Do you know what GREsGATEs or the EMBL subject interview are?

If not, probably you are on the lucky side of PhD student admission system: Karolinska. Indeed when it comes to the requirements to enter a PhD study, KI is an outstanding exception compare to many other prestigious institutes, where grad students face tough admission tests and only who scores highest gets into the program.

The issue of measuring students knowledge before, during and after a PhD was at the heart of a Doctoral Student Association meeting at Medicinska Föreningen, the medical student union of Karolinska Institutet, last week. The discussion was organized and coordinated by Arash Hellysaz, PhD student in Neuroscience, to who goes all my gratitude for having rise such interesting topic.

Post-editing note May 6th: Credits for the discussion groups within DSA goes to Ranjita Dutta Roy, the secretary of DSA and coordinator of the workgroups. She has put a lot of work and effort into these initiatives, to which goes all my thanks.

As I tweeted from MF room, I we soon realized that people were very attracted by the topic.

Why do we discuss this? At KI all you need to have to be eligible as PhD candidate is a university education (in any field, for what it matter) and a “certifiable” level of English, meaning you may need no document at all, as long as you come with the “evidence” that your english is ok.

On top of this, the administration grants exception based on “special grounds” (p12 Rules for Doc Education).

A provocative question at the meeting was: “Is it because of this ‘loss system’ that frictions occurs between supervisors and students?”. Is rising the bar of eligibility and putting all candidates on tight competition will benefit the institute, the students and the supervisors?

PhD everywhere in the world wants their title to be associated with great competence, incomparable skills and top education available in a particular field. This will make them more suitable for high-profile job positions within and outside academia.

Test or not test? In my view, Karolinska would gain so much introducing sever examination for students: it will increase its prestigiousness, have more prepared candidates and obtain numbers for a data-based education policy.

Should PhD students in Karolinska picked based on fair and tough tests? If so, what will we be tested?

But, as I see it, KI couldn’t bother less about your background education. This because KI delegates so much the choice to principal investigators (PI) and/or supervisors. The PI scrutiny is all it matters.

However, it appeared to all that in some cases people taken from the street are put on a PhD project for reasons not linked to merit, knowledge or measurable capacity to perform best. Why would a PI do so? The real “problem” is that PhDs have characteristics much more similar to employers then students. The student is chosen by the PI who has all the right to have the last word on who to get into her/his lab. For what it matter, a PI may find way more suitable a “middle knowledge” person with good manual skills then a social-incompetent nerd.

If a test has to be introduced, it may be rather used as complementary information to the PI, that in combination with few weeks of direct supervision of the student in the lab, will make her/his decision.

Though KI doesn’t set the bar for PhD admission too high, it is expected that the education provided will be on the very top scale. Should knowledge tests be introduced during and at the end of the PhD? Some argue that what you really have to acquire at the end of your PhD are not hard core knowledge, but capacity to be open mind, acquire critical thinking and other soft skills. In other words: the time for you to learn text book is up.

PhD students faces the half-time control (2 years post admission) and the Gran Finale thesis defense. Cases of students that failed these tests are relegated as metropolitan legends in Karolinska. This may indicate that either PhD here are all good or the system is not measuring things the “right way”, whatever that means.

It’s clear that the goal of these tests at KI are not to block those people inadequate to persuade their studies, but rather giving feedbacks to students and to make the best out of what you already got. Excellent or not.

In this sense, during a PhD career, students are often asked to point her/his learning outcomes. Is this sufficient? Blind exit polls from KI seems to suggest the contrary. These numbers are not confirmed, but it appears that though students perceive their education as very good, external examiners seem disappointed by PhD performance at their thesis defense. Who’s right? Which of the two scale should taken in account evaluating the Institute education performance?

To add layers of complexity, Karolinska delegates the PhD education to each of its 22 departments, where students are required to perform very differently. In some, text book examinations have been introduced, clearly to face the incredible variety of backgrounds and gap-in-knowledge that students have once recruited.

Which brings us back to the point: tighter the net? The word to you.

10 Responses

  1. Great post! About centralized testing, one of the biggest problems is that professors at KI is extremely allergic against any actions from “central KI” that decreases their autonomy. Afterall, KI departments are supported by the overhead paid by the professors, and they have the right to be unhappy when you use their own money to find ways to “control” them. However, I believe that some professors would still appreciate some central help with picking out the best students for them.

