How to Choose the Right Type of Literature Review (Full Transcript)

[00:00:00] Speaker 1: should you do? It's a question as a professor I commonly get asked. Should I do a narrative review, a scoping review, systematic review, critical review, indicative review, systematic review? It can really be bewildering and seem mystifying when you're just starting out, but not today. We're gonna help you wade through the different types of lit reviews, take the confusion out of the mix, and help you stay focused on what it is you're really trying to achieve with your literature review, and from there, pick from a menu of options which type is gonna best get you to where you need to go in your research journey right now. So this is gonna be relevant to you if you're just starting out, maybe you're grappling with a lit review, you don't even know what type of lit review you're doing. It's also gonna be helpful if you're in the thick of doing a review and you've been getting stuck, because sometimes what you'll come to find is getting stuck in your review means maybe you didn't have clarity on an earlier foundational step that you just kind of glossed over along the way. I see this especially happening commonly with people who, in absence of feedback, or maybe not getting the support they need, outsource research judgment to AI that takes them down a rabbit hole. For those of you who are new to the channel and to these live sessions, I'm Professor David Stuckler. I've been a professor at Harvard, Oxford, and Cambridge, and this channel is about providing you the support that I wish I would have had when I was just starting out. Because frankly, I learned the hard way. I learned by making mistakes, and that's a way to learn. But the faster way to learn is to ask somebody who's been there and done it to show you how. You'll get to the same place just a lot faster. Now I've provided that benefit one-to-one to my students in the Ivy League over the years. Now I want to make that open access and available to everyone, and that really is the goal of FastTrack and our programs, to make that implicit research logic that's often handed down, explicit, and available for all. As ever, we're going to take your questions. We have several that you've submitted at the end of the session. If you have some as well, I always check the chat in the comments, so do drop them there. You can also see them off-site today, so I do apologize in advance if we have any hiccups, but I've decided this year rather than wait for the perfect recording setup, I want to protect this time that we have every Friday at 4 o'clock Central European Time. That's 9 a.m. US Central and 3 o'clock UK, so that you can consistently know I'm going to show up and provide feedback to you on your research, as well as cover themes that you suggest throughout the week you'd like me to cover. So think of this really as your time. So with that, what I'm going to do is I'm going to pull up a whiteboard, and let's dive straight in. I'm going to say, yeah, and I can see Israt here. Israt's asking how to choose a good topic, and we just covered this a few sessions ago. So guys, we have a lot of resources on my YouTube channel, and if you go through the live sessions, you'll see a lot of live sessions on topics. You'll also see videos that are pre-recorded videos using our two-stage method for finding topics, identifying gaps, and helping to confirm if it's a winning topic using Pico models, nearest neighbors, other tools that we use inside our FastTrack systems. So check that out, because it'll go into more depth on the topic choice than I can get into today. But we do have some questions at the time at the end, Israt, so I may be able to come back to that. And Akari, hey, welcome. We're Truly International. If you're on Team Replay, by the way, give us a like. That helps the algorithm find people who might benefit from this message and wouldn't otherwise get it. So the algorithm helps us to reach and serve other early-stage researchers like many of you. Okay, let me go ahead and share my screen. I love pulling up whiteboards because I find the people who I work with tend to be quite visible. Visible online, but they tend to be quite visual in terms of how they think and approach research. So first, I think it'll be helpful to just demystify the types of reviews. So some of you, if you're doing the most basic kind of literature review, the more traditional one is sometimes called a narrative lit review. And this is what you may have been more familiar with as an undergrad. You might have had a topic or theme and you start googling around for stuff and cobbling it together. And sometimes in a way you're just winging it to put together an argument. That's well and good, but at the PhD level, a lit review is much bigger and more formalized. Some of you get asked to do a literature review at different points in your journey for different purposes. So the most common starting point is someone's trying to figure out a topic and their supervisor says, oh we'll go do a lit review and come back to me. And so they're doing the lit review to try to figure out their their topic. Not necessarily to publish, but to just get a feel, kind of the lay of the land, understand what's out there. Other purposes is it can be required. It's part of a research proposal. It's a chapter inside your PhD. And whichever reason you've been pulled into doing a literature review, and you're almost always gonna have to do a lit review, it's helpful to know which style you're doing. So narrative is one that's really, it doesn't really have a built-in structure. So I'm going to put in some attributes. It has no built-in structure. You're gonna have to make the structure yourself. And I think this is why some people get lost. Because they just dive in and they're kind of muddling around, finding some articles, reading that, finding other articles. And sometimes then they're finding, oh I'm not finding enough articles, or I'm finding too many. And because there's no built-in structure. And you have to define that structure yourself. And so in any literature review, you need to be aware that there is kind of a funnel-like structure. This is also almost why you need to define, understand why are you doing this literature review for yourself. Is this for my general knowledge? Is it for my supervisor? Is it to find gaps? Is it to find topics? Typically what you're trying to do with a literature review, is you're trying to justify, sorry for all the caps guys, justify the reason that your study needs to exist. And so that's probably the most basic reasoning for a literature review. So what that already means is, lit review is not what a lot of people think of it as. A lot of people think it's just a summary. But it's not. It's actually a strategic argument for something. And I think the lit review is a bit of a misnomer. But for that reason. So think