Box-Plots with Statistical Details - all you need to know

A Box-Plot is a powerful type of data display, which is useful already with a sample size of only 5 observations.
And In this video we’ll explore all aspects of the box-plot, starting with
- a single value, the Median,
- progressing through the simplest box-plot possible
- and finishing with a notched box-violin plot with two types of confidence intervals and statistical groups comparison.
At the end of this video you’ll be able to understand every little detail about this plot.
If you only want the code (or want to support me), consider join the channel (join button below any of the videos), because I provide the code upon members requests.
Enjoy! 🥳

[@oussamakadded978]

Just when I thought I know everything there is to know about boxplots haha. I didn't know about those notches!. Clearly a lot of work went into making this video, thank you very much.

[@yuzaR-Data-Science]

Glad you enjoyed! Thanks for watching!

[@MohamedIbrahimEid]

Thanks for you explanation, you have demonstrated in very interesting way. The way you prepared and explained is really amazing. Thanks again and really very appreciated, you call me back to R while I was using Python for a while.

[@yuzaR-Data-Science]

I am glad my content is useful ;) Thank you for watching!

[@WilOspinoC]

Thank heavens for letting me know your channel and content. The previous-me guy would've never belived how excited I'd become for box-plots. Before this video box-plots were the most boring and bland graphics ever. Now they are part of my top 3. I truly never got to understand the notches and violin part, but now the've become part of my arsenal. Thank Yury you R the best.

[@yuzaR-Data-Science]

Glad to you liked it! :) I was feeling the same before I gave box-plot a chance. Now I absolutely loved them. What are your two other favorites plots?

[@zane.walker]

Although I have used boxplots for years, and violin plots in more recent years, your video gave me a greater appreciation of these very effective representations of data. Much appreciated (and great graphics!).

[@yuzaR-Data-Science]

Glad it was helpful, Zane! Appreciate your feedback! If you liked this video, you might like the newest, I improved a bit on the editing side of a video I think. Anyway, thanks for watching!

[@fishfish20]

You're work is s blessing.

[@yuzaR-Data-Science]

Thank you! 😊

[@luisroca835]

Such a great video I found here! Congrats! Perfect editing!

[@yuzaR-Data-Science]

Thanks a lot for your support Luis! Really glad it is useful for people other then my students! :) I plan to make more content.

[@saifalshehhi281]

Thank you very much

[@yuzaR-Data-Science]

You are welcome!

[@TheMISBlog]

Great Explanation , thanks

[@yuzaR-Data-Science]

Thanks mate!

[@mrigankasinha]

What an explanation 👍👍👍...

[@yuzaR-Data-Science]

Thanks a lot, Sinha!

[@SeverSava]

Hi Yury! I really enjoyed your presentation. I was looking to understand the reason for illustrating data points as they are on a Violin Plot. I do not understand the reason for the way they are spread. Also I don't understand how to interpret a data point that sits outside of the density line but still inside whisker's length. Is that an outlier or what? Also this was the first time when I heard of the mirroring thing when doing the Violin Plot. I understand the density line is mirrored, but what about the data points? From the plots I saw until now, I couldn't draw this conclusion. It always looked like the density line is mirrored, but the data points are spread in a manner that I can't really figure out . Can you comment a bit along this topic, please? Thanks a lot for your time! :)

[@yuzaR-Data-Science]

Dear Sever, thanks for valuable feedback and great question! The spread of the points is just for the visibility, otherwise they would overlay each other. The spread of the points is not distributed similarly to the violin in this package, although it could, and is not doubled. The doubling and distribution of the points similar to the density curve (or double curve - which is then the violin) was introduced solely by myself for the educational reasons (and has nothing to do with the package), because most of non-statisticians freak out when they see the violin-plot and start to run away, if I try to explain it or even just say "distribution" :) Thus, the point outside of density plot is definitely not an outlier, but simply jittered from the other points in order to reduce overlapping. So, there is no pattern or meaning of the spredding of the data on the final plot. The density is mirrored, while the points are not. I hope I could clarify your questions :) Cheers mate!

[@SeverSava]

@@yuzaR-Data-Science so when you say jittered, you mean they are spread randomly just for having a visual representation, but there's no correlation between the place of a point and any data point in your data set. On a scatterplot each point on the graph stands for a data point from the dataset. It's not the case for the Violin Plot, correct? Am I getting it right?

[@yuzaR-Data-Science]

Correct Sever, it's just a representation of the particular group. That's all. While on a scatterplot, there are two variables, which both have values, and the point should meet both variables, so that there is only one exact place for every particular point. However, sometimes, for the visability purposes people use "geom_jitter()" also on a scatterplot. Cheers! ;)