To give districts breathing room to prepare for complex tests on the Common Core standards and to free up money to do so, State Superintendent of Public Instruction Tom Torlakson is recommending the suspension after this year of most state standardized tests not mandated by the federal government. He made the recommendations in a lengthy report (download top item) released Tuesday.
If adopted by the Legislature, the report’s dozen recommendations would represent the most sweeping changes to state testing since the adoption of the State Standardized Reporting Program or STAR system in the late 1990s. They would start the transition to a new accountability system, based on different priorities, perhaps with fewer tests or with subject tests not given to every student every year – the operating principle of the current system. Torlakson suggests alternatives in an appendix.
Districts would have to continue offering some tests: math and English language arts assessments in grades three through eight plus grade 11, along with science tests in grades five, eight and ten. All of these are required under the federal No Child Left Behind law.
But the state would suspend – and perhaps drop permanently – standardized tests that it has chosen to administer in second grade and suspend most high school end-of-the-year math and science tests, along with social science and history tests currently given in middle school and high school. Within a few years – education officials can’t say how soon – high school tests, say, in Algebra I or Geometry could be replaced by new tests, aligned to Common Core, that the state, perhaps in partnership with other states, would design. Or the Legislature could decide that end-of-year tests in history would be a local district option, no longer tied to a state accountability system. The first priority, said Deb Sigman, deputy superintendent of the state Department of Education, would be to design tests for the Next Generation Science Standards that the State Board of Education is scheduled to adopt this fall. The state would also create new tests for English learners, based on standards adopted last year.
California’s high school exit exam (CAHSEE), a state-imposed test that all seniors must pass to graduate, would continue – but its days are numbered. Torlakson offered several options for replacing it in the report. The state would offer Algebra II and English language arts in 11th grade, along with the Early Assessment Program test that the California State University uses to determine whether incoming students are prepared for college-level courses. But the expectation is that the new 11th grade assessment, based on Common Core standards, would become the key measure of college and career readiness – however the latter is defined.
The heart of the new testing system would be Common Core tests in English language arts and math that California and member states of the Smarter Balanced Assessment Consortium are designing. They’ll be administered in third through eighth and 11th grades, starting in spring 2015.
The new standards, stressing problem solving and critical thinking, are demanding, and the new tests, designed to show if students have those skills, will prove challenging. The bulk of questions will still be multiple choice – they remain cheapest to administer and quickest to take, both practical advantages. But they will also include questions requiring short written answers and multiple-step problems, called performance tasks, that require students to apply their knowledge and to explain their thinking. They’re designed to be taken on computers, although districts without the equipment and capacity will be given paper-and-pencil tests for up to three years. Torlakson said Tuesday that he is confident most districts would be computer-ready, since the new tests could be given over a two-week period in a school computer lab. (Nonetheless, credibly comparing scores of districts using old and new technologies will be a major challenge.)
Tests to drive classroom instruction
Torlakson and Sigman, who’s a member of Smarter Balanced’s governing board, are true believers in these types of assessments, which they say will drive how teachers teach. That will be the big benefit, Torlakson said in an introductory letter: “The concept is simple and powerful: if our assessments require students to use problem solving and critical thinking skills to perform well, those same skills are more likely to be taught in our classrooms day in and day out. The goals we set for our assessment system have profound implications for our students and our schools.”
Torlakson is recommending that the Smarter Balanced tests serve as the model for the successors to the California Standards Tests in history and end-of-year tests in high school math and science that would be put on hold after this year. But, as he acknowledged, these assessments would be lengthier and costlier. They’d also come with state-designed interim tests and a library of practice performance tasks that would guide teachers’ instruction.
Suspending more than two dozen end-of-course and grade-level tests, plus tests in Spanish tied to current math and English language arts standards, will free up money that can be put toward developing new assessments or providing training for teachers in Common Core standards; the Legislature will decide how it will be used, said Assemblywoman Susan Bonilla (D-Concord), who’s chairing the Assembly Budget Subcommittee on Education Finance. It’s not clear how much that will be; all told, Sigman said, the state is paying $54 million annually to administer the STAR tests and $11 million for the high school exit exam.
One cheaper option
One criticism of the current state and federal accountability systems is that too much focus has been given to math and reading standardized tests, leading to a narrowing of the curriculum. So there’s also talk about giving more emphasis to history, science, technology or the arts, through tests or other measures.
One alternative to cut the time and expense of testing is to test a sampling of students or to administer sections of a test to different groups of students. This option could provide reliable district data, but not classroom or important subgroup data. Individual students would not get test results, and data from this method, called matrix sampling, could not be used for teacher evaluations.
Creating assessments with a goal of promoting high-quality instruction would mark a change in direction from the current system’s goal of strictly measuring the results of instruction. Doug McRae, a retired educational measurement specialist living in Monterey, thinks this shift is misguided.