of it as a strategic argument. And the way to think about this is a lit review is going to follow a funnel shape. So we're gonna have to get into some basic comments about a literature review. But it follows a funnel shape, in the sense that you need to go one level up. And sometimes you need to go broader up here. So you're gonna read a little studies that are a little bit wider. By the end, you spew out this justification for your study. Which might involve setting out what a gap is, and maybe your research questions. That's going to bridge over to your methods later on. Now, this narrowing structure of the funnel, where you go broad, to more precise, to ultimately very specific about what you want to do. That's where you have a narrative arc in your lit review. And so this is what a traditional narrative summary is not systematic, doesn't have a built in structure. And it's not aiming to be reproducible, either. So you're going to go around and read different things in the literature. But it's not like just just looking at the chat, Akari does this one way, and then Isra would be able to reproduce exactly what verbatim what Akari did. Hope that that makes sense. So, so this is very, I can see Flint who's asking, is this a live presentation? Yep, we're live, aren't you? So shoot questions away. These are always, always live. These are not pre-recorded, like the rest of my channel. So these are two features of a narrative literature review. Often, these are, I find, harder to publish. And they're often the preserve of experts in the field, especially in natural science. So sometimes people want to hear more your opinion, and you're thinking about the literature, just because you're an expert, and you have a perspective. And that expert opinion is quite valuable. But if you're just starting out, I find these are very hard to construct. So I'd say these are the most common, but they're also the most difficult. Because of this lack of built-in structure, they're the hardest to publish, largely because they're not reproducible. And they're often, often they're invited by editors. If you are going to do this kind of literature review, you need to be very clear about what you're trying to achieve with it, and where you're trying to get to. And you need to make sure you build in a structure by having a funnel. And if you know where you want to end up, in terms of justifying your study's existence, it's much, much harder to get lost. And it makes it clearer, because as you start summarizing literature, you can start evaluating it, saying, does this help me justify the need for my study? If not, it's interesting, but it doesn't go along. And that will solve one of the biggest problems of this kind of lit review. If you guys have any questions, if you're doing a narrative lit review, let me know in the comments. Just comment narrative. Yep, I'm doing that. If you don't know what kind of lit review you're doing, hopefully by the end, this will be much clearer. But I'd say the vast majority of you who don't know what type of lit review you're doing, are probably doing a narrative literature review. Okay, the next kind of most common type really falls into two kinds of categories, and they share a lot of similarities, is scoping and systematic review. These are two separate types, but I lump them together because they're quite similar in practice. And the difference here is that they are going to have built in structure, and they are going to be reproducible, they're going to be easier to publish, and they're typically not invited. So it has these features. So these will go so far. And the systematic review, you actually have to pre register your method, like you would do with a trial study. And your methods have to be incredibly robust. Systematic reviews will often also have something called a formalized quality assessment, where you not just evaluate the literature and draw conclusions, you actually show how you've systematically gone through and assess, well, these studies were weak, these were strong, and try to balance your conclusions on the basis of the strength that you've estimated, often using a tool of the literature. So I like these a lot, because they just by following steps, they actually have guidelines and step by step protocols, they force you to follow a final process, just by following the steps that are laid out. That said, if you need something quick and dirty, say your supervisor just wants you to find some gaps, go get familiar with the literature review, well, here narrative is okay, but you just need a quick and dirty, maybe summary to get to explaining why, why you've got a gap. But if you want to publish, these are definitely the way to go. I have a personal bent, I prefer systematic reviews. Because of the pre registration, if somebody ever asked, you can turn a systematic review into a scoping review, but you can't go the other way around. And systematic reviews just have a little more gravitas, they're a little more respected. So I always prefer starting there. I've had several researchers, one you can find on my channel, if you go into some testimonials about our programs. Her name is Nehal, a doctor in Ireland. She started off doing systematic review, reviewers asked to convert it to scoping, no problem, converted to scoping during the revision, got published like that. The main difference for them comes with pre registration. And I won't get into this, you can see this in other videos, and what tool you use to define your topic. By the way, guys, I highly recommend you use tools to define your topic, it just is kind of an extra check. And I find researchers I work with like checklists and gates, so they know if they've done it right. But yeah, scoping review is going to use something called PCC. And systematic review is going to use Pico. Don't worry about these outcomes. Now, you can Google them later, and you can find them in other videos on my channel. So then there's a third type here, that sometimes people ask me, what's this? I don't know what this is. And then just not knowing the terms, people get worried about things that are thrown out there that they see or don't know. And they sometimes see something called an umbrella review. And this is a very specific type of review, called a which is basically a review of reviews. So this can be really highly used when there's dense, dense literature. And there's maybe a lot out there. And maybe there's a bunch of scoping and systematic reviews already. And you want to review those reviews. And that's why it's called an umbrella, because it's funneling in all these different reviews. And so what's different here is often in your systematic and scoping review, you're intentionally excluding reviews and only going for original evidence. Umbrella review does the opposite. It only wants the reviews themselves. So it's a review of reviews. I won't get into that. This is