How To Automate Your Literature Review ETHICALLY Using ChatGPT (Prof. David Stuckler)

hey professor Davis suckler today I'm going to show you how you can automate your literature review using chat GPT but unlike other videos you may have seen I'm going to show you how to do this in a completely ethical way and I say this as a professor because I want to make sure that you feel confident you know sometimes with these tools it can feel like cheating and because in many cases you are but I'm going to show you how to do this without plagiarizing and without uh running a foul of any of the rules

of your University or of publishing wherever you are in your journey as a student all the way through to being a researcher or Professor like myself now I'm also going to show you kind of the right way to do things and how you can supercharge them with AI I think it's really important that you learn the right way to do things going forward so that you can then augment with these new tools that are available that frankly I wish I would have had when when I started where you are uh Flash Forward now I've post

400 articles and peer rreview journals I could have done this a lot faster using the tips here finally of course is YouTube if you stick around I'm going to show you a secret tip for your writing that's going to transform the way you think and help you feel really good about your literature review and know what you need to do doing the first past yourself so let's dive straight in we're going to cover the four stages of writing a literature review starting with defining your research question now you need to at least have a topic

in mind I've got an example here from a student I've been working with who asked hey you know Chris what's your research question uh I don't know I'm interested in school impr Improvement plans exactly so many of you might be in that very same place you need to turn this into a clear question and Define boundaries so that you don't get lost to say take a topic like I don't know uh feminism I want to understand feminism well you're going to run in circles and get lost feel frustrated and confused you need to connect that

to an outcome to study a relationship like the impact of feminism on income inequality that's going to put boundaries and parameters so that you limit the space of your search so you don't get get lost and many of you just want to get this done in in a pragmatic and reasonable amount of time so okay so School Improvement plans and you can see even if you just start without that outcome chat gbt is going to help you so it's going to list out help me find a research question it's going to list out a series

of potential questions um such as this one how effective are School Improvement plans and enhancing student academic performance in underperforming schools that's great it's immediately uh really connected the outcome and defined also here the population uh under performing schools kind of your your tools that you have available in refining your research question you can Define the population here US underperforming schools as a population outcomes here's student academic performance um what you're looking at sometimes the intervention or exposure here at school Improvement uh I really like this question and ultimately working with a student that's what

we went with but you can see there's also a many of other questions that you can look at and some of these you might want to integrate into your review anyway so keep them in mind when you've got this and you've defined your research question normally now you would move to the second stage which is developing a search strategy to go find literature on This research question except with this new breed of tools we can short circuit that process and we're actually going to go to step three and invert the order and then come back

to step two so step three is we're going to develop an outline and a structure and instead of conventionally getting material and developing structure we're going to use support of the power uh of chat gbt to go create a structure and outline Force we might refine this later but it's going to be a first start and help focus our search so let me show you how to do that okay so scroll down and let's say hey chat GPT let's go with the first one can you write an outline for a literature review with a structure

and potential themes ever willing to help certainly uh it comes up with a structured outline it's really important that you put structure and potential themes in your outline so uh it gives you working title an introduction it's going to set up you need a little place to put your methods if if you don't have a method that's really the search strategy I said in the Second Step we're going to talk about that in a second um and it gives you a series of outlines section one an overview maybe historical context section two really getting into

the meat of the article the effectiveness of these School Improvement plans uh with subsections there section three challenges and limitations this is great this is going to save you a ton of time you can adjust this outline say hey chat gbt calibrate this I need this to be around just thw essay if you're doing it for a school project if you're doing this as a paper you want for a peer-reviewed article maybe you want to save towards a 3,000-word structure you might want to eliminate some sections you may even find as you go through and

I'm going to show you an example in a second that you're going to tweak some of these bullet components of the outline um so remember this is not cheating right CET GPT hasn't done the work for you this is essentially like you've had a conversation with me or another professor to give you a framework tools and methods from within which you can roll up your sleeves now and do the work so how are you going to do that like I said next step you need a search strategy and the starting point that I recommend especially

if you're doing this for the first time is to head over to Google Scholar and we're going to do targeted forensic searches on each of these outline themes so what we need to do is we're going to copy and paste this outline we're going to bring it over to word and then we're going to move to Google Scholar start searching and use our strip method a method we've developed to really help you in your literature review but while still avoiding plagiarism let me show you how so back here to chat GPT and we're just going

