
Members of the State Board of Education who favor replacing the three-digit Academic Performance Index with a “dashboard” of measurements highlighting school performance can count on the backing of Gov. Jerry Brown.
The K-12 summary (pages 22-23) of Brown’s proposed 2016-17 state budget, released last week, stated, “The state system should include a concise set of performance measures, rather than a single index.” Brown said the new federal Every Student Succeeds Act creates the opportunity to design a “more accurate picture of school performance and progress” than in the past.
But whether the state should or even can switch, under the new federal law, from a single index like the API to a more complex school improvement system will be a potentially contentious issue this year. Both approaches to accountability, the dashboard with multiple measures – such as test scores, high school graduation rates and an indicator of student readiness for college and jobs – and a single index compiled from a mix of factors, have strong advocates.
The state board is in the process of drafting a new school accountability system and will continue the debate at its meeting on Wednesday (see agenda, Items 1 and 2). A state consultant, attorney Julia Martin, will outline how the state board may have to tailor its concept of school improvement to meet the requirements of the new federal law, which Congress passed and President Barack Obama signed last month as the successor to the No Child Left Behind Act.
The API rates schools and districts on a scale of 200 to 1,000, based on standardized test results. The Legislature established it in 1999. But with the passage of the Local Control Funding Formula and the transition to the Common Core standards in English language arts and math, accompanied by a new set of state tests, the state board suspended calculating the API two years ago. While the board hasn’t yet voted of get rid of it, members, led by President Michael Kirst, appear inclined to.
The Local Control Funding Formula, the school funding and accountability law that Brown proposed and the Legislature adopted in 2013, calls for a broader measure of school performance than the API, at least as currently constructed using test results. The funding law set eight priorities for school districts. They include student achievement, implementation of academic standards like the Common Core, parent involvement, student engagement and preparation for college and careers. The API is one of two dozen metrics that the law says districts must use as yardsticks of improvement. Test scores would continue to be a key measure in a new accountability system but not, Brown suggests, through the API.
Which metrics to use?
The state board’s challenge is to decide which of the many metrics in the funding law should be used in a statewide accountability system that will determine the schools and districts that need targeted assistance and the chronically poorly performing schools that require intervention. Examples of limited assistance might be helping schools better solicit parents’ ideas or reducing suspension rates for a subgroup of students.
Kirst, board member Sue Burr and the state board’s executive director, Karen Stapf Walters, advise Brown on education issues, so it’s not surprising that the governor would endorse the board’s basic approach to school improvement. But Kirst said Brown’s statement in the budget, recommending a “concise” number of metrics, provides new guidance to the board. And the governor has made clear his view of the API, he said.
Kirst argues that boiling down achievement into an API number diverts attention from the key components – graduation and attendance rates or percentage of students admitted to college without need for remediation – that might comprise the future dashboard. But others counter that the API provides parents with a simple comparison of schools’ performance that they can use when deciding where to enroll their children.
A school list without an index?
How the new federal law is interpreted could decide the issue. The law is generally in sync with California’s approach to school accountability in requiring that states adopt multiple measures of performance: high school graduation rates, measures of English learners’ proficiency in English, test scores on standardized tests in math, English language arts and science, plus one other non-test measure of student achievement that states can choose.
But the federal law also requires that every three years states identify the lowest-performing 5 percent of schools – about 500 schools in California – and monitor plans for improvement. The Every Student Succeeds Act doesn’t mandate using an index to rank the worst schools, but California will find it challenging to create a list without an index, said Christopher Cross, an education consultant from Danville who served as an assistant U.S. secretary of education for educational research and improvement during the administration of President George H.W. Bush. “It will be a tough road for California with regard to the 5 percent, and I don’t know how they can do it (without creating an index) in a way that is acceptable to the federal government and parents, too,” he said. “The creation of an index can factor in other measures, but in the final analysis it needs to lead to the identity of those schools that require extra supports.”
Brown has criticized judging a school on test scores alone. In 2011, he vetoed a bill that would have expanded the API to include multiple indicators. Calling it a “siren song of reform,” he wrote in his veto message, “Adding more speedometers to a broken car won’t turn it into a high-performance machine.” A year later, he did sign SB 1458, a simplified version of the bill that calls for expanding the API to include factors other than test scores for at least 40 percent of a school’s score.
The latest statement reflects a further change in Brown’s thinking. And he said in the budget summary that his administration would support removing “outdated accountability components.”
But the API is interwoven in dozens of laws on the books, like SB 1458, and programs whose supporters will be wary of changes. An example is the Open Enrollment Act, a 5-year-old underutilized law that requires the state annually to list the lowest-achieving 1,000 schools based on API scores. The law allows parents of children in those schools to transfer to better-performing schools in other districts that make spots available; few districts, however, have done so. Threatened with a lawsuit, the state Department of Education last month reluctantly reposted the list of schools from 2012-13, the last year of the API, with the caveat that parents should be wary of using an old list.