a little bit rare, and I don't recommend it. But I want you to be aware of what it is. Then you've got a fourth kind of miscellaneous mismatch category. And I think people get really confused on this that they sometimes see where there's lots of quirky terms for things like there could be a critical review. There can be a integrative review. You guys might even come up with some other quirky names. They're much less common. And so if you search, say, take one of the big databases like Web of Science, which is a general repository, and just search for all critical review, or all integrative review, you'll see the numbers overall in those databases of what's like the universe of everything being published is much, much smaller. These are sometimes niche to certain fields. So critical review can sometimes be more in critical theory. Sometimes people say critical review, if they really want to emphasize the nature that they're critiquing the literature. But you should really be critiquing the literature in a narrative or a scoping or systematic review anyway. So I don't find critical review says that that much. Same thing with integrative review, they want to emphasize that maybe they're integrating dispersing narratives or different fields, but you should be doing that kind of synthesis in your analysis elsewhere. So what I find is that these are the main types, but you have some other names that are out there like these, and you probably might be aware of others, maybe we drop those in the chat. If I haven't covered one, there's a bunch of these quirky ones that are less common. But typically, I find these are actually in structure, just a narrative lit review. Now, the second thing we have to understand about the lit reviews that you superimpose here is how you do the analysis itself. So in a lit review, in a way, if you're doing a qualitative study, your data are from interviews, or a quantitative study, you have numerical data, maybe from a survey, or some data you've collected. Well, in a lit review, your data are the research articles themselves. So you have kind of two universes to evaluate these articles. And guys, I can see my video might be lagging a little bit, but I hope you can hear me. If you're having trouble hearing me, do do let me know. It's one of the perils of being off site. But you have two universes for your analysis to differentiate. And it basically is kind of like a qualitative quantitative split. And so one is kind of a qualitative analysis. And another is you're going to remember what you're doing the literature reviews, you're treating your articles as your data. And another is a quantitative analysis. And I clearly can't spell quantitative, you'd be amazed how difficult until you've taught yourself, even just in front of a whiteboard, just talking and writing at the same time, there's different neuro neurological circuitry, and it doesn't always work. Um, so these are the two kind of buckets. And that's where you get and qualitative analysis, sometimes you can get a thematic analysis performed of literature, you can get even a discourse analysis and look for how people talk about different things, the language they use across fields. In the quantitative analysis, you can get lots of different things going on. You can get something called a meta analysis, where you take different papers, pool them and actually do your own quantitative analysis of the data. And you can do also something called a network analysis. This is common in medicine, where you might say, there's studies that compare a treatment A to B, and studies that compare treatment B to C. And now I want to compare A to C with the idea that there's transitivity across them. And this is often used in kind of big clinical comparisons, when they say we want to look at the best first line treatment for depression. We've got all these pair wise comparison studies that we want to look at across the universe of all of them. So different quantitative analyses you can do. Sometimes this is called here for systematic uses called systematic swim, or systematic review without meta analysis. And that is more this kind of narrative synthesis that that's qualitative. And it really encodes the difference between the two. But yeah, so sometimes people then get confused in the lit review when they're like, Oh, I'm doing a meta analysis. It's like, well, if you're doing a meta analysis, you are doing a systematic review in your analysis of the meta analysis. So you got two layers. So there's layer, how do you collect the data for your review? And then how did you analyze it? So there's nothing to say you can't do a narrative literature review, and combine it with a quant analysis, or a network analysis, it just would be a little bit quirky, and you probably would get destroyed by reviewers. If you're trying to be that formalized, but you're not using a spell this wrong, this is not reproducible. And you're using a not reproducible method is kind of problematic to pair narrative literature review up with these types of analyses. So almost always you see narrative literature review, paired with a qualitative type of analysis here. And you'll see scoping and systematic review paired with one or the other. And then there's other tools out there. This network analysis, I give you the specific meaning that's used in medicine, I often do this sometimes bibliometric analyses here that are done that analyze co-citation patterns. I've done some of these myself, I've lost favor of them because I don't feel like they create thick insights in the literature review. But you can also, these are also some qualitative tools of literature, you can do networks and analyze who are the key authors and the nodes, where, in which journals is there the most activity at the moment. I just lay, I've done them myself, I just have found that they don't add enough to justify the extra effort. But don't let me deter you from doing that. Sometimes there's a discussion of choosing the right tool for the job that you have. And for the questions that I've personally been asking our researchers, this is something that just complicates the paper without actually adding something really exciting or significant. Okay, yeah, my narrative, okay, I'm glad you guys can hear me. That's, that's the biggest thing. The video is usually the thing to go, go the fastest. So sorry, guys, but off site, I wanted to join you, no matter where I am in the world, if I can. So with that, so we've got one here. So Maria asked, what's the difference between a lit review and a narrative review? Exactly. So the, basically that traditional narrative lit review is your most common review. If people don't know what kind of lit review they're doing, they're almost always doing a narrative lit review. 