In her review of Daniel Koretz’s new book on testing in schools, Diane Ravitch reminds us of Campbell’s law:

In 1979, the psychologist Donald Campbell proposed an axiom. “The more any quantitative social indicator is used for social decision-making,” he wrote, “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.”
Ravitch applies this to high-stakes testing in school, using a story from Soviet Russia to bring the point home:
The classic (and probably apocryphal) illustrations of Campbell’s law come from the Soviet Union. When workers were told that they must produce as many nails as possible, they produced vast quantities of tiny and useless nails. When told they would be evaluated by the weight of the nails, they produced enormous and useless nails. The lesson of Campbell’s law: Do not attach high stakes to evaluations, or both the measure and the outcome will become fraudulent.
High stakes testing in schools is pernicious, Ravitch writes:
The children from elite homes are convinced by their test scores that they deserve their high status; their scores demonstrate their superiority. And children of the poor learn early on that they rank poorly; their test scores confirm their lowly status.
Source: New Republic