Speaker 1: A literature review is a summary of the existing research on a particular topic. It's typically done at the beginning of a research project and I did one for my undergraduate thesis, for my master's thesis and for my PhD thesis. And in this video I'm going to answer all of your literature review related questions. The first thing is how do you start a literature review? Well, to start a literature review, surprisingly or not surprisingly, you need literature. Where do you find that literature? Well, there's a ton of places. The first place I would go to is illicit.com. This is a new AI tool which allows you to ask a research question and get all of the papers related to that question. For example, here I can say how effective are conditional cash transfer programs? It will go away and search more than 125 million academic papers and here are the first four abstracts here. And here are all of the different researched peer-reviewed papers and that means that experts in the field have looked at these papers and said, yes, they are true. They are something that is a valuable contribution to the research field. So that's why you should be reading them. And we can go through and see that we've got a little summary and we just click through all of these and we can go and read them individually. That's one way, semantic searching. The next thing you can do is use Litmaps. Litmaps creates a map of literature for you to search. We can go in and create a map. Here I've created a map from one of my peer-reviewed papers that I wrote during my PhD and you can see I get a nice map of all of the other stuff that I need to read. You can do this with a single seed paper or you can put in a load of different papers in this tab in Discover to find out a load of different papers that you need to read about. Then you can also use something like Google Scholar. This is old school. This is like OG science and research. You'd go in, you just type keywords. For example, charge transport in OPV. So I'll click here and then here are all of the different papers that I should consider reading. Clearly, you don't need to read all of them but we'll get into that in a minute. But this is where you start. You start by searching the literature. You can have a look since 2024, since 2023 and this is the foundational activity for any literature review. Get comfortable searching the literature and you'll become a power user of all of the literature that you're about to write about. Before you start reading any literature, you need to have a literature review outline to work with. So this is the general structure of nearly every literature review for any field. It goes like this. First of all, we start with an introduction at the top. This introduction gives background information about the research field that you are investigating. It's in a reverse pyramid shape because this is the very, very broad step. This is where we're just sort of like looking at the overarching umbrella of our research field. Then optionally, we can talk about background and methods that are used to look for the research that we're going to talk about in the literature review. For example, you may want to say we looked at these databases, we looked at these sort of questions and background is the background of the field that you're specifically interested in. So we're going a little bit deeper, which is why it's the next step down on the inverse pyramid. Then we need all of the main text and this is all of the literature that you found searched by either theme. So you sort of group it together as like, this is a group of research that I can talk about because it's under one theme. Here's another theme or here's another theme and you've put research under that. So in here, you may have one, two, three plus themes under which you will talk about literature or, which is very uncommon I think these days, but you may be lucky that you may be able to sort this based on time, which means initially these people did this and then they did this and then they did this and that's how you structure your literature review. So you say they did this first, here's all the literature in the initial stages of that research, then they did this, here's the next stage of research, the evolution of that research field, here's the next stage. So it may be theme or time, it's completely up to you which one you use, but most people use theme. Once you've outlined all of the main themes and you've talked about the literature under that theme, then you need to have a discussion to bring it all together. This is where you're looking at all of the research themes and you're talking about your specific research question. Why are you doing this research into this literature and how does it help you sort of like answer the research question or the interest you have in a particular research field and why you're looking at the literature in the first place. And then you're looking at conclusions. Based on all of the stuff that you've read, all of the individual themes, all of the chronological studies, all of the papers you've included in this literature review, what conclusions can you make specifically about the current state of the field? And that is the general structure of nearly every literature review ever produced. Now, there's an easier way to do it obviously. What I like to do is go to ChatGPT and I just say, create a literature review outline for a study about and then whatever I'm interested in. Here I've got an example where it says, the effect of climate change on plants. And as you can see, it says introduction, background and here it says I want basic concepts of climate change. Then it says general impacts of climate change. Then we want