22

I’m currently doing my masters in chemical engineering in a relatively unknown university here in the Philippines. For my master’s thesis, I am planning to theoretically model transport processes of spray drying of nutraceuticals. Neither my advisor nor any other professors here are experts on the subject.

I wish to know:

  • Is it still possible to do publishable or even outstanding research by learning the state-of-the-art of the area, specifically about transport modelling, by myself?

  • How can I approach this dilemma? (I just need the extra boost in confidence since I am wading through unknown waters here.)

I do have experience on research and writing papers as I have experienced it during my bachelors. Also, I do think people in my current department will be able to understand my research since the topic that I want to pursue largely deals with basic chemical engineering principles. Although, I'm quite in doubt really in my current course of action since none of them are experts on theoretical modelling. I didn't publish anything from my bachelor's work, and I am still relatively a neophyte at research.

  • 10
    My immediate reaction to the title question is "As opposed to what?!" But that reaction relies on quite a bit of experience doing research at all. – JeffE Aug 5 '16 at 3:39
  • Welcome to Academia SE. I edited your question a bit to remove the poll aspect of your question (“I'm looking for people who is/was in a similar position and succeeded”) as such questions are not a good fit for this site. Please check that everything is still according to your intentions. – Wrzlprmft Aug 5 '16 at 7:54
25

At some point in a research career you will need to learn directly from papers. Even as a practical programmer, when I wanted the absolute state of the art in an area, I would read academic papers, not wait for the material to show up in textbooks. A researcher often ends up knowing more about their topic than anyone else in the world, not just in their university, and must learn by means that don't depend on access to more knowledgeable people.

That does not mean your proposed course is a wise one at this stage in your career.

As I see it, student research has three objectives:

  1. Learn how to do research
  2. Collect evidence you can do research, in the form of degrees and letters of recommendation.
  3. Contribute to the collective body of knowledge.

You need to decide the weight of these objectives for yourself, but generally the earlier you are in your studies and career, the greater the importance of the first two objectives.

The risk you are potentially taking is sacrificing the first two objectives at a relatively early stage. If your professors do not understand the subject of your research, they may be less able to guide you. It may also be harder for them to judge the quality of your research, which they need to do in order to grant degrees and write recommendation letters.

If you already have really good research skills, the first objective is unnecessary for you. However, if you had that much research experience I doubt you would be asking this question.

You might consider a compromise: Pick some research direction that fits with the research at your university but moves in the direction you want. For example, do a theoretical modeling project that models something your professors do know about.

  • I'd argue (at least in my own discipline, mathematics) that establishing an up-to-date knowledge base is also critically important to "do research". – paul garrett Aug 6 '16 at 22:27
  • @paulgarrett Yes, a good knowledge base is necessary. I think that can be achieved by independent study, at least for my field, computer science. – Patricia Shanahan Aug 6 '16 at 23:33
  • Patricia, I am willing to believe your claim that the relevant knowledge base can be achieved "independently" in CompSci, and maybe other things. But the way "mathematics" has evolved is much like the stock market (in the U.S.), namely, if one plays by the rules (no insider trading), one will never make money (because everyone else has the same public information), and, by the "gambler's ruin" principle, one eventually goes broke. :) Although many people will not care, I do think it is the case that in contemporary mathematics one cannot get up to the ... [cont'd] – paul garrett Aug 6 '16 at 23:37
  • ... [contd'] starting line without some mentoring/insider-information. In the U.S., the "REU" bit wildly misleads many people, not just kids... Srsly, if one can generate publishable papers in 8 weeks after one's junior year (3rd, in the U.S.) without having any abstract algebra or modern analysis... "why would a PhD take 5-6 years..." All of that. A very salient point is that "math does not become wrong/old"... which is part of its appeal to me... but is violently different from most other human endeavors. – paul garrett Aug 6 '16 at 23:41
  • I commend you guys for giving out such helpful answers. Although, I do want to add though that my advisor, a professor here in my university (in a sense that she has attained the highest rank of being an academic), wanted to branch out to very unfamiliar areas to her and to the department as a whole. As per your arguments here in the thread, you've said that I need an experienced researcher, one who can basically point me to the right questions. Would that statement still hold irregardless of advisor's expertise? – Nicotinamide Aug 9 '16 at 1:53
11

As already mentioned, it is quite common that an experienced researcher learns a new topic from scratch – usually to somehow combine it with topics they are already familiar with. One way to do this is to give this extension as a thesis project to an advisee, whose main task is to learn about the new topic and connect it with the group’s existing work. It’s something that happens quite frequently in our work group and may be a good approach for you.

The advisor then learns about the topic through status updates, theses, writing publications with the advisee, and possibly visiting conferences with the advisee. On the other hand, the advisee learns to research through the advisor scrutinising their work – though it may sound a bit trite, a central aspect of research is to ask the right questions and to do this, your advisor does not need to be the foremost expert on your topic. The advisee can also learn about writing, since it does not depend on the subject that much and good papers should be understandable by a broader audience than just experts on that particular topic.

However, this approach may not be suited for everybody and every topic. A few points to consider:

  • Is your advisor willing to do this and does this fit his style of advising?

  • How large is the topic that you would have to learn? On a related note: How important are method knowledge, specific subject knowledge, and a wide overview important in your field (or with other words: How strong is specialisation in your field?)? For example, in my interdisciplinary area, new ideas and research often come from people combining concepts of different fields, on which they are not the foremost experts.

  • Is there a connection between the topic and your advisor’s topic that you can build upon?

  • Do you have some connection (e.g., via your advisor) to an expert on the topic, who can give feedback on your work every now and then?

