An ethical problem arises when statistical practice creates a conflict involving harm, honesty, responsibility, or fairness. Incentives can reward favorable conclusions and punish inconvenient ones. section 1.4
A successful pricing system?
A company claims a new pricing system increased revenue by $70 million with statistical significance. A competitor’s retreat may explain the increase instead, creating pressure to present a convenient conclusion diplomatically. section 1.4
When findings become inconvenient
A cereal company commissions research on diabetes risk, then pressures the researcher to re-evaluate unfavorable findings, search for different variables, or bury the study. Cherry-picking supportive studies can make weak evidence look persuasive. section 1.4
