Campbell's Law

Donald Campbell's principle that 'the more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.' Closely related to Goodhart's Law but emphasizing the corruption mechanism: when people are rewarded or punished based on a metric, they will manipulate the metric rather than improve the underlying reality. Test scores rise through teaching to the test, crime stats drop through reclassifying crimes, and corporate metrics improve through creative accounting.

When to use it

When metrics show improvement but reality doesn't feel better; when designing performance evaluation, incentive, or accountability systems; when high-stakes decisions are based on a single quantitative indicator; when 'gaming the system' is suspected.

How it can help

For any metric you use to make decisions, assess its corruption vulnerability: how easy is it to improve the metric without improving the underlying reality? The easier it is, the more it will be gamed. Design measurement systems with this in mind: use multiple independent metrics (harder to game all simultaneously), audit for metric manipulation, and periodically change metrics to prevent gaming infrastructure from accumulating. The most important practice: never use a single metric for high-stakes evaluation—the higher the stakes, the more metrics will be corrupted.