How to detect Hidden Correlations in R using Linear Models

I feel that spurious correlations get more attention in data analysis than hidden correlations, though both pose similar challenges.

A hidden correlation means that while there is a relationship between two variables, we don't see it directly because it is hidden by another variable.

In our example, we see no effect of study hours on the score in an exam, which is counter intuitive. Rather than throwing our data away, we look for an explanation. Some students drank alcohol while preparing for the exam. Creating a model that controls for alcohol consumption, we now see a significantly positive effect of study hours on the score in an exam.

Have you found similar examples? What are your experiences? Looking forward to you comments!

You can find data and code here: https://github.com/fjodor/spurious_hi...

Contact me, e. g. to discuss (online) R workshops / trainings / webinars:

LinkedIn:   / wolfriepl  
Twitter:   / statistikindd  
Xing: https://www.xing.com/profile/Wolf_Riepl
Facebook:   / statistikdresden  

https://statistik-dresden.de/kontakt
R Workshops: https://statistik-dresden.de/r-schulu...
Blog (German, translate option): https://statistik-dresden.de/statisti...

Playlist: Music chart history
   • Music Chart History