If you don't believe in the measurement tool

I was rereading the recent report on Teacher Pay for Performance that many are citing as their "I told you so" moment.  But then I came across the paragraph below.   The researchers used survey data to ascertain teacher beliefs about the program.   One important finding, is that teachers did not appear to believe in the measurement tool:

Participating teachers generally favored extra pay for better teachers, in principle. They did not come away from their experience in POINT thinking the project had harmed their schools. But by and large, they did not endorse the notion that bonus recipients in POINT were better teachers or that failing to earn a bonus meant a teacher needed to improve. Most participants did not appear to buy in to the criteria used by POINT to determine who was teaching effectively. Perhaps it should not be surprising, then, that treatment teachers differed little from control teachers on a wide range of measures of effort and instructional practices. Where there were differences, they were not associated with higher achievement. By and large, POINT had little effect on what these teachers did in the classroom.

If teachers did not believe the measurement tool effectively captured who was the worthy of the bonus, why would they change their behavior?