Simpson's Paradox

I show an example of Simpson's Paradox. This is when an observed relationship is reversed when a third lurking variable is brought into the picture.

Kidney Stone data referenced from: ^ C. R. Charig, D. R. Webb, S. R. Payne, O. E. Wickham (29 March 1986). "Comparison of treatment of renal calculi by open surgery, percutaneous nephrolithotomy, and extracorporeal shockwave lithotripsy". Br Med J (Clin Res Ed)

[profrobbob]

Thank you. I hope to hear good news about your exam grade soon:) I am still in awe that I am helping people in so many places with just a camera and a computer. This is so awesome!

[profrobbob]

Thanks for watching and I hope you'll come back to learn more!

[Mupke]

my exam is next week.. thank you from the netherlands!

[profrobbob]

Good luck on you're exam:):):) Thank you for watching!

[profrobbob]

Congratulations!!! That is what I am hoping for all my viewers. Keep up the good work:D

[normanb]

Rob, you are amazing. This basically saved me when doing my stats homework.

[anubhavnehru]

Thanks, Prof, but can you please explain to me the reason why when we club the values the whole scenario changes, in this case, Treatment B has higher Success % in 2-way table

[lukavongola]

Hello! I have an AP stats question...and I was wondering if you could maybe explain a question from
The Practice of Statistics
Book ...maybe through a message? >.<
It is 6.30? Please help, I would appreciate it!

[profrobbob]

At least I can be your go-to YouTube statistics teacher now!...and "you" could start a study group with your class after school and you could all watch my videos together and pretend I was your teacher...lol...you should try it:D

[annacristinapozzato]

Thank you for the video but may I ask you what can we conclude? Is it better to keep the data aggregate or separate it by the factor? Thank you

[joefagan]

Cracking example - but where's the intuition. I've learnt that Simpson's paradox is there - and there are hundreds of examples - but you have been proper lazy about showing me how/when/where. No intuition imparted whatsoever. I know nothing new.
Show me how to creat my own exhibit of Simpson's paradox.
Show me the exhibit of Simpson's paradox with the fewest total number of treatment. Is it possible to exhibit Simpson's paradox where the fewest integer number of treatments <44?
If not, why not?

[mrphuckbud]

great vid. not what i thought simpson's paradox was haha. stats class failed me