[00:00:00] Speaker 1: Welcome back, everyone, to this week's Fast Track Live. I've got a special guest with us. I'm really excited to bring on, gonna introduce him in a second. And listen, if you have ever struggled with a literature review, if you've gotten stuck, felt like you were drowning in papers, didn't know what to do next, didn't even know where to turn for help and support, well, you're in the right place. This is gonna be a doozy today because by the end of this session, you're going to get our four-line search logic grid that's helped me personally publish in journals like the British Medical Journal and the Lancet. You're going to have tools that you can implement immediately that are going to save you tons of time. We're going to go through a live example of actually developing a search strategy for an effective literature review that you can put into practice. And we're also going to cover different types of reviews so that you can make sure you're doing the right review for you. And I encourage you to really follow along actively. We're gonna have several points of interaction where we're going to ask you what you think and in the true spirit of community, you get what you put in. And the more actively you pay attention and engage, the more you're gonna walk away with. So it's not really going to be a passive session at all. Many of you are on Team Replay. We have researchers from all over the world. I've mapped it out with Google Analytics, over a hundred countries joining us, which is tremendously exciting. Again, because of time zones, we try to vary things. So I know we're gonna miss out on some of our researchers in Australia and New Zealand, for example. If you're on Team Replay, let us know in the comments below, hit Team Replay. And if you like what you see, do give us a like because that does help the mysteries of the algorithm. To reach more people helps us serve others who might not otherwise benefit. And that's really what this channel is all about. I created Fast Track because when I was a researcher as a graduate student, I made about every mistake you could possibly think of. And if it wasn't for real help, real mentorship along the way, I would have never been able to get to where I am today, having been a professor at Harvard, Oxford, and Cambridge. And with that, I'm really delighted to bring on one of my personal mentors. He's joining us today, Professor Martin McKee from the London School of Hygiene and Tropical Medicine. Martin, great to have you with us.
[00:02:22] Speaker 2: Thank you, David. So, just say good morning, good afternoon, good evening, wherever you are.
[00:02:29] Speaker 1: And just to say, Martin, I mean, almost needs no introduction. He's been the president of the British Medical Association, published a thousand, no exaggeration, over a thousand papers in peer-reviewed journals. I don't know how you're- Many of you- I've had the good fortune of joining you on at least a hundred papers. So, yeah, I always learn from Martin and I hope you will too. And just Martin, if anybody ever wants to reach out to you, by the way, what's the best way for them to reach you? Email. Email. Email. I'll pop our contacts in the bottom of the chat so that you can follow up with us. I personally read and reply to every comment. So with that, Martin, I wanted to put you on the spot. We always start this program with a quick tip. And I like those tips to be about productivity. And one that I just released a YouTube video about is on procrastination. I wanted to get your best tip on have you ever felt any procrastination in yourself and how have you dealt with it or seen it in the researchers you've mentored?
[00:03:32] Speaker 2: Yeah. So, you know, you said that, like with you, you know, published a lot of papers. And I used to say to people that the commonest reason for failing to publish papers was the inability to put them in envelopes. This was in the old days when you actually sent your manuscripts to the journal by post. But often, you saw that people just hold on to things forever, and they hope that there'll be some inspiration, something will come to them. And you just need to get on and do it. And I think, really, I have a list of things that need done, because I get far too many emails, far too many requests, which I'm delighted I will help, but just bear with me. I may not be as efficient as David in responding to them. But it's very easy just to put something in sort of what I call that too difficult trade. And I think, well, I'll get back to that and then other things take over. That's one thing. I think just really making sure that there's a momentum there. One of the issues is that when you get reviewers' comments, with some of my colleagues, we've got a paper back, what day is today, Tuesday, last week, on Thursday or Friday. And to my mind, with that, really try to get them back within 24, 48 hours at the maximum. If you've got a number of authors on them, you may need to do a bit of consultation and make sure everybody's happy. But really, unless you have to do a massive amount of new work, that should never, ever take more than a week. And often, the sort of revisions really can take a day or two days at the maximum. So just really keep the momentum going. The other thing is that we have a saying in English, the perfect is the enemy of the good. And whenever you send any paper to a journal, almost always you will get some comments and some revisions. Now, occasionally, and it has happened now and again, I've got feedback to say, no, it's fine as it is. I had one memorable paper which the reviewer loved so much that they said, I salute the authors, which was very nice. But that doesn't happen very often. So you're going to get revisions anyway. And whatever you do when you think you've just added that critical bit, you'll find that the reviewer wants something else. So you're going to have to change it anyway. So as long as you feel it's good enough, send it in. And then, in a way, the reviewers will help you, hopefully help you. Not always, but hopefully help you to refine it just to get it up to that standard to be published. MARTIN SPLITTINGHALSEN Martin. MARTIN SPLITTINGHALSEN Those are my tips.
[00:06:11] Speaker 1: Keep the momentum going. Keep the momentum going. keep things moving. I love that because if it's just sitting on your desk stagnating and one of the great things to do is you could reach a stopping point, hand it over to a co-author so the paper is always in motion.
[00:06:25] Speaker 2: Oh and just on that, sorry David, I would add on that that one of the things that we do because I work with a number of people in different time zones, so make maximum use of the time zones. So you know, you work it on your day, you send it to them. I do a lot of work with colleagues in Australia, New Zealand, on the West Coast of the US. So you work behind the US generally. So just keep that going so that you send it at a time that they can work on it. And if you do it properly and you've got people all around the world, you can actually have the paper being worked on for the whole 24-hour cycle.
[00:06:58] Speaker 1: Davey Fletcher Definitely, really smart. Really smart way to just optimize the real 24 hours in all of our days. I like what you said too about perfect being the enemy of the good. I think procrastination is sometimes a symptom of perfectionism in disguise. And I've seen, this is really something that I find afflicts very smart people. They're used to producing at a very high standard of work and the problem is they can kind of circle and circle and hit diminishing returns on a paper and not even realize it. And so I encourage people sometimes to aim for 80% of what would be good enough, uh, because otherwise it won't get done. And intentionally we have a tactic sometimes to even sometimes in a, in a perfect paper, leave something out that's obvious. So the reviewers eyes fall to that point that you know you can address, but you have such a good paper. The reviewers sometimes feel they have to find something and can create a bit of a mess for you later on.
[00:07:58] Speaker 2: And the other thing that you might want to do nowadays, I'm not sure how well this is becoming accepted, but if you, first of all, I would say that if you are using artificial intelligence, you do need to be careful that you don't put it into any algorithms that actually breach the confidence and put it into the cloud somewhere, because if you've got innovative research, the journals will not like that. But if you do have access to some of the packages that don't do that, that keep it within your university or wherever, you can get that to, you know, just ask it to pick out obvious flaws and identify problems that way. So that's just worth thinking about, but you've got to do that carefully.
[00:08:34] Speaker 1: Definitely. We're going to come back to AI because we've got some questions and thanks to all of you who have submitted video questions, we're going to get through all of the ones that we can today. And I said, Australia can join us in here. I stand corrected. We do have super Kuja joining us from Australia. So welcome, welcome. Good day. I'm going to pivot now to our mini-training on lit reviews, and while I'm pulling that up, Martin, you've guided, mentored many literature reviews, what are some of the common challenges you see some of the researchers you've dealt with face in their literature reviews? I'd like to be sure if we cover a bit of that in our training.
