This report from the Pew Charitable Trusts says that many states are reconsidering the costs of Common Core testing, and a small number have withdrawn from participation in the two federally-funded tests, PARCC or Smarter Balanced Assessment Consortium.
“But as controversy over the Common Core has challenged some states’ commitment to the standards, a number of states have decided to withdraw from PARCC or Smarter Balanced or to use alternative tests, raising questions about the cost of the tests and the long-term viability of the multistate testing groups, which received $360 million in federal grants to develop the tests. The federal grants will end this fall, and it is unclear whether the testing groups will continue past that point.
“What gets tested is what gets taught,” said Joan Herman, co-director emeritus of the National Center for Research on Evaluation, Standards and Student Testing at UCLA. “To the extent that the assessments well represent the spirit and meaning of the standards, the spirit and meaning of the standards will get taught. Where the assessments fall short, curriculum, instruction and teaching will likely fall short as well.”
Federal law prohibits any officer or agency of the federal government from attempting to influence or control curriculum or instruction in the nation’s public schools. It is axiomatic that “what gets tested is what gets taught,” so it is surprising that the U.S. Department of Education funded these two testing consortia; private foundations should have done it.
In the article, several “experts” are quoted about the minimal costs of switching to the new tests, but at least one of them points out that the low-ball estimates rarely include the costs of new technology and additional bandwidth.
At a time when many states are cutting education budgets and increasing class sizes, some states will find it challenging to increase their spending on assessment.
Unmentioned in the article is the issue of computer-scored testing. Students in theory will answer questions by explaining why they answered as they did. Computers will evaluate their “deeper thinking” as well as their essays. Les Perelman of MIT has demonstrated that robo-graders are unable to tell the difference between sense and nonsense so long as the sentences are structurally sound. Yet millions of students will be judged by computers that are unable to discern irony, wit, creativity, humor, or even fact.
Whose idea was it to put all testing online? Dumb idea. In my view, which doesn’t count as much as Arne Duncan’s or Bill Gates’, most tests should be written by teachers (they know what they taught) and graded by teachers (so they can discover immediately what students learned and did not learn).
Diane Ravitch's blog 