Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

as things for you. Um, a bit of a disclaimer before we start. There are many ways to tackle. I see a free and to mind is just one perspective from it. There is no definitive guard, although there may be some things that a lot of people may recommend. And if you do have any questions on this, you know that may relate more to your own specific course. Please do contact, of course, Lead to get the information from the horse's mouth as it were. So we're going to be covering four different things. Gonna be covering what? This I see A is how to approach the main body and then how to approach your abstract and your lay summary. So what is it? Well, essentially, you're writing the findings from the data you collected in your lab report in the former of an academic paper with an abstract handle. A summary because your writing also also of up, sort of a mini, um, research project. It may vary between courses, but overall, you're looking at a 1500 words for your main body. 350 word abstract and a 500 word lay submarine with no more than seven figures or tables. Unfortunately, with this one, there's normally know referencing limits. Okay, so you can make sure your accurately citing or your statements throughout and emission maybe any sort of excuses with that. It is quite a big one, and this one is 21% of the overall BSC. So I can understand that there might be quite a lot of worry about this one. But what a good thing you do get from this is that you get a good experience of how to write up a project and now come really handy and then module free when you're writing of what is even a bigger portion of your overall BSC mark. So let's move straight into the main body. So when you're submitting your, um, I see a free you're going to submit it in this order your introduction first, then your methods followed by your results. And then finally your discussion, and when you read it, that makes a lot of sense. You want to introduce the topic? Explain what? Why, it's important. What what's going on with the research currently? How did you investigate your current question? What did you find. And what does that mean? Okay, that makes perfect sense. But when you write it, it can actually be rather different from that, because it isn't always the same thought process. So I would suggest that when you're wanting to write up your I see a free, you want to start with the ones where they're they're going to write about the other stuff. So what I mean by that is, if you want to introduce a concept, you won't need to know what concepts you have to introduce. And the only way you're gonna be able to find that out is for doing your results. Because what you find in your results then lead you to what you want to introduce at the start. So when you want to go through your I see a free, it's really good to start off by doing your results section first, then your methods, then your discussion, and then finally, maybe finish it with the intro. Or you can swap these two around. And that's because from writing your results section, you can work out what you're then going to discuss and introduce, and you can also then work on your methods in between all that because that's sort of a nice, standardized, descriptive path of your I C A. So it makes a lot of sense to start with your results and then move through it, Um, in that way. So let's start with talking through how you do your results section first. And some of you may have seen this sort of statistical decision tree on the pre BSC course, and it is quite useful. And it does help with a lot of sort of queries of what statistical tests should I or shouldn't. I use a few things to sort of note from this because there's maybe a few sort of language that people aren't familiar with or quantitative. That just means when a measure of value, it expresses a number. If something is a paired test, that means it's the same population before and after and intervention. And finally, if something is parametric, that means it's normally distributed. So let's say you collected all your data from your lab from whatever your course leaders given you. What's your first step? So step one is you want to work out whether your data is parametric or non parametric and there's three different ways of doing that. Firstly, you could put it all into a histogram, and you could visually inspect it and see if there's a nice, normal and Gaussian distribution. But that isn't really recommended, Okay, because it's not that sort of scientific. It's not that accurate, and it is a lot of subjectivity with doing it that way. So there's two other sort of ways you can go about doing it. Firstly, you could do. Emilson is a common growth Smirnov test, and there's a lot of statistical software that can help you with these tests, which I'll cover at the end of this mini section. But this is really useful if you have greater than 50 and sort of data points. So if you've got and 50 people and you want to work out whether that data is distributed normally or not, you can use a common growth spurt of test. However, when you're working with much smaller samples, Shapiro Wilk is more accurate. So if you're dealing with any free or four samples a time and you're looking at the difference between those, um, you want to use a Shapiro Wilk test, what those tests will give you is a p value. And if you get P value out of those tests, which is less than 0, 00.5. That means your data is significantly different from a normal distribution, and consequently you can imply that your data is non parametric. Okay, so that's sort of your first step. Once you've worked out, whether it's non parametric or parametric, you can then go down your statistical decision tree and find your appropriate test to perform again. If you get a P value less than nine point, not five, that is normally sort of indicative that there is significant statistical difference between two care. What's importantly, when you're presenting that data, any Parametric data has to be