The Manhattan Institute came out with a report Tuesday that looks at gains in science achievement among a small group of Florida fifth-graders. It was somewhat cheekily promoted earlier this week by co-author Jay Greene as offering evidence that claims of narrowing the curriculum under NCLB were unfounded. I say cheekily because Greene specifically cited statements by Common Core co-chair Diane Ravitch and Fordham president Checker Finn as examples of the kind of claims this research would debunk. But the study’s findings don’t back up Greene’s spin. Not even close. The study found that, for students enrolled in a school receiving an “F” grade under Florida’s A+ assessment system, science proficiency increased about a .08 standard deviation. The authors claim that there is “some evidence to suggest” that that gain was enabled by increases in student math and reading. But as their strongly qualified language signals, the evidence offered up is far from convincing. And there are numerous other possible explanations for the increase in science proficiency—including the fact that performance on the science test carries stakes within Florida’s A+ system and that NCLB now requires science testing. Even though performance on science tests are not “high-stakes” under NCLB (as scores don’t count toward AYP), the scores do count toward the grades schools receive within the state—and all of this testing undoubtedly affects the emphasis given science in Florida classrooms. So science testing is “high-stakes” in Florida. This could very well explain the increased science proficiency Green et al. identified. The Manhattan Institute study might have been more interesting if they had looked at a subject entirely overlooked by Florida’s A+ system and by NCLB, such as history. That would’ve been difficult to do but it would’ve really given us something to talk about.