How to Spot a Dead-End Research Topic Early (Full Transcript)

[00:00:00] Speaker 1: Hey, Professor Stuckler here. One of the worst tragedies I see in early-stage researchers is when they work incredibly hard on a research question, read hundreds of papers, write multiple drafts, even do a technically sound analysis, only to realize far too late in the process that their topic was actually dead on arrival. And here's the worst part. On the surface, most of these dead-end topics seem okay. They're not unworkable. You can do them. They're feasible. They're just quietly unpublishable. There's no real contribution, no meaningful debate, and what that means is there's no realistic path to publication. At that point, with a dead-end topic, it doesn't matter how well-written or crafted or technically executed it is. It's like the British say, you can't polish a turd. So in this video, I'm going to show you where these dead ends creep into the research journey and how you can spot some of the most dangerous ones before you lose months or even years of work. So if you're choosing a topic, trying to narrow your research question, or just unsure whether your current idea is publishable, this video is for you. I'm also going to share with you two topic validation texts that we actually use inside our FastTrack Mentorship Program that you can implement today to avoid this dead-end topic trap. If you're new to the channel, I'm Professor David Stuckler, and I've published over 400 papers in high-impact peer-reviewed journals and been a professor at Harvard, Oxford, and Cambridge. This channel is all about helping you publish in high-impact journals and avoid some of the dead ends that I fell into myself multiple times. It's not about how bright you are. It's just that a lot of this stuff, this real-world practical advice, is left implicit. You're left to figure it all out on your own. So I'm sharing with you the support that I wish I would have had when I was just starting out. So let's dive straight in. Getting your topic right is probably the most important thing you can do as a researcher. I commonly say about 95% of your ultimate publishing success comes down to your choice of topic. I would rather see you have a great topic but a weaker, less polished analysis than an incredibly weak topic that's masterfully done. Because the second topic doesn't have wings. It's not going anywhere. And that's where I see even brilliant students fall into this trap. They often leave topic selection to kind of vague, intuitive process. Or maybe they inherit something from a supervisor that they never really probe, test, or validate to make sure it's publishable before going too far. So sometimes this can be okay if you just want to tick a box for a PhD. But it can absolutely be a dead end if you're looking to catapult your career and get on that proverbial fast track by publishing. And it's really true for academics. Publishing is like money in the bank. You need to do it well. You need to do it routinely and get in the habit of getting into high-impact journals. And it all starts with your topic. So not all dead ends here in research are the same. So some can happen early when you're still vague and unfocused. That's okay. That's frustrating. That's fixable. I'd rather intervene early and have a correction here. The most dangerous dead ends happen when you, and this is where it's slippery, is when your topic is clearly defined feasible, executable, but not publishable. That is the one that costs people years and leads to the most frustrating situation I see when they have a perfectly executed manuscript that gets desk rejected over and over and over and they don't know what's wrong. And I hate to bring the bad news to come back and say that this just never had a chance from the very beginning. So let's go over how you can avoid this yourself. And we'll go through how people actually come up with topics themselves so you can see where in the flow the problem happens and how you yourself can course correct before going too far. So you've probably been here. A lot of people start in a topic neighborhood, not with a crisp well-defined topic. So they might say something like, oh, I'm interested in AI and education or I want to study mental health. I'm interested in gender inequality. I think that's a really good starting point to start with something you're passionate about. And I often do see people drift over time away from their passion. So I think it is really helpful because you're going to be working so hard on a topic. You're investing in yourself and your knowledge and also signaling to the field what you're capable of in a space. So make sure you start from your passion. Don't drift from that. This broad topic space, that's not a mistake. That's a normal starting point. To get there, we typically use our convergence method, something you can see in another video, to help you land on this topic space. But for now, I'm just going to assume you've already got it. So you're in the right neighborhood. It's important to know a neighborhood is not a paper and it doesn't mean you've got a publishable contribution. The problems can start to arise when people try to go from this broad neighborhood into executing something without checking whether the topic can actually carry a paper. But usually here, they're going to hit a dead end much faster and just realizing that the topic's unworkable. They're going to drown if they're doing a review. They're just going to drown in papers because it's too broad and not well defined or they won't really be able to make clear methodological decisions to implement the topic. Yes, this is a dead end, but it's not as dangerous as the dead ends that come at the next stage of topic development. So here's at step two where things start to feel like real research. You've got to, at this stage, define the boundaries of your topic and not just describe it in a broad topic space. One tool you can use for doing this is a framework that we call PICO. It was born out of medicine and systematic reviews, but it works especially well for really any field, especially well for quantitative research. And the PICO framework really is the elements of a