The State Board of Education suspended the Academic Performance Index for a second year on Wednesday, moving one step closer to the 15-year-old accountability system’s expected demise.

State board President Michael Kirst and other members have made it clear that they intend to replace the API, which calculates a three-digit number based primarily on a school’s or district’s standardized test scores, with a new system in which test scores would be just one of many measures of student achievement and school performance. The extra year will give the board time to figure out what that system should look like and to discuss statutory changes needed to make the transformation happen.

In adopting the Local Control Funding Formula, the Legislature established eight priorities that districts must address. The priorities, which will be core to a new accountability system, include school climate, student engagement, access to courses leading to college and careers and the implementation of new academic standards, such as the Common Core State Standards, as well as measures of student achievement.

By a unanimous vote, the board adopted the recommendations of an advisory committee, which had struggled with attempts to integrate non-test factors, like graduation rates, into the API. The  Public Schools Accountability Advisory Act committee decided instead to urge the board to abandon a single index and launch a new accountability system no earlier than fall 2016.

From a practical standpoint, the state board had little choice but to suspend the API, which the Legislature permitted it to do for two years. This spring, school districts are taking for the first time the Smarter Balanced tests of the Common Core standards in math and English language arts. It could be several years ­before Smarter Balanced scores can be used to judge school and district performance. The state board hasn’t set a timeline, nor has it set statewide target scores for the new tests.

Additionally, the Legislature ordered the state Department of Education to stop giving all other tests based on state standards, while the state creates a new generation of online tests that, like Smarter Balanced, measure more complex learning skills. It will be at least three or four years before the state produces a new social studies test and a science test aligned with the Next Generation Science Standards, which the state adopted two years ago.

Over the next year, the state board will decide whether the API should continue in a diminished form, with less clout, as one of many elements in a new accountability system, or disappear.

While the API presented a one-dimensional view of a school, it also had the virtue of simplicity. It offered parents a composite number on which to measure a school’s performance. By requiring the calculation of separate API numbers for student subgroups, it exposed gaps in achievement for low-income students, special education students and Hispanic and African-American children.

Several board members acknowledged that highlighting the achievement gap must not be lost in whatever system replaces the API.

“Whether we think it was the best way to look at a school, there has been progress identifying underperforming subgroups,” said board member Trish Williams. “I heard from families that once they understood the API, they could look at their neighborhood school and decide whether it was where they wanted their child to attend.”

Lupe Aragon, a parent leader from 20th Street Elementary School in Los Angeles Unified, reiterated that point, urging the board not to eliminate a system that is understandable.

Brian Rivas, director of policy and government relations for the advocacy group Education Trust-West, said the state board should be cautious about adding so many measures that a new system loses the focus on equity and “muddies the definition of what we consider a successful school.”

Stephen Blake, a senior adviser for another advocacy group, Children Now, endorsed the concept of multiple measures but said “stakeholder groups had expressed anxiety” over an indefinite period for suspending accountability measures.

Board member Patricia Rucker acknowledged this concern, saying while it is important to “hit the reset button” on a flawed API, the board should set a clear deadline for moving ahead with its replacement.

To get more reports like this one, click here to sign up for EdSource’s no-cost daily email on latest developments in education.

Share Article

Comments (19)

Leave a Reply to Paul Muench

Your email address will not be published. Required fields are marked * *

Comments Policy

We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.

  1. Tom Richardson 9 years ago9 years ago

    I totally agree with Don. The SBE is proposing to eliminate the decile ranking of schools. This should be strongly opposed, though those on the lower end will celebrate. It's amazing that a school can significantly improve API scores but not increase its decile ranking. In fact, some have seen their ranking lowered even when their scores rise. In a sense, that's good since it indicates that scores across the state are rising. At the … Read More

    I totally agree with Don.

    The SBE is proposing to eliminate the decile ranking of schools. This should be strongly opposed, though those on the lower end will celebrate.

    It’s amazing that a school can significantly improve API scores but not increase its decile ranking. In fact, some have seen their ranking lowered even when their scores rise. In a sense, that’s good since it indicates that scores across the state are rising.

