Research / Behaviors

OB-05Strong evidenceDone by organizations

Rates Simulation difficulty before judging results

Scores each Simulation's difficulty (for example, with the NIST Phish Scale) and compares results only at similar difficulty.

Why it matters

Difficulty predicted behavior in replication research while training did not. Without difficulty, a falling click rate may just mean easier emails.

How to observe it

Difficulty rating recorded for every Simulation; trend lines reported by difficulty band.

Evidence

Supporting (2)