90% of the time. And so it really gets confusing, because the name, there's two dimensions going on. It's like, how are you analyzing the literature? And then what it was the way and setup you're using for defining your study and collecting the literature. And narrative is basically this one of not following a reproducible method. I'm just going to kind of Google Scholar around and find some stuff and put it together and analyze it myself in a narrative way that's not reproducible. Hope that makes sense, Maria. But thanks for asking. So with that, guys, yeah, we got time for a couple questions. I'll keep the whiteboard up in the background. I'll go through the questions that we got this week. I can see binary asked ones here. And binary says for an emerging topic like AI and education, how do we do we decide between a scoping review, Prisma based systematic review or critical review? So I think you know, my answer here, based on what I said, I strongly recommend a systematic review. strongly recommend systematic review. Alina asked case study, case study is not in the world of a lit review. Case study, you're evaluating a specific case. So that's, that does not let her literature review. So case study often, you get returning to the world of medicine could be a rare patient comes in presenting with some particular conditions that you want to describe, because that could be of interest to a lot of doctors out there. So you describe that case study in the environmental field, Alina, I know you work in the environment, maybe there's a very unique, innovative recycling program, and you want to showcase that. And so you did a case study evaluating that program that you think has broad relevance for for people who work in the circular economy area that I know you're active in. So that's not a lit review. It's actually an empirical analysis of some form of something that happened in the world. And that's why it's a case study. But good questions. Good questions. Funchal, I don't know if you had any questions, I see we have some other questions about how to use AI in academic writing and avoid plagiarism, and how to choose a good topic. So we'll come back to those guys, if you can be more specific about your questions. That always helps. But Mainari, I hope I've answered your question there. And you've got some more clarity, as I promised, in making your decision. I do see that you mentioned critical review that often is one that causes people confusion. Alright, let me go through the questions we got this week. And I'll keep an eye on the chat for whatever you have. But yes, so we've got Felix writes, Felix wrote and says I need help. I'll copy this in the chat. So you guys can see it. He needs help finding a research gap on the role of the central bank in corporate governance and the stability of individual banks, evidence from the Nigerian banking sector. Okay. All right, let's think about this for a second. And this is something you're not going to solve on your own. Just ruminating or reflecting you've got to roll up your sleeves and start reading. So here would be a case where you want to go do a maybe mini literature review for yourself, going out and trying to understand what's been done and what's not been done on this topic. One of the things when you say the gap, so if you go back to a few lives ago, we talked about our publishability formula. I just want to recall that for a second. So your publishability formula is going to be a function, it's a multiplicative function. So publishability is a function of the strength of your gap, times the value that you can add, times the alignment your study has, the clarity you have and the fit of the journal. And if any of these things are zero or small, it's like you multiply anything by zero and the publishability is going to be zero. So if you do everything fantastic but you submit your journal, it's just not a fit. Imagine you, Alina did a great analysis on the environment and she submits it to a biomedical engineering journal. It's fantastic but it's not a fit. Zero publishability there. And what you've done in setting up your topic here on your gap, is you've made your gap very small. By doing something to say you're providing this evidence in Nigeria, well why do we care about Nigeria? Nigerians care about Nigeria but typically if you want to publish in big journals, high-impact journals, it's got to have a broader general relevance. So I worry already structurally in setting up your topic, you've taken a gap and set up a value of your study that might be considered marginal. Unless you can describe why it may be Nigeria has a very unique banking system that has some very unique features that maybe other countries can learn from. Now if you've got that element that others can learn from what Nigeria is doing, then it has broader relevance. So it's not to say that Nigeria isn't important, it's just this is just part of the publishing game that you've got to be aware of. So in terms of your topic, so you need to know what's the debate here? The role of the central bank in corporate governance? Well what I know here is there's a lot of discussion on independent central banks that are independent from government decisions and that's talked about, but I don't know. So you need to look up the literature and look at this, but this is almost like two questions in one. What's the role in corporate governance and the stability of individual banks? These are kind of different questions. So if you want to see like how does the central bank stabilize individual banks? Seems like a different question to me than when dealing with corporate governance. So I would separate those. I would try to get clarity on why Nigeria is important and can maybe answer something that others haven't been able to. And you need to go out and actually connect this to the existing literature. So you need to figure out like what's been done before and what's not been done before. So take us right up to the edge of knowledge and figure out what's missing. And the only way you're gonna find that, the only way you're gonna find that is by looking at actual studies on these topics. So to show you that very briefly, what I'd be doing, I'd be going into Google Scholar and I would be going to impact central bank on corporate governance. Just get a sense of what the evidence is. So here would be a study. I'm already worried because like look this has got nine citations. It doesn't look like there's a very big debate going on here. This is what I was worried about. I don't know. You'd have to trace out the mechanisms for me on how the central banks are affecting corporate governance. My understanding was just more that I'm not an expert in this, I'm not an expert macro economist, but they're just setting the interest rates, especially if it's an independent central bank. Sure that changes the climate of lending for corporate actors, but why is that directly engaging with corporate governance? So I think there's some part of the logic here that that's being lost on