    At the high school where I teach, we raised our Decile ranking from a 2 to a 4 in three years. That is very difficult to do statistically. We are an 80% poverty school with a very high minority population. Our district did not celebrate our improvement. Neither did the state. No TV announcements. I and others felt slighted for sure. Celebration should be built into the evaluation process state-wide … for both improvement in scores and improvement in Decile ranking. We need to insist upon it.

    This Decile ranking is very revealing when it is disaggregated into ethnic, economic, ELL and Special Education groups. The results should be revealed to the public and guide a school district in allocating funds and resources. These results are available for the public view on the web. I am curious about Don’s comment that SFUSD hide this disaggregation from the public It’s published public information.

    We need to insist that the Decile Ranking remains in the system. It is the most revealing component.

  2. Don 9 years ago9 years ago

    The API led to data fraud. For this reason the dashboard approach is much better than the single indicator and the folly of the San Francisco Unified API over the years illustrates why. Until 2012 and 2013, SFUSD annually celebrated its stature as the highest performing urban school district, throwing central office parties coordinated with media praise. And why not celebrate success? The answer is that there was nothing to celebrate when the single 3-digit … Read More

    The API led to data fraud. For this reason the dashboard approach is much better than the single indicator and the folly of the San Francisco Unified API over the years illustrates why.

    Until 2012 and 2013, SFUSD annually celebrated its stature as the highest performing urban school district, throwing central office parties coordinated with media praise. And why not celebrate success? The answer is that there was nothing to celebrate when the single 3-digit indicator was disaggregated by subgroups. On closer inspection, non-Asian minorities dramatically underperformed statewide peers. Among the four major ethnic groups, African Americans, Hispanics and Asians underperformed the same statewide ethnic groups by -78, -48 and -32 points, respectively. Only SFUSD whites outperformed their statewide peers. It was only for the sheer size of the generally higher performing Asian student population in SFUSD that API was high.

    This was a fact that district insiders were obviously well aware while simultaneously touting their success to the public. Ridding the system of the gross oversimplification that is the API and taking a broader approach would help to prevent such abuse of the data that allow school leaders to celebrate failure disguised as success and hoodwink an unsuspecting public.

    Replies

    • FloydThursby1941 9 years ago9 years ago

      They should publish the academic API in addition to the feel good measure so informed members of the public can interpret the data as they will.

  3. Tom Richardson, Ed.D. 9 years ago9 years ago

    I have a Ed.D. Degree in Performance Evaluation and Improvement from USC. In that, we students learned much about the evaluation issues that plaque any business or organization. The well-known Dr. Richard Clark was the dean of that program 15 years ago. I don't know where he is now ... but he is the man for help you through this process. The API measures actual academic performance. Such a measure should … Read More

    I have a Ed.D. Degree in Performance Evaluation and Improvement from USC. In that, we students learned much about the evaluation issues that plaque any business or organization. The well-known Dr. Richard Clark was the dean of that program 15 years ago. I don’t know where he is now … but he is the man for help you through this process.

    The API measures actual academic performance. Such a measure should always be the “main score.” The problem was that the tests were inadequate and focused upon lower learning skills. Such a score is comparable to profit, market-share, targets achieved.

    Other measures such as students’ economic status, attendance, suspensions and expulsions; the availability and use of curricular resources, teacher education, administrator education, etc.; and parent involvement are good predictors of performance, with occasionally some odd excepts. They however do not measure performance. They are demographic and process variables.

    It seems to me that the people who are doing the thinking for the SBE are not fully in the know of how performance, process and demographics different in nature and should be used differently. They all should be reported.

    The CBE should seek out performance experts who work in fields very different than public education. If they do not, they will not excel in the measures that they will use.

