I keep getting struck wondering wether it's the measurements that are the problem, rather than the measured outcomes. Maybe it's harder to measure progress than looking at some test results. Either way, this interesting report adds another question mark to the idea of using high stakes testing to make high stakes decisions with teaching careers. If we can't adequately measure progress with the brightest students, taught by the best teachers, that doesn't say a lot about the whole ill-concieved enterprise.
The University of Houston researchers who conducted the study found that students in these programs were more likely than other students to do in-depth coursework with top teachers and high-performing peers. Yet students who barely met the 5th grade cutoff criteria to enter the gifted programs fared no better academically in 7th grade, after a year and a half in the program, than did similarly high-potential students who just missed qualifying for gifted identification.
"You're getting these better teachers; you're getting these higher-achieving students paired up with you," said Scott A. Imberman, an economics professor and a study coauthor. "To our surprise, what happened was very little."
Here's the paper.