Statistics Lesson - SPSS - Normality Tests | Ton Van Voorden | Skillshare

Statistics Lesson - SPSS - Normality Tests

Ton Van Voorden, MSc.

Play Speed
  • 0.5x
  • 1x (Normal)
  • 1.25x
  • 1.5x
  • 2x
2 Lessons (10m)
    • 1. 12

      3:28
    • 2. 12

      6:50

About This Class

Back again with some Normality test. Important to check before doing analyses!!!

If your dataset is quiet small (number

Transcripts

1. 12: Hello. Didn't storm afforded your statistics teacher and this video? I'm going to explain something about no melody tests and why they are important. Firstly, for a lot of statistical techniques, it's important that not the data itself, but ever just off the data. It's normally distributed, and most of the time it's not necessary to check it, because if the end the sample size were bigger, larger as 30 the means off the data automatically normally distributed. But sometimes you have quite small, data said, and it is important to check this information. So what I just said, there's two options to possibilities. Ascension off. No melody. Oh, and it's bigger, equal a study. Then you will be good, but and it's smaller, a sturdy, and then you need to do it just to check if you, um if you apply, if you violate tests to check whether you violate this assumption or not, if you don't violate assumption. If you don't violate the ascension, you still go to go. Which test. But if you violate it, you have to perform a bootstrap analysis and for almost all so called aromatic death, you have to use in the in those cases, the bootstrap results and I will explain the bootstrap. How'd works in another video? So firstly really important to know the end off your simple as to be bigger study? Notice if you have, um if you compare different groups like a Novara t test, every group has to be the sample size of every group wedding. There has to be higher than 30. So no, only the all the data itself. If you have, for example, 2020 people from Spain and 64 Netherlands and you compare Spain against Netherlands, then 20 will not comply to the and it's bigger, equal and charity. So you have to assess a normality test in this case. In the next video, I'm going to explain to you how to do the normality test. Thank you for watching this video. 2. 12: Hello, is this storm of order, your statistic, teacher? And this year's about how to do a normality test within your favorite computer program back . It's SPS s first What's video I showed before, which is about when to use a normality test. And when you don't have to use it anyway, in this video, I'm going to show you how to use it. So let's see here. We got some data within SPS s. Let's say I want to do a test with and what I need to use wanting to eat a burger valuable sport on temperature temperature. And on this three variables I want to do in normality test, I goto analyze descriptive statistics explore and I put this say fables. Inside I pressed a controlled Kato letme or you can also select them one by one, and I selected into the dependent list. Now the next step is click on float and employees you see either normality plots with tests . You have to click it on. If a press OK, get so much data and get explore, I get case process summary descriptive. I don't need to use it for now. The together test of normality on what you will see. It has lost heaps of more data. I never use it. It's because he requests the plots together with normality. Test only thing you need our test off normality. Let's go into detail. What you see over here is wanting to eat a burger. You see test of normality. You see the common goal goes me in office, the most frequently used for really big data sets. If you and it's bigger is 1000 you can use the ship. People will, but notice that you actually don't need to use the normality test, and that's it. Zero. And it's 1/2 of Jesus. It's zero is they is normally distributed, and it's one is. Data is not normally distribute. So what you hope in this case, and that's with most of the assumption that it's not significant. So how to lie down, for example, for wanting to eat a burger you can write down De just just reminded why that it's a D In between bracelets, you put degrees of freedom equals 0.0 for eight, and the B value equals points toe 00 This is higher than 00.5 so there's not enough proof to reject the A zero. So in this case, we assume that is, there will be full and we can assume that the data is normally distributed. If you go to the below one for the temperature you got. Why, down off course D 88 again equals 0.100 I just wide out this piece because if you're white animosities, you just have to want it down. But you don't need to. I know what it means. And the P equals point 00 This is lower. Then the offer off the open 05 normally used often So in this case, would reject a zero would conclude that data will not be normally distributed if the inside's is always also smaller as 30. Um, in that case, we have to do a bootstrapping analysis, and I'm going to explain that to you in the next video about boots happen. I'm gonna give you another example. And that's the example. You have to do an analysis about wanting to eat a burger. But now you're curious. How did how the averages are wanting to eat a burger in a different countries, So not way. Don't want the normality assumption on three different variables like last time. But now we wanted on only wanting to eat a burger but separately for the different countries and are four countries from death case. We will get for results click analyze descriptive statistics, explore steps out of same I removed the temperature ever It's minutes of sport and now the country I added toe effect a list so I will not together results from the country. But I would get a retail from wanting to eat a burger. Split it up in the four different countries like you see at this brought this burden is still clicked on. I can click. OK, you can get a lot off output Just need legs neglected, please and he will see same kind of output with for all the for different countries. In this case, you can conclude looking to the numbers. Of course you have to widen down. Well, if you're writing it teases if you look to the numbers that began as you that everything is normally distributed because all the values are higher as 0.5 and noble stepping needed so we can just do all the techniques like you didn't before. Thank you so much for watching this video