That is not a solution, said Bill Lucia, president of the education advocacy group EdVoice. The Open Enrollment Act remains the law and the school list must be updated this year to avoid litigation, he said last week.
Others, too, with an interest in the API will be closely watching the transition to a new school improvement system, and the ramifications of changing or abandoning the index.
To get more reports like this one, click here to sign up for EdSource’s no-cost daily email on latest developments in education.
Comments (18)
Comments Policy
We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.
Nina Bishop 7 years ago7 years ago
API or dashboard; why on earth are we STILL giving the annual test? Is this supposed to be improvement? Silly me, I think it's a waste of taxpayer money to test every single kid, in every single grade, in every single school, in the whole United States, every single year. Why not reasonable grade span testing like the NAEP? We've had it since the Johnson administration, it's not reinventing the wheel nor will it costs … Read More
API or dashboard; why on earth are we STILL giving the annual test? Is this supposed to be improvement? Silly me, I think it’s a waste of taxpayer money to test every single kid, in every single grade, in every single school, in the whole United States, every single year. Why not reasonable grade span testing like the NAEP? We’ve had it since the Johnson administration, it’s not reinventing the wheel nor will it costs us since we already have it and it gauges progress. That said, it would mean fewer tests and that would impact profits of testing and curriculum companies. Geez, what a conundrum.
Replies
John Fensterwald 7 years ago7 years ago
Nina: There is grade-span testing in science and Smarter Balanced testing in only one year in high school: 11th grade, along with annual testing in grades 3-8. In March, Superintendent of Public Instruction Tom Torlakson will present further recommendations on standardized testing. It will be interesting to see what he recommends.
Todd Maddison 7 years ago7 years ago
Love all the angst over designing methods of determining how well our kids are being educated, usually combined with "we need to make sure our kids are college and career ready". Meanwhile, the one standardized test used universally to determine whether kids are college ready - the SAT - continues to show declines every year. But we ignore that because it's inconvenient. No, we need more ways of measuring school performance, preferably ones that show everything getting better … Read More
Love all the angst over designing methods of determining how well our kids are being educated, usually combined with “we need to make sure our kids are college and career ready”.
Meanwhile, the one standardized test used universally to determine whether kids are college ready – the SAT – continues to show declines every year.
But we ignore that because it’s inconvenient.
No, we need more ways of measuring school performance, preferably ones that show everything getting better despite the drop in SAT scores….
CarolineSF 7 years ago7 years ago
Jamie has it right.
Test scores should be used for nothing but determining how the student is doing so as to offer support as needed. Nothing else. No judging schools, teachers, blaming, shaming, rewarding, remunerating — nothing else. By now haven’t we learned that the latter motivates all kinds of dishonest behavior that hurts schools and students rather than helping them?
Replies
Paul Muench 7 years ago7 years ago
We’re doing basically the same with LCFF. Which seems just as reliable as using test scores.
Greg Austin 7 years ago7 years ago
John, one challenge is that any index for statewide accountability has to be based on common measures, which should include school climate. But beyond requiring expulsion and suspension rates, the LCAP legislation leaves the indicators for school climate very much up to the district (i.e., “School climate, as measured by all of the following, as applicable… (C) Other local measures, including surveys of pupils, parents, and teachers on the sense of safety and school connectedness.”).
Replies
John Fensterwald 7 years ago7 years ago
You’re right, Greg. Any measures that can be used for state and federal accountability must be comparable across schools and districts. That’s not to say that local goals and indicators, like surveys and parent engagement goals, shouldn’t be a priority for a school and district via the LCAP. Chronic absenteeism and suspension rates are indicators of school engagement and culture — evidence of issues to examine and perhaps focus on.
CarolineSF 7 years ago7 years ago
Don’t chronic absenteeism and suspension rates also correlate with poverty? Is this another case of blaming a school that’s overwhelmed by the needs of its high-poverty student population?
John Fensterwald 7 years ago7 years ago
Usually so, but rates vary among schools with similar demographics. There may be things to learn from them. That’s what data can reveal.
Jamie 7 years ago7 years ago
Why don’t we just rate schools based on the home prices in the surrounding area? Basically the same thing.