me, but it's probably clearer to you. So I would encourage you to think that part through. And again, if you don't have kind of clarity on what you want to show, why you're interested in this, it's gonna be very very hard to get the next step of defining your gap and what you want to show, which is really critical for a good topic. Okay, I'm gonna take a couple other questions in between as I've got a few others on the list. We've got a long question here from Moctar who says, I'll try to read this out to you guys, and let me show, it won't show the full thing here, but says, thank you for your invaluable guidance throughout the PhD journey. Your advice has deeply shaped my approach to research and publishing. Awesome, glad to hear it. As an early career researcher aiming to build a strong high-impact publication record, love that you're doing that. He says, I met a crossroads. Should I pursue a PRISMA-based systematic review to establish scientific rigor and credibility, or opt for a critical review to argue for the necessity of this interdisciplinary paradigm and material science? Well again, a critical review is just gonna be harder to publish. So I always lean to a systematic review because it's just easier and it can accomplish the same things as a critical review. So is there a reason you wouldn't want to do a systematic review? Because I tend to take that as the default good option, just because of its easier publishability, especially for material science. So yeah, you're asking which approach would you recommend in established authority? 100% systematic review, unless you see a lot in your field with critical review on the tin. Sometimes critical review can be very niche in certain fields, but I haven't seen that necessarily in materials science. So especially because some of those lit reviews, like I said, are invited and they're because the editor might know someone senior in the field or that person's even on the editorial board. So you just come cold with a lit review. It can sometimes get a tough ride. Minari also saying SLR would be the best choice. That's what I think, too. So Puncho, Puncho asked a great question. So the introduction part of the paper I want to publish will be a narrative review. Excellent. Yes, 100%. That introduction of a paper is typically a narrative review and it is a strategic argument in your introduction for your study. It is basically you're going to be driving your gap. And so a successful introduction is that same funnel structure of going from the kind of introduction, why are we having this conversation now? Why is this topic or area so important? To what do we know and don't know on this topic? To kind of gliding right into your research question or hypothesis and what you're going to do. So very much that that kind of funnel broadening to narrowing to your specific study, so that by the end of it, your study should feel almost inevitable, like it needs to come into existence. That's how your reviewer should feel reading it. That's how your editor should feel reading it. It is a strategic argument in the introduction of your paper for the existence of your paper. So yep, 100%. Okay, got another material science researcher asking the same question as Moctar here. No, I think this is Moctar asking again in a different light. But yeah, Moctar, I answered that just a moment ago. Thanks for the question, though. It's a good one. And Funcho says, yeah, yeah, 100% Funcho. So really glad that you asked that. I think also, right, when people do the lit review of their paper, they're not asking which type. It's just the lit review for the paper. But it does help to know what type of review you're doing to make sure you build in the structure yourself and you know the purpose of that review. You know where you want to end up so you don't get lost. Yeah, so briefly, again, so this is what we went over, John. So the core differences is the scoping review is going to have a built-in structure, it's going to be reproducible, it'll be easier to publish. It won't have a quality assessment, that will be like a systematic review. But these are two in the structured category. Scoping review is going to follow this PCC model. If you want to read more about this, you just Google PCC scoping review, you're going to define here your topic using the population concept and context. And see it here, it's saying it replaces PICO. I still prefer systematic reviews for the reasons I said before that you can always convert systematic review to scoping, but you can't go the other way around. But both these are great because of the built-in built-in structure. The narrative review is just harder to publish. So the way I would think about this for your master's, if say your master's is just the deliverable of your master's thesis is the lit review, then I would make it a scoping review. If though you're planning to say go do a data analysis or something else, I would do a narrative literature review, which follows a funnel I talked about, if you just jumped on, rewatch this part of the video, and just kind of delivers, like I was saying to Funchal a moment ago, the strategic argument here for why your study needs to exist, for why that empirical part of your master's needs to exist. So at the master's level, that's how I would think about that, that trade off there. John, I hope that makes sense. Let me let me know if you have a follow up question. Veronica asks, Greetings, Prof. How can we pull out literature review and applied linguistics and language studies? Sometimes we're stuck as to review to use. I think you mean you don't know which review to use. So these lit reviews, think about it, the analysis inside the literature review, qualitative or quantitative, doesn't matter which field you're in. Those are kind of your broad tools for analyzing literature. It's going to fall in one of those two camps. Same thing, the tool you use, are you going to go look in Google Scholar, basically, you're going to be in narrative review land. Are you going to use a formal search of databases, you're going to probably be in a scoping or systematic review territory. Are you going to be looking at just existing reviews, you're going to be in umbrella review territory. But that is field agnostic. Right? So the type of review really, I think you got to step back for a second, Veronica, and it's not which studies you're looking at is going to necessarily dictate the review. I would put it as more, do you want to publish this review? Are you using this review just for your own knowledge? Is it is it just a quick review at the introduction of a paper to justify the existence of your article? Why? Get real clear, why what is this review for in your research journey? Does a systematic review apply if you have to do data analysis? Good question. Hamza12456. So if you want to do data analysis, is the data analysis you're wanting to do of the articles