Advisor pushing for dead end topics

Question

I am new to this forum. I am encountering a problem in my doctoral studies that I haven't been able to find in an existing thread so I wanted to ask here. Any advice on how to navigate this would be very welcome.

I'm 2.5 years into my doctoral studies (geography) working under two co-advisors. Their areas of expertise broadly overlap with my dissertation topic, but the topic is my own, and I know the study area better than both of them. One advisor has been pressing for me to conduct a certain methods for social data collection that I knew wouldn't work (research fatigue is high in this area, distrust towards scientists without existing connections in these communities is a huge - if not insurmountable - obstacle). I tried to explain that these issues would make this type of social data collection very hard to pull off. I suppose I failed at this communication, because he pressed for this method all the same (with the rebuttal "science is hard"). Because he's an advisor I assumed he knew something I didn't, and I wrote up a dissertation proposal with this approach being used to answer one of my research questions (in my program a "dissertation" consists of three publication-ready papers). I emailed a few groups engaged in work in these communities. The reply I got was what I expected: they strongly advised against attempting to collect household surveys for the reasons I had tried to explain to my advisor before. I just got this reply today, I have not yet discussed it with my advisors.

This does not sink my entire research proposal necessarily, but only a portion of it. There are other methods of data collection and analysis I'm still working on to answer the other research questions. Furthermore I spoke to my other co-advisor earlier in the fall (not a social scientist) asking what my recourse is if the social data collection component falls through for the reasons I feared. I was told not to worry necessarily, that we could figure out a way to work around it.

That being said, I'm still worried. I'm worried that the advisor who pushed for the an approach that was ultimately dead-on-arrival doesn't fully understand my research. He's suggested approaches in the past that have turned out to be dead ends too. I'm also worried that I wasn't able to recognize this as a problem until I am 2.5 years into my program. The heavy initial course-load (full graduate course-load for the first two years) and first few research dead ends had me seeing the individual trees, not the forest. I'm sure a lot of this is my fault; I should have communicated better, I should have been more clear on what the expectations were. But now that I'm waking up to the fact that this is a problem, I'm not sure what to do about it. I have (excuse the lack of modesty) a good topic, and I think there's still a lot of value in seeing it through. I suppose I will have a better idea of what to do about this when we all meet next week, but an outsider perspective would be really valuable to me right now.

Thank you in advance

Answer 1

One piece of useful advice my thesis supervisors gave me was:

We're going to throw a lot of ideas at you. You need to figure out which ones are worthwhile and which ones aren't.

If you know that something isn't going to work, then you need to figure out how to communicate to your advisor that it's not feasible. If you've tried and didn't succeed in making your case, try again using a different means of communication. Remember that the ability to make a convincing argument is crucial to success in research.

Answer 2

A few (hopefully) practical measures:

• Talk to your co-advisor about this, asking him to accept your position and, assuming he has, to call for a three-way discussion of this matter.
• This is not always possible, but if you can - schedule the dead-end work last, i.e. after other endeavors about whose success you're more certain. Why do this? More opportunities to change the advisor's mind; time for him to learn to better value your judgment; or perhaps even time for you to theoretically reconsider your opinion.
• Regrading your "failure at communication" I: Write up your objection - Write either a methodological survey paper, or something less formal such as a blog post, arguing the shortcomings/dead-endedness of the collection method.
• Regrading your "failure at communication" II: It sounds like you feel ashamed/afraid/embarrassed to bring up the subject again with your advisor after it was supposedly settled. Try to accept the danger of appearing foolish/obstinate etc. and have another go at convincing your advisor, being better prepared for it, perhaps with material you can pull from some folder you've compiled etc.
• Consider whether you can't do research on the failing methodology. You could employ both collection method A (the poor one you disapprove of) and collection method B (a better one), documenting your preparations for both and the process of collection, more closely than you would do when interested in just the results. Then, besides a paper about the actual subject matter, you could write a methodological paper about the comparative failings of the method you disapprove of. In other words, make lemonade from the lemons.
• Consult your advisor's other graduate students, current or former, about this matter. Maybe they've had similar issues and can share their experience of how they handled it. Hey, maybe one of them has had the exact same issue regarding data collection, which would help you even more.