to copy all of this and uh drop it let's pull up a Word document here we go we've got we got our structure here in our Word document and now what I'm going to do is go straight over to Google Scholar so let me open up a new window Google Scholar and let's pop in school Improvement plan sip Effectiveness in the US and what you can see we're going to pull up a bunch of different articles I found one here that's worth noting and what we're going to do is we're going to click on this

article and the next thing you need to do you need to start saving these articles so if you don't already have it you need zoto and you'll see a little connector if you set this up correctly save to zoto so click on save to zoto it's going to ask you which Library well hang on wait a second I need to create a library create a dedicated library for your project so come in here create a new collection so by if you're not using zotero but you're using cat GPT you really need to have a conversation

with yourself you're using Advanced tools but not using the basic one it's like don't even have a cell phone and you're trying to make a phone call and and you're still using landlines this is fundamental you need zoto okay I digress so um back here zoto uh is free I I'm 100% committed to these free platforms committed to open access uh because I I really want to reduce the barriers so that everybody in the world wherever you are can do high quality research so here we go create let's just call this sip okay and zot

is going to save it the item right here click on my library I'm going to pick the library I want it in sip bam done it's going to come back to zotar I've got it here later on when you get the PDF you can upload it there we've got a full video on getting started with so tarot you need this okay so now that we've got this here we're going to avoid the risk of plagiarizing because we're going to site everything I'm not going to go through the full article but I'm going to show you

how to use our strip method so I can already see here several components of this abstract that are relevant I'm just going to pluck these bits out and we're going to move this into our working document here in word and you want to make sure you get the reference here so you avoid any risk of plagiarism um so going do this automatically but I'm just going to pop this in here for now so I've got it here and if you look in this abstract you're going to do this with a key you're going to forensically

go through your article and pull out key bits but I can see different points that fit in different sections so this one here zooming in says that the empirical evidence based on school Improvement plans is limited great that's not just me saying that that's somebody who in 2023 said hey on This research question there's limited limited evidence so what's that going to do that's going to help Focus the purpose of the review to say that there's a gap and I'm just going to put this here and drop this and say this is Mariano at all

2023 and U we're going to rewrite this later we're going to rewrite everything later we're just dropping the evidence in the right place in the outline um same thing here I can see here ah there's even less on the intersection of school equity and planning oh that's really interesting and tells me hang on wait a second maybe that's a really important part of the these studies on academic outcomes it's going to be a subsection of academic outcomes maybe I'm going to look at the equity side of it maybe it's going to get its own bullet

point but for now this Equity Point here is really important and I'm going to drop this Mariano study here to just say hey even less literature focuses on this point about Equity um which is really important the median student might not reflect underprivileged students for example um so we'll come back to that so you're going to repeat this process on each of these bullet points until you hit saturation that is when you go into Google Scholar you search for more articles and you're not getting anything new you've captured it all there's nothing left to say

and that sat satation often will hit when you've hit maybe 5 to 10 articles for each of these bullet points that's often enough but it's going to depend on how mature your field is so again this is not unethical because you are doing the research and heavy lifting yourself but CAD GPT has lowered the barrier made it a lot faster enhanced your power as a human uh to hone in on the value unique value added that that you're able to contribute so that is step three one other thing I just want to flag for you

I am committed to free tools there are other tools within which apart from Google schol that you can search for example one of these tools is called site which will search the literature for you and this can be quite helpful but it is a paid for tool for the vast majority of you Google Scholar is more than good enough and that's my official recommendation um there is a place for these tools but again in the spirit of Open Access like zoto and Google Scholar I want to really promote on this channel free publicly available resources

that anybody can do this anywhere in the world no matter what resource you have available to you so you've gone through you've harvested your literature you've popped it into your reference manager you're ready to go to start writing the fourth critical step where chat GPT is going to help you is on editing and it's going to save you a ton of time but you do need to know our secret formula for writing well it's going to make chat GPT more effective and it's going to help you really flesh out this outline fast we use a

pier system that we've developed and adapted and Pier is kind of like a hamburger uh where each of the elements appear p e r are kind of like the bun the meat and the other bun of the hamburger p is the point it's a topic sentence it's the main point each paragraph should make one point following our one point Rule and that's usually the first sentence of your paragraph then you have the ease which is kind of like the meat the cheese and these are evidence example explanation and finally the last component is the bun