[00:09:10] Speaker 2: So one of the biggest issues is not having a good question and not being clear what your question is. You get so many people come and say, I want to look at something, and looking at is not a reason for doing a literature review. I think you will be talking about frameworks like PICO in particular, but it has to be a question that can be answered. And then the second issue after that is, is it a question which is to be answered in the general? And like, does drug A work better than drug B? And that's a straightforward question, you know, leaving aside the inevitable genetic variation within humans. But let's basically, if it works in one, if you're doing studies in different places you're likely to get the same sort of results and the more you do and the more literature you bring in once you've adjusted for anything in the design you know you're going to get more and more to wherever the true result is but there are other issues where that's not the question and the question is what works in what circumstances and that takes you into the area of realist evaluation you need to think in a different way about doing a literature review about that because you're not asking, well, you want to know that it can work. That's the important thing. If it can't work anywhere, then, you know, why would you be interested? But there are so many things in relation to context. Like if you've got a, you know, system for payments and you do one study in Manhattan where you can go to an ATM or everybody uses Apple Pay, it's going to be different from if you're going to be in, you know, in a forest in the center of a number of low-income countries where you don't even have a cash-based economy, it's going to be very different. That's an extreme example, but there are many others.
[00:10:54] Speaker 1: So yeah, Martin, I think that's really important, and I'm glad you primed us to think before diving into this about the topic. Just kind of our good topic triage checklist here that Martin's specifying, you need to have real clarity on a few things. You need to know what your gap is in the field that you're trying to fill. What's missing, almost that the research has brought us to here, and now I'm going to try to bring us to here, and that missingness in the field is the gap, and that where you're bringing us this kind of space here where you're going, that's your value added. So the gap really connects to the value add, and one of the big reasons papers get rejected is the editors will say, or the reviewers will say, well, what's the novelty? What's the contribution to knowledge here? And if that's happening, it's often kind of an original sin, something earlier in the process of not having clarity on your gap in value added of your paper. And that links too to the clear kind of research question you want to formulate that's going to come into contact with that gap. And often we encourage you to find something, it's a very good practice, whatever your field is, whether you're doing a literature review or not, is to find a, what we call a nearest neighbor paper. That's kind of the paper that's the closest to yours in the research field. And that, again, helps you establish, well, the research field got to here with this nearest neighbor paper. You're often going to want to cite that in your introduction and have clarity on it. And that's also important so you don't duplicate something that's been done and sets that baseline or point of reference so you can establish where you got us to from there. So yeah, Martin, thanks for that. These are things that you really need to make sure you have and this is going to come back into focus as we take some of your questions from today. One of the challenges I see researchers in literature reviews is often their topics are too broad and so they are drowning in literature or they're too narrow and they can't find any studies and so one of the tools that can help give you some control knobs to dial in your topic a bit and this works across fields although it's really embedded in clinical medicine but it's become much more prominent in management social sciences now is a PICO model and PICO here Martin I mean you know PICO intimately can you share with us a bit I'm going to put you on the spot back to grad school here about PICO and why it's valuable. Oh, Martin, I think you're on mute.
[00:13:15] Speaker 2: Yeah. Sorry. Yeah. So P is for patients or population. And that basically means in what group of people with a particular disease or gender or it could be anything, any characteristics, socioeconomic, whatever, but often it will be with a particular disease or condition or something like that. And it may be that you're looking in a particular country or in a particular setting. So who are you actually talking about? It can be patients as well. So population or patients. The intervention is what are you actually doing to them? What is the intervention? Often it will be something like a medicine, But it could be that you're doing something in a population where you're not doing things to individuals, where you're doing things to a group of people. So that would be where you're looking at cluster randomized trials, where you have, I'll give you an example of a randomized control trial we did, the HOPE-4 study. We were looking at integrated management for hypertension. And it was a package of interventions where you're looking at the use of mid-level health workers, simplified guidelines, combination therapy, and so on. Now, you couldn't do that easily with just randomizing one patient and then the next patient to something else. You could do it in different villages or communities, and then you randomize there. So the intervention can be a complex intervention like that. It can be giving them a medicine or giving them a form of physical therapy or something like that. But you need to be clear what it is. You need to be absolutely clear about how to define it. If it's a medicine, dead easy, just need to make sure you've got the right dosage, that sort of thing, well, maybe a bit more complicated to you need to make sure it's absorbed in the right way and the other, but that's more detail than you need. But if it's a complex one, you need to understand not just which combination of things, but who is delivering them. So for example, you might have mid-level health workers as we did, but they may be different types of middle-level health workers in different cases. The C is the comparison. What are you comparing?
[00:15:21] Speaker 1: Wait, wait, wait, Martin, I just wanna come back on the intervention because it doesn't just have to be for a randomized controlled trial. So we've done these looking at, for example, exposure to populism or other, it's not just an intervention, it can also be used as exposure to, I know you've done some work, Martin, on social capital. We've had researchers looking at different artificial intelligence interventions in the classroom and how those affect grades or student engagement and performance. So also be aware that this can be what somebody in the studies exposed to, not necessarily in the context of a trial itself.
[00:15:57] Speaker 2: So basically what you're saying, I'm absolutely right, David. So this is the difference between experimental methods and an experiment. You do something to people and a non-experimental, you're having people who have had something done to them. And that can be people who are living in different places where a policy interview is, I mean, classic one the United States is where a lot of this is done because you've got 50 states plus the District of Columbia and therefore you've got a lot of potential for policies that are introduced in one place and not in others or you can do it before and after and there are a lot we might get on to talking about the different methods I think David you've got a huge amount here so it can be experimental or it can be non-experimental and it can be as you say it can be populism, it can be a change in the years of schooling, for example, where they might increase the compulsory, the school leaving age by a year or two, our work on minimum wages where the British government brought in a minimum wage and we were able to compare people who were just below the minimum wage and therefore benefited by the legislation and those who were just above, who were earning very little more but didn't benefit from it. So you can have all sorts of of things like that.
[00:17:12] Speaker 1: So, so exactly. So as you guys, most of you watching long have a topic. And if you haven't done this exercise, if you feel good about your topic, fantastic. But if you have any sense that you're struggling a bit or getting lost, I'd encourage you to look at the components of your topic based on a Pico, just break it apart, almost like an engineer. You want to learn how to design a car engine, you have to almost disaggregate, disassemble the engine and put it back together. Try to take apart your topic and think about what's the population in your study that you're doing right now. the intervention, because the vast majority of studies that you're doing will have at least one of these components defined. So that's really, really helpful, Martin. I just want to get everybody active. And if you want to share with us in the chat what your PICO is, that inspires not just us, but the entire community of researchers watching on. So we'd love to see what your PICOs are using this model. Let's finish the PICO, though, Martin. What's the C for?
[00:18:06] Speaker 2: A comparison. Comparison, so what is it that you're actually comparing? And in a placebo-randomized controlled trial, it's a active medicine with a dummy, or it may be comparing two different types of treatment. If there already is a well-established treatment, then it may be unethical to just give them a dummy treatment. So you're comparing with whatever is a standard treatment. And I mean, again, in non-experimental methods, it's usually where you're looking at something that's being done. intervention is a new policy is brought in, put into place, and you're looking at comparing people who have been in an area where that new law or policy has been introduced and the ones that are not, or it could be that you've got some measure, you've mentioned populism for example, if you have some, well again you can do before or after, you're going to have the election of a new, the election or whatever way coming to power of a different political party who is pursuing populist policies, and you can do a before and after, or you might be able to compare different places, some of which have regimes that are more populist than others. You can do the same with things like trust and a whole series of other things.
[00:19:20] Speaker 1: I like this, Martin. Also, think about the comparison, even if you're not doing an RCT, can really help you get clarity be on something the economists love to talk about, identification, especially you get into something about, think about causality, you have to think about that question of what would have happened if, what the counterfactual can be. We got a couple of questions on this, I just want to cover it for a second, because Sana asks, can comparison be for countries?