It goes without saying you shouldn't try all of these at the same time (but some of these do fit well together).

Answer 3

If you have not collected the data, and figured out your proposed method will not work, do not use that method. If you have used it, and the usage of it revealed the difficulties, report this in your study so that it can benefit others. This can make this part of your thesis weaker but not the whole thesis, I guess. You still can make the other two questions stronger.

Answer 4

Not all advisors are up to the task, and sometimes they are wrong. I've personally encountered situations where advisors were wrong and guided (or misguided) their students down a failing path. Depending on your university, and your financial situation (are you beholden to a particular advisor for your grant funding?) it might be necessary to find another advisor. You indicated you already have two, perhaps a 3rd?

That said, remember your goal is to get the PhD. Share this feedback from the outside groups with both of your advisors and discuss it with them. I'd start first with the one more likely to agree with you, then the other. If he/she still insists, then politely toss the ball into their court: "Professor, I don't understand how this (survey stuff) is going to work given the limited time and resources, can you please show me how we can do this." I used "we" on purpose - it draws them in as a collaborator not an opponent in an argument.

Part of the PhD process is learning how to manage people, your advisors in this case.

Also consider the possibility your advisor is correct. It's not easy for the student to call out the master, and you have to be pretty darn sure if you do. If the advisor is wrong, find a way for him/her to gracefully back down rather than challenging them directly.

Been there, done that.

Answer 5

Some further practical points in addition to @einpoklum's answer, in particular the advise of having another go at convincing the advisor (and too long for a comment)

Directly trying to convince your advisor (who is convinced of the opposite) may just result in more communication chaos. So instead, I'd

• ask the advisor for an appointment to discuss (again) the choice of method: Ask the supervisor to let you first present your point of view so that once they know your understanding of the situation they can convince you of their point of view.
• present your arguments
• keep in mind that practical arguments are not necessarily scientific arguments. IMHO, they should be marked as such and then practicality considered separately from scientific arguments*.
  Practical arguments against doing something share the disadvangate for you with negative results that there's always the latent question lurking of you being lazy or not good enough. So make sure your presentation of arguments will not look as trying to avoid hard work.
• Now listen to their arguments.
• Actually, you should also prepare a list with the advantages of the method. Doing this in preparation for the discussion would put you into a much stronger position, and it is actually an important working technique to be able to identify advantages of methods you dislike. Take the role of the Devil's Advocate.
• The important point for the discussion with your advisor is: you need to be willing to be convinced! Be curious how they come to the opposite conclusion you came to.
  Otherwise, no communication to solve the scientific/practical problem can happen.

---

* I have in mind several situations where computational feasibility was a question. I found that quantifying the effort goes a long way: computation time of 1 h (single occasion) may be unpleasant in an interactive session, but is certainly feasible in general. A final calculation taking a couple of months on a desktop computer would IMHO be quite inside what can be expected for a PhD thesis (I mostly deal with calculations where a "sneak preview" will give you a rough idea what is going on and then lots of computational effort goes into refining the results). So would be a couple of months of unpleasant final data collection given that pilot experiments/a small pilot survey shows that this is actually the way to go with the data collection. Whereas a computation that would take several years of Rosetta@home would clearly be outside a single PhD thesis and there would be no discussion whether this is feasible or not.

Answer 6

Building on cbeleites’s answer - ask for a couple of recent examples relevant to your work where your advisor’s approach was used to good effect.

It may be that your advisor has more insight on how to do this properly, and you can learn something from the examples. In that case, your advisor would likely be able to rattle off a few references off the top of his head. Studying those examples should give you some idea of how to do what your advisor is advising (which, incidentally, may not be exactly what you’re objecting to).

On the other hand, it may be that your advisor is too busy / is distracted / etc, and makes the same suggestion to all his students. In that case, it might be harder for him to come up with relevant positive examples, and you can then suggest a few relevant positive examples you’ve found that follow your preferred methodology. (Naturally, you’ll have to do some work to find those examples first.)