themselves. But if you have an actual data analysis, I wouldn't mix article types here, I wouldn't have a systematic review, and then do an empirical evaluation of something innovative. I would just have a traditional narrative lit review at the front end of your paper, and then have your empirical data analysis component on paper. Unless you're doing some kind of quantitative analysis of literature, and you're doing the review to collect the articles using that to create a data set. And then your data analysis is based on that article data set you constructed as more meta analysis, or one of the other types or a bibliometric analysis, one of those things we carved out a moment ago. And then then, yes, systematic literature review makes sense. Veronica saying like that, that makes us cool. Please, please that help. And Elena saying, Veronica, thanks for asking that question. Guys, I love you asking these questions. Because just think about that. Elena is a good example. There are other people who have these questions, but might be too scared to ask. And so by I'm not saying you are Elena at all. But I'm just saying a lot of people out there are sometimes just a bit scared to ask questions. That's totally okay. It can be a bit of imposter syndrome, you feel like, oh, people are gonna see I don't know stuff. Well, you have to lean into not knowing stuff. That's where you're doing this training. And in fact, the whole enterprise of research is not knowing stuff. And you're doing the research to figure it out. So you actually have to get very comfortable with not knowing. That's kind of the name of the game. So that can be a very discomforting, especially when you're starting out, because you came from a world of grades, there's a right or wrong answer. And how you perform is you personally being judged. And everything changes once you start, start doing research from how you proceed to the rulebook, how you get evaluated. And so that means also kind of updating how you respond to feedback. Don't take it so personally treated like a gift. Realize that when you feel imposter syndrome, you're being stretched, you're actually growing. And that's a good feeling. It's like when you go to the gym, yeah, it's painful to train. But you're growing and you get stronger. And in time, you get addicted to that feeling because you feel really great. And research is like that. So just lean into that growth and lean into that discomfort. Because I love things that are a little bit uncomfortable, because that also tells me not a lot of people are going to be able to do it. So intrinsically, it can often have more value that you've been able to achieve something that's hard that other people see, try and run away from. So yeah, just treat it like the challenge that it is and lean into it. Okay, we take a couple other questions that I had from the list. So Joshua asks for feedback on research cover letters and quantitative or qualitative analysis. I mean, okay, how much time do we have? Joshua? Great, great question. But guys, keep the questions focused for me. But let me point to you where you can get this. So of course, guys, I'm biased. I love our research community. Because we have dedicated drop in sessions, we have dedicated qualitative drop in sessions, workshops, actually, you know, let me just show it to you. I mean, we've got something going on. Going on pretty much, pretty much every week here. In all these areas, but let me just pull this up. So you guys can can see what I'm talking about. And I really think you guys love this. And give you kind of a micro tour. So here's where we are. Okay, here's where our community lives. Small but growing 224 members. And you can see you head over to our workshops, we got stuff going on all the time. So right, we had a quant workshop, dedicated stuff on quant. Next week, we had a really fantastic editing workshop with Peter, whoops, we've got systematic review workshop, and it's everybody working on systematic reviews. Next week, back, let me head over here. Two secs. Next week, we've got lined up. Go on to the next week. Okay, this is getting buggy. Go to the next week, we've got a really great qualitative workshop that I'm excited about. So right, what's great is it just closes the feedback loop. So you get feedback a whole lot faster. And I found that feedback really is critical. But feedback is not alone, you need an actual system. So for all things, like if you're getting stuck on, well, I don't know how to write a cover letter. Well, I'll take one example, we have a really nice navigating the submission process shows you how to live submit on this one example, help you avoid scam journals, really good cover letter guide and template in our videos of training showing you how to write good cover letters in your field to optimize the chances that you don't get desk rejected. So I can't possibly go through all this now. But I'd encourage you to just check this out. It's just a quick I'll put a link here $10 quick trial. You can you can cancel anytime and see if it's a good fit for you. We'd love to see you on the inside. So yeah, I can't answer all that here. But I've got a lot of really valuable free resources. It's just here, there's actual hosting costs that we got to cover, even though it runs on a nonprofit basis. So yeah, check it out here. Okay, other questions I've got lined up this week. uzman has a project thesis. Okay, this this looks like this could be quite challenging. uzman says, please, the screen is not really clear. Okay, sorry about that. Joshua, I could maybe zoom in a little bit more. I'll go to this next one here. uzman says he's doing a project thesis for BA mission. The topic is theological studies of factors influencing sustainability of an SIM mission field after missionary withdrawal, a legacy of ballistic mission field in Chicago. Um, so um, I mean, it's really hard to understand what's going on from your title. This is hard for me to make sense out of, honestly. But for the BA, usually I just do, I keep it very, very simple. Now's not the time to do huge project, you want to do something as contained if it's an undergrad thesis. My undergrad thesis was just a lit review. I think mine, back in the day, it was I was just looking at some of the causes of homelessness, or I think it was the the impact of homelessness on mental health. And I was just summarizing literature. It sounds like you're trying to do a case study here. But, um, I don't know why a theological study of factors influencing sustainability. And why missionary withdrawal is important. And I don't know what the ballistic mission field isn't Chicago, it just just too much stuff going on here. I would just even just look at this, make it so much simpler. Just look at, look at maybe maybe you want to look at why the missionaries withdrew. And from this, this mission, and how can you better stabilize missions? I don't