Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I think you meant to write "reverse causation"? Correlation is by definition symmetric and directionless. If the researchers make no causation claims, they don't need to be concerned about the direction of causation.

I haven't followed the field, but it's hard for me to imagine published research in serious academic journals where causation claims are made without at least some discussion about the direction of causation and hidden confounders.



It is likely that the poster was referring to a negative correlation. While incorrect terminology was used, correlations do have direction.

https://stats.stackexchange.com/questions/173276/is-it-corre...

https://study.com/academy/lesson/negative-correlation-in-psy...


Negative correlation also known as anticorrelation does not mean there is a "direction". Simply, it is a scalar number and not a vector. Don't add to the confusion of terminology by using a colloquial definition of "direction" to mean it can be negative or positive. We could have defined correlation to be in [0,1] and your sense of direction would be very confusing.


Thank you for the clarification, you are absolutely correct that its positive or negative rather than having a direction.


Probably, not my first language.

> If the researchers make no causation claims

But sentences like "kids using their phones more than x hours a day are on average sadder" very much imply one.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: