Computer generated transcript

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

Oh, I was ready to start presenting for who? For how, how hi guys. Can you see and hear us at all? Are you able to see us? Hear us? Let us know in the chat, please. Metric. Yes. Um I can have a look at my phone and see if we're alive. We always having technical difficulties, but I'll just message sure if you guys can hear us and see us. Thank you for coming today. Hopefully, we should have an interesting session for you guys. If not, we will try and get the tech sorted. Yeah. Yes. Ok. So thank you very much, Larry very much. Appreciate it. Brilliant. Shall we get started then? I don't think anyone else is joining us, especially when it's so sunny outside. It's understandable. Let's get to it. Ok. That's the final, final test, final. Well, yup. Mhm. Can everyone see that? I can see it? Ok, perfect. Um Yeah, hopefully if someone can't, they will let us know at some point. Ok, so, uh, hi guys, my name is Praveen and I've got Asher here with me, you know, last session and uh we're, we're from Cardiff Un Women and Surgery. Society. We on committee. And today today is our second session in our research and academic skills teaching series. And we'll be walking you guys through an introduction to basic statistical skills. Um So just a brief outline of what we'll go through today, we'll start by discussing statistical analysis and why we do it. Um And then we'll discuss the different types of data, the types of stati statistical methods and the various statistical tests you might come across. Uh And then we'll finish by talking about how you can interpret statistics. We'll have time for AQ and A because we know it can be quite a confusing topic but do stop us at any time. If you guys are a bit confused about what we're talking about, you would like us to clarify, I'll be monitoring the charts so you have to stop us at any point. Um If you need us to go over anything and as always, the slides will be sent to you and it'll have everything we talk about in the notes. Um Yeah, so just sit back and watch uh just a brief uh intro to as to who we are. So Cardiff Universities, Women and Surgery Society, we aim to encourage and educate female medical students to consider a career in surgery. But these sessions are for everyone. And right now, elections are running for our 2025 26 committee. So if you're from Cardiff University, consider nominating yourself and running. It's always nice to see some fresh faces and uh it'll look good for your portfolio. There's a lot to be done. OK. Let's get started. Let's go straight into it then. So before we start, we thought, before we start going through different methods, we thought we'd look about what statistical analysis actually is and try and define what it actually is. So statistical analysis is when you use mathematical tools and techniques, tap it again. Sorry. There's just a lot of animation today and you use it to analyze data, you then use that to identify any patterns. You use those patterns to draw conclusions, then identify the significance of any findings. You then make inferences about a population and then make informed decisions. So I thought I'd give an example very random. I thought I'd leave the science behind and be very random. So if you tap it again, oh, why is it worth the animation for a bit? Keep just keep tapping animation. OK. There you go. OK. No worries. Just keep me. Keep, just keep typing, keep getting, put it all on the screen. So, oh, we didn't do a run through. But how you up for being cute, you know, you're trying to work out. Oh, your research question is what paint color should I buy? Um For adults in London, you know, you do a big survey of people in London and you find out that the color most people want to buy in the age is 18 to 25 are green and it's blue for 40 to 65 you then draw your conclusions and you conclude that these two colors must be really popular. So then what do you do? You identify the significance of these findings? So this maybe it'll be significant for you, maybe it won't be significant. But if you're a paint company and you want to increase your sales, it is significant to you. You also might hear things like P value which we will go into later, but I thought I'd just put it there um to sort of start thinking about it. But is it significant to me as A B and Q? Yes, I want to sell things that are popular. So then you make inferences about a population and you infer that your sample of people is big enough for what you want to do and it should be true across, not just London, but you know the southeast of England. So you decide that in areas of younger people, you will sell green paint areas of older people. You sell blue paint and you will advertise this more. So that's how you make your informed decisions. You do your test, you do a book survey, you get loads of data, you analyze it, you find a bunch of patterns, you find out, you know, you draw some conclusions here. I said these paint colors are popular. You identify the significance to you and your research question. So do my findings actually answer the question I had? Yes, they do. Maybe you do a test, for example, I do all of this and I find out no color is completely popular. And that might mean that it's not, this actually isn't significant to me at all. I make inferences about our population. So I've decided that everyone in South England is sort of similar enough so that these results are signi uh what's it called applicable to them as well? And then I make informed decisions. So I've decided I'm gonna sell that. So I thought I'd just give a little example here. Hopefully, I haven't confused people a bit. Hopefully, that makes sense. So, next slide, please. So why do we do any of this at all? If it's just, you know, quite a lot of effort. Well, it helps us um understand and make sense of data because if we did, if you were here last week and we talk about different research methods, we talk about different types of study design and it's a lot of effort and it's a lot of work and you get a lot of data. But what's the point of it if we don't actually understand the data and understand what it's showing us? So that's why we do stats. We also use it to identify trends in data. So, is there something that's particularly common? Is there, you know, it helps us understand is there cause and effects going on? Is there causation is there correlation, what is actually happening between two different variables? And also it helps us summarize and organize data because sometimes you'll do a big study and you know, you'll look at the results and it'll be really complicated. But statistics makes it really easy to look at something and think. Oh this, this is what they found. So for example, you read a study, there's loads of texts and you see actually 70% of people um improved when you gave them a different hypertension medication. And you're like, OK, that makes sense. That actually summarizes the data and makes it easier for you to understand. And that's why we do stats because it just helps and to show what we've done. So we thought we'd start by talking about the different types of um which is what we do statistics on. So you guys might have already come across this. Uh We divide data into quantitative and qualitative. So quantitative data is data that's expressed in numbers. It's also known as numerical data and essentially it's quantifiable. So data takes on distinct measurable values. So numbers um and this is the data that can undergo statistical analysis. Um typically like this is the one when people talk about statistical analysis, they're talking about numerical data. Um And it's further subclassify interdisc and continuous, which will go into a bit more depth. Now uh on the other hand, we have qualitative data which is data that's expressed in groups, so also known as categories, which is why it's also known as categorical data and the categories are mutually exclusive. So one data point can't belong in more than one category. Um so, and for the most part, they cannot undergo stat statistical analysis because there's no numerical relationship to work on. There are statistical methods that you can run on on certain types of categorical categorical categorical data. But for by and large statistical analysis is o on numerical data and qualitative data is also classified or subclassify inter nominal and ordinal which we'll talk about now. Um So quantitative data, like I said is data that can be organized into groups called categories and it can't be counted or measured like you would with numbers. Uh And it's typically collected through surveys, questionnaires and observational methods. And there are two types like we said, nominal and ordinal. So nominal is data that is organized into groups without any inherent ranking. Um So it's unordered, uh the categories are unordered. Um So for example, this would be like ethnicity, hair color, uh sex, for example, like none of these tho those they have categories, but they're not, there's no inherent right.