given as the mean value plus or minus the standard error of the mean or you can use standard deviation. Um, and any non Parametric data is given as the median plus or minus the inter quartile range. Okay, so that's really where that first step really sort of helps you distinguish where you're going to go down for the rest of your results section. So we've found out it's non parametric or parametric we've performed are statistical tests. We know that one group is greater, less the same. We now want to present that data as a graph. Now, if your data was initially parametric, you can present it as a bar shot. And if it's non parametric, you'll be wanting to present it as a box and whisker plot. So, for instance, this is Parametric data, and we've got it as a nice simple bar chart on the right axes and nicely labeled, um, on our two cohorts label rather well. One thing you'll notice is that there are error bars in here, and whenever you give a value of an average, you must always include it's variance. So if using parametric, you must always give it with the standard error of the mean. An average by itself is rather worthless, so always make sure you're including a variance measure with it. Make sure it's not exclude any outlines. Okay, if you have some values which really just don't sort of fit within that average, certain Softwares will put them as individual points. But make sure whatever you do, you're not just excluding them because they don't fit your trend. It's not sort of good science. You know, outlines may be there for a reason. I need to sort of explain that as part of your project. So please don't feel, um, like, just because it doesn't fit whatever you're trying to prove and that you should be excluding them, please leave them in. One thing you'll notice here and one thing you'll see in a lot of papers and which are published is that any sort of statistical difference in the graph is often denoted using an Asterix. And normally one Astra expired itself. You know, two p is less than not point, not 51 thing that they have done on this graph from the right, as they have used four Asterix is to determine. And that was a much greater statistical significance. But that potentially isn't, you know, entirely necessary, indicating that it is just pay less than 9.95 should be sufficient. People can do that as well again. These are all sort of little variations people can choose to make. If you are using a proportion in your Y axis in your graph, make sure it goes from 0% to 100% because that way it will prevent you from sort of misrepresenting just how big the difference is between your two cohorts. You know, a 5% difference will look rather big if you're only going up by not point, not 5%. Um, so make sure you're using an appropriate scale to show the whole picture. Access does need to be labeled of appropriate unit as well. And if there's units on, sort of, you know, normally now and make sure to include what they what they are written out as in for In your legend, coming onto the legends, your legend should allow that figure that graph to be sort of interpreted, understood in isolations. So, in other words, if I took that graph on the right out of the whole paper, which was explaining it, I would know what was going on in that graph. Okay, so if you're sort of struggling with that, make sure you have a look at some all the years sort of projects or maybe just look at your go through put bed, look at a few papers and you sort of get the gist, um, of what they're trying to say with their legends and how clear they make the graph that they're trying to illustrate. However, not all graphs are perfect, so there are a few issues with this one, so just sort of a rhetorical question. Yet is there any issues with the graph on the right? And also you'll notice in the legend that they've used a lower case n Okay, So why why are they doing that? Okay, because this is this is actually really good graph, but there are few issues with it, so firstly, it's quite a minor thing. But they could have probably stated what typographic is. Whilst it might be obvious to person who's made it just simply stating that this is a simple bar shop would have a good bit of information to this graph from the right. Furthermore, although it might be perceived as common knowledge especially, you know, if you're familiar sort of reproductive sciences or endocrinology, um, putting what LH stood for in the legends Luteinizing Hormone would also be a worthwhile addition. And this is also another little point again, which may have seemed initially like an error. But if you ever seen sort of a lower case n anywhere and sort of research that tends to mean that the data is from the same person or subject. So in this case, those Leydig cells were from the same mouse before and after exposure to LH. If there was a capital N, however, that would mean that they came from several different mice or if they were from several different people. At least that's just another little sort of and the temperature that we use within a lot of publications. But that isn't just the results. So while she may have your figures which you want to select, those which clearly demonstrate your results and ones which structure your findings in a logical and clear way, you also need to write about them. This is simply a description, okay, so if you imagine your your sort of introduction and discussion on your you know, explaining, you're expanding and you're giving reasons for things, this is sort of a much lower skill. You're simply just describing what is going on. That's it. No explanation and forward. You're simply describing what's going on within your results. It's important to note that you can't say that something has increased or decreased unless you're statistical tests. Has proven significance. So unless something is p less than no 0.5, you must use the words rose or fell to describe how your sort of averages are changing. Okay, so then, um, sort of rushing to say something has increased if P is greater than 