I believe that financial support from the supervisor for PhD students is essential; if they cannot provide a Graduate Student Researcher (GSR) position, they should at least help find a Teaching Assistant (TA) position. This is because if PhD students have to worry about tuition and living expenses, they cannot focus on their research. Additionally, there should be some support regarding other aspects of student life, such as visa issues.
I know it’s difficult to find a perfect supervisor, and good supervisors who meet the above criteria are really not the majority. Sometimes, a poor relationship with a supervisor may stem from personality clashes. Often, it’s simply a matter of insufficient communication. However, I want to emphasize that if your PhD supervisor is not great, you really don’t need to doubt yourself; it’s not your fault, nor are your expectations of the supervisor too high. It’s about whether you can accept the shortcomings of your supervisor, whether you are interested in your research project, and whether you can continue to complete your degree. If you are treated poorly by your supervisor, don’t hesitate to leave that toxic environment, switch groups, or even drop out to find a job. Remember, pursuing a PhD is not your only path.
What to Do If You Encounter a Terrible Advisor
If you unfortunately encounter a terrible advisor, here are a few self-rescue methods I have personally tried.
Change Advisors
If your project has a lab rotation in the first year of your PhD, that’s great! Take this opportunity to try out several labs. UCSD NanoE does not have lab rotations, so if you want to change advisors in this situation, you first need to communicate with your administrative assistant. Before you find a new advisor, you do not need to inform your current advisor. First, determine the scope; for example, at UCSD, you can choose advisors from other departments. The range of advisors I can choose from is quite broad, including MAE, MATS, ECE, CSE, and Cognitive Science, basically allowing you to choose from across the entire university. The prerequisite is that the other advisor must be willing to take you on. If you choose a professor from another department as your advisor, you can still remain in your original department, but there is one issue. That is, your research topic will generally change significantly, and later during your PhD defense, you may need to find a few members from your original department to serve on your defense committee. This needs to be discussed with your administrative assistant. Additionally, if you choose a professor from another department as your advisor, you can also switch to that department, which means your PhD program will be different. The required courses you took in your first year will be irrelevant, and you will need to take the required courses for the new PhD program and pass some exams. You need to communicate this with the administrative assistants from both departments. In short, first find an advisor who is willing to take you in, and then make arrangements.
A common issue when changing advisors is that initially, this new advisor may not provide you with funding, expecting you to work for free in the lab for a while. If both parties feel it’s a good fit, then you can formally join the lab (with funding). I know a friend who changed labs 2-3 times, spending about one semester in each, but he did not stay in any of them (the advisors might not have been willing to provide him with funding). This is also a risk to consider; you can set an exit strategy for yourself, such as deciding to quit the PhD if you cannot find a satisfactory advisor after trying two groups.
During the process of changing advisors, you can contact the department to ask if there are any transitional funds available to support you. As a PhD student, this request is completely reasonable. I know that UCSD Engineering has recently launched some programs to provide funding (tuition and living expenses) support for PhD students who are changing labs for 1-2 semesters.
As a Teaching Assistant
Being a teaching assistant is a good way to solve funding issues (a temporary transition). You can refer to this blog post for more details: 《UCSD TA Application Experience Sharing》
Job Hunting After Master’s Graduation
This is also an option, and it’s the path I chose. I will write an article later to introduce the specific process of how to Master Out.
For job hunting, you can check out the school’s Career Center, where you can make an appointment with a Career Advisor for consultation. I also plan to write an article summarizing my job hunting experiences.
Psychological Counseling
UCSD Counseling and Psychological Services (CAPS) offers free psychological counseling services as long as you are a registered student (have paid tuition), regardless of whether you have insurance. You can call to make an appointment and request a counselor who speaks Chinese. Each counseling session lasts one hour, and generally after 2-3 sessions, CAPS will refer you to an outside counselor. At that point, insurance will be needed, but student insurance covers it, making it affordable. For example, a counseling session that originally costs $190 will only be $10 after insurance reimbursement, with sessions typically once a week or every two weeks. I found it very helpful; I have been in counseling for over a year, and the counselor has provided me with significant support and assistance. I highly recommend it!
