Avoiding the Rabbit Hole of Bad Science
The scandal surrounding the recently published so-called empirical evidence for precognition has caused psychologists to reconsider the way they analyze their data, and highlights the importance of replication. What can empirical subfields of computer science learn from this?