9.95. And again, as I've mentioned before, state your exact average is with your and from your data with variants, measurements, asshole time. So you think something is mean 15. Make sure to include what standard error of that mean is with it. And so, as I mentioned before, there's a lot sort of clever statistical software which can help you with all that. So we don't think you need to go away and completely understand how a superior world test works. These are all built into statistical packages and that are out there and a free for you to use. So ones that are available for imperial R. S, P. S S and Stata. The S. P S s is very sort of simple to use, and there is sort of a description on how to use it in the pre BSC course. It's the one I use and it is a very good introduction and to doing statistical tests through software. Um, Stata is also free, but does require a bit of coding knowledge. And maybe you know something that people might want to explore but probably requires a bit more patience to get to know it. There are some free to use ones which students do, like using so very popular. One is graph pad in prison because this is what produces very sort of pretty looking graphs, which quite aesthetically pleasing and that is free for a 30 day trial. So although it isn't free for Imperial, and you can just make an email and I'll go up and use it for 30 days and for those people who really want to, you dive into sort of coding and how to do stuff, you know via coding packages, and you can use our, which is really good free software, which can do a lot of advance stuff, but maybe a bit complex for this first license. So we've covered the results section. We've got our figures, we've got our legends with them and we've wrote them up, and we've described what's going on now. Let's move on to our methodology, which is, as I said, sort of the second thing we we probably want to write about. And that's again because we're doing a lot of description we're describing are subject. Our patient's the sample were describing are dependent variable. So what we're measuring and how we measure it. And we're also describing what statistical analysis we performed. So what do you want to talk about when you're bringing up your subject or your patient? So if there are any relevant past medical history or core mobilities, this is where you want to give them. Okay, If they're a smoker, are the obese Are they diabetic? This is where you want to sort of be giving that information to your reader so that you're being honest with the type of patient you have, but also to potentially highlight any further points for discussion. You also if you have any any exclusion or inclusion criteria. So if, for example, you ran the experiment and certain PCR things failed, okay, that might be something that you know. You're not gonna be able to include certain data for, um, any previous investigation findings so often. You'll just be examining one part of the picture. But there may have been other sort of, you know, hormonal measurements taken beforehand. Um, any viral swabs or anything that you think is relevant to how you've got your sort of sample, And you also want to describe you know what is your independent variable? So, what are you changing? Are you giving a hormone stimulus? Are you making a dietary adjustment? Are you giving a drug which you're expecting to see a change for? So what you actually doing to this subject, which is making you then record something and write about it, And then what exactly that is? So are you measuring? You know, weight loss with that? Potentially. So you're dependent variable. So this is how you're measuring the effect. Um, So again, you just want to describe how you measured the changes, So that might be what equipment you used. Okay. What time of the day you did it? Potentially what the agents were used. And often, if you're using reagents, okay, what the manufacturer was and where that manufacturing produced it. Um, Are there any known method? Illogical standards. So sometimes the w h o has sort of very rigid sort of specifications and how you should measure things or maybe nice as a sort of guidance on what should be used and what shouldn't. And therefore you may want to quote that. And if you followed it or if you didn't, if you're using a microscope, what was your magnification? How many fields did you used? Was it an oil immersion? Um, and also any equipment used to measure it. So again, if it was a microscope, but also if it was a specific machine or flow cytometer that you used to really important sort of give that information here. Okay? And so finally see your statistical analysis again. You're gonna basically just sort of describe what I just told you that. So what normality test that you use or statistical test? Was it a student's T test? Was it a kraske out Wallis test? What did you do? Your alpha value. So that's another name for your P value. In most instances, you'll be using your point, not five. Normally, if you're using one above that which probably won't be relevant for the majority of you in your I see free, it's when you're using a particularly low powered test, But all of those tests that sort of given in the pre FBs see course are pretty high power. So you should be using anything above no 10.95, and and also here you often want to mention the statistical software that you use. So sometimes people do throw in like they use Microsoft. Excel for part of it. Unless you're doing your exact statistical tests and you're graphing in it, it's often just standard to just give what exact statistical software you use. Like S P E S s. Um And then finally, what did you do with anomalous data? Which, as I just told you, you shouldn't be excluding it, so you should mention that because it gives you sort of real good, um, backing to your data that you're not sort of cherry picking your results. So we went through our to first sections. Nice, straightforward. We've got our steps. We've got our sort of description that we're going through, but what