|
|
|
|
|
|
by aliceryhl
1762 days ago
|
|
|
This is a classic example of Berkson's paradox [1]. If the job applications you receive have a positive correlation between on-the-job performance and competitive programming performance, and you use competitive programming performance to select who you hire, then you can very easily end up with a negative correlation among the people you actually end up hiring. When you don't include the people you rejected in the statistic, the correlation is useless. [1]: https://en.wikipedia.org/wiki/Berkson%27s_paradox |
|
|