know. But I think you got too much going on is my initial reaction here. So yeah. Yeah, guys, if there's something Joshua, if there's something here specific that you wanted to see, let me know. And I will go back and show you that. Okay. Elena asks, What if I have a PhD thesis with a narrative lit review, but now I want to publish an article based on it? How should I change my narrative lit review into a systematic one? That's really tough. You almost have to redo the whole study. So that is unfortunate. I would honestly, I wouldn't don't want to deliver frustrating news. But you can't really reverse engineer that effect. Whoops, sorry about that, guys. I got disconnected for a brief moment there. Hope you're still with me. But yeah, Ellen, I was just saying, you can't really reverse engineer that. I know that's frustrating. I don't like to deliver that news. But you basically in a systematic review are gonna have here's I searched these databases with these criteria, and I got these articles so that anybody can follow the steps and get those articles. There's not really a way to reverse engineer a formalized search and get exactly the final set of articles in your set. I just don't think it's gonna work. So I think I would just I mean, the cool thing is you've done the analysis already, I would just start over and do the search. And maybe your analysis will come out the same, you'll get a broadly similar set of articles, maybe you'll get a few more. I find often researchers are surprised when they get out of a filter bubble. And when you're searching Google Scholar, you are in a filter bubble. And when you get out of that, you will find articles you didn't even know existed, you'll often tap things in other adjacent fields you didn't even think to search. When I say filter bubble is that the algorithms of Google are producing things that it thinks you're going to like based on your past search histories. So you're systematically being blinded to stuff that might be highly relevant, even though it is distant from what you've looked at before. So yeah, hate to deliver bad news for you. But you could try to publish it as a narrative lit review. But like I said, in general, it's not impossible. It's just harder to do. In my experience. Okay, I know. Sorry, Helen, I'm glad you asked that. That it's really helpful for people who are just starting out now. You see, I would go the other way. So I would have done systematic review first. And then if they they insist on you having narrative review, well, you can always turn that into a narrative review quite easily. You've done it, you have the funnel structure there. You just had a different tool for getting the literature review in so you can turn systematic review into something else. You just can't go the other way. So, oh, don't worry about asking twice. It's okay. Um, says you got a practical question. Um, well, the quality assessment tools. Sometimes there aren't quality assessment tools out there. Insider secret here, I don't, in the first pass of submitting to journals, usually I don't actually include a quality assessment, even though I know the reviewers are probably gonna ask for one. There's a couple reasons I do that. So one of the reasons I don't put the quality assessment in initially is that you're gonna have to add about another month to the process, which we find start to finish takes about three months, investing about five to 10 hours a week, and the extra month is just enough to where it can get overwhelming and that I find sometimes people drop off and never get it done. So there's that pragmatic component. The second is that you sometimes in a paper want to leave something obvious out. That's not a big deal. So the reviewers can just be like, ah, you need to do a quality assessment, and then they feel like they're smart. They came up with something, but it's great. It's almost like I just kind of left this little hole here and you could walk into this hole. Imagine if that hole was not there, then they're gonna dig around, try to find other holes that are more annoying to deal with. So there's that. Usually I find they're not gonna reject it just because you didn't do a quality assessment because they know they're gonna reject when they think, oh, this is really bad. But if they say, oh, you need a quality assessment, they know, well, you can just do that. The third reason is because, and I've seen this happen, they do a quality assessment with one tool, and then the reviewer happens to look at different tools and says, well, I want you to do it with that tool. And you end up having to do the quality assessment over again. So you might as well just wait, if you get a quirky reviewer, to see which tool they want you to use. So for those three reasons, I don't do it, even though I'd say about 70% of the time the reviewers ask for it. And that's also because a lot of them are using AI to do peer reviews now. And the AI spots it and it's like, oh, a systematic reviewer has to have a quality assessment. So they ask for it, even if it doesn't necessarily add very much. I guess that's the other component. It usually doesn't add much. It's kind of a box ticking thing. But if you don't have a tool off the shelf, sometimes you just make one. You make one. You just say, these are criteria of good quality studies in physical models. And look for some somebody else who said, this is what makes good model robustness. This is what makes good reproducibility. And do some tick boxes around that. So you could invent something. Joshua says, general view, really interested in research as a young professional. Yeah, again, I'd encourage you to check us out. There's others out there. But we've got a complete ecosystem and a lot of research groups you can join. So check us out. Again, not really anything to lose. See if we're a good fit and you like the way that we approach research. I mean, you can see kind of my ethos and my approach and you'll find naturally gravitate to some people of different styles and different systems. That's totally okay. But I just want you to be getting kind of the support you need to thrive, whether from us or from someone else. You have a Chitty says, I have a project thesis using discrete descent. So I don't really know what your question is, guys, try to ask me a focused question. I do tend to be quite liberal and want to respond to everybody. If I can, I get a lot of questions coming through in the chat. But um, do do let me know. I'm going to go to earlier in the chat, where I got a question here about using AI and academic writing to avoid plagiarism. This is a good one. This is something that's provoking a lot of anxiety. And I don't think it needs to provoke that much anxiety. Look, guys, just declare how you used AI. Obviously, don't use AI for references. Make sure you're heavy on citations. But if you're worried about it, just cite it, then you're on