For four years, schools in nearly every state have been working to put the Common Core State Standards into practice in classrooms, but few have put them to the test鈥攍iterally. This year, that changes.
The 2014-15 academic year is when nearly every state must have assessments in place to reflect the common core, or other 鈥渃ollege- and career-ready鈥� standards they have adopted. And unlike last year, when many states were allowed to cut back on their regular tests because they were field-testing new assessments, this year鈥檚 achievement results will be a cornerstone of states鈥� public accountability reporting.
The specter of college- and career-ready assessments has loomed large in education leaders鈥� minds for several years, since it comes with a volatile mix of novelty and risk.
Schools will be held responsible for how well they鈥檝e imparted the new standards, even as skills such as reading complex text and demonstrating mathematical reasoning are new to many students, and as teachers are still figuring out how best to teach them. States face big drops in proficiency rates if the new tests are, as expected, tougher than the previous ones.
Even as educators steel themselves for those results, questions swirl about how well the tests will measure the standards they鈥檙e based on, and the skills educators value most.
Two dynamics further complicate the question of assessment.
Some states have moved to choose new tests after backing out of shared assessments or reversing their common-core adoptions.
And national uneasiness about the time and money spent on standardized testing, and about the decisions based on it, is increasing.
鈥淲e鈥檙e hoping that districts in Illinois, and everywhere, wake up to the problem with all the state testing, because this year it鈥檚 coming home to them as it never has before. It鈥檚 craziness,鈥� said Cassie Creswell, who is helping organize opposition to testing in Chicago, where her 3rd grader attends public school.
Alignment Concerns
With next spring鈥檚 testing season already on the horizon, measurement experts worry that many states risk giving assessments that don鈥檛 fully reflect their academic standards. Allowing only months for a new or tweaked test鈥攐r using an existing test for new standards鈥攅rodes the likelihood of good alignment, they warn.
鈥淲hen you鈥檝e developed a test with one goal in mind, and that target is changed, you鈥檒l have a misalignment between assessment and instruction, and that鈥檚 not good for anybody,鈥� said Stephen G. Sireci, the director of the Center for 91制片厂视频al Assessment at the University of Massachusetts Amherst.
State education leaders are all too aware of that uncomfortable truth as they transition to new tests. Iowa adopted the common standards鈥攚hich cover English/language arts and mathematics鈥攂ut chose to keep using its own state test this year.
Brad Buck, the director of K-12 education in Iowa, said that a recent study showed 鈥渨eak to limited alignment鈥� between the Iowa assessments and the common core, a claim the test鈥檚 maker disputes.
鈥淲e recognize we鈥檙e in that middle ground between the assessments and the new standards, and it鈥檚 not an easy place to be,鈥� Mr. Buck said.
Some vendors, from the start, have been tracking the common-core initiative and work by the two major state assessment consortia developing aligned assessments, so they might be able to offer tests that capture the standards relatively faithfully, said Gregory J. Cizek, a professor of educational measurement and evaluation at the University of North Carolina at Chapel Hill, but it鈥檚 no sure thing.
鈥淵ou could buy an off-the-shelf test, and it could be terrible because it鈥檚 not aligned [to the standards], or it might be aligned. We just don鈥檛 know yet,鈥� he said. 鈥淎t the same time, you can鈥檛 measure the common-core standards without asking kids to construct arguments and cite evidence. And you can鈥檛 do that with just multiple-choice questions.鈥�
A state would be well advised to demand research that shows the testmaker has performed an in-depth study detailing the degree of alignment between the state鈥檚 standards and the test it鈥檚 considering buying, Mr. Cizek said.
鈥淎 vendor should have鈥攁nd the state should demand鈥攅vidence of alignment,鈥� he said. 鈥淲hen a vendor says, 鈥楬ere is a test that measures the common core,鈥� a state鈥檚 first question should be, 鈥榃hat is your evidence?鈥� 鈥�
It takes many months of development to design a test that uses performance tasks, evidence-based essays, and multistep math problems to probe students鈥� skills in a nuanced way, experts say. States that buy new tests or quickly revise current ones, they warn, are likely to end up with assessments similar to the multiple-choice exams that have been so heavily criticized in the wake of the federal No Child Left Behind Act and its emphasis on test-based accountability.
鈥淭he closer a state is to scrapping the consortia tests and doing something quickly, the more likely those tests are to be closer to that end of the continuum,鈥� said Derek Briggs, a professor of research and evaluation methodology at the University of Colorado-Boulder.
The two main groups of states that used federal money to design common-core tests, the Smarter Balanced Assessment Consortium and the Partnership for Assessment of Readiness for College and Careers, or PARCC, have worked for nearly four years on their tests. But measurement experts say the question of how well those tests are aligned with the common core won鈥檛 be settled until research on that issue is completed. Both groups plan to complete such studies in the coming months.
PARCC spokesman David Connerty-Marin noted that not all the alignment work has been left until the end of the test development.
鈥淭here鈥檚 a significant difference between this test and tests states typically purchase,鈥� because the PARCC test was 鈥渂uilt from scratch鈥� for the common core, he said in an email. In addition, he said, the consortium enlisted teachers and other educators to review items for alignment as they were written.
As a result, 鈥渢here should be a lot less mystery鈥� in determining PARCC鈥檚 alignment to the common core than there would be when states try to determine alignment of their own tests, or an off-the-shelf test, to those standards, he said.