we want to cover in the discussion. So overall, you want to put your results into context. This is where you've just been stating them. You've just been given fact Now you want to say why? Why is that important? So what? What is the point in all of this? Well, you want to compare your study for the wider literature, and this doesn't mean that you want to sort of force your data to support what somebody else has said. But you're comparing to say if there is a difference, why is there a difference? Is it because our subject has slightly different characteristics? Is it because we measure things in a slightly different way? Or is it because our study was just so small? The observed effect wasn't visible? You then want to look at your own results. And maybe this isn't in that exact order. You can sort of mix these up, but maybe explain why your results have occurred. Is there a physiological mechanism? Why the results you have found occurred? Okay, Is there something going on? You also want to explore how any confounder may have affected the result on what you could therefore do in the future. To minimize that effect, it's likely your study won't be perfect, and your course leads will know that, but they want to see how you can approach that how you can sort of reason it and how you can come to a balanced conclusion as a result, and any limitations is going to be addressed here and also how they may be addressed in future studies. And you're finishing all of this with a nice conclusion, which hopefully just sort of summarizes your main finding. Often, people have a tendency to really sort of short change themselves. Hopefully, you will have found something to really make that sort of clear within your conclusion without overstating it. So, for instance, this is just sort of a little example of a paragraph and that I wrote because examples can be quite useful. So, um, this study found the sperm morphology improved following weight loss sufficient enough to reduce the patient's bm I from 36 to 28. Okay, so basically what I'm they're giving is I'm giving what we found and as well as what are dependent variable measured and then give going to give a reason why I think that occurred. So, given the deleterious effects of reactive, oxidative species have on spur metha, genesis and the increased levels of them in obese males, is potentially unsurprising sperm for actors, improved put weight loss. So here I am, giving a pathological physiological reason why my my results may have occurred, giving further backings that these results are just a fluke. There may be an actual underlying reason for them, So this is supported by a prospective cohort study by Beasley. It's out which found that in males who lost greater than 10 kg in in one year experience, the 32% increase in normal sperm morphology. So here I'm comparing it to the wider literature. I'm saying it supported it. This was found in a much larger study, therefore giving further backing to my results. However, whilst it is promising semen premises have improved following this inexpensive intervention, future should should impact should examine. Sorry. The impact weight loss has online birth rates and couples with male obesity's evaluates direct impact on fertility outcomes. So with that last sentence, not only am I giving the reason why, I think that's important. So what of it all? Why is my result actually relevant by more so giving how this could then be further advanced in future studies? Some, That's sort of where your discussion is is going And once you've sort of done your discussion, This is why I normally like to come back to the introduction. And the reason why is because you've just discussed a load of important information and you've discussed your results. We without sort of the prick correct context and understanding. You know what sort of going on within the field that you're looking at? It can all be a bit sort of worthless, and it isn't sort of in that much context. So this is what the introduction really helps Set the C. And it allows to make sure that nothing, you know, particularly new um, he's discussed later on. You touched upon it in some little way in your introduction to begin with, because this is where you're setting the scene for your narrative. Okay, you're setting the scene for your overall story, why your research is important and what you did and therefore what you found. This is probably very reference heavy, and it can often be quite difficult to write and can get to be a bit of sort of an art to it. So make sure you're sort of reading other instructions. See what's a common theme to them and do spend a bit of time on this. It's not going to be the entirely easy things. You probably got a lot you want to get on paper and getting that all within. You know, your overall word current and 1500 can be quite difficult, but that is all normal. So don't feel like that's, you know something that you should just be struggling with, because everybody will be struggling with that as well. So in your first part of the introduction it tends to be you want to explain the condition or problem and that your overall addressing then you want to sort of talk about how much of a problem is why we reading it if it only affects wanted a billion people, something that may be less relevant than something affects one in four people in their lifetime, and you really want to use those exact figures as well. Don't be afraid of quoting exact figures from reliable sources. And then why are we looking at this specific issue? So if you're looking at weight loss interventions rather than you know M IVF has a fertility treatment, why is that a good thing in your other paragraphs, and you then want to discuss the concept that you want to touch upon later in the discussion. So if you're touching upon, um, sort of an antibiotic, okay, touch upon that. If you're touching upon why you're using flow cytometry. Um, rather than a different method, maybe you want to approach how you're measuring the effect here, Um, and