Скачать или смотреть Why your data (probably) doesn't need to be normal

There is a common misconception that data should be normal if you want it to be usable in various tests and models. "Normal" here meaning bell curve-shaped.

This video has the basic structure:

1. Introduce normal distribution
2. Cases where normality is expected
3. Reasons why normality is not strictly required in most cases
4. Brief discussion of normality tests
5. Methods of transforming data to get more "normal" data
6. Conclusion and space fugue

I need a minimum # of views to post links to some files with code/spreadsheets (R, Python, and Microsoft Excel) for anyone who is interested. These will be posted here later.

This is my first video, so I'm experimenting a bit, and any feedback or suggestions would be greatly appreciated!

Video timestamps:

0:00 (Introduction)
1:10 (The Normal Distribution)
3:41 (Z-tests)
7:20 (p-values)
9:06 (Ahoy!)
9:28 (t-tests)
11:53 (Space!)
12:26 (Linear Regression)
13:59 (Only the residuals...)
15:05 (Central Limit Theorem)
16:08 (The Actual Problem)
16:32 (Lilliefors Normality Test)
18:50 (Q-Q Plots)
19:46 (Data Transformation)
23:14 (Reddit Example)
23:55 (Overfitting and Model Validation)
25:40 (Conclusion)
27:00 (Credits)