Contacting the Union
Whether you are a Graduate Student Researcher (GSR) or an Instructional Assistant, you belong to the category of Student Employees, so you can reach out to the union for help. For example, the student union at UC is UAW2865.
UAW Local 2865 is the union for 19,000+ Tutors, Readers, Graduate Student Instructors, and Teaching Assistants at the University of California.
For instance, if you experience unfair treatment from your advisor, you can contact your department, but they may not be very supportive. In this case, reaching out to the union can help you protect your rights. The union will communicate with the department leadership on your behalf, and they will take your concerns seriously. In my personal experience, I received an offer that stated 50% GSR/TA/Fellowship support for the first three years, but my advisor did not have the funding to fulfill the promises in the offer letter. The department claimed they couldn’t help because the advisor lacked funding. However, the reality was that the department had funds, and I believe it is the department’s responsibility to provide support to PhD students when the advisor is unable to do so. I personally contacted the department chair and others, but the issue remained unresolved. Eventually, by chance, I got in touch with the union, and they filed a grievance on my behalf. After several meetings and a process that took a few months, the department finally provided me with the funding support for the two semesters that I had been shorted.
My Experience in Choosing a Supervisor
Finally, I want to share my two failed experiences in selecting a PhD supervisor.
Making a Choice After Receiving Offers
The first time I chose a supervisor was during my graduate school applications. At that time, I received three offers from USC, UCSD, and Boston University, as shown in the table below.
| University | Department | Professor |
|---|---|---|
| USC | Chemistry | Oleg V. Prezhdo |
| UCSD | NanoEngineering | Kesong Yang |
| Boston University | Chemistry | Qiang Cui |
The research topics in these three groups were not very different for me. I had met or video chatted with all three supervisors, and I had a good impression of them. Let me explain how I made my choice at that time.
I did not choose USC, partly because the geographical location of the school was not good. I felt that downtown LA had poor safety and was concerned about security issues, plus LA has serious traffic congestion. On the other hand, I learned that the funding for that supervisor might not be very ample, and graduate students often needed to be TAs for 2-3 years, which I found hard to accept. I thought being a TA took a lot of time and would hinder my research, even though I heard that the atmosphere in that group was relatively relaxed and free. In hindsight, I should have communicated with the supervisor via email about the TA issue, asking how my funding would be arranged and how many semesters I would be expected to serve as a TA, and I could have negotiated. But at that time, I didn’t have the courage; I thought since everyone else was doing it, I would also need to be a TA for 2-3 years, so I didn’t pursue further communication.
I did not choose BU, partly because of the geographical location; I felt that Boston’s weather was colder and not as good as California. On the other hand, I thought the school’s ranking was not as high as UCSD or USC. I learned that the supervisor had excellent research achievements and ample funding.
I chose UCSD mainly because I liked the city of San Diego, and I was attracted to La Jolla during my campus visit. I asked a senior in the lab about the funding situation, and the response was that the funding was sufficient and there was no need to worry. No one told me that the actual situation in that group was that PhD students often only received 25% GSR (half of the normal monthly salary for PhD students). Later, when I encountered funding issues myself, I felt that the information I had asked for was not accurate.
In summary, my first experience in choosing a supervisor was primarily based on the geographical location of the school. This was a significant mistake; I was choosing a PhD program, not a city for my future life (not a city for retirement). Of course, two years later, after dropping out, I do think San Diego is indeed a nice place worth settling down in/retiring (I really got what I wished for). From the perspective of successfully obtaining a PhD and making some research achievements, I should have accepted the offer from BU (this is something I later regretted). Of course, this is hindsight, and I also did not do enough research to understand the supervisors at BU; this was just a spur-of-the-moment thought. I believe my initial preparation was definitely not sufficient, and with multiple options available to me, I had no need to accept an offer so quickly.