[00:19:46] Speaker 2: Yes, it absolutely can be and often is, you've just got to be a bit careful, you know, there's saying in economics, ceteris paribus, which means in Latin all else being equal. So you've then got to just be a bit careful that the countries are really comparable in every way or as many ways other than the thing that you're looking at. Now there won't be but you may be able to adjust for that and that's where it gets a bit tricky and that's where you know in general I think the time series ones where there is a change and you're keeping the country fairly constant apart from that change, that's where that's most useful. But there's some, I don't know how much detail you want to get into, but there's a particularly useful method called synthetic control analysis.
[00:20:33] Speaker 1: We try to cover the range from beginner through to advanced.
[00:20:36] Speaker 2: Too detailed.
[00:20:40] Speaker 1: I mean, the randomized control world in a way seems for me much simpler because it takes away all these confounders. When you get into the real world, suddenly things get quite messy from the artificial setting of the lab. Our Australian colleague here, Super Kujo, asks, is there a variant of PICO for systematic reviews?
[00:20:58] Speaker 2: Well, PICO is for systematic reviews. Yeah, yeah, it is.
[00:21:03] Speaker 1: But we recommend it and we found it works well, not just for systematic reviews, but for any type of research project in general, because it forces you to have very well-defined parameters. And I think one of the challenges people have with literature reviews is not not having well-defined topic. And once you break it into these parts, it's gonna help us later to diagnose, is the problem, why are you getting too few studies? Why are you getting too many? It's gonna help us lay a foundation that will help you develop a search strategy, which we're gonna come to a little bit later. So yeah, and this really works for any research topic in our experience. Okay, lastly, Martin, very important. We gotta complete the PICO, the very important component of the, oh.
[00:21:46] Speaker 2: Outcome, what is it that you're trying to achieve or you're hoping that the intervention will be doing? So will it be survival? Will it be death? Will it be a change in health status? Will it be a higher level of education? Will it be behavioral change? Something that will change as a result of the intervention. And it needs to be related to the intervention. So there needs to be a plausible reason why it should happen. You don't, I mean, there are so many examples people use in teaching like the correlation between the migration rate of storks to Sweden and the birth rate, that sort of thing. Going back to the old idea that babies are who come from storks, leaving them under bushes, that sort of act. So, you can find these spurious correlation. I mean, there's no biological plausibility there whatsoever. So it has to plausibly relate to the intervention. And also there needs to be some sort of mechanistic link. and maybe I can illustrate this with an example. There was a lot of concern by some people that the evidence from randomized controlled trials of wearing a mask during the pandemic was really not very good from randomized controlled trials. But if you think about it, the intervention was on the person who was wearing the mask, but the outcome was that you wanted to prevent the virus being spread to others. So the outcome was on the people around them. So trying to do a study like that at a randomized method was extreme, well, virtually impossible. There were cluster randomized control trials where they got mask wearing in every different villages and Bangladesh in this case. So you can't do things like that. But for an individual, you know, it just didn't, when you thought through the logic of it, it just didn't make sense.
[00:23:32] Speaker 1: Yeah, you make a good point, Martin, too. In thinking through the logic, we do use a tool called Directly Acyclic Graphs. for another day, we have a full training on that you can find on my channel. But that also is another really helpful tool to set up your causal logic and make sure your study is capturing the right population. That one's particularly complicated because of the externality of the mask benefiting others, not those direct individuals. But yeah, this outcome, so this This outcome can often be, it could be a clinical outcome, it could be things at a macro level, it could be things like GDP, could be things like income, could be educated grades if you're looking at student performance. The outcome is almost, almost all of your PICOs will need to have an outcome attached to it. It could have a set of outcomes, but typically that's what you want to see. All right, one last thing I want to mention that I think is easy to lose sight of, particularly in a literature review, is that literature review often kind of funnels like, functions like a funnel, not funnels like a function, in the sense that sometimes I see people kind of at the end of their literature review is where they need to be, but their end is at the beginning. And I think we're going to see some examples in the submissions I have from you today. What I mean is that sometimes you go broad in searching for papers in order to be able to spot what are gaps, what's missing in the field. So sometimes I see people really defining very narrowly a population and maybe that's where they want to be at the end of their literature review is the gap that they found and that's what they want to roll out the red carpet for their next research instead of research agenda. But if you start your literature review saying I want to look at diabetes in young women in Madagascar who are ethnic minorities, you're probably in your literary review only gonna find one or two papers and not have anything to review. So you almost have to zoom out a little bit and that's gonna give you, be able to help you make some contrast to say, well, we're actually missing these studies on young women with diabetes in Madagascar and that's where we need to go. So sometimes your perfect literature review, I like to think about as this funnel approach, which might be a step broader than where you're thinking is right now. And so, especially if you're not finding any papers at all, you probably need to take one of these PICO control knobs and dial them a bit out. So for example, if you had this Madagascar population of, I can't spell, young women with diabetes, well, maybe you need to relax that to all with chronic diseases, or maybe all women, or the entire population, or maybe even look at an entire continent or region, if that makes sense. So that's why PICO helps you dial in your topic.
[00:26:29] Speaker 2: And just on that, David, one of my concerns about people doing literature reviews is often because we've got so many checklists for systematic reviews, I mean, we have just PICO, I think is fine, but there are some that are very mechanistic and people just work through them. I really don't think you can do a literature review without knowing what you're talking about, what you're researching. You need to understand the topic. You need to read around it as well, because otherwise you will miss the obvious. And, you know, in that, I mean, lots of other ones, I mean, my immediate thing would be, well, maybe it would be relevant. I mean, you might look at the, Madagascar's an interesting example, because of course the population of Madagascar is largely from Asia, actually, just because of the way people move. So there may be, you might want to look at Sri Lanka or something like that, or some Indonesian islands or something like that. You need to think a little bit laterally.
[00:27:21] Speaker 1: This is also where it really helps. I see a lot of researchers using AI to try to help hone in on their topics. There's just no substitute for real human feedback. I mean, when you ask Martin about Madagascar, I wouldn't have known the comparison with Sri Lanka. And so really getting that real feedback from a mentor is where a lot of the real magic and creativity happens. There's just no substitute for it in this day and age. Okay, I think this is a good backdrop though, as a basis on lit reviews. Let's take a few questions. I can see we've got some comments from the audience, but I wanna go share now some of the videos that several of you brought up. Some of them are in health, some of them are in finance, and we'll take your questions along the way too. So let me read out a couple of the questions that we got at this phase. So I've got Dr. Jason Piarkowski says, I've been tasked, I'm gonna put it in the chat, I've been tasked with a systematic review of a medical administration research paper. Despite having done a master's in health leadership management studying research courses, I still feel overwhelmed with research and literature searches, especially systematic reviews. Okay, well, let me take a quick crack at that. Firstly, on my YouTube channel, We've got a hundred percent free and valuable step-by-step systematic review course. It's fantastic. A lot of researchers have been telling me that they get great results from it. And we also have available to people who join some of our mentorship communities, a full step-by-step course with feedback along the way. I might give you a sneak peek at that later. But yeah, it's a little bit hard to say. I mean, my first thing here is again, if you're feeling overwhelmed and getting stuck, I would ask Jason, have you defined your PICO? have you got the topic dialed in? Because I would say many people get to later stages and they're getting stuck and it traces back to an original kind of inconsistency or incoherence in the topic. Martin, how would you diagnose a question like this?
[00:29:24] Speaker 3: The first thing is that I'm not quite sure what this question is about because it's a systematic review of a paper.
[00:29:29] Speaker 2: So, you know, systematic reviews are lots of papers. So again, I'm coming back to what is the actual question here. But I mean, we need a lot more detail on that. So, I don't know quite what the question is that it's trying to answer and that's where Pico would help. What is the intervention and what is the, you know?