potentially you want to explain potential reasons why your condition you're seeing is occurring. So for diabetes, you might want to see what a sedentary lifestyle, maybe a high glycemic index diet. Um, and potentially the people are just living longer and therefore using this, you can then potentially highlight why you're in intervention. Why, you think it would produce the change that you're going to see in your in your results and just sort of touch base with it? Uh, and then finally in your final paragraph, and you then just want to give your aims of your study as well as your hypothesis, which would just be one line. Um, and there's been a you really want to quote actual statistics throughout your interest, so even really other paragraphs, if you can put some actual statistics from other well cited papers. To really make sure you stand out and give some good evidence as to why your paper is important. So top tips overall, sort of your main body sort of. Think of your story that you're giving throughout. Okay, Make sure to keep your sentences eat out. Ultimately, succinct, clear and affected. Often people can go through a tendency to be quite rambling. You really want to ask yourself of each sort of statement? So what? Why is that statement important? Is that getting across the meaning that I want to give and finally use your feedback and tips from the previous I see a one assessment to help you because they all do feed into one another. Okay, so that's sort of the main body. Now let's talk about the abstract. So your intro you want to give a brief background, um, give it the path of physiology and your aim and hypothesis, and some pertinent stats be here. So this is even more concise, even more brief and just giving a very basic background your methodology again. You're keeping it short and simple. What happened? What was your subject? What measurements did you take? And what stats did you perform? Your result is probably gonna be bigger, and this is where you are going to report. Report your findings and don't be afraid of again giving your exact average with its variants. That's very important. If you imagine you're looking at a paper and put med, you mainly read the abstract before then going into the actual paper. So it's really important to give your actual values here because that's sort of what your main findings are. So it's really important to give. Your exact average is with their variants and P value, and your discussion is again going to be much shorter literally. You want to include your interpretation of your results, strength of your study, maybe a limitation if you have room okay, or maybe where there is room for improvement in future studies. And finally give your overall conclusion. Okay, any future research you could pop in within that limitations and within that conclusion. But just make sure that it is sort of touched upon. Okay, so that's sort of your abstract. It is quite hard getting all that. We've been freedom from 50 words, but where you really want to focus on is your resulted. Make sure you don't sort of avoid giving your exact values, because this is really that's really the important part of your abstract. So the lace summary fire park. So overall you want to keep it simple. But, like really simple. This is if you imagine your average sort of newspaper reader. This is what this is sort of part of who you're looking at, someone who reads BBC news on the tube into work in the morning. So you want to keep it really simple. Make sure you really examine if there's a way of explaining that, or if there's a way of simplifying it down into more layman's terms. Explain any terminology. So one example that we sort of had in Reaper was motility and morphology now fertility that could nicely be summarized into how the sperm move. A morphology can be simplified into how the sperm look okay, and that's quite nice and a good example for you as well to explain. That's how you're putting into just sort of more layman's terms. Your classic intro methods structure that I talked about earlier that's really less relevant here. You're wanting to tell a nice sort of narrative and focus on how it sort of flows with the reader rather than how it flows scientifically in terms of how we structure our intro method results. Your rough structure for this and this is again very rough, is you want to give context to what you were doing, the strategy of how you approached it, and then your results, discussion and significance can all sort of meld into one with each result you get, rather than being just in one specific part. Avoid using numbers here. I know I spoke very strongly about sort of giving your numbers in the abstract as well as in the interim. Here it's less important what numbers you've got and, more important about what trends you saw. That's more easy to understand and part of the lace summary rather than actual values, which don't mean anything by themselves. Your discussion and your significance really focused on what your findings mean and what they mean for the person who's reading it. What could have affected your findings, which you'd maybe like to control and what implications, um, of the recess for the future and what future direction and made this research be taken. Some people may have tendency to say. Okay, well, I found that X increase. Why? That doesn't mean that X cause is why. Okay, there may be an association there, but you shouldn't start overstating your findings here. Okay? So still keep on using May possibly could might, um, just as you would in your discussion. You don't want to overstate your findings. So it's important to continue that within the lay summary, even if you are sort of simplifying it down. Finally, keep it conversational but still formal. So don't be tempted to use birth instead of, however, make sure you're still keeping that sort of formal academic turn with. However, um, I believe I got one more point, if you can, might be quite hard, but get some of know medical knowledge to read through your summary