  4. Doug McRae 9 years ago9 years ago

    Suspending the API for 2014-15 is not fresh news at all. When the SBE approved suspending API for 2013-14 for Elem and Mid schools) more than a year ago, de facto it suspended API for 2014-15 since there was no Base API data from 2014 upon which to calculate a 2014-15 API. So, the suspension action by the SBE two days ago was really just a ministerial action confirming action de facto taken some time … Read More

    Suspending the API for 2014-15 is not fresh news at all. When the SBE approved suspending API for 2013-14 for Elem and Mid schools) more than a year ago, de facto it suspended API for 2014-15 since there was no Base API data from 2014 upon which to calculate a 2014-15 API. So, the suspension action by the SBE two days ago was really just a ministerial action confirming action de facto taken some time ago. I see the action this time was pretty generously covered by the print media; I guess the general public doesn’t follow the saga of the API closely [grin, I don’t know where the emoticons are on this machine].

    More pertinent is the issue of the 2015-16 API. The recommendation from the PSAA approved by the SBE two days ago was very ambiguous on this issue, only saying “new accountability system no early than fall of 2016.” It didn’t say whether the new system might retroactively use 2015 data [SBAC and other] to produce a 2015-16 base-to-growth system whether is it called an “API” or not. ACSA (supported by CSBA and others) has asked the SBE not to use 2015 SBAC scores for a 2015 Base API (or its successor, I assume). Due to the lack of validity for 2015 SBAC scores in general, ACSA is on strong grounds in making this request. But the CDE and PSAA, and now the SBE, has danced around and/or ignored the request. No one knows how the 2015 SBAC scores might be used, due to the waffling or stonewalling thus far.

    Students and parents absolutely deserve to know how test results are going to be used before they take tests. Students and parents have justification to generate major pushback if teachers or schools do not reveal how test results will be used in advance. That’s just fundamental common sense. Going up the food chain, schools and districts absolutely deserve to know how statewide test results will be used, especially for accountability purposes, before the tests are administered. This requirement is even built into current statutory language for accountability data. Not to have a strong communication from education leadership in CA on this issue is failure to recognize what the trenches deserve to know before statewide testing begins, and as of this week the statewide testing in CA has begun.

  5. Andrew 9 years ago9 years ago

    To embrace the new anticipated multiple measures, I've decided to open a new charter school, Hard Rock Charter. For students, I am going to gather a few of the rocks that I see out my window in my yard, and I will put the rocks in classrooms and call them students and will call what happens in the classrooms "education." Since attendance will be perfect, I will win … Read More

    To embrace the new anticipated multiple measures, I’ve decided to open a new charter school, Hard Rock Charter. For students, I am going to gather a few of the rocks that I see out my window in my yard, and I will put the rocks in classrooms and call them students and will call what happens in the classrooms “education.” Since attendance will be perfect, I will win on that measure. Expulsions will be non-existent, so I get another win. I have decided to declare the classrooms occupied by the rocks as AP and college-ready, so I win on that measure as well. I am declaring all the senior class of rocks to have graduated each year, so graduation is 100% and I win on that measure. The rocks have wonderful emotional stability indicative of the best school culture, so my winning streak continues on the basis of culture. I am on the way to having the highest rated school in California. If only I can avoid or do away with standardized testing, and I may well luck out in that regard.

    Replies

    • Manuel 9 years ago9 years ago

      Andrew, have you decided how you are going to produce the student records that CALPADS requires for your school?

      You could be audited and might end up, well, between a rock and a hard place…

      • Andrew 9 years ago9 years ago

        My new charter school will have strong STEM emphasis and achievement, Manuel. All the students will get straight A+'s in geology. Maybe you would consider being a co-founder? Are you good at rationalizing? I think we better make it an online charter school. That will save us a bundle on facilities and associated costs. Since we'll be pulling in full ADA or something pretty close … Read More