“Based on many many years designing and developing large scale K-12 testing systems, I can say that the two purposes cited above are mutually incompatible,” he wrote in a critique of the report, which he said had “a strong anti-accountability theme.” He said it would be unwise to suspend the high school end-of-year tests, which have provided useful trend data and closely measure what students are taught.
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.
Lucia Rey 10 years ago10 years ago
When will the final decision be made as to whether or not CSTs will be delivered the 13- 14 school year and who is making that decision?
John Fensterwald 10 years ago10 years ago
The Legislature must decide this month which CSTs, beside those required by the federal government for accountability purposes under No Child Left Behind, to test next spring. Lawmakers will decide that through AB 484, sponsored by Assemblywoman Susan Bonilla. There are ongoing negotiations between legislative leaders and the governor’s office on which tests to suspend, which to continue and which to end. Stay tuned.
navigio 10 years ago10 years ago
I had a teacher tell me just yesterday that because they are still not sure whether the CST they take will be done away with they still have to teach X, where X is apparently something the CST tests for, but CC does not.
woe is me..
Cal 10 years ago10 years ago
So let's see. California is denied a NCLB waiver because of our refusal to use value added testing. Suddenly, Torlakson is all "Whoo hoo Common Core full speed ahead! Stop testing!" and of course, if we stop testing, how can the feds be mad us for not doing value-added testing if we're all Common Core whoo hoo because we can't be on board for common core assessments *and* be doing value-added testing, right? It's not … Read More
So let’s see. California is denied a NCLB waiver because of our refusal to use value added testing. Suddenly, Torlakson is all “Whoo hoo Common Core full speed ahead! Stop testing!” and of course, if we stop testing, how can the feds be mad us for not doing value-added testing if we’re all Common Core whoo hoo because we can’t be on board for common core assessments *and* be doing value-added testing, right? It’s not possible.
So why does anything see this as anything other than delaying tactic by Torlakson–a good one, mind you–to get us our NCLB waiver?
Never mind that a whole new suite of tests is obscenely expensive, and the funding will end six months before the due date. Never mind that the state doesn’t have the money. Who the heck believes they’ll be done on time?
I am stunned at how many people are taking this Torlakson maneuver seriously, rather than a brilliantly cynical ploy to use everyone’s pretense to take CC’s claims seriously to buy California some time.
Meanwhile, here’s Ze’ev worrying about 8th grade algebra again, because lord knows with all the idiocy going on the really important thing to worry about is whether or not functionally innumerate 8th graders are taking math they’ll never use and don’t understand.
Ze'ev Wurman 10 years ago10 years ago
Comparability seems low in priority for Torlakson, #11 of 12.
Specifically, he recommends to “conduct these studies, a representative sample of students across California would need to take both tests, at approximately the same time in the same school year.”
Will we find the money to do it for grades 2-11, in both ELA and math, a year or two after much of CST has been discontinued? Why am I doubtful?
John Fensterwald 10 years ago10 years ago
Ze'ev: Supt. Torlakson's report includes recommendations for comparability. I don't know how that will be done by ending them when the new Common Core assessments are implemented. Good point. Bea: The Smarter Balanced governing board has received push-back from states over the time issue, which is why the latest plan apparently is to reduce the number of constructed response questions and performance tasks. For those schools with only a language lab, the administration of the new … Read More
Ze’ev: Supt. Torlakson’s report includes recommendations for comparability. I don’t know how that will be done by ending them when the new Common Core assessments are implemented. Good point.
Bea: The Smarter Balanced governing board has received push-back from states over the time issue, which is why the latest plan apparently is to reduce the number of constructed response questions and performance tasks. For those schools with only a language lab, the administration of the new tests could be a problem solved immediately with pencil and paper, I suppose.
el 10 years ago10 years ago
As kind of a side note, one of the things that could happen in the middle of an online exam would be loss of internet connectivity or a browser crash. I hope whoever is building this program has figured out how to detect and handle these situations. A crawling internet can also happen. A friend of mine took an online course via the community college and had that happen during her (timed) final exam. She had … Read More
As kind of a side note, one of the things that could happen in the middle of an online exam would be loss of internet connectivity or a browser crash. I hope whoever is building this program has figured out how to detect and handle these situations. A crawling internet can also happen.
A friend of mine took an online course via the community college and had that happen during her (timed) final exam. She had enough points to pass with a decent grade, so no one had to ask or answer any hard questions.
Ze'ev Wurman 10 years ago10 years ago
Bea, Thanks for pointing out my lack of clarity on the EAP issue. Torlakson did suggest to keep Algebra II and the summative mathematics (and the 11th grade ELA) are "fully developed and implemented" whatever that "fully" may mean. But not necessarily until it is agreed upon by all stakeholders. First, it is unclear when -- or even whether -- the UC, CSU, and CCC will aall gree on the SBAC as sufficient for non-remedial admission. I … Read More
Thanks for pointing out my lack of clarity on the EAP issue. Torlakson did suggest to keep Algebra II and the summative mathematics (and the 11th grade ELA) are “fully developed and implemented” whatever that “fully” may mean. But not necessarily until it is agreed upon by all stakeholders.