All of these things can only be decided who knows your field or supervisor, respectively.

  • She actually agreed and liked my proposal in doing theoretical modeling on spray drying since one of her projects involve spray drying of nutraceuticals from fruit extracts. I have had some experience on theoretical modeling in some of my courses (transport phenomena and process control) but none of this scale. – Nicotinamide Aug 5 '16 at 9:15
  • Note that probably none of us can make a useful statement about your particular topic, as we are not familiar with it. This is something only you and your advisor can decide. – Wrzlprmft Aug 5 '16 at 9:18
  • Yes, I am aware of that and I am also fairly confident that I can see this thesis through. What I am afraid is if I can produce publishable results/papers since I'm basically on my own in doing it my way. Some of the projects in my university are largely experimental and thus most of the professors here are experimentalists by training. I'm doing the opposite. – Nicotinamide Aug 5 '16 at 9:38
  • Not only "experienced", but "well-informed" researcher? – paul garrett Aug 6 '16 at 22:28
  • @paulgarrett: Well-informed with respect to what? I fail to understand what your comment is aiming at. – Wrzlprmft Aug 7 '16 at 7:38
9

I'd stay away from that approach.

Reason: while the papers you consume will contain the cutting edge information, they will not contain the mundane basics of the field, i.e., things that "everybody" knows by practice, just by being around other people. Things that are too trivial to write down.

Ask yourself how important it is that you do a topic that is not common at your university. If you have such strong reasons for it, are they good enough for you to move to another uni (maybe abroad) where there are experts on it? If not, are there really not any experts around which have their own topics you can fit in with?

  • This answer raises very important points!!! All those unspoken things that "everybody (meaning "insiders") knows" are not easy to see in "the literature"... – paul garrett Aug 5 '16 at 21:17
6
  1. Yes, it is possible to learn everything by yourself without expert help, but it will be very inefficient.

  2. I'd pick another subject. Science is full of researchers and entire research groups that are doing qualitatively good work, but are just not at the cutting edge because they have an insufficiently developed overview of the field they are working on.

My prediction is that you will spend too much time learning new things, not enough time actually researching. You'll write something that might be good enough in the end (if you're good), but very likely it will not be anything groundbreaking, and probably will be something that has already been done before or turns out to be trivial.

My suggestion (as you obviously want to do something special) is to pick a normal topic, on which your supervisors are experts, and then try to expand from there if your initial steps are successful.

5

Beyond learning the state-of-the-art (something that university prepares you for), research worth publishing, especially outstanding one, requires a few other skills. For instance (from my experience):

  1. Can you identify the key relevant papers in the area where you plan to do research? An expert can point you the right direction.
  2. Can you identify the implicit rules in the area that you'll need to follow? Experts might not be able to identify them, but will at least tell you if you're violating them.
  3. Theoretical modeling picks some assumptions and approximations and develops the consequences of those. A good student can do the second part. But it can take more experience to understand well the first part and the implicit rules to follow for a paper (in relation to (2)). Otherwise, you might do perfect work and have its starting point challenged by reviewers.
  4. For outstanding research, can you identify the questions and approaches the community sees as worthwhile?

In my field (programming languages, across computer science and maths), some of those are things I learned during PhD and that aren't written or collected in many places, they're just folklore to experts and are transmitted from advisors to students in a system that resembles in some respects a medieval guild (without any malice implied). I understand that's common in research across fields: that's one reason this website exists, but it's of course not a complete advisor ;-).

0

Definitely Yes!

The way people do research is changing so rapidly that by the time you finish reading this answer it will probably be out of date. I'll make the intro short and go straight to what you need to solo your (theoretical) research project.

1. Time. When you work in a team of experts in your field, you get things done much quicker. If you work alone, everything takes much longer.

2. Money. You need money to buy time, mainly. If you need to pay off a student loan or work part time to pay the bills, you will probably not have enough time to finish reading all those articles and books in time.

3. Google Scholar. You can find full texts of most of the papers you will ever need there, and an abstract of virtually every paper.

4. Access to book torrent trackers. Legally, this is a grey area, but it puts everyone on the level playing field, whether you are a student from Harvard or some local college in India. Hey, there was a time when it was illegal for women to read. It still is in some countries.

5. Stack Overflow. Present your research results with working code, making it easily reviewable, replicable, and implementable for industry and other researchers. Leveraging the current tech will catapult you to the top.

6. E-mail spam. Once you have the results you suspect someone might be interested in, find them and let them know. These potential targets can be universities you want to apply to for a PhD program, companies that you want to work for, or journal editors that you suspect might be interested in publishing your work. Send a few emails per day, wait for their response, improve your letter (taking into consideration the responses of previous targets), keep sending your cold e-mails until you get a positive response.

There are so many tools out there for you to do the research independently nowadays, all you need is passion, a lot of time, and a little bit of money. Hey, Einstein figured out his theory of relativity in his patent shop without computers, Internet, Google, Stack Exchange, etc. etc..

Two more things I would like to add. The recent wave of progress is leaving a lot of "old school" researchers behind so you may not get a warm welcome when you cut a bunch of corners with your state-of-the-art tech and get their job with a higher salary just by punching a few keys on a keyboard and clicking some links. Another thing is, a lot of people are doing the same thing. People from South America to Europe, to Africa, to India, to Asia all of a sudden have all these tools to do research that for centuries was only accessible to the western elite. State-of-the-art doesn't stay state-of-the-art for too long so you got to keep up with it or you will be swept aside by the same force that you exploited on your way up.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.