One of Eli Goldratt's favorite expressions he used many times was, "Show me how you measure me and I'll show you how I'll behave." One of the primary reasons he used this expression was to emphasize the importance of selecting the "right" performance metrics. One of the metrics he wrote about was the predictable behaviors exhibited when manpower efficiency or equipment utilization were used to judge the performance of a manufacturing organization. Whenever either of these performance metrics are used, the predicted outcome is excessive work-in-process (WIP) inventory that clogs processes which in turn causes extended processing times and late deliveries of products to customers. The bottom line is that performance metrics do motivate behaviors. But guess what...this same phenomena happens even in school systems.
In 2008, The Atlanta Journal-Constitution broke the first of what would be several stories highlighting suspect test scores in Atlanta Public Schools and other Georgia districts. From what I understand, the educators were responding to pressure from the school system's leadership administration to show improvement in test scores or face discipline, or less pay, or even termination. Some even received bonuses for higher test scores.
So like many manufacturing companies, this school system used test scores as their performance metric, which by itself is not a bad thing. But when you tie test score performance to a person's livelihood, you get the result we're seeing here in Atlanta. That is, teachers and administrators actually changed answers on standardized tests to get better scores. Once again, show me how you measure me and I'll show you how I'll behave.