You're top students and don't understand that a sample of two is a poor way to reason about social and human problems at scale? Across large population of individuals with different traits?