Joe Willhoft, the executive director of Smarter Balanced, said that once its alignment studies are complete, the consortium will have time to revise parts of the test that are not well enough aligned before administering it in the spring. Even still, alignment is best thought of as ongoing, he said.
鈥淲ithout a doubt, there are some aspects of this alignment work that will be tighter than others,鈥� Mr. Willhoft said. 鈥淚t鈥檚 never something that鈥檚 all done, it鈥檚 perfect, and we don鈥檛 have any more work to do. Each year, we expect to be refining it.鈥�
The Testing Timetable
The obligation to administer college-readiness tests has its roots in two key places, both linked to the U.S. Department of 91制片厂视频: the work by PARCC and Smarter Balanced, and the waivers the department offered to excuse states from key requirements of the No Child Left Behind law. States that belong to one of the consortia or have an NCLB waiver had to commit to having tests of rigorous standards in place by 2014-15.
The consortia work dates back to 2010, when the 91制片厂视频 Department awarded $360 million in grants to the two groups to make tests tied to the common standards that emerged that year from a project led by the National Governors Association and the Council of Chief State School Officers. Wielding voting power in either assessment consortium鈥攚hich nearly every member state chose to do鈥攔equired states to administer those tests in 2014-15.
鈥淭his is the year that the consortia are expected to deliver on the things they promised,鈥� said Mr. Sireci. 鈥淭his is the year we are going to see what the quality of those assessments might be, and what kinds of results we鈥檒l get back.鈥�
But in the past year, the standards and testing landscape has seen big shifts. Three states鈥擨ndiana, Oklahoma, and South Carolina鈥攔eversed their common-core adoptions. They have had to hurry to craft new standards鈥攕ome of which draw heavily on the common core鈥攁nd revise or buy tests. States that kept the common standards but opted not to use PARCC or Smarter Balanced, such as Florida, Iowa, and Kansas, are revising their current assessments, or engaging vendors to design new ones, as the clock ticks down to spring testing time.
In 2009, every state was using its own test. Within two years, all but five had agreed to use shared exams being produced by the consortia. But 11 more have jettisoned those tests amid mounting worries about time, cost, and political flammability鈥攖he tests were increasingly seen in some quarters as federal overreach into local school affairs.
It鈥檚 not only those 16 states that plan to go their own way with tests next spring, however. An 91制片厂视频 Week analysis showed that 24 states鈥攕ome still members of PARCC or Smarter Balanced鈥攈ave chosen to use another test, or are still deciding what to use.
Even though consortium rules required 鈥済overning鈥� states to use the resulting shared tests, and likely prolonged many states鈥� commitments to do so, there were no consequences attached to changes of heart鈥攗nless, that is, a change of heart occurred in a state that had secured a No Child Left Behind waiver.
First offered by the 91制片厂视频 Department in 2011, those waivers have been awarded to all but seven states (and withdrawn from two, Washington state and Oklahoma). In exchange for leaving behind some of the more onerous requirements of the federal law, waiver states must give 鈥渉igh-quality assessments鈥� of 鈥渃ollege- and career-ready standards鈥� in English/language arts and math in 2014-15. States could meet that requirement by promising to use Smarter Balanced or PARCC tests.
Nearly every state that made such a promise in its waiver application and then changed its mind found a letter in its mailbox from the federal department asking for detailed testing plans in lieu of the consortia exams.
No state has yet lost or been denied a waiver because its testing plans didn鈥檛 meet federal muster (although one state, Oklahoma, did lose its waiver after reversing its common-core adoption). Typically, when a waiver state has decided against consortium tests, the federal department has sent a letter of inquiry, asking the state to detail its plans to 鈥渁dminister annual, statewide, high-quality assessments aligned with college- and career-ready standards.鈥�
According to the department, 13 states have received such letters; several others didn鈥檛 because they had included satisfactory testing plans in their applications. It鈥檚 clear from the department鈥檚 decisions that it is willing to grant flexibility to states that have chosen non-consortium tests.
Securing Approval
Alabama, Kentucky, and Virginia, for instance, never got those inquiry letters because their plans to use non-consortium tests were detailed in their original waiver applications. Kentucky is using Pearson-designed common-core tests in grades 3-8 and a suite of ACT tests in high school. Virginia is using its own Standards of Learning tests. Alabama chose the ACT Aspire suite of tests, which it used in 2014 and plans to use again in 2015.
Michigan鈥檚 legislature barred that state from using the Smarter Balanced exams this school year, but it still got a waiver extension after explaining to the 91制片厂视频 Department that it will modify its current test. Georgia got an inquiry letter when it dropped out of PARCC; the state submitted an 85-page description of its testing plan, and its extension was approved in July.
Tennessee received a letter from the federal department after its legislature approved a law requiring the state to use its current test, the TCAP, in 2014-15. The state submitted details of its plans to align that test with the common core, and it is awaiting a federal decision.
States are feeling intense pressure as they seek approval of their testing plans from the 91制片厂视频 Department. Officials from several states contacted for this article refused to discuss the matter, citing political controversy over testing and the fear that their waiver extensions might not be approved.
鈥淚t鈥檚 crazier than ever,鈥� one state official said. 鈥淣o one around here will be willing to talk to you about this.鈥�
States are also mindful of another source of testing pressure coming down the pike: new federal guidelines for peer review of their standards and tests. Those criteria, due out soon, shape how experts chosen by the 91制片厂视频 Department decide whether states鈥� standards and assessments meet requirements in federal law.