[00:29:52] Speaker 1: Yeah. Thanks. Okay. I've got a video here that I want to share. This one's from John and John says, and we'll pull up this video in a second, he says, I would like feedback on my draft systematic literature review to improve it until it's published. Awesome. So I love the motivation. I mean, that's everything we deal with in a fast track is trying to help you publish and thrive. That is what we are all about. And uh, due to hang on a second, let me see if I can make the screen a little bit bigger of John here. So hang on, here's John and we'll pull, pull up John's paper in a second. Let's see if I can get all of us on the screen. Hang on. There we go. Okay. Floor over to John. Let's let's hope you guys can hear this.
[00:30:35] Speaker 4: Hello. My systematic literature review is looking at the financial intermediation of savings and credit cooperatives. I've attached the file. It was reviewed and I got feedback on how to improve it and I would like more guidance on how I can do this. I was told my major challenge is problematization, that the contribution of the paper is not clear, it's not clear why we need to do a systematic literature review in that area right now, and And the actual study is like a report of what has been done and doesn't have a real contribution on theory, you know, a framework in the area. So I'd like to get support in that area.
[00:31:43] Speaker 1: Okay. John, thank you for sharing that with us. It actually traces back to some of the things we talked at the very beginning on your topic of establishing the gap value added, the research question, the nearest neighbor paper. So I'll be looking for that as we pull up your draft. Yeah, Martin, any reflection so far? We'll pull up the draft in a second.
[00:32:01] Speaker 2: Yeah, so I think I haven't had a chance to read the actual draft. I've just seen the beginning of it when we were doing just before the program went live. But, you know, savings and credit organizations are considered to be quite a valuable means of basically borrowing money from people who have money to lend and then pulling it and then redistributing it to people who need it. Like any bank, really, it's that sort of a function, but at a very local level. So they're small, they're community organizations. So the question is, are we clear about which ones, what size? Because I think we're talking about ones at a local community level. We're not talking about multinational banks, which technically might fall within that definition, but that's probably the easiest one. But then, what is the, first of all, what's the population? Because in some places, these will be the only means of getting money, other than money lenders and people who are extorting very high interest rates and things like that. In other places, there'll be a very active banking network. So you just go to the bank and get things. So, you know, but this would, so I think that's the second thing. Then what are you actually trying to achieve? What's the outcome? Now, I think what these can do and what they've been found to do is they can create a sense of community empowerment, which would be one perfectly legitimate thing. What another would be that they could encourage the emergence of small and medium enterprises, family, you know, new startups, things like that. Another would be that they would be a mechanism for avoiding, say, catastrophic expenditure, either through a health crisis or an environmental crisis or something like that. So what are you actually trying to achieve with them? Because they do a whole series of things. And then what is the intervention? Well, the intervention is, well, I think in your review, it was financial intermediation. Now, of course, the financial intermediation is the process of gathering the money, the process of distributing the money. So I wasn't quite clear. I mean, those are just the standard things you do. Are you talking about how well they do them? In which case you'd need to get some sort of a scale. I don't know that that's particularly helpful because these organizations have to have financial intermediation. That's how they function. So what is it about that that is the issue? So the main thing for me is what are you trying to achieve with them? Because they can achieve potentially quite a lot of generally quite good things. Whether they do that or not depend on what the alternatives are, and you're comparing them, what are you comparing them with? Are you comparing them with nothing? Are you comparing them with money lenders? Are you comparing them with local organized crime networks? That may be the place in some cases. Yeah. Yeah.
[00:34:52] Speaker 1: Martin, see, I think one of the things I've really tried to do with our training is, I mean, you can hear just the high value support that Martin's providing in thinking, and what What implicitly you're doing, what I've tried to do is break down the thought process that you and I have had so that we can replicate it and share it with many others. If you see what Martin's been doing, he's been immediately connecting John's topic to what are the big debates Martin has background knowledge in this area on. He's already asking intuitively, what are they trying to achieve? He's asking about outcomes, connecting John's topic to a big outcome that's important to a lot of people, such as catastrophic expenditure. He's asking about the C, the comparison. What are the alternatives? So, Martin may not even realize it, but just his reaction initially to the topic was giving suggestions around setting up a better, richer PICO that's going to fill a debate, that's going to speak to a gap, an important research question that a lot of people are going to care of and knock out of the park this critique that you got about, well, what's the novelty? Why is this important? If you look at the paper, and I know you haven't looked at this, Martin, oops, this is one of the papers we're going to next. And just look at the abstract. So on this kind of, if you've ever gotten a desk rejection, you need real human feedback because you can't decipher what's going on. You can't go do surgery to a paper when you don't know what the issue is. So avoid that temptation. Get feedback. But you can see here already in the title and the PICO, we're missing clarity about that outcome, which you very helpfully diagnosed. And in a way, this abstract reads, if you take a second to read this, you might even want to pause the video. It's almost like it starts in the middle of the story. It doesn't kind of, you know, right, the classic joke, a man walked into the bar. Well, you got to start with somebody walked into the bar and set the context of scene, the background. Why is this important? Why are we having this conversation? And it's just kind of says, well, we're going to provide an overview, but that's not really a research question. It's not telling us the first part of your abstract, well, what's the gap that you're going to fill? What's the value added? You just dive straight into doing it.
[00:36:56] Speaker 3: And there is a big literature on that.
[00:36:58] Speaker 2: This is not my field, but just from general reading, one is familiar with some of these things. But there is a real problem of access to capital in many low and middle, well, access to capital. The microcredit. The microcredits, things like that. And what happens as a result of that is that people borrow money from unscrupulous lenders and then get into debt bondage and they never get out of that. And that's a real problem. So this is a means of community helping itself. But here we see, sorry, go ahead.
[00:37:28] Speaker 1: Oh, sorry, Martin. Yeah, I was just gonna put this on the board as you were talking. So if we were gonna make an eco for this, because I think you have a way to kind of reformulate the research question a bit, because you're saying kind of the comparison would be those unscrupulous lenders.
[00:37:45] Speaker 2: Well, one of them might be, but it depends where you are, because there may be a banking system, but although the banks are there, many people can't access them. I mean, that's a big issue in the United States at the minute and many countries where local banking branches are closing down and it's all going online where you have problems with digital access, a lot of people even in high-income countries do not have access to data on their mobile phones. They may not have smartphones but more of a problem, they actually don't have access, they can't afford the data that you need to do some of these things.
[00:38:18] Speaker 1: So here in a way, what you've got, I mean I've seen this done for rural farmers trying to get access to capital, but, um, right now in your topic, all you've really got is the intervention is the kind of, uh, the savings and cooperatives. Yeah, but it's not connected. You haven't defined for us what the population is, what the comparison is, or what the outcome is. I think if you do this, once you have this, you'll be able to make a Pico linking sentence, um, to say, does access, for example, to, um, I don't know, certain types of savings cooperatives, I don't know what they're called here, help avoid catastrophic health expenditures or something where people are having to sell everything they have to pay for diabetes treatment or something. Yeah, one potential outcome. Yeah. And then the systematic review of, I don't know, let's say longitudinal studies, if you want to go further, to track this or interventions, RCTs, or whatever you're looking at.