well-defined topic. So P is for population and that population, it can be firms, can be documents you're looking at. It's often people. It can be ethnic minorities. It can be a geographic region. Sometimes this can be a context like a setting in hospitals. There are different refinements to this PICO, but these are the nuts and bolts I want to go through. I is an intervention or exposure. Often, especially if you're doing quantitative research, you might have a left-hand variable in your model or an X. And the core of a lot of social science, even natural science, is we're looking at the effect of something on something else. That's what this I category is capturing. Comparison. Implicitly, there are sometimes comparisons being made when you develop something called a counterfactual. You want to causally describe what's happening in the world. I won't go too much into that here. This is not always necessary to define, but it can be helpful to think about, especially if you're wanting to do causal research. And O is outcome. You're looking at the effect of something on something else or a Y variable in quantitative research. And that's really important to describe as well to really complete the boundaries of your project. Imagine it's like if you were trying to play football, but you construct your boundaries very narrowly. Well, that's going to be a very boring football game to watch. Not much traction is going to happen there. But if you define your boundaries very vast on a football field that's very open, it's also going to be a very boring game to watch because nobody's going to score. Anyway, the point is you need these boundaries to help you define what's in and out and really get the scope of your contribution, right? Two other elements I want to add here that can be useful. If you're doing quant work, especially you might want a T for time period to be added and a D for your research design. Just a quick interruption from this video by today's sponsor. Me. I want to share with you something that's incredibly exciting and you're going to love. If you want to work with a real researcher, a real professor to get feedback on your work and to save time and optimize your success from publication, click the link below because what I'm doing is I am taking 10 researchers who I'm going to work intimately with and go so far as to offer a personal publication guarantee. That is, if you show up, you work with me, you do the research, not going to leave you hanging. I'm going to work with you directly to the end until your paper gets to the finish line. Again, keeping that small and intimate and only opening that up for a select few researchers to have the opportunity to work with me. If that's of interest to you, you want to work with a real person, not AI, click the link below. Let's jump on a call and see if you could be a good fit. So here's the trap. Well, firstly, a lot of people gloss over this step to get this crystal clear clarity on their topic. So I do encourage you to use a Pico. There are other models out there. We just find in our fast track mentorship program that this is particularly helpful to make sure that you've got the logic clear that helps prevent you from drifting later, which is another kind of dead end drifting from your topic into something else. And you wake up and find, wait, this isn't even what I wanted to do. But what's more dangerous here is that even if you can't get this defined, people sometimes relax and feel some relief. But that can be dangerous if you don't take the next steps I'm going to share with you, because just because now you've got a Pico that can be technically feasible, you've got a workable topic still does not mean that it's publishable. And this is often where people start. They've got a clear topic. They run with it. They go the distance and they hit the dangerous dead end to only discover later that they're getting desk reject after desk reject. So here's what you have to do. Critical step. And we call this our, I mean, this step saves months or years, and it's so basic and so obvious, and yet people don't do it. And that's our nearest neighbor paper test. You want to take a validation step to find the paper that is conceptually, can also be methodologically, closest to yours. So look for a paper, just go into Google Scholar, find a paper that's closest to yours in the topic, the method, or the population, or other elements of your Pico. And you need to ask a brutal question. What does my paper do that goes over and above this one? What do I do to move the field forward? Implicitly, you're defining the gap here in your value add to the field. And this first critical test is sometimes what we call a duplication test, because at the worst, you're just duplicating a paper that's already been done. Or you might find an honest answer when you look at this is that, well, yeah, my topic is similar, but it's only kind of slightly tweaking something. Maybe it's the same idea, but I'm just doing it in another population or another country. It's just I'm replicating with minor tweaks. And that's really where the value is. You need to know what that value add is before you start. It's here where if you take this pause, just take this breath before going too far, that it can dawn on you that my value add is so weak that even if I can deliver on the topic, it's structurally unpublishable. Not that it's wrong. It could be you're not tapping into a main debate, or so it's peripheral, or it just doesn't have value add. And this problem is getting worse with AI. So I see people going to AI to get a, they might go to AI with a topic neighborhood. And AI then recommends some weak, unvalidated topics. It often recommends things that won't pass this duplication test. And so I've seen researchers run. I had one doing a project on physical activity and sleep. She had implemented the paper, and we said, I said, hang on, let's do our nearest neighbor test. And we found three papers that did exactly already what she wanted to do. And if she sent out the paper, and it actually got reviewed, it would probably go to those reviewers who would say, well, you're just, you haven't even decided this properly, which is a problem, and