Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Lior539
on June 16, 2023
|
parent
|
context
|
favorite
| on:
Annoying A/B testing mistakes
I'm the author of this blog. Thank you for calling this out! I'll update the example to fix this :)
hammock
on June 16, 2023
|
next
[–]
The example is fine, you are calling out confounding variables. Just call it confounders, instead of Simpson paradox.
keithwinstein
on June 16, 2023
|
prev
[–]
FWIW, the arithmetic in that example also has a glitch. For the "Mobile" "Control" case, 100/3000 is about 3% rather than 10%.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