the side of angels. And that's allowed. Elsevier's guidelines for publishing, as with many other journal families, are totally okay with it, as long as you haven't had AI do the original parts. But you should just be using AI anyway, to maybe edit your language lightly. So don't have AI draft large blocks of text for you. That's not a great idea. That said, you can use AI to say, hey, help me cut words, help me say this in a more in a leaner way, or get it, you can get it to edit writing you've already done, and that's fine. Watch it closely, make sure it has not deviated from the precise scientific meaning. It does that a lot. I have some other videos on my channel where I show how AI edits will actually change the scientific meaning in subtle ways. So you really need to check that any anything that AI has edited or touched has not deviated from what you wanted to say. If English isn't your first language, I know that can sometimes be hard. But AI really has been a game changer for non-native English speakers in its ability to produce better written text. But still, if I highly encourage you to do the first pass, and again, coming back to happily repeating myself, just err on the side of transparency of how you've used AI, and you'll be fine. Israt, so I'll give you a very brief overview of how to choose a good topic. So first, you need to get in the right topic neighborhood. The example we had before about the Nigerian banks, it was kind of in two topic neighborhoods. And so I was encouraging it to pick one. That's phase one. And so usually want to look for something where there's a good debate. And turning back to that Nigerian example, I didn't see a lot of citations, I didn't see signs of healthy debate. So I didn't love it. You want to make sure it's something feasible, you already have some background knowledge, or you can get that knowledge very quickly. You can do a topic, you can do something, you're not setting yourself up for a randomized controlled trial, that's going to take a decade, right to something that you can actually achieve, and something you're passionate about. So that's our convergence method. You want to get in that sweet spot. That's phase one of choosing your topic. Phase two is getting more fine grained. And you need to and again, I've got a live session just a couple weeks back, look for it on my channel where I went over this in detail. But you then need to make sure you're not duplicating something that's already been done, or duplication test, you need to further forecast your impact in a more granular way like I've done. And you need to find specifically this idea of a conceptual nearest neighbor paper. That's the paper that is the closest to yours. And that's your benchmark for calibrating this paper got us to here, that is the edge of literature. And from that paper, my whole idea of my topic is to get us to here. I do recommend running your topic through one of the tools, like a Pico tool, just to make sure it's crisp and well defined, and very, very clear. I haven't done that today, because that's not really the theme. I wanted to help you guys navigate the confusing array of literature reviews that's out there. And I hope that makes sense. And exactly. Dot org has nailed it. AI is a tool not in the driving seat. Yep, you are at the AI accelerates, but you got to steer. And so if you are going the wrong direction, AI is going to get you there faster. That's why AI lowers the cost of being wrong faster than it lowers the cost of being right. And I see people getting into huge messes. It creates system problems because it bypasses that you go so fast, you zoom past the normal error correction and friction you would have early on. And then you wake up and you find yourselves three months into a project that's completely unviable. That's one of the AI failure modes. And you can see some more videos on my channel about common AI failure modes that I want you guys to, to avoid. Um, you know, a chitty discussion follows a common template. Check out I've got a how to write a perfect discussion in my how to write a paper playlist on my channel. So check that out. It'll show you the ingredients for the discussion section of paper, as well as discussion section of a discussion chapter of a thesis. It'll have the same ingredients. So check it out. And Joshua says, some paraphrasing tool is seen as a content with a report. So I've got a great video on my channel on how to paraphrase with AI, I show you two methods that work really well, and we'll keep you on the side of angels. So yeah, check that out. Any you can do it with any LLM, but just follow the guidelines I have on that video of basically I can't remember the title now, something like how to paraphrase with AI. But guys, fun session. Thanks for bearing with my connection. I decided even with that connection, I did not want to miss you. Every Friday, at this time consistently we're here. If you made it to the end and you do want to participate yourself, I would encourage you to submit a video question to us. I love video questions, because I can just better understand and so can people watching on what your question is what you want help with. You can also upload a research document for me to look at and provide feedback on if you have your PICO model for your topic, send it in if you have a draft systematic review, you want me to look at a section, send that in. Even if it's messy, I like messy stuff, because I can see through the mess to see diamonds in the rough. That cliche is really true. I'm used to looking at a lot of messy drafts. Everybody starts in that place. That's the whole process of science is bringing order to disorder. You go from a messy state of the way we think about the world to getting clarity and order. It's a great feeling when you get that in place. So submit your video question. If you are interested in working with us, I've got another QR code at the top. It'll help find which program is right for you and show you a bit how we think about research. If that resonates with you, we'd love to see you on board. With that guys, I wish you all a very good weekend. I'm glad, Emmy, that you got some nuggets and takeaways. That's great. Joshua says, send the link to the videos. They're all on my channel. Just go to my channel, and you can look at the playlist tab. So YouTube, just go YouTube, put my name, Professor David Stuckler, and then you can see the lives, and you can see the playlists, and a ton of value there. A ton of value. I'm trying to give you 100% free what other people would want you to pay for. So yeah, that's a great question. I think we'll do a dedicated session coming up here on using AI and academic writing, but if you want to see my latest thinking on that, you can see that on my channel. Guys, thanks for sharing your Friday afternoon with me. This is your time, and I will see you all, same place, same time, next week.