I don't have a thesis topic!

[standing_ovation]

I want to apply to PhD programs in Biostatistics, but I've been hearing that I should already have well-developed thesis ideas. I don't. There are areas of statistics that I'd like to learn more about, but I can't say that I have a specific thesis idea. I feel I have a lot more learning to do before I could have an intelligent thesis topic. I have an MS in Statistics, but there was no masters' thesis, so I didn't really gain "expertise" in any particular topic.

 

I've also been hearing that I should email potential advisors and tell them about my ideas, which freaks me out since I don't have a thesis idea! It seems like a terrible catch-22. How can I become savvy enough to have a viable thesis idea unless I get to take PhD-level classes? And how can I get to take PhD-level classes if I can't get into a program until I have a thesis topic?

 

Other than this (major) problem, I think I'm a good candidate – very high GPA and GRE scores, brand-name education, first-author publications, work experience, great references. But at this point I'm starting to feel thoroughly overwhelmed by the prospect of coming up with a thesis topic.

 

I'd really appreciate any advice. Thanks.

[victorydance]

Wait you have a first author publication in a journal and can't come up with a basic thesis topic? That makes no sense at all.

 

It doesn't have to be incredibly detailed. Just something that is original, can be tied to other recent publications in the field, and you can ask a couple questions about in your SOP.

[standing_ovation]

The publications aren't in stats. They're in various other fields. I was involved in these projects either as an independent investigator or as a statistician. There was stats involved, obviously, but the papers aren't about stats.

[victorydance]

What were your favourite classes, what are your favourite papers in biostats? 

Edited September 3, 2014 by victorydance

[GeoDUDE!]

Wait you have a first author publication in a journal and can't come up with a basic thesis topic? That makes no sense at all.

 

It doesn't have to be incredibly detailed. Just something that is original, can be tied to other recent publications in the field, and you can ask a couple questions about in your SOP.

 

 

Why doesn't that make sense? Perhaps you have published a paper, but in general publishing a paper (say from your undergrad thesis) doesn't necessarily mean that you came up with the idea on your own; thats what advisors are for! Often times, young STEM majors are largely 'technicians' doing all the programming, mathematics, data collection, statistical analysis ect, but the advisor often has to come and give context to the project. Even on your first (and maybe 2nd) first author publications during a PhD this is true, where the third (and often final) publication tends to be derivative of the first two but totally original with less input from the advisor.  In the stem field, and you are a professor, and your student spent many hours working on a project, it is immoral to not give your student first author. Even if much of the intellectual direction is controlled by you (the professor). This is an unwritten rule about being a good advisor. What is important, is that you can develop original research topics (fully developed) at the end of your PhD. Not before you even start it.

 

Long story short, young scientists who publish first author even in high impact journals still probably need help coming up with a well developed topic. 

 

That being said, it is important to show POIs and the adcomms that you are thinking about potential topics. Perhaps even reading papers in potential fields. But your PhD is likely going to be funded by an NSF grant, and complete that grant you will. You cannot research elementary particles when your grant funds you to research the behavior of quadriplegic cannibals with ebola. 

 

What is important is to show promising thoughts, not well developed tops for your dissertation. That would be absurd for a very strong majority of PhD students. 

[victorydance]

Why doesn't that make sense? Perhaps you have published a paper, but in general publishing a paper (say from your undergrad thesis) doesn't necessarily mean that you came up with the idea on your own; thats what advisors are for! 

 

Well I think being involved in a paper from inception to publishing, regardless of whether or not you were the theoretical mastermind behind it should give you the tools to come up with a general thesis topic. And of course, no one said you can't get your past professors or letter writers to look over this part of the SOP to help you, I certainly have.

 

It doesn't have to be incredibly detailed nor some ground-breaking idea because that's not what is needed for a SOP.

 

That being said, I think it is probably the most difficult part of writing a SOP. Perhaps not showing that you know the literature and setting up the idea, but succinctly detailing how you would go about solving that puzzle with proper knowledge of the field definitely. It's something that I am kind of struggling with myself on my SOPs. I have a topic because I spent the better part of my senior year doing research on it, but detailing it in an SOP can be challenging.