[00:39:17] Speaker 2: Something like that, Martin, just trying to synthesize what yeah, but I think this is where you just, you just need to think through a little bit as to, again, it's about the plausibility and so on. So the population in low and middle income countries, first of all, they're incredibly diverse. So you need to think about, well, how might you break that down, you may want to say, urban and rural, and peri urban areas may be different. So the peri urban slums in many countries, it may be that there's a more organized system there and people can move around whereas in isolated rural areas they can't. So you need to think through exactly what it is you're trying to apply it to. I think as I said we've already said the comparison what you compare the alternative sources of capital are going to be different. An issue there is going to be just in terms of generalizability the question of land titles and Hernando de Soto a Peruvian economist has written a lot about this So that's the question as to whether people actually have assets that they can use as a collateral for loans and so on, and what are the consequences. And we did a paper some years ago looking at Ethiopia, looking at something similar to this. And it was whether or not families had one or two oxes, cattle. Because once you get down to your last ox, basically that's a real problem because you lose that if the loan goes wrong, then you really are in trouble. So you need to just think, understand the sort of pathway by which any of these might actually work. So just a matter of thinking it through.
[00:40:55] Speaker 1: That's really helpful. I just want to share one resource, John, and those of you who stick to the end, I'll share this with you as well. I'm going inside our research collective community at the moment to our systematic review course. And one of the very first starting points is how to define a good past topic. And for each, we break the whole process down in step-by-step chunks, but we've got a whole worksheet that captures a lot of this logic we've been sharing with you today. And these templates just, I mean, our researchers say 40% in our internal data show that they saved seven to 10 hours a week, about 20% say they would have never finished without it. And this worksheet will really take you step-by-step through the process of defining your topic. And so I think in your case, I think this is really going to help you, and this is going to conclude with your final validation of getting your PICO well-defined and clear, and give you tips to troubleshoot some of the common issues that you might be facing. Again, send a comment below, just put topic triage, and that will signal to me that you'd like a copy of this worksheet that we have available. Okay. I'm going to take some others. We have, so John, again, big thank you. I know it's always kind of a bit confronting to share your story. Thank you. I'm gonna pull up another one here, but in the meantime, I wanna take Daniela's while I'm pulling up one from Funcho. So I'm gonna pause Funcho for a second and then let's pull up Daniela. Martin, can you see this on the screen still?
[00:42:26] Speaker 2: Yeah, I'm looking on one. You're seeing a bit of my head.
[00:42:30] Speaker 1: We've got too many things open at once.
[00:42:31] Speaker 2: I've got my cameras on one on my laptop, but I'm going to use the big screen for this.
[00:42:36] Speaker 1: So this is in our Facebook group as well, guys, 100% free to join. And we give priority to people who ask questions here too. So Danielle says, I'm a recently accepted PhD student, I've been following the guides, which is an open access version, systematic review course, and ask, in PubMed, is it better to combine mesh terms and keywords for each concepts? Okay, so this is getting into the nitty-gritty of actual searches. So I just use MeSH terms alone and then some databases don't support complex search strings like Scopus, limits the number of Boolean operators, how can I insert consistency and coherence across databases and am I making my search too specific by using both MeSH and keywords?" That's an awesome question.
[00:43:15] Speaker 2: That's an awesome question. It depends on the question. That's an awesome question. Yeah, but it depends on the question that you're asking I think on that because it may well be that you do need some specific keywords within the MeSH terms are just too broad for for what your question is, in other cases it won't be. So I think, I just say it depends.
[00:43:32] Speaker 1: It depends. I don't like relying in PubMed on MeSH terms alone. So typically the best practice is to use MeSH terms and keywords. So treat the MeSH term as just another keyword variant of one of your keywords. By the way, many people struggle to figure out what their keywords are when doing a literature review search, and it comes straight from your PICO. So you've got your PICO well-defined, you're gonna pretty much know what those nouns are gonna be, the main nouns that are gonna be in your topic and keywords. So coming back to our example, the keywords that you would have here, right here, you would have the savings cooperatives and variants of that.
[00:44:05] Speaker 2: Yeah, because you won't have that as a mesh term. Yeah, I mean, that will not be a mesh term.
[00:44:10] Speaker 1: This you might have a mesh term though, on medical debt, maybe.
[00:44:13] Speaker 2: No, no, you will, but not on savings cooperatives.
[00:44:16] Speaker 1: No, but not on savings cooperatives. Yeah, no, you will, yeah. And if you do RCTs, that would be, there's a common, a previously validated filter. We have that available in the course. You can also Google Scholar for papers, but yeah. So just to say, so yeah, we recommend that. For some databases, I don't recommend searching, well, there's a line, friends don't let friends publish an MDPI, but I don't find it that helpful to search those individual journals like that in your search. You're just adding complication that doesn't have a payoff, really. But Scopus does. You have to go into the advanced search of Scopus. It's the same as in PubMed. You need to go sometimes a kind of rookie error on these searches is to just search. You need to make sure you're going into the advanced builder and building this up step-by-step. If we have time, we might get to one of these later here. If you can see this, I think the screen is getting truncated. There we go. Okay, let me come now though to, let me come to Funchal's question. And let me play this. Hopefully everybody can hear this. Thank you, Dr.
[00:45:27] Speaker 5: Skotla, for this opportunity. This is my first ever manuscript. And so, anything you think is appropriate is what I need. Beginning from the title, the structure, which journal do I publish this article in?
[00:45:45] Speaker 1: Thank you. So yeah, thanks for sharing that with us, Funcho. and here you can see what Funchal shared with us as a working title. Yeah, thoughts, reflections, Martin, on this?
[00:46:00] Speaker 2: Well, I mean, this seems to be a reasonable question. In educational interventions by pharmacists, we're talking presumably about professionally qualified pharmacists as opposed to community drug servers, just drug sellers, clarify that. On medication-related, I mean, I think, first of all, you've got three outcomes here. So medication-related knowledge is one thing, which will be relatively easy to assess. Adherence, again, relatively easy to assess. But biological suppression, well, you need to have — all of these are getting increasing levels of sophistication of collecting the data. So I would maybe think about — first of all, you need to have some sort of a causal or conceptual pathway for all of those and show how they're related to each other and among people living with HIV. I mean, I think that's pretty good, actually, as a question.
[00:46:57] Speaker 1: I think this is actually pretty good. And you can immediately see, those of you watching, I hope you can see how you could break this into a very clear pico. I hope you can see how this is a well-defined topic. And what's nice about this, too, by specifying the outcomes, Funchal can adjust. If he's finding tons of papers on adherence and not enough on on virologic suppression to do a review, you could then adjust. I've even had researchers who find a whole bunch on each of these three buckets and end up splitting them into three reviews, answering three separate questions. So I really like this a lot. I would just encourage you to look, now you need to establish the gap and see what the nearest neighbor systematic review is. So go make sure you find, again, that worksheet I just shared a moment ago will kind of force you into doing that step. Find the systematic review closest to yours so you can again establish what your contribution is over and above that existing paper. We won't have time to scope that out now, but definitely for the first paper, this is great. This looks like it would work very well for a systematic review. I think this might be a good point also, Martin, to maybe pause and share a little bit about the different kinds of literature reviews, because that's something people get confused about. What's a scoping review? What's a systematic review? What's a narrative lit review? What's an umbrella review?
[00:48:14] Speaker 2: Yeah, but just on this, I mean, one of the considerations here is that in most journals, you're looking for a paper of about 3,500, 4,000 words. By the time you've reviewed the literature on each of these outcomes, knowledge, adherence, theories of change, and the biological bits for the viral suppression and so on, you're going to be using up a lot of words. And then when you try to relate it to the broader policies and things like that, you're going to have a very big, this is a lot of material to fit into one paper.
[00:48:47] Speaker 1: Yeah, so it depends, I think, a bit on the, there's some piloting. If there are only 10 interventions that have been done that meet your net, you might then very, have to kind of have a broader set of outcomes. But if you've got hundreds, you might have to narrow it down and only pick one of these outcomes. And that's, again, where these control knobs and dialing in, in the piloting phase of your search comes in. Thanks for sharing this with us. Martin, we've got more questions. I just wanted to briefly make sure we've covered this point because I know this can cause sometimes some confusion. How, Martin, do you tend to steer people in terms of lit review, systematic review, scoping review, umbrella review?