is the repeating sentence while I'm going to talk about Theory I'm going to show you not just tell you how to do this and in fact it's one of my pet peeves personally that professors kind of treat students like a banking system I'm open your mouths I'm going toit information keep it and regurgitate it uh I I I try to of level the playing field and think of us as intellectually curious equals working together to advance the collective body of knowledge research to make an impact like so many of you who follow this channel really

want to do with your work let me show you coming back to our outline how you're going to do that so let's take this uh Mariana point about equity and how could we flush this out so uh my sense I'm going to make something up here okay don't make things up but uh something that's probably the case based on what they said and how you can flush this out so if we're going to make this into a full-blown paragraph I would probably start something like this despite I know a wealth of evidence on the impact

of sips on academic performance much less scholarship exists on its uh Effectiveness for underprivileged or resource deprived students right that is the point of the paragraph that I'm going to make the second I'm going to maybe cook this a little bit more using cooking anology I really don't want to drop in raw food I want to kind of develop this a bit more I'm going to explain it I'm going to say it is I'm fleshing out this idea it is highly likely that um the uh effect of Si P for median or average students um

is different than that on uh uh lower income or ethnic minority groups you know and then here I'm going to deploy kind of like dropping in a little nugget of evidence I'm deploying ammunition notice one thing I'm going to do here I'm not writing the paragraph to say Marano and colleagues in 2023 XXX it's quite a boring way to write you'll sometimes see people who are novices Mar colleagues did this Luigi did this I'm going with Italians because my colleagues are all Italians so Mario did this right no you don't want to do that make

your point and then drop your evidence like to back that point up so we're going to do this here and say okay one study by Mar or and it was actually a Content analys One content analys by Mariano and colleagues well side it 2023 um reviewed sips in uh tier one US schools some things up you need to go through and get this evidence into the outline um um and found that less than 2% performed an evaluation of Effectiveness and of those none included an evaluation of uh Equitable effects clearly uh more research is needed

to understand how sips impact on vulnerable groups Okay so something like this right so you've got your first draft and you're going to do this use our Pier system here right like what what we've got the point here the uh explanation and the evidence and the repeating or linking sentence with spelling errors um here and then what we're going to do is chat GPT you can now take this whole section in editing and this is where you're going to tap the power in an ethical way of chat GPT because you've done the first dra yourself

using our Pier system using the outline um and drop this back into chat GPT we're going to ask chat GPD to edit this for us so I'm going say here um please edit this paragraph um that is in the literature review above in an academic tone featuring clear and succinct writing style okay and we're just going to drop this in um you do need to Prime GP chat GPT to let it know what you want it to do and if we come down here we're we're going to see this is a little bit a little

bit wordy but I I I like it there's uh despite this extensive research so some of this is just wordsmithing of it that's okay I would be careful sometimes it says startlingly low rate I'm a little bit British in my writing I like to let the data speak for themselves and not try to embellish or add accents so some of this is what we'll turn into a little bit of what I call Purple Pros just using more complicated language than is needed use Simple language when possible I always encourage my students to try to explain

things so that their grandmother can understand things like alarming this tone just isn't right and this is where CH gpg goes a little bit off but you'll see it it did really little change to the structure um and this can help you save time especially if English is not your first language I recommend tools like grammarly but chat GPT here is going to save you a bunch of time editing so that's it guys my student who had failed to do his literature review three times before tapped the power of chat GPT and got this review

and school Improvement plans done in under a week and in the right way that he could feel proud of and all I feel proud of that he using cat gbt learned faster mastered the basics and now is gone from chapter 2 which was literature review and already breezing through chapters 3 four and five it's almost like getting past this first Hurd of literature review unlocks the floodgates for everything that comes next it's so often the sticking point that I see for so many students who just get frustrated don't know what to do and throw in

the towel and give up don't let that happen to you that's exactly what this channel is about I do believe the future is going to be that AI will surpass our abilities as humans and we need to learn to coexist and not feel threatened by it but use chat gpg in the right way and we're going to continue to deliver content like this on this channel for you because if you're not using it you are at a competitive disadvantage compared to your peers uh so uh again stay tuned for more comment like this you've requested

it from me I'm delivering it to you by the way uh do join my Facebook group if you like this video give me a like subscribe and in my Facebook group we can communicate directly we have live chats where professors other research students and peers are collectively working together as a community of practice to help Advance our own research and kind of lift the tide for all boats uh do hope you join hope to see you there and I'm going to look forward to seeing you in the next video