First, it is unclear when — or even whether — the UC, CSU, and CCC will aall gree on the SBAC as sufficient for non-remedial admission. I have seen how tedious the process was with CSU, and how UC refused to join in for a decade. I have deep doubts whether either of them will be happy to switch to whatever will come via SBAC, although I am sure they will face enormous pressure to cave in.
Second, I don’t really know how the SBAC tests will be administered — only as an 11th grade test, as is implied, or as a high school test taken whenever the student is ready. Sounds trivial but it is not. For example, the majority of students pass EAP conditionally, i.e., needing another year of “high school math.” How will it work for students taking it in grade 10? In grade 12? Moreover, the pass rate for those who take Algebra 2 in grade 11 is only 7% unconditional and 22% conditional. It is *much* higher for those who take EAP with one course beyond Algebra 2 — 22% unconditional and 67% conditional. The SBAC 3rd year HS content is defined *below* current Algebra 2 and that is when those students are supposed to take the “11th grade” test.
Finally, how long do you think the current pipeline of students ready to take Algebra 2 or higher by grade 11 will continue if you break it in two places — Algebra 1 and Geometry?
A well thought out process would *continue* to maintain current CST for at least couple of years in parallel to phasing in the SBAC. This would allow SBAC calibration and maintain the usefulness of historical achievement data. What Torlakson is proposing is to make a clean break from the past (with even a bigger hole in HS) so comparison with old data would be impossible. This way nobody can track the changes in student achievement as compared to the past. Not too surprising, I may add.
Bea 10 years ago10 years ago
@ Ze'ev -- thought Torlakson said that the EAP would remain until/unless there is a new 11th grade assessment tied to the common core and agreed upon by all of the stakeholders. @John, it would be helpful to know how much time is forecasted to be required for the new adaptive assessments. If it's anywhere near the hours that current STAR tests require, it will take far longer for a school site to complete their assessments … Read More
@ Ze’ev — thought Torlakson said that the EAP would remain until/unless there is a new 11th grade assessment tied to the common core and agreed upon by all of the stakeholders.
@John, it would be helpful to know how much time is forecasted to be required for the new adaptive assessments. If it’s anywhere near the hours that current STAR tests require, it will take far longer for a school site to complete their assessments class by class in a single 30-station computer lab.
Ze'ev Wurman 10 years ago10 years ago
$54 mil for about 5 million tested students comes to $11/student. Dirt cheap in comparison to the upcoming "smarter" tests at somewhere between $30 and $50/student official estimates, and probably closer to $100/student once costs of actual scoring of open-ended items, and the cost of the supporting technology in schools, are fully accounted for. But, hey, we are a rich state after prop. 30 was passed. What's another half a billion for testing? But it's worth … Read More
$54 mil for about 5 million tested students comes to $11/student. Dirt cheap in comparison to the upcoming “smarter” tests at somewhere between $30 and $50/student official estimates, and probably closer to $100/student once costs of actual scoring of open-ended items, and the cost of the supporting technology in schools, are fully accounted for. But, hey, we are a rich state after prop. 30 was passed. What’s another half a billion for testing?
But it’s worth considering what we are giving up. With just a single test in the HS, and no second grade test:
– We are giving up on the ability to catch second graders with reading problems before they hit third grade. Data shows that after third grade the probability of fixing reading problems drops dramatically. (Yes, some school will still administer diagnostics. Many won’t.)
– We won’t offer any standard end-of-course test for high school math courses. Sure, we can trust all the schools to work diligently and blindly trust their scores. Right.
– The one thing that California is ahead of the nation is the Early Assessment Program that allows 11th graders to be certified as ready for CSU without remediation. Bye bye, EAP.
– And the big elephant in the room. Today we know that 2/3 of the cohort take Algebra 1 by grade 8. Tomorrow? How will we know how many take Algebra? We won’t. But Torlakson is happy for us not to know — it would show the abysmal drop in such students in the near future.
Hey, but “we will have less testing.” Wonderful.
Welcome to the new Lake Wobegon, where all our children are successful and above average. As long as they are untested.
John Fensterwald 10 years ago10 years ago
Good question. State is still surveying districts. CDE says that the computing and bandwidth requirements from Smarter-Balanced are not as difficult to meet than the state had expected.
As I mentioned, drawing comparisons between scores of districts that do computer adaptive testing and districts with paper and pencil will be very difficult. Even harder if schools within districts choose either option.
el 10 years ago10 years ago
Do all elementary schools have computer labs now, with enough bandwidth to administer the tests? I am curious if anyone has started surveying and quantifying that yet.