Let’s just say that I am developing a simple experiment. I want to find out which one of my kids that my dog likes best. Here’s the experiment: both of my girls stand 20 feet away from the Fletcher, the dog. They aren’t aloud to call him, they are just aloud to stand there and see which one he comes first. Pretty simple and probably unscientific but it will work to prove my point. Would anyone out there say that my experiment was flawed if I put a pork chop in one of the girl’s pocket? Adding an unrelated incentive would make my entire experiment invalid. Is that what your saying?
OK, then, let me change my experiment. This time, both girls standing 20 feet away only in front of one of the girls is standing outside of my invisible fence – you know, the kind that shocks the dog when he tries to run through it. What? You don’t think I’m validly determining which girl the dog likes best. I can’t believe that!
They do it on the PSSA! Rewarding schools for improving performance as well as punishing school districts who aren’t successful is the same experiment. If the test is to determine what students know and are able to do, don’t positive and negative incentives based on scores invalidate the results? If I’m handing out five dollar bills to every student who scores proficient or if I ‘m closing schools that don’t meet minimum standards am I not putting pork chops in some pockets and using electric shock on the others?
What happens is that the most important thing becomes the test. It doesn’t matter how we got there as long as the scores are there. All actions then are determined and justified by the test. The ends justify the means. And the means usually include cutting time on other academic subjects and can extend all the way to removing administrators.
Not much of a valid test environment.