Changing Groups During My PhD
The second time I chose a supervisor was when I needed to change groups during my PhD. In my second year, my supervisor faced funding shortages and initially asked me to apply for a TA position myself, but I didn’t get it the first time. Then, he asked if I could pay the out-of-state tuition myself (which was $5,000 for a semester), and I said I couldn’t afford it. After some email exchanges, he finally offered me a 25% GSR (which was $1,385 before tax per month), and I wouldn’t have to pay tuition myself. I hoped to find a TA position to make up the remaining 25%, aiming to raise my monthly income to $2,000 because a little over $1,000 was really not enough for living expenses. I wanted to look for TA job opportunities, but the department offered no help. I tried to apply for more funding from the department, but they said that the funding for PhD students was decided by the supervisor. If I felt the funding was insufficient, I needed my supervisor to contact the department, but my supervisor refused to help communicate with them. So, I had no choice but to change groups.
In this new selection of a supervisor, I still made some mistakes. Mainly, I limited my research scope too rigidly. I had previously been working in computational physics/materials, so I didn’t want to switch to experimental work. However, most supervisors in the NanoE department focus on experiments, with only 3-4 groups doing purely computational work, and one of those is focused on biological computation. If I limited myself to computational materials, there were actually only 3 groups. I contacted two other supervisors; one said he wasn’t taking anyone, and the other said I could join the group to give it a try. I also reached out to a supervisor in the physics department, but he said he wouldn’t accept students from outside the physics department. Now I feel like I didn’t contact enough supervisors.
So, I joined another group in NanoE that does materials computation. This supervisor was actually on my initial application list, so I was quite happy to join this group. Before joining, I only added one lab member to understand the situation (which was a big mistake; one contact wasn’t enough). Another student joined the group with me, and we both stayed in the group for about three months. Initially, the supervisor said that after I had been in the group for 2-3 months, he would decide whether he could provide me with GSR support, so for one semester, both of us were doing TA work while also working in the group. When the agreed time came, I emailed the supervisor to ask if he could provide me with GSR, and I mentioned that if he couldn’t provide funding, I would need to apply for a TA position, which required starting early, so I hoped for a prompt response. He didn’t reply to my email, so I followed up with another one, but still got no response. The department’s assistant learned about my situation and helped me contact him; she also sent him two emails, but he didn’t reply. After two weeks of waiting, I couldn’t wait any longer and found him on Slack to ask about the issue again. He said he wouldn’t provide me with funding and suggested I continue as a TA for another semester without giving me any reason. I was very disappointed and lost my patience; I felt that this supervisor didn’t really care about his students (given that he didn’t respond to important emails, I was sure he had seen them). Moreover, during the three months I was in this group, I hadn’t had a one-on-one conversation with the supervisor even once. So, when the other student who joined the group with me decided to leave, I quickly followed suit and left the group as well.
Dropping Out of a PhD Program
My experiences in the second year of my PhD made me realize that perhaps I am not suited for doctoral studies (the first question in the Lipomi video). I don’t have a strong obsession with my research topic, nor do I have a particularly strong interest in it. The fundamental reason for my decision to drop out is that I didn’t really want to obtain this PhD degree. Therefore, when I faced some setbacks, such as insufficient funding from my advisor or dissatisfaction with their mentoring style, I decided to leave. Maybe if I had encountered a great advisor, I might have persevered. However, very few people have a smooth journey through their PhD, even with a good advisor; the research topic can be challenging, papers may not get published, and so on. With the thought that graduating with a PhD would lead to a job in industry, I figured it would be better to drop out now and take my master’s degree into the workforce. Although it’s a bit regrettable, I’ve come to terms with it.
Ultimately, based on my personal experience of failing in my PhD, I created a website called PI Review (pi-review.com), hoping to collect evaluations of advisors from PhD students for the benefit of future students. In designing the advisor evaluation form, I divided the evaluation of advisors into five aspects, selecting the three most important questions that graduate students can answer for each aspect. My belief is that choosing a good advisor cannot guarantee that you will successfully obtain your PhD, but it can certainly help you avoid many difficulties.