[00:49:31] Speaker 2: Well, I mean, yeah, I mean, basically, it comes back to the question and how much literature you're likely to expect. So an umbrella review is where you're looking at already existing reviews. If there's a lot of reviews out there and what you're trying to do is pull them together, that's where you would do an umbrella review where the questions are still quite poorly defined and you don't know much about the topic. A scoping review is the most important. That's where you're saying what the literature is saying, what are the key messages, how important is context, for example, in that how much, you know, what other factors influence the relationship between the intervention and the outcome. A straightforward systematic review is clearly best where you've got a very clearly defined intervention and you know what the outcome is that you're looking at. A narrative review can be helpful just in terms of pulling together what's known. You've got to be careful with it, of course, because there is a risk of bias in it. But it may be adequate for telling a story. You're saying that something is important, for example, Or sometimes you don't actually need a formal systematic review to, I'm thinking of an example where we were making the point that, co-chair of the Lancet Commission on Corruption, where we're trying to make the point that corruption was an important topic that had been overlooked. And we did a narrative review on that to say, well, there's evidence that it's a big problem. There's evidence that it's been overlooked. And in that case, we were comparing it actually to the Me Too movement with sexual abuse and so on, where everybody knew it went on, but people weren't talking about it. So we were pulling in analogies and things like that. And a narrative review can be very helpful on that. So you're making a point, and you're raising it as an issue. And then having done that, you might go on to see what are the topics that come out of that. You then can do a scoping review. If you find that there's lots of existing reviews, you can do an umbrella review. if you find that there are lots of individual studies, but they haven't been reviewed, you do a systematic review. So again, it always comes back to what it is you're trying to answer.
[00:51:44] Speaker 1: Mad Fientist That's really helpful, Martin. And you kind of outlined a really nice algorithm for making that choice. Just in very kind of broad terms, my bias on the table, I recommend almost all researchers at the very beginning start with a systematic review, just because it ingrains really good research habits using with proficiency, a reference manager, it helps you sift through literature and forensically find separate wheat from chaff, separate important details from not important details, you have to overcome writing effectively, many, many, you have to define a topic very cleanly, many, many micro hurdles you have to overcome. And the payoff of that is you're going to have a review that's publishable. And you're going to have a burgeoning kind of growing mastery of your field.
[00:52:29] Speaker 3: So I just mentioned maybe two others.
[00:52:32] Speaker 2: I mean, one is a rapid review, because the difficulty is that, I mean, I have seen too many systematic reviews take on a life of their own, where people are doing them for a year, a year and a half, and then by the time they're published, it's another year. By that time, the question has been answered or it's become irrelevant. Now, I'm co-director of the European Observatory on Health Systems and Policy, so we often do rapid reviews where a minister wants an answer over a weekend a week or something like that. And there you are, you know, you're cutting corners, but you're doing it in the knowledge that you are cutting corners, you're recognizing the knowledge is contingent. And then there's an integrative review, where you might be looking at a complex topic, and you're deliberately trying to integrate insights from a whole range of disciplines. Because one of the problems with a lot of the social interventions that we're trying to understand is that they're being looked at by quantitative and qualitative sociologists, by anthropologists, by other different disciplines, and often they do not read each other's literature. So an integrative review is a way of pulling this together. So those are just two others. So I think just be aware that there's a portfolio you can draw on.
[00:53:40] Speaker 1: Excellent. Excellent. And guys on my channel, I've got a dedicated video going a bit deeper into different types of reviews with a guide on how to choose the right one for you. We're going to take a couple more questions. I know we're going to come up to the hour we might run just a couple minutes over, but But I want to take Ibrahim's and Qayyum's, I see Ibrahim's in the chat, we've got a submission from Daniel, Saima, and also Ruth. So also want to be respectful of Martin's time, graciously taking time out of his very busy day to join us. So here's Ibrahim's question. I think I'm going to try to pull up Ibrahim's video in the background while you're reading this, Martin. Oops. Oh, hang on a second. So let me get that up and see if I can get Ibrahim's here. I was having trouble hearing the audio, so I'll have to hear, see if you can get the audio. Yeah, I'm having trouble with the audio. Okay, let's see if this will, maybe. Oh, no. Abraham, we can't hear you so well, but okay. But I think we've got the gist of the idea here. I want you to guide me on, for a systematic literature view on my topic, and integrate, There's no fluvial system investigating special temporal variations. River depth with, oh man, it's a long one. Case study of river regassa. Um, I think guys now with aided by our PICO discussion, you guys will be able to see if you can diagnose what is likely going to be, uh, the first immediate issue, uh, with this topic. And, uh, what, what, uh, give people a second to think about what do you think Martin?
[00:55:15] Speaker 3: Uh, okay.
[00:55:16] Speaker 2: Uh, I'm not sure where the river regassa is. not one I'm familiar with but it sounds interesting anyway. You know this is way outside my level at my area of competence. I'm not beginning to pretend you know that I know what's going on but I can you know I can see that what it's talking about is you know with presumably this is to do with flooding and changes in water flow and so on which is obviously becoming much more important with climate change. Yeah.
[00:55:47] Speaker 1: Yeah.
[00:55:47] Speaker 2: I'm not really, really sure. I don't know where to start.
[00:55:51] Speaker 1: Well, initially, I think this fits with the funnel model we developed before. This is going to be maybe the case study you want to do, having done your literature review, to answer some remaining questions in the field. Maybe you want to find what literature has been done, doesn't apply to the specific river Ragazza, which has something very interesting to say about fluvial systems. The challenge here is, I think this is too narrow. If you go look for literature, this PICO is so, I mean, you've defined very, very tightly your topic to where I don't think you're going to find a lot of studies already done on this. So you need to dial your knobs out to maybe look at all fluvial systems and define a clear outcome that's widely interested. Martin already mentioned an outcome just naturally gravitates there to flooding. And so maybe you want to look at the factors that mitigate or exacerbate the risks of fluvial systems flooding more widely than the River Ragazza.
[00:56:50] Speaker 2: So I mean, I think, you know, again, without knowing exactly what this is about, I mean, I can just think through that this may be important, particularly with climate change, you've got the instability in many mountain ranges at the minute, you've got the rocks are becoming unstable because of the loss of ice and various other things and cracking, and you've got a lot of debris coming down which can be swept into the rivers, which then along with drying out, I mean the Rhine for example was unnavigable a few years ago for a while, could lead to the loss of of navigability of some of the, and this is just off the top of my head because you know nothing about the subject.
[00:57:33] Speaker 1: You guys are seeing Martin in action. Why Martin has mentored effectively and produced so many researchers who went on to become prominent professors because just naturally gravitates in his thinking, well, what's the big debate? What's something I know that a lot of people care about? How can I connect this topic to that, to widen the international audience and relevance of it? Because with your research, it is helpful to swim in a big pond and take on, maybe find a small niche, but in a big pond that answers a big research question, that's gonna give you a much bigger boost for your career, your visibility, because your research is almost like planting a flag out there in the world and saying, I'm an expert on this. And so I do encourage you to not be shy about asking big questions, which Martin, you've always encouraged us to do.
[00:58:22] Speaker 2: No, but I would say I can see, I'm sorry, I still don't know why I should have Googled where River Riyadh is. But I mean, these issues I can see are clearly going to be important in the parts of India and Bangladesh in particular that are being fed by the melting glaciers in the Himalayas and then in the Andes in South America. Those are the two big areas off the top of my head where this is a big issue. We've got massive amount of heating, particularly in the Andes at the minute. So I can see the logic. But this is not my area at all.
