Testing For Normality - Clearly Explained
HTML-код
- Опубликовано: 17 мар 2020
- In this video, I will provide a clear overview of normality testing data. Testing for normality is an important procedure to determine if your data has been sampled from a normal (Gaussian) distribution.
There are two main ways that are commonly used to deduce whether data have been sampled from a normal distribution: analysis of graphs (eg, Q-Q plots and frequency distributions) and performing normality tests (eg, Shapiro-Wilk test).
HOW I CREATED THIS TUTORIAL (AFFILIATE LINKS)
Laptop amzn.to/38bB7JG
Microphone amzn.to/2OFn1sd
Screen recorder & editor techsmith.z6rjha.net/BQQkL
RUclips SEO www.tubebuddy.com/SHTeach
Software (Microsoft PowerPoint 365 ProPlus)
FOLLOW US
Website toptipbio.com/
Facebook / toptipbio
Twitter / toptipbio
AFFILIATE DISCLAIMER
Top Tip Bio is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to www.amazon.com. We will earn a commission from Amazon if a purchase is made through the affiliate links. - Хобби
Once again you have found a way to simply describe something that can be difficult to comprehend. Your explanations and videos are truly first rate.
Thank you so much for such an informative and useful guide. I write my bachelor thesis and try to find out if my data is normally distributed. Thanks to your clear explanations, now I know exactly how to test it!!👍🏼
I find it the best video currently available on RUclips👍🏼👍🏼👍🏼
fantastic explanation. the entire normality confusion is cleared now. i wish this channel comes up with more statistical chapters.
SUBSCRIBED !
simply put, you are great. keep up the outstanding job man
Great explanation, thank you so much!!
Good video saving this for a reference point to anyone looking into BI Data Analysts prep kit I'm making
Nice vid! keep up the good work.
Thank you very much for this explanation !!!
Tremendous explanation. Thanks.
Thank you very for your fantastic explanation!
Fantastic explanation!
Very clear...thank you!
That really helps.. thank you so much
Very helpful. Thank you
absolutely fantastic. really interesting point about power 8:40
When p-value is bigger than 0.05 we do not reject the alternative hypothesis. The only thing we are observing is whether or not we reject the null hypothesis, therefore only thing we can reject is the null hypothesis if p-value is below our significance level. Otherwise great vid.
I had the same reaction. We can't reject in both depending on the p-value, just reject the null or fail to reject the null because we don't have enough evidence to reject the null with that level of significance.
I concur: we can either reject or fail to reject the null hypothesis.
That's the end goal, is that what you mean? No pun intended
Wonderfully explained
thank you, it's easy to understand
Your videos are absolutely amazing!! How do you prep your video? Do you do it in powerpoint, and do you use graphpad to make these graphs and figures? How do you also lay your graphs/figures on top of each other?
Thanks very much :)
For this I made the graphs using GraphPad Prism and present them in PowerPoint. I add and remove data sets from the Prism graphs to make different 'layers' and animate them in PowerPoint.
You can see some links to software I use in the video description.
Thanks!
Steven
This is nice, short and knackig :). Thanx!
You're welcome!
Excellent explanation👍👍
very clear, thnaks a lot
Thank you for this video
Hi can you please make a video on ROC Curve and AUC curves using Graphpad. I appreciate your efforts.
Thank you very much!
Testing for normality? More like "Terrific video that you gotta see!" 👍
Now I'm definitely curious about the specifics of the normality tests, but I bet they're rather complicated...
Great explanation it helped me a lot with my data interpretation, thank you so much . Parting from here, would be great to have something like how to chose the proper statistical analysis for the data we are interpreting. It is yet very confusing
instablaster...
Great explanation! Thank you
You are welcome!
Very nice and easy to understand, thanks
Very welcome
note: if p>0.05 you not accept the null hipothesis, just fails to reject it. it is not the same.
what does that even mean ? if p < 0.05 we reject null hypothesis and if p > 0.05 we retain the null hypothesis statement. It's that simple, please don't confuse the world.
Hello Really i appreciate your video.
I have a question!!!!!
I have a negatív value in the X axis
What shall I do please!!!
Great video
Thanks ❤
Thanks 🙏
Great explanation, easy to follow and understand
Thank you Sir
thank you
Thanks
Good statistics course
great video, thank you
You are welcome!
thanks!
Thank u
I wonder what is the smallest number we can use the normality test.
Perfect
If the data lets say score of student is not normally distributed then what will we do? Will we use non parametric test like Mann-Whitney?
Hello. It depends on what you want to do. If you want to compare two sets of continuous data that are not normally distributed. You could try and transform your data (eg log transformation) to see if this improves the distribution. Or you could use a non-parametric test, in this case, a Mann-Whitney test
Great explanation! But I have a question. Suppose I am using likert scale to level of agreement in my study, can I use demographic variable to assess the normality of my data?
Thanks Kiera.
That depends on the type of data you have. If the data is a continuous variable (e.g. age, height, weight etc), then yes you can assess the normality of this data.
If the data is categorical (e.g. gender [male/female]), then no.
Hope that helps,
Steven
@@StevenBradburn thanks.
Appreciate you videos a lot.
Thank you!
Thank you :)
The video sound is pretty good, beyond my imagination
Is it safe to say that when the data are not normal, we use nonparametric tests? Thanks for reply.
Yeah! Use nonparametric if it is not normal
Thanks, what does it mean if the q-plot shows normality but skewness/kurtosis does not
It means the same. Because qq measures normality and skewness measures lack of normality.
Hi,process capability aim to achieve by consuming 50% tolerance
When the dats are LSL to USL range we can get P value
But we are fixing control limits how can get P value
How to calculate the p-value? Please could you make a short video on that?
Hi Mohammed,
Why statistical software do you use?
Thanks
Steven
@@StevenBradburn Hi, I don't use any software right now. I'm looking at the basics now. But will later be using in MATLAB.
Why normality tests? Why dont we implement poissonity test? What makes normal distribution privileged among other distributions?
I.....finally....understand 😭
Memerlukan lebih ramai orang jadi sebarkan video ini lebih banyak
FUCK YEAH STEVEN THANK YOU BRO
It is not correct to either accept one or the other hypothesis! There is also the option that you can't conclude any correllation in the data.
Analyses don't assume that the population is normally distributed though. Is that the argument you're making?
Who knew Lee Dixon did statistics?
Thanks Jurgen. Hopefully my statistics is better than my Arsenal performances
The so-called “normal distribution” is just a special case of all common unimodal distributions. It is not a big deal in the foundation of statistics.
My answer to your quiestion are you normal or not is :if you have listened all these ramblings about normality - you are definitely not normal ,wich is not necessarililly a bad thing...
Very helpful thank you.
Thank you very much!
Welcome!