U.S. Secretary of Education Arne Duncan has just made the challenge of transitioning to the Common Core standards less burdensome for about one in five schools in California.
Duncan on Tuesday announced that schools that do the field test for the new Common Core assessment next spring can get a one-year waiver from also giving current state standardized tests required by federal law.
For California, that means those schools can turn next year to teaching the new standards exclusively without focusing on the California Standards Tests in English language arts and math – and the weeks of test prep that many students and teachers must endure preceding the paper-based, multiple-choice assessment. The schools that take the field test would use the CST results from the previous year for accountability under the No Child Left Behind law. The waiver could complicate the state’s ability to compute a school’s and district’s Academic Performance Index (API) score.
California and 44 other states that have adopted the Common Core standards have agreed to begin giving the new tests in spring 2015. California has a leading role in shaping the tests for Smarter Balanced, one of two consortia of states creating the federally funded tests in math and English language arts.
This past spring, 1 million students in 26 states, including 200,000 students in 1,400 districts and charter schools in California, took the Smarter Balanced pilot test, whose purpose was to test the validity of the new questions. Next spring, double that number – 20 percent of students in California and other Smarter Balanced states – will participate in the more extensive field test, a dry run of the full assessment they’ll take a year later, according to Deputy Superintendent Deb Sigman, who also serves on the executive committee of Smarter Balanced. Criteria for school participation in the field test aren’t yet available, Sigman said.
This will give many schools their first try at a computer-based test; they’ll also get some limited test results back. One purpose of the field test is for Smarter Balanced’s designers to determine score results for classifying students at proficient and other levels of achievement.
A capacity issue
It’s unclear how many schools in California have the number of computers and the Internet capability to administer the field test. Fewer than half of districts responded to a Smarter Balanced survey last year gauging their technological capability.
Evan Marwell, CEO of Education Superhighway, a San Francisco nonprofit that is working on the schools’ technology gaps, estimates that 27 percent of the nation’s schools have the bandwidth to handle the type of assessment that Smarter Balanced is designing and twice that percentage can handle a slimmed-down version – basically moving multiple-choice questions to the computer, without the complex, media-rich questions that the consortium is designing.
Duncan’s willingness to eliminate double-testing next year for grades 3 through 8 and grade 11 will provide relief for students facing the prospect of two tests of six to eight hours each – perhaps longer for Smarter Balanced – and for teachers faced with teaching two sets of standards, one state and one national. While there is considerable overlap, the Common Core standards, with an emphasis on conceptual thinking, will create additional challenges for teachers.
In a statement, State Superintendent of Public Instruction Tom Torlakson praised Duncan’s position on testing flexibility as “good news for students and schools as we make the transition to new assessments based on California’s career- and college-ready standards. Double-testing would have taken up time students need to learn, and could have made our move to new assessments even more challenging.”
Torlakson has already proposed suspending most state standardized tests not required by NCLB starting next year and to use the savings in test administration to design new subject-matter tests aligned to Common Core standards, starting with Algebra and Geometry.
Reprieve on teacher evaluations
In the same statement, Duncan also announced that the federal government would push back by a year, to 2016-17, the requirement that growth in student test scores be used as a component in teacher and principal evaluations. Teachers and education advocates had complained that it would be problematic, if not unfair, to judge them by the initial year’s scores under a new system of Common Core standards and tests. Duncan’s extension would permit compiling three years of student results before use in evaluations.
California won’t be affected immediately, since the requirement applies to states that have received either a Race to the Top grant or a waiver from sanctions of No Child Left Behind. Duncan rejected California’s waiver application last year, and the earliest the state could apply again would be for the 2014-15 year. The extension could affect the nine California districts that are seeking a district waiver from NCLB through their organization, the California Office to Reform Education. They’ve agreed to use scores of standardized tests as a minor element in teacher evaluations. CORE expects to hear within the next few weeks whether its waiver application, taking effect this fall, is approved.
To get more reports like this one, click here to sign up for EdSource’s no-cost daily email on latest developments in education.
We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.
Karen Andresen 10 years ago10 years ago
To me the cup is all the way full. Get rid of STAR and be able to adjust to the new test without it counting. Implement a real evaluation system to involve peers in evaluation. Thank you Mr. Secretary.