[00:58:53] Speaker 1: Thank you for sharing this with us, Abraham. We've got another video question here. I'm just gonna pull up briefly, if I can get the source up one second. Share, no. How do I add a scene? Media. Bear with me, guys. So we're gonna add a local video. And this is from Abdul Qayyum. And I hope this is all right. All right, sometimes, guys, the audio is sometimes tricky for us, but we're trying our best to capture what you're saying. Okay, let's see if this will, this will work. So it's Dr.
[00:59:28] Speaker 6: I'm coming from Bangladesh, Dhaka, Bangladesh. Actually, I was preparing to do PhD in Israel. So I was preparing a research proposal for my fellowship. So I have done it. I have two specific questions regarding the research proposal. How I can use sci-space to make a research proposal without violating the research ethics? No.1 No.2 I saw your video of GIFT methods for preparing a research proposal, how to utilize this method more efficiently. And I was very, I'm very excited to join your workshop in USA at 12 p.m. ETC. I didn't get any, any...
[01:00:41] Speaker 1: Okay, okay. I hope you could hear that some, Martin. I'm going to share my screen and actually pull up the proposal here that Abdul shared with us. Hopefully you can see this. And Abdul asked about our GIFT method, and our GIFT method is one on research proposals that has four components at its core that are the essential ingredients to an effective research proposal. And it's been in every successful grant proposal that I've used. It's probably been in there in a stealth way, Martin, I don't know if you've come across our GIF method, but the G is for the gap. You have to have very good clarity on the gap. The I is for your idea or intervention, what it needs to come into contact with your gap. The F is for feasibility. You've got to prove to us you can actually do what you're proposing. And the T is for a timeline. You also have to show that you can deliver the goods in the time that the funders, or in this case, your PhD program wants to see you finishing. So I'll be I'll be looking at the proposal for each of those elements and try to make suggestions where you can fix it. So Martin, yeah, I know I'm hitting you cold with this one as well. This has some of the things that I spot immediately looking at this remind me of the earlier paper that we looked at actually that I had said started in the middle of the story. This one does a little bit better about setting the context, but yeah, anything immediately jump out to you, Martin. I know we're putting you on the spot with this.
[01:02:12] Speaker 3: You certainly are. Any other topics that you want to think of that I know nothing about?
[01:02:19] Speaker 1: So while you're looking at that, so immediately when I see the research proposal, I like to see, make it very clear for your readers what the big question is. I find exploratory work, while it's not bad, it's just often a lot more difficult. When you're exploring, it's almost like there's no playing field. You haven't set a boundary of what's in and what's out and it can get very big and overwhelming. So especially if you're just starting out, I recommend just strategically to avoid big exploratory topics. Is that strictly necessary? No. it can be helpful to you to save some time. And then I'd like to see what the, this is trying to say the, I think the question here is, the extent to which amalgamation of Western fashion and traditional attire affects consumer perceptions and augments the commercial viability of both fashion paradigms in an increasingly globalized environment. So I guess I'm trying to say does, how does the fusion of Western and traditional attire affect consumer perceptions? To me, that's an outcome that's quite, quite broad. And again, Martin was saying, what do you wanna achieve? What do you wanna show? That's in the outcome. I think it would be even helpful to lead with a little bit what you want to show and achieve. Go ahead. I wanna get to the idea and the method and feasibility.
[01:03:46] Speaker 2: Yeah, so again, just this is off the top of my head. I'm not knowing anything about this area at all, but things that stand out to me would be to look at issues of cultural appropriation, because there are... And also maybe to look at the historical idea of taking ideas, of taking fashion ideas. If we look at the history of global interchange going back, I mean, really at least 1492, but around both with the movement across the Atlantic, but also around European explorers going to India by sea and so on. You have had lots of examples where fashion devices, you know, fashions and so on, did actually make that move. And, you know, there are lots of examples of where some Asian fashions were taken to Europe, and also where people, there's a big literature and certainly people, Europeans going to India and adopting Indian fashions and so on. Now, part of that, there were a whole series of motivations. So, I think looking at the historical, reading the historical literature on that would be quite useful to understand what issues have arisen to put it in this historical context of the exchange of ideas. And I think the second, then, would be looking at more contemporary issues of cultural appropriation, because that is a bit of a backlash against that. So those would be my first bit of reading on both of those, just to get it in the frame, coming from a position where I know nothing about it. Because there is an argument that you can have the leading Paris fashion houses taking ideas which people in low-income countries say, well, hold on, that's ours. We've been making that for years. Maybe.
[01:05:31] Speaker 1: So I like that a lot, Martin. Again, Martin's, as ever, connecting you to bigger debates that are international in nature to say, what does this case study about Bangladesh tell us that's of interest to the whole world? And here I can see that I want to unpack your passions for a bit because I can see you say something that you want to advocate for a more inclusive fashion industry. So in a way that kind of signals bell going off for me that you think something is not inclusive. You think there's something wrong with the fashion industry at the moment. I'd like to see that almost come out a little more clearly. I think behind this, that's kind of veiled in the way you've written this up is you're concerned about something. And that could be the cultural appropriation or misappropriation Martin's talking about. So yeah, I'd like to come out in a research question so that everybody immediately gets why is this important? Why are we having this conversation now? Why do we want to stop what we're doing and pay attention to your study or invest in it? The other thing here, you say you're going to conduct a comprehensive analysis of this. It's kind of unclear how you're going to do this. I think you need to give your introduction room to breathe. We follow a peer system on writing more training on that on my channel, but each paragraph makes one point. And this is kind of smushed a lot together. You need to give these points room to breathe. And especially this point about the analysis you're going to do and start painting a picture for us of what exactly you're going to do. How are you going to go and do that? Now I know you get into that a little bit down here when you get tried to get into the significance of the research. I'm scanning this. We've got a preliminary lit review, but all this here needs to be about driving and explaining very clearly what is the gap in the literature, which is completely absent at the moment from scanning this. It won't go through all of this, but you have a lot of background, but you have less in telling us what specifically you're going to be doing. Even in your methodology, it's not painting a picture that's very clear what you're going to do. I would try to simplify this, this message is still very thin and having read it, I still don't know what you're going to be doing. So I guess the second part of your question was using size space ethically on the research proposal. And I guess my point here is that I don't think you're in a position, I think you've got to get the fundamentals and the nuts and bolts of the proposal right before you can use AI effectively. If you just try to superimpose AI onto this, it will start polishing something that doesn't have the fundamentals right, and it will just amplify some of the issues that are already there. It won't solve them for you. So I don't think it's, the question you're asking about how can you ethically use it or not, I mean, I guess first, I don't think size space is the right tool for where you're at. And second, I think you've got to get the fundamentals of your proposal right before you begin to integrate powerfully AI to save you time.
[01:08:25] Speaker 2: Well, I also, I think this is a classic example of where you need a whole systems approach. We do a lot of work on systems analysis because it's not just the production of the fashion. Clearly, there is a very, very large textile industry in Bangladesh, we all know that. The problem is that those are mainly ready-made for export, low cost, low wages, and so on and so forth. So is there, I'm not sure, again, the person who put this in will know, the extent to which there is the training and support for design and high-level design and fashion and so on in Bangladesh? Are there enough people there who have stayed there and not emigrated? I mean, the real issue here is that you will probably have a lot of the high-end fashion which is being sold in the West, probably made in Bangladesh anyway, but being designed by designers in the West, made by low-income Bangladeshis, and the profits, the markup, will be accrued in the West. So that, I think, is one of the real challenges that you want to face. For somebody who's actually trying to start up in Bangladesh, you're going to have huge numbers of obstacles. You're going to have tariffs. You're going to have access to capital that we've already talked about. You're going to have to have those relationships for, particularly, marketing, because you're going to have to deal with the big global marketing brands to do it. So you're going to have to look at all of the bits that are needed to take an idea to fruition, to design it, to manufacture it, to export it, to market it. And all of those are not just the production of the good. Again, just off the top of my head, but I try to think from a systems point of view. What is it that you need to have in place for this to work? And it's much more. But maybe it's in there, so I haven't thought about it.