direct effects of climate change on plants. So you can see we've started broad and we're getting narrower and narrower as the literature review goes on. And then we've got different themes. So we've got indirect effect of climate change on plants. So altered pests and disease dynamics, that's a theme. Changes in land use and habitat, that's a theme. And then we've got other themes underneath. So this is how you can easily structure and get a first kind of draft of the structure of any literature review that you're writing for nearly any subject. It's just amazing. And as you can see down here, the last one is conclusion, summary of key findings and then final thoughts on the importance of further research. So this is how we can use ChatGPT to structure our literature review outline. Nice stuff. Once you've got all of the literature you need to read and you've got a structure under which to put that literature, then you need to just write. You type out all of the stuff in your literature review. Before you do that, you may want to have a look at something like explainpaper.com that allows you to quickly understand peer-reviewed papers. Peer-reviewed papers are notoriously hard to read. They're dense, they're thick in academic language. And here, it's a really nice way to just get the simple summary. And I think this is one of the most powerful ones, explainpaper.com. All you need to do is highlight a certain area and over here, it will say, okay, explain your explanation. As a middle schooler, we can move this up and down and then we just click explain. And underneath, it will tell you the undergrad explanation of what you've just highlighted. A really great way, particularly if you're early on in your academic career, if you're undergraduate, if you're in high school, this is a great way to unlock all of the power that's behind the horrible language found in peer-reviewed academic papers. Once you understand what's actually in all of this, you've collected them into themes, you need to write it. There are a few tools that you can use. So you can use jenny.ai, that's an auto-writer for research papers and literature reviews. You can use yomoo.ai. And that is another sort of like auto-writer for peer-reviewed and papers. But to be honest with you, the best thing you can do is sit there with a Word document, with a Google document, Google, what do you even call that? Google Docs? Google Word? I completely forgot. Anyway, you know what I mean. You sit there with a word processor and you start typing. You put in your structured headlines and then you say under each one, what literature you're going to mention and you start fleshing it out. It takes ages and ages and many, many revisions. Make sure that you get someone you trust or your supervisor to look over it as you're writing it. Maybe each chapter or each theme that you write, you get someone to look over it and then at the end they look over everything all together. It's a really, really long process. It takes such a long time. For my thesis, it probably took a good few weeks to get all of the information into a sensible structure and literature review. So here we are, here's one of the themes. Overview of photocurrent generation in organic photovoltaic devices. So that's just one of many, many themes in this thesis and depending on what stage of study you're at, it could be long, it could be short but let's talk about that next. Okay, how long should a literature review be? Well, there are no hard and fast rules but I like to think about it like this. Is there enough in your literature review to provide enough context to what you're doing and what you're researching? Is there enough context for you to understand the problem that your literature review is looking at and addressing and also, is there enough data in there to talk about the up-to-date research and where the current state of the field is? That's really what we're looking at but here's some rules of thumb. So if you're doing it for an assignment, one thing I recommend that you look at is about 3,000 to 10,000 words. That's normally good enough to get an overview. For example, in my undergraduate thesis, it's only about seven pages. There's not much in there. There's some fancy diagrams, there's lots of references but ultimately, it's about seven pages. So it's not much. So 3,000 to 10,000 words is all you need for a small assignment or an undergraduate thesis whereas for master's and master's theses and PhD dissertations, one thing I recommend is you look at what's normal for your field. In some fields, it's like 10 pages. In other fields, it can be up to 40 pages but ultimately, as long as you have enough information and literature to be able to provide context to your problem and you provide an up-to-date representation of that research field, then you've got enough in there. Like I said, I like to use just the guide of what is normal for my research field before I start writing my thesis so I can say, okay, normally it's about 20 pages and therefore, I need to fill 20 pages worth of stuff and that is a good starting point for almost any literature review. So there we have it. That's the introduction to literature reviews. I'd love to know what you think and also, I have got so many videos on this very channel about literature reviews with AI, how to find literature using AI tools, how to write it in seconds using tools that are available online. I'll put all of the links below in the description so you can sort of build on the knowledge that we've gained in this video but if you really want to go look at a powerful video, go check out this one where I talk about how to write an exceptional literature review using AI. You won't be disappointed. Go check it out.