and identify any areas of confusion. So, for instance, you know, I got my mom doesn't never You never, um, work sort of in any sort of academia or science was really helpful. Just seeing how she sort of interpreted what I was saying, which made complete sense in my head, but may not in hers, and that's what I thought was really sort of good points to take away. And so if you do have a resource like that, or even if it's just a friend in a different course, which is a medicine, it's really important if you can get them to give you a few sort of pointers. Any areas that they weren't confused on and ask them to sort of give your story back to you and see if you're on the same lines with them. So overall, that's sort of it. Sort of it. I know it's been that's quite a short talk, but hopefully that's sort of underlines what you need to sort of do sort of crack this. I see a. These slides will be made available so you can go through it. Hopefully, all the steps make sense, and I'm happy for anyone. Ever email me and ask for any questions. I'm perfectly happy with that. Okay, Brilliant. Jack, thank you so much. I'm just giving the audience an opportunity to ask questions. You should be able to message in the chart. And also, could you please fill in the feedback form for Jack? That would be fantastic. Um, I haven't actually asked the Q and A people to join until 7. 50. So we've got a little bit of time to, um, answer any questions that you might have, So please, go ahead. Jack is very friendly, so you shouldn't have any problems. Um, So go ahead and write your questions in the chart. Um, Jack, can you see the chart? And yet I can I've got teams up properly, so I can. Okay. Great. I'll just let people go ahead and type. Okay? Any tips on how to describe the methodology in lay terms? Uh, that's a good point. So it might be. It obviously depends on what course you do, and sometimes it can be quite hot. Now, that doesn't mean you don't. If we think about how I said to to sort of describe your methodology previously sense of what equipment did you use? What magnification. Obviously, that's sort of without saying that's what needs to be simplified down. If you want to say you look at something under a microscope, you analyze something doesn't necessarily mean you have to say we use this. You may want to say we analyzed it looking for X. If you were looking for a specific DNA signature methylation signature, Um, you're sort of wanting to really simplify that gap between it and with something like the methods. That's perfectly reasonable to expect. Okay, um, you're maybe not wanting to spend too much time on that as well. Okay, so whilst you may be important to say, this is how we got this result, it's more going to be what that result means. That's sort of more important in your lace summary. Okay, um, what was the hardest part of writing? Uh, I see, for I see a fruit. So, um uh, in my opinion, I think it was Well, there's two things. Firstly, is getting the first draft done, because that can be quite hard because you don't really know what you're doing and whether that's the best way to say something, and that's completely reasonable. You, where you want to get it across will not be the way you you feel like it's done best the first time. But that doesn't matter the whole point of a first draft. That's just getting it to exist. Um, as long as it does that that's a That's a great first draft, and you'll spend many times redrafting it. Another hardest part was that I I messed up the data analysis I did with a week to go, and so I had to rewrite my whole thing. That was that was my own mistake, which I'm hopefully making sure none of you guys have to buy. Making sure the results stepped were very clear. Um, so then someone else in the street. So if you have two columns of data placebo versus intervention and if the placebo column is Parametric. But Intervention column is non parametric, how would you present both from the same craft? So that's a really good question, and one that I have been Ask what I have been asking myself recently, one of one of the projects I'm doing in my spare time often from what I've I've seen within a lot of literature, is if you have one sort of say, section or cohort that is Parametric and the other non parametric, you tend to want to use the non parametric data and just use it all as non parametric. So, for instance, if you're saying you're, um, intervention column is non Parametric. You want a whole non parametric graph? So you want to put your placebo column into, you know, median, I QR into a box and whisker plot, and then you do the test that you would do for a non Parametric data set. And the reason why that works is that statistical tests are sort of based on a certain level of assumption. So if you do the student's t test, um, it is assuming Parametric distribution, um, which means that it's a less rigorous test, and that's okay as long as you've got Parametric distribution. But if you haven't, you need to use a more rigorous tests and the more rigorous tests of the non Parametric tests because they're not assuming a normal distribution. So if you've got two columns one in Parametric and non Parametric, just consider it all to be non Parametric data. I hope that makes sense. That's a good question. Um, any tips for using S P S s. So that's a very good question as well, and there are a lot of sort of forms online which can help you. And in my experience, I think the previous C course was very helpful for us, and it does take you through it step by step using a data set. Um, and so it can be quite easy, cause it is exactly what you will be doing, which is sort of a nice graph and a nice, you know, um, statistical test that you'd want. Um, but that's why I say you want to start with your results section first and go through that and and try it, you know, quite early on. Um, remember, with SBSF, you can't get working. You do need the imperial VPN, which is quite easy to set up. And that took me a while to stop. Get working on my