[01:10:20] Speaker 1: I think it's also interesting, Martin, if you start tracing the social life of these fusion garments, where are they produced? Is the cultural hybridization coming? Is it happening on the supply chains? At what level? Anyway, you raised a whole lot of questions. I mean, Abdul, I hope this is giving you some food for thought, taking the topic forward, but I would really encourage you to get, at the very minimum, the gap very, very clearly stated up front.
[01:10:46] Speaker 2: Yeah, and actually there is a book that's been written, which is somewhere in my bookshelf behind me, of which there are thousands of books, but, and it's something like the social life of a T-shirt. And it is, I think if I recall correctly, about the production of t-shirts in Bangladesh and all of the things that need to be in place. So I'm sorry, I can't remember the exact detail, but that book has been written, which looks at all of these issues.
[01:11:10] Speaker 1: But yeah, but again, trying to connect into your topic into a big debate into the field will amplify your impact. Because it would be a shame to do a very, I mean, we see this a lot. I often say 90% of your success comes from getting your topic right. You can do a detailed, perfect technical study, But if the topic is small or narrow, few people are interested in it, it's not going to have wings. It's not going to publish well. So, that's definitely a theme coming out from today. Martin, I want to respect your time. We've got a few quick questions. If you need to shoot off, I'll just hear from you.
[01:11:44] Speaker 2: I'll have to go at quarter past.
[01:11:46] Speaker 1: Quarter past. Okay. Let's take a couple last quick ones here. This is coming from Daniel. put it in the chat who's been using our FastTrack AI mentor that's freely available in ChatGPT. He's saying he's using ChatGPT to help rephrase or improve clarity. It's been very helpful, but he's starting to feel unsure and worried that professors might think he didn't do the work himself. That was entirely generated by AI and wants to make sure he's following university guidelines where it's still considered his own. Is there a way to use ChatGPT in a safe, responsible way within academic integrity rules? We could do a whole session on this, but great to get your thoughts on this, Martin.
[01:12:18] Speaker 2: I think this is a very rapidly changing area. And I think, I mean, I'm using, well, first of all, I should say, we're currently writing a book on AI in the health field from the European Observatory. So we're sort of getting deeply embedded in it. Somebody, Trish Greenhalgh, who you know, David, said on one of our WhatsApp groups the other day, that the question of whether AI will replace doctors is not a good question. It will not replace doctors, but what will happen will be doctors that use AI replace doctors who do not use AI. And I think the work that we're doing with, as you know, Ricard, who our colleague that you and I are working with, is really coming into this area of the interaction between the human and the algorithm. And you need good algorithms, and you need good humans who know their topic, and you need to carefully work with it. So I think that there's no easy answer to this. If you just delegate everything to AI, you're going to get complete nonsense, and you won't know it. But if you are, if it's something that, you know, going back to the size space, for example, I mean, what I will do with size spaces, if I already know from my previous reading, that there is, you know, that, that I need a paragraph on something, and it's not critical to the the paper, but I just need a paragraph, and I need a few up to date references to illustrate it. It's very good for doing that. And then I read the papers and make sure they really do say, I mean, it brings in a lot of junk as well, the size space, I'm afraid a lot of low-quality stuff, but within that you will probably find good things.
[01:13:47] Speaker 1: So use it for very specific things and Validate it and check it and so on And and I say err on the side of transparency Yeah, journals require this disclosure, but um, there's nothing to hide just so yeah Yeah, that's gonna change more and more.
[01:14:03] Speaker 2: This is going to happen. I mean people are going to do this Yeah, and why not and then the way a spell check check. Grammarly is another AI tool which many people use. I use it because I used to when I got my colleagues and maybe who had more difficulty writing, I used to spend hours going through their manuscripts and now I use Grammarly and I just check one at a time and it just saves me a huge amount of time. It doesn't always get it right either. This is the thing.
[01:14:34] Speaker 1: Yeah, Martin, very last question. I know I haven't gotten to all of yours and I see several of you put questions in the chat as we went along. I'll come back later and answer those. I'll leave, Martin, my contact details below and as well, if you're interested in working more closely together, I've got a link that I'm going to put here on the screen where you can book a call with a member of our team, see if you're a good fit to work together with us to accelerate your publication. But yeah, last, last question here, Martin. This is about a common challenge I think researchers, especially in developing countries, face about trying to overcome article processing charges and the ways they try to deal with it don't always – might have some false starts. So an editor had a special issue that didn't have APCs, so Hiro Saima contacted the journal editor. He replied they need to analyze the title and abstracted the manuscript to prove they they could publish it on their subscription-based model. Subscription-based model is great. We often encourage our researchers to search, map out the journals they want to submit to. We have a whole template for doing that. So that's good. But then she says, when I sent him my title and abstract, he didn't reply, and then didn't reply after follow-up. What do you think of that, Martin?
[01:15:49] Speaker 2: You know, it happens to us all. So we've got a paper that we, not to do with this, but we made fairly minor revisions, And then the editor was off on leave for several months. And it just got stuck there without being sent out to the reviewers. I mean, editors are very busy people. Some journals, really, it's one of the things for the editorial boards I'm on. I really make a point that you need to speed these things up. And yeah, I think that's not really acceptable. And that's a problem with the editor. They should get back to you. I mean, I know that it's a lot of work, but it's still not. This is important for people who are writing. And the whole publishing model's got really problematic anyway, but that's for another day.
[01:16:30] Speaker 1: That's, I think, another day, Martin. Maybe another session on the future of publishing or the demise as we know it. I saw Nature, for example, just converted over to a fully open peer review model. We've been seeing more what appear to be AI-generated peer reviews coming through from our researchers of publishing. So a lot more to be said. First, Martin, just a huge thanks for being generous with your time. This is, I always learn a lot from you. I hope all of you in the chat feel the same. We're gonna be doing more of these. If you wanna submit your video question in the future, that you can find the link. I'm gonna put the link up here very quickly to submit your video questions so that you can take part and we'll go into detail about what you shared with us. And again, if you're interested in working together in a more intimate way, do get in touch And we'll see if you could be a good fit. We have a range of different support programs from our step-by-step courses that save you a ton of time through to directly working together and publishing. So yeah, I encourage you to get in touch. And Martin, are you happy again to leave some of your contact below in case anyone wants to touch?
[01:17:45] Speaker 2: But I may not be as good as getting back to you as David is, but I'll just say.
[01:17:51] Speaker 1: Might be as bad as some of the editors, but Martin, I think you can be forgiven for this, how busy you are. And always remember, Martin's a great example of this. There's no ill will, there's no hostility, it's just I've seen, Martin, the state of your inbox every morning. It's amazing anything gets through to you, through that chaos. All right, everybody, that's a wrap for today. We will see you next week.
Transcription
by a Professional Transcriber & Editor
Captions (FCC/SDH)
by a Professional Captioner
Transcript Editing & Proofreading
by a Professional Editor
Translation (text/audio/images)
by a Professional Translator
Subtitles
by a Professional Subtitler
Video Description (AD)
by a Professional Describer
Transcription Services
Add new speaker
AI Insights
Summary
Copy link