Karen Andresen, Teacher Librarian, Petaluma Junior High school
Richard Moore (@infosherpa) 10 years ago10 years ago
So we are pushing back results until 2016-17. So Obama and Duncan will be gone as the damage reveals itself. Surprise, surprise.
All this to get pennies from the Feds. 44 States should be ashamed.
Doug McRae 10 years ago10 years ago
". . . without focusing on the California Standards Tests . . . and the weeks of test prep that many students and teachers must endure preceding the paper-based multiple-choice assessment." John, why do you refer to test prep practices as something that teachers MUST endure preceding paper-and-pencil multiple-choice tests? Those practices are not mandatory by any statute or regulation, and in fact in many cases may run afoul CA's current statutory language prohibiting teaching … Read More
“. . . without focusing on the California Standards Tests . . . and the weeks of test prep that many students and teachers must endure preceding the paper-based multiple-choice assessment.”
John, why do you refer to test prep practices as something that teachers MUST endure preceding paper-and-pencil multiple-choice tests? Those practices are not mandatory by any statute or regulation, and in fact in many cases may run afoul CA’s current statutory language prohibiting teaching to the test. Those practices are quite simply ways folks try to scam the assessment and accountability system. And practices to try to scam these high profile systems will almost assuredly continue with open-ended constructed response and performance task testing formats . . . . we’ve been doing writing samples for large scale tests for more than 30 years, essentially performance tasks, and folks find ways via formula writing practices and other mechanisms to try to shortcut writing instruction and in effect scam the scoring rubrics for writing samples. Such shortcuts need to be characterized as part of the human condition based on an attraction to take the easy way out, rather than blamed on some sort of invisible mandate inherent in a particular version or format for tests.
Widely read and well respected journalists like you ought not fall into use of rhetoric like “must endure” that falsely attacks a particular form of testing. Don’t get me wrong . . . multiple choice test items have both good and bad attributes, as do constructed response and performance task testing formats. In a perfect world, all types of test questions should be fair game for those responsible for test making as well as those taking the tests and those using the results.
On a more substantive level, whether it will be wise for California districts to apply (through the state, I presume) for Duncan’s 1-year waiver from double testing in 2014 is a very open issue. It depends on the nitty gritty details for the waiver as promulgated by the feds — do districts/schools have to use one test or the other for all students all grades both content areas, or will they be able to isolate selected grades (or classrooms), grades, and content areas for SBAC tests? We dunno yet.
What information will be generated by the SBAC “field” tests in 2014? The initial SBAC plans indicated that standards-setting to determine cutscores for proficient (and other levels of achievement) would be done by August 2014, but there are serious questions whether SBAC has developed and tried out a sufficient number of test questions via their spring 2013 pilot tests to construct a complete “final form” field test for spring 2014. So, we dunno yet.
And what STAR tests will be required in 2014? That will presumably be determined by AB 484 now making its way through the dance of legislation in Sacramento. AB 484 now has the SPI’s recommended suspensions [all STAR tests not currently used for NCLB or EAP] but there are alternative suspensions that make better sense and save greater amounts of testing time and dollars, particularly suspensions that include use of STAR end-of-course tests for NCLB requirements as well as CAHSEE graduation requirements to eliminate perhaps 80 percent of current CAHSEE test administrations. So, again, the answer is . . . we dunno yet.
Finally, as the post mentions, what will be the effect on APIs for both the 2013-14 cycle as well as the 2014-15 cycle? That will depend on score equivalencies between STAR tests and SBAC field tests, and when those equivalencies will be available. So, we dunno yet. We need answers for a lot of unknowns before folks can make rational judgments whether they want to apply for the just announced Duncan waiver on double testing in 2014.
navigio 10 years ago10 years ago
Paul, do you really expect STAR-SBAC equivalency recommendations? Do you think they can be accurate? I know people who have given up on the idea of an SSC for the next few years, partly because test scores will be meaningless for the next couple of years and partly because LCFF takes away a huge chunk of the funds the oversee.