Replies
woodstocksez 7 years ago7 years ago
If you think the correlation is meaningful (and no doubt plenty of people do), you are already free to do that. However, what you've really done is rated/ranked student outcomes and the correlation you posit arises because of demographics far more than quality of schools. (One of the teachers at my sons' elementary school lives in a high-income area with "good" schools. However, she laments the lazy/poor teaching she has encountered there. … Read More
If you think the correlation is meaningful (and no doubt plenty of people do), you are already free to do that. However, what you’ve really done is rated/ranked student outcomes and the correlation you posit arises because of demographics far more than quality of schools. (One of the teachers at my sons’ elementary school lives in a high-income area with “good” schools. However, she laments the lazy/poor teaching she has encountered there. She’d likely tell you her local schools are ranked highly – very high 900 API scores – because of the home environments from which the students come, rather than any magic worked by the schools.) It would be interesting to devise a system that measures school quality largely independent of other factors. I think that’d be a very hard thing to do; perhaps it’s not feasible. But it is important to be clear about what any of the current school rankings/ratings are really able to measure. Almost all of the debate on this subject ignores or glosses over the difficulty in teasing out the distinct effects associated with the various influences on student performance.
Jerry Heverly 7 years ago7 years ago
{Pretend}I'm looking for a new home. A big consideration is the schools--are they good enough to get my kid into Harvard {they all want their kids to go to Harvard}. The state gives me a chart of several measurements, not one API-like number. What's to prevent some enterprising realtor from designing a simple algorithm that churns the various state categories into......(ta da) one number? This realtor then circulates her algorithm via some kind of real … Read More
{Pretend}I’m looking for a new home. A big consideration is the schools–are they good enough to get my kid into Harvard {they all want their kids to go to Harvard}. The state gives me a chart of several measurements, not one API-like number. What’s to prevent some enterprising realtor from designing a simple algorithm that churns the various state categories into……(ta da) one number? This realtor then circulates her algorithm via some kind of real estate publication. In a trice we have the new API.
Replies
woodstocksez 7 years ago7 years ago
Sure, that can be done. And, so what? Withholding the institutional imprimatur still stands and that carries value. Other entities have always had, and often exercised, prerogative to interpret things as they like. Ultimately, no one can force anyone else to stop being simple-minded.
Ed 7 years ago7 years ago
@ Don: The continued gyrations about how to measure schools is more about how to construct a measurement that can't be used to distinguish between the quality of teachers. If you keep mixing things up and making it sound hard and confusing you can, as you identify, ensure the status quo. Don't ever forget that most of the agonizing by adults who are continuously trying to build a better education "moustrap" is about … Read More
@ Don: The continued gyrations about how to measure schools is more about how to construct a measurement that can’t be used to distinguish between the quality of teachers. If you keep mixing things up and making it sound hard and confusing you can, as you identify, ensure the status quo. Don’t ever forget that most of the agonizing by adults who are continuously trying to build a better education “moustrap” is about how to preserve comfort for the adults.
Replies
woodstocksez 7 years ago7 years ago
Or, it could be an attempt to be more sophisticated about discussing something that does not lend itself to easy encapsulation in a metric or two. Right?
Todd Maddison 7 years ago7 years ago
No, it's pretty much about making it more difficult to identify poor performers. I've designed compensation plans for a living - for divisions of billion-dollar companies in the past. The absolute first rule of that is to design something that is as objective as possible - no gray area that is subjective, or subject to interpretation. Because people whose pay depends on it will ALWAYS interpret things in their favor, right? I love the quote … Read More
No, it’s pretty much about making it more difficult to identify poor performers.
I’ve designed compensation plans for a living – for divisions of billion-dollar companies in the past. The absolute first rule of that is to design something that is as objective as possible – no gray area that is subjective, or subject to interpretation.
Because people whose pay depends on it will ALWAYS interpret things in their favor, right?
I love the quote ““Adding more speedometers to a broken car won’t turn it into a high-performance machine.”
No, it won’t, but it will quickly tell you when you DO NOT HAVE a high performing machine…
woodstocksez 7 years ago7 years ago
It's not hard to design a system of evaluation, used to determine compensation or anything else, based on "objective" factors. That doesn't mean the system performs well, though. In fact, it may be that no system of evaluation based solely on "objective" factors can do the job well or even adequately. Your automobile analogy is inapt. For instance, measures of student achievement, such as API, are, at most, a proxy for measuring quality … Read More
It’s not hard to design a system of evaluation, used to determine compensation or anything else, based on “objective” factors. That doesn’t mean the system performs well, though. In fact, it may be that no system of evaluation based solely on “objective” factors can do the job well or even adequately.
Your automobile analogy is inapt. For instance, measures of student achievement, such as API, are, at most, a proxy for measuring quality of education provided. Other factors influence that achievement, some of them more important than anything a school does or doesn’t do.
Don 7 years ago7 years ago
Our state education system can continue to devise new school accountability measures, but local school districts and their teachers and students will continue to do much the same as they have always done and in 5, 20, 10 years from now student achievement will be very much the same as it has always been.The only difference will be that we will have devoted a great deal of time and money to figure out new ways to measure what never changes.