    1. I agree with you. Departments known they are greatly supported by PIs money, as happens everywhere in EU to different extends. But in such situation there is a great conflict of interest in giving great education to students and interfering to the PIs business.
      I think the only way out, if these tests will ever have to be set to action, will be to give them as a “supportive tools” BOTH to PI and the student: one will know better who’s employing, the other will focus her/his skills and possibly plan a better course program to fill the knowledge gap.

  2. ”A PI may find way more suitable a “middle knowledge” person with good manual skills then a social-incompetent nerd…” That’s tough, but I really think this is the core of the problem, because these socially incompetent nerds are often self-critical and don’t know how to stand up for themselves…but if people would measure the competence rather than the self-enhancement things would be different!

    Great reflections, by the way! Looking forward to more discussions on the topic.

  3. Thank you Ranjita.
    Again, even if we introduced test… what would we test? Will be able to prevent “socially incompetent” people to gain access to KI? And if so… is there a good reason why people with more knowledge skills then “soft skills” should be prevented to access KI?
    We need both.
    I think we should still leave the last word to PIs.

    1. I still think (1) general foundations in the topic (such as immunology, oncology or bioinformatics), (2) scientific reasoning and (3) lab skills (when applicable) should be tested first of all. This could for instance be tested in the form of interviews and lab rotations.

      I also agree with you, however, that the PI should be able to decide who he wants to take finally, but only after these things have been dealth with on a central level. I also think that the relationship between supervisor and student is important, but it’s important that he defines what he expects from the student and knows how much the student knows in order to be able to build on that.

      1. “it’s important that he [the supervisor] defines what he expects from the student”. I think we are over doing it here. I’m sure that all intelligent and responsible students in KI knows that they are on the “grill” of the supervisor. They are full of anxiety to perform at their very best and reminding them that we expect that you “stay here over weekends, read and study hard, make experiments and brings results every week” is a bit too much. Anyway, such conversation _already_ happens between PIs and students. So.. should this “conversation on expectations” be written down?
        Maybe you are suggesting something very interesting: a working contract between PI and student, where duty, responsibility and rights are clearly written down. Something similar to any other “normal” employee/employer contract.

  4. Ranjita, they made me notice that you are the big brain behind such interesting discussion and an acknowledgement to you and your work should have been done.
    So I updated the post right now, and I’m sorry if I miss that before. I look forward to meet you soon 🙂

    1. Haha 🙂 I was definitely not the big brain, I was just throwing out questions, and you guys were the ones that did the brain work!

      About the conversation on expectations, I agree that most students are on the ”grill”, but ”grilling” can be either a way to dominate or to actually supervise. I think the advantage of writing it down is to have a more directed ”grilling”. That way we don’t have to lose time due to misguidance (which will happen anyways because we cannot possibly know everything…but it shouldn’t happen with the purpose to push the student down). I don’t think writing down ”expectations” nescessarily means having to work 24/7, I think it’s important that the supervisor thinks of the student’s well-being and respects that he/she needs to ventilate from time to time to be productive in work and in life. Maybe this is something that also needs to be made explicit, but I think it should be of high importance at a medical university.

  5. You can say it loud and clear, PhD students are grilled for all the duration of their studies which far exceeds reasonableness. I agree with Riccardo & Ranjita, that PIs expect their students to “stay here over weekends, read and study hard, make experiments and brings results every week”… in other words 24/7 work or slavery. There are also other issues concerning the conditions of the doctoral education which varies between groups and even in the same group. A classical case is when funding comes from outside KI (and certain rules can be by-passed) and a student ends up in full control of his/ her PI, like a puppet with strings. Also, it happens that certain students are misled in the beginning of their studies thinking you are going to do one thing but one fine day your boss comes & changes everything. All this leads to unhappiness, severe anxiety and unhealthy mental conditions that certainly will not make a student productive. Society tends to put all the blame on the student and this makes it even a bigger burden. Communication problems are also prevalent between student & PIs. One more point is that many PIs themselves do not know how to supervise a student. They just did a PhD themselves & published papers. This only means that they produced results not a certification that they can manage a student!! There is a wide acceptance in academia that PhD students have to work to the point of getting sick. I wonder how a burnt-out students can then continue to excel! What is the point of doctoral education then? Students need more protection from being abused. I also agree with the employee/employer contract-style, where everything is clearly stated and discussed in presence of the education director & possibly a student association representative.

  6. The GRE is a terrible indicator of success in a PhD program. They need a new GRE similar to the MCAT exam.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.