Doug McRae 10 years ago10 years ago
Navigio: STAR-SBAC equivalency data have been promised by CDE folks based on analyses to be done on data from kids taking both the STAR tests and the SBAC field tests in 2014. The equivalency studies are not included in the SBAC scope of work for the feds, but rather are the responsibility of individual states if they want equivalencies between their previous tests and new SBAC tests. But, obviously, any equivalency studies (whoever … Read More
Navigio: STAR-SBAC equivalency data have been promised by CDE folks based on analyses to be done on data from kids taking both the STAR tests and the SBAC field tests in 2014. The equivalency studies are not included in the SBAC scope of work for the feds, but rather are the responsibility of individual states if they want equivalencies between their previous tests and new SBAC tests. But, obviously, any equivalency studies (whoever does them)have to have access to valid and reliable SBAC scores from a strongly representative sample of students and include scores from all relevant subgroups. Whether or not the SBAC field tests will supply valid and reliable data from a strongly representative sample of CA students and for all subgroups for all grade levels in 2014 is becoming very questionable. SBAC only piloted 5000 test questions this last spring, rather than the 10,000 previously indicated, and typically only half of piloted test questions have empirical characteristics to qualify for inclusion in field testing. Only 2500 qualified test questions spread out over 14 tests (grades 3-8 and 11, for E/LA and Math as separate tests)are not nearly enough to populate valid and reliable field tests that meet all the blueprint requirements. Also, the piloting this last spring did not address the measurement needs for special populations like Spec Educ students or English learners. All this information adds up to significant concern whether CA will have STAR-SBAC equivalency information shortly after the SBAC field tests are conducted spring 2014. The equivalency information will be needed for any APIs for the 2013-14 cycle for schools or districts participating in the SBAC field tests next spring, and then for the 2014-15 API cycle for schools and districts taking SBAC tests for the first time spring 2015. Doug
el 10 years ago10 years ago
It seems to me that norming the tests together is something of a waste of scarce time and resources. Why not reset the clock on the growth targets etc? The whole point of completely changing the curriculum is to completely change the curriculum and test wholly different things. If the school does well with the new tests and the new curriculum, what does it matter how they did with the old. And similarly, if the … Read More
It seems to me that norming the tests together is something of a waste of scarce time and resources. Why not reset the clock on the growth targets etc? The whole point of completely changing the curriculum is to completely change the curriculum and test wholly different things. If the school does well with the new tests and the new curriculum, what does it matter how they did with the old. And similarly, if the school does poorly with the new curriculum, what does it matter how they did with the old?
Doug McRae 10 years ago10 years ago
EL: Not to be overly technical, but equivalency studies are not the same as "norming." But the main answer to your questions is the argument that Common Core does not constitute a whole new curriculum, but rather a change in emphasis how target content is to be taught. Once local superintendent said -- Common Core content is not a whole bunch different than CA's 1997 content standards, but the expectations for how to … Read More
EL: Not to be overly technical, but equivalency studies are not the same as “norming.” But the main answer to your questions is the argument that Common Core does not constitute a whole new curriculum, but rather a change in emphasis how target content is to be taught. Once local superintendent said — Common Core content is not a whole bunch different than CA’s 1997 content standards, but the expectations for how to teach to those stadards is indeed a whole bunch different. Personnally I’m rather agnostic on the topic of old CA standards vs the Common Core — seems to me the CC E/LA standards provide some nice tweaks to the expectations in the CA 1997 standards, while the Math CC actually set somewhat lower expectations from a content perspective especially through the middle school grades [CA’s 1997 standards expected Algebra by grade 8, the CC expectations by comparison are less ambitious]. Clearly, the CC favor focus on concepts rather than skills. AS a test designer and developer, I’ve been trained NOT to bias tests toward one approach to instruction or another; rather the view testing folks have to take is that the test needs to fairly measure WHAT is learned rather than HOW it is learned. Also, I think most folks agree with the proposition that different kids learn things in different ways, and it is the chore of instructional leaders and teachers to match differing methods with the needs of individual kids. If indeed the Common Core constitutes a whole new curriculum, then you have a good argument to reset the clock on everything and ignore continuity of assessment and accountability (i.e., API) data. But, I don’t think Common Core constitutes a whole new set of content to be taught, and there are many needs for continuity of both assessment and accountability numbers. Thus the need for equivalency translations from the old to the new. Finally, from the public perspective, invariably the first question asked when new tests are installed is “How to the results compare to the old tests?” Doug