Education

More on High-Stakes Testing and Science

By Sean Cavanagh — July 09, 2008 1 min read
  • Save to favorites
  • Print

Yesterday, I wrote about a new study that purports to show a link between high-stakes tests in reading and math and gains in student achievement in science. The study examines test results in Florida, and it gets at a crucial question in education these days: Is science being pushed out of the curriculum to make way for reading and math? The authors suggest the answer is no.

In the spirit of bringing some outside scrutiny to that work, I called David N. Figlio, a professor of economics at the University of Florida who has studied the effects of testing and accountability systems on academic performance across subjects.

Figlio told me he approved of the study’s general methodology (in fact, he’s used similiar methodology, which is cited by the authors). But he found the study’s overall conclusion about a link between high-stakes reading and math tests and improved science performance “less compelling” and “a stretch.”

The authors suggest that the pressures brought about by schools having received an F in reading and math could have sparked overall academic changes that led to improved science education. But Figlio said other factors could just as easily be at work. While science was a “low stakes” test in Florida at the time of the study, meaning schools faced no penalties for poor performance in that subject, those scores were still publicly reported, he pointed out. The public pressure to do something about weak science scores—rather than the penalties brought about by reading and math scores—may, on its own, have been enough to compel Florida schools to do better. Also, schools’ low test scores in reading and math may have simply prompted them to focus more on test-prep across all subjects, he said, rather than on making substantive academic improvements.

Despite those concerns, Figlio said the overall message conveyed by the report is “significant.” While you could debate the precise effect of reading and math testing on science, the study suggests that, at the very least, high-stakes testing has not undermined science instruction to the extent some people had feared.

The fact that you don’t see science scores falling in Florida, he said, suggests that schools are not “sacrificing low-stakes subjects” to meet the demands of reading and math instruction.

A version of this news article first appeared in the Curriculum Matters blog.