computer. Um, so similar to my last question, How would you go about explaining factors which may have affected your results is simply saying our results made, but influenced by factors X, y or Z, um, or would you also need to explain the mechanism of these compounds may have influenced results in later. So I agree. That's yeah. It's a very good point to sort of, say how they may in the discussion and properly explaining it with either, you know, backing up from, you know, different publications. Um, but I agree that later may have may be more difficult, but you're wanting to a quite a similar, similar approach you're wanting to simplify. Okay, So, for instance, if you've noticed that, um, say someone's heart disease has improved because they've lost weight. But simultaneously, they're diabetic. Control has also improved. So you can't sort of tell whether it's exactly because of just the weight loss or diabetic control, or it's all in there. Well, have you got any other papers around you suggesting, um that they may be a confounder? Now, that doesn't mean in your lay. Summary have to say that this paper may have also said diabetes management improve this, um, and give the exact name. But what it may want you to write in your lace summary is that other papers have found X has an effect on why, and consequently, you know, we we can't fully distinguish that weight loss is having this effect. But we may and and therefore we would like, you know, to perform further studies on. This is perfectly fine as long as you're explaining it in a nice, simple way. So it doesn't need to be your your explaining diabetes as well in that amount of detail. But make sure you're bringing up that. You're sort of you've got two things going on. You're not exactly sure you would like to do this, And that's sort of enough. So I think it is enough to sort of state it in that way without going into a team of difference. I hope that makes sense. Um, so we have audit data for Aisa. Do you think that would pose any different challenges? Um, other bs. CS uses lab collector data, So yeah, Order date, I presume. Um, you sort of meaning like, you know, if you've got patience and then they sort of come through the records and found either outcomes or something that's happened, so that does pose other challenges. In some ways, it's obviously a less controlled environment. There's probably gonna be greater hydrogen, a itty within patient's potentially. And also, you may have not been there when they've collected the samples, so you may have less of an idea of what's going on. So that's certainly something to consider. And if you are confused, make sure the message your costly, but the same things sort of apply. One thing that's very important for that type of data is your demographics table. Okay, so if you've got an intervention and a control group and there's sort of, you know that they're two different people, sort of two different people, so they're they're not in both. Make sure you're highlighting relevant demographic data quite nicely. So their b m I What's the average age of each group? Um, you know what? Their smoking status, Um, you know, how long were they living with the condition? Maybe whatever you think is relevant for the condition that you're identifying, make sure you include it as part of your demographic stable to make sure you're being quite nice and open with what data you're actually using and who you're talking about. And hopefully that makes makes sense. The demographic data is probably gonna be the most useful for that. Um, so another question And do we need to include references in abstract and lay summary. So I recommend asking your course needs to make confirm that in my course, and I think I think most but I could be wrong is you don't need to include it in the abstract or the lay summaries. Okay? And I know that if some people are published, that is sort of different. And some some some publishers do require that. But I don't think with this. I see You need to worry about that. Um, so another question, when writing a figure legend, how do I keep it concise, but still provide enough info about the graph? That's a good question. And obviously, you know you're not gonna have to write the whole method section each time you write a figure. In my opinion, what I think is the most important thing is to say this is a graph of So if it's a simple bar chart you want to say of, You know, uh and you know, for instance, LH and of the effect of LH on Leydig cells and before and after, for instance, um, so that tells you what the graph is. You then might want to describe the graph. So obviously a box and whisker plot is the most easy to understand graphs you might want to say, you know, middle line represents median without a box representing I QR and and tails representing range. So you're then describing what the graph actually looks like. Um, so in a bar chart, that would be, um, columns and me and plus or minus S e m. And then you may want to give abbreviations and what your statistical significance is as well. That's sort of enough. Okay. Through a graph, I would say that's probably about all you're looking for. Any other questions? Yeah. Any other questions? Um, otherwise we might end it here, and then basically, the BSC specific breakout rooms are included within a link. So I'll just get you all to fill in the feedback if you can. So I'll just post the link again in the chart. That would be fantastic for Chuck. It's a brilliant lecture, but it would be lovely to hear what, um, people thought about it. And also, at 7. 50 um, we have the q and A, um, section, which is bs. See specific. Um, so you'll get B S C specific advice, and that happens on medal. So if you click this link, it will take you there. But please, please do fill in the feedback. Um, I think that's everything. Unless anyone has any more questions, go ahead and ask all jacks here. I think I think everyone's out of questions. I was just asking all the ones that I could think of. I'll end the recording here, then, um