you're just duplicating. And I'm seeing this slipperiness happen with AI that helps people get past the early frictions where they'd hit an early dead end and go down the slippery slope of a topic that is actually a dead end topic. They don't know it, only discover it later. So suppose you've done this and you find, oh, oh no, my topic actually is a dead end. Then the common reaction is to narrow to something that feels safer to find a gap. So they take the Pico control knobs and narrow the population, or they start narrowing the outcome when they sense that, okay, there's no gap here. And so they narrow to something that feels safe, narrowing the scope, because maybe that feels like the contribution is a little more precise. But here's the paradox. When you over narrow here, instead of strengthening your gap, strengthening your value add to the field, you often drift further from the main debate and make your topic, yes, feasible again, but too weak to matter. So common way I see this, if you do a review paper on barriers to maternal healthcare in Zambia, well, you'll run into a problem if there won't be enough papers to review, but you've also shrunk the audience who is naturally going to be interested in your paper. Now, if you are going to do a case study and shrink to safety, you just need to justify, it could be say you're doing a quantitative study or a natural experiment in Zambia, then you just need to be able to justify why is the wider field going to care about Zambia. And it's not that Zambia is unimportant, it's just unfortunately there is a western-centric bias to a lot of top-tier journals, and they want to see that a wide general readership is going to be interested in your findings. So if you are going to take this approach of trying to shrink to safety, just be careful that you haven't shrunk the debate and made your paper so peripheral, even if you have a gap, it's such a weak gap, so it's so peripheral as to be irrelevant. The final step I want to share with you that's really important is how you can actually forecast your impact. So you can go in with realistic expectations and validate not just the publishability, but maybe even the ultimate citability of your work. Try to gauge its impact. Remember, you're investing a lot of time and energy into this topic space. Instead of diving straight in, I mean, take that, you know, five, ten percent of your time to really establish, cement, and validate the topic. There's this important principle in science that we need to talk about for a second called the Matthew effect. It's basically the rich get richer, and that is absolutely the case in science. There are studies out there that show those who win early grants have much more success than those who are very close, just miss the cutoff, and didn't get a grant success, but get success. And there's also an important halo effect where people who are at top institutions, maybe work with supervisors who are more successful, also tend to be more successful. And this inequality in the academic world is something I'm very passionate about breaking down. I want to democratize this implicit knowledge that was passed on to me at elite institutions. Now I'm trying to pass it on to you, but this Matthew effect is very, very real. And so what can commonly happen on the topic step is you might be a great well-intentioned researcher, but perhaps you have a weaker supervisor than somebody who has a world-famous rock star supervisor. And that famous supervisor is going to be right at the center of debates, have a thriving research agenda, and often will hand their mentees a topic that is going to be a winning top. Where if you have a supervisor who maybe, maybe I see several who have supervisors who haven't even published in top-tier journal themselves, it's going to be very hard for them to hand you a topic that is going to hit a home run or be a fantastically winning topic. Not everybody wins a supervisor lottery, but we can short circuit that. I can help you. And this is where our second important test comes in. And this is where it's a calibration test, and you're going to forecast your impact. And it comes back to that critical nearest neighbor paper. And that's because the closest predictor of your paper's impact is going to be the impact of papers that are similar to it. So look back at your nearest neighbor paper. When you look into Google Scholar, you'll often see how cited was that paper? Did it get picked up in a debate? Or did it just disappear quietly? Remember, the median paper that's published today in the first two years gets a median citation count of zero. That's just the median paper. I want you guys to be a lot better than the median. So you want to look at how well your nearest neighbor's paper citation, how well it's getting picked up. And if you just see crickets, just silence in the literature, it's probably a sign your topic is going to be a dead end, just like that other paper you're looking at. So if you see your surrounding literature has no traction, no debate, very little audience, it's not bad luck. It's just bad topic selection. Listen, guys, this is the failure mode and research that I hate to see because it's not about motivation. It's not about intelligence. It's not about skill. It's just making the wrong decision early and doubling down on it and realizing there's just no juice that you can squeeze from a dead tree. For me, it's one of the most tragic errors. It's not that the topic was impossible, but exactly because it was possible, but irrelevant. Again, to repeat, once you get the topic right, everything just becomes so much easier. And if you don't, again, you can't polish no amount of effort is going to save it. Listen, if you want our help implementing these tests and more, we keep our best stuff inside FastTrack where we actually work together. If you want that help before you invest months into a dead end topic, I'd encourage you to click the link below, see how we work in research and if it resonates, and if so, apply to join us. But if you take nothing else away from this video, remember, choose your topic like your PhD depends on it because it really does. See you in the next video, guys.