Edited September 3, 2014 by victorydance

[juilletmercredi]

I don't think you have to detail how you would go about doing the project.  You just have to come up with the idea.  My applications also asked me to propose a potential dissertation topic.  I think I spent about a sentence or two on it.  Nobody expects you to have a detailed and nuanced perspective on what you might study; they just want a general idea of what you're interested in and want to see whether you are at least able to formulate an idea.  I think this skill has little to do with whether you've published or not, and has more to do with your undergrad education and research experience - it should've at least prepared you for this much.

 

Also, remember that nobody is going to hold you to those 1-2 sentences from your SoP.  I did something completely different from what I proposed in my SoP.  I realized 1) the resources just weren't at my current university to do it and 2) I didn't really want to do that anyway, as I discovered new interests in grad school.  The idea doesn't need to be what you actually are going to do; it just needs to be something that is, potentially, in a perfect world, feasible to do as a dissertation project.

[Catria]

I wonder what are the biostats standards to this end...

[victorydance]

Yeah I don't know.

 

My discipline places a great deal of emphasis on the SOP. Particularly your ability to craft researchable questions and lay out a framework to solve them. My discipline expects you to spend up to one half of your SOP due to this endeavour alone. And my letter writers have been really drilling this part of my SOP drafts. 

 

I don't know what the norm is in biostats.

Edited September 5, 2014 by victorydance

[ginagirl]

I definitely don't think anyone expects you to have a thesis idea at this point. In some ways, having a too-specific research direction may be detrimental to your application process if a program doesn't feel they can support you in that particular area. All a program would expect from you at this point is an interest in the field of biostats and maybe just some general area that you are interested in, whether it might be more theoretical or more applied.

The publications aren't in stats. They're in various other fields. I was involved in these projects either as an independent investigator or as a statistician. There was stats involved, obviously, but the papers aren't about stats.

In response to this, any statistical work you do is the work of a statistician and is relevant. As a biostatistician you may be expected to work on any number of projects in a variety of fields. That is the nature biostatistics. At this point it's about deciding if what the work you had done previously is at all what you are interested in or if you would like to explore a new area of application. A program would be interested in hearing about that too.

As an aside, I'm not sure who exactly was telling you that you needed to propose a thesis topic either in your SOP or to faculty. Unless the program asks you specifically, I would say that you don't need to force one if you don't have one. I'm sure some people do have grand plans and address it in the SOP or to faculty, and that's fine, but this is about YOU and what YOU have to offer the program. It's their job to decide if what you are bringing to the table is what they are looking for. There is no sense in forcing/faking some idea and then possibly eating your words after they admit you to the program when you can't back it up. Not everyone can offer a thesis topic at this point, and I think that's fine.

Edited September 5, 2014 by ginagirl

[bsharpe269]

You arent supposed to have a thesis title chosen and email professors saying "I want to study this exactly, would you be inerested in advising me?" If you were at that point then it might even hurt you since PIs need grad students who are open to a variety of topics  and flexible enough that they can mold their ideas to fit within the PIs interests and grants. At this point though you should be able to say that you are interested in a specific subfield(s) and would be interested in learning more about X,Y, and Z types of topics.

 

For example, I am interested in a whole subfield of concepts right now and have tons of research experience. I know enough about my interests and subfield that I can tell you that I am more interested in the theory side and less in the computational side. I could list all of the well know PIs in my subfield, tell you who is speaking and leading the big conferences and then I could divide these people in those who fit with my interests and those who dont. Within those who fit, I can give examples of the projects they are working on that I would really enjoy working on too. However, I dont think that I could come up with thesis topic idea (well maybe ideas, but probably not really good ones) that these PIs should explore next. Over the first few years of my phd, I hope to get my PI's help in guiding my interests towards a specific area before I have to do this. I think that this is above where you should be when applying to PhD programs. You don't need to be ready to start your dissertation yet.

Edited September 5, 2014 by bsharpe269

[standing_ovation]

Thanks everyone. This calms my nerves a lot. Sounds like I can go with giving a little information about my areas of interest within statistics, even if they aren't exactly a topic yet.