        My new charter school will have strong STEM emphasis and achievement, Manuel. All the students will get straight A+’s in geology. Maybe you would consider being a co-founder? Are you good at rationalizing? I think we better make it an online charter school. That will save us a bundle on facilities and associated costs. Since we’ll be pulling in full ADA or something pretty close to it for each student, and with negligible facilities cost, and since we’ll be stiffing teachers on salaries and benefits, we will have to decide how to get all the extra money into our own pockets. We’ll need to allocate some money for freeway advertising billboards replete with photos of beaming photogenic students and a link to our website. But that leaves a lot of money for us, and we will have to decide how to get it to ourselves. There are several options. We can create and own a for-profit CMO that supplies our non-profit school’s “magic” curriculum and other services, and we can have the non-profit school pay our CMO a bundle to provide those things. Or we can have the non-elected self-appointed board of the non-profit school pay us handsome salaries and bonuses. Our school will probably eventually be shut down, but given the glacial pace and gentleness of such interdiction, we’ll be long and comfortably retired.

        • Manuel 9 years ago9 years ago

          Andrew, you seem to know exactly how to run a charter to maximize profits. Are you sure you are not in the racket already? OTOH, they caught up with Chavis, didn't they? But you are right, they can't get to them all so... by Jove, I got it! We just simply close the school, change its name, and re-charter it in some other district! That's the ticket. You are a genius and can finally quit all … Read More

          Andrew, you seem to know exactly how to run a charter to maximize profits. Are you sure you are not in the racket already?

          OTOH, they caught up with Chavis, didn’t they? But you are right, they can’t get to them all so… by Jove, I got it! We just simply close the school, change its name, and re-charter it in some other district!

          That’s the ticket. You are a genius and can finally quit all that commuting to “the city.”

  6. Andrew 9 years ago9 years ago

    It is important to determine the new multiple measures and publish them quickly so that those so inclined, including certain for-profit charter management organizations, can develop their plans to game and con the new system. Some of the measures proposed sound ripe for gaming.

  7. Don 9 years ago9 years ago

    John, can you tell me if the SARC will still be around and whether SBAC will be reporting grade level mean scores like with STAR? Regarding the article, I think I was thrown off by the end your first sentence - "moving one step closer to the 15-year-old accountability system’s expected demise." By that I thought you meant no API replacement since the API is effectively a done deal. And you backed … Read More

    John, can you tell me if the SARC will still be around and whether SBAC will be reporting grade level mean scores like with STAR?

    Regarding the article, I think I was thrown off by the end your first sentence – “moving one step closer to the 15-year-old accountability system’s expected demise.” By that I thought you meant no API replacement since the API is effectively a done deal. And you backed up that interpretation when you said “whether the API ( in another form) should continue… or disappear”.

    It seems to me, with the clock ticking, delay beyond two years more problematic and Smarter Balanced results likely to significantly underwhelm, the decision revolves around whether no measure would be more politically palatable than one with diminished emphasis on test results and more emphasis on other already controversial factors like attendance, school climate or suspension rates. At least that seems like the obvious enigma facing the SBE.

    Coming from a parent’s perspective I doubt many will put much stock in such non-academic factors. For better or worse, the old index was used to gauge a school’s student body achievement for the purpose of school choice. Short of that, I suspect most parents will find a kinder gentler dashboard of multiple measures as insightful as the nutritional breakdown of a Big Mac.

  8. Don 9 years ago9 years ago

    John, I'm not sure whether your article excelled in describing the API issue in the vein of -"if you're not confused you're not paying attention"- or whether your article missed the clarification mark. Your title implies only a delay, but your first sentence says the API's demise is expected. You spend the rest of the article talking about the API development and delay with only a brief reference back to … Read More

    John, I’m not sure whether your article excelled in describing the API issue in the vein of -“if you’re not confused you’re not paying attention”- or whether your article missed the clarification mark. Your title implies only a delay, but your first sentence says the API’s demise is expected. You spend the rest of the article talking about the API development and delay with only a brief reference back to your initial contention which was – ” Over the next year, the state board will decide whether the API should continue in a diminished form, with less clout, as one of many elements in a new accountability system, or disappear.” Also, when you say, “in a diminished form” you are implying that a new more multidimensional API is inherently an API of lesser value and import.

    I understand there’s no clarity on what will happen over the next year. This article makes that crystal clear.

    Replies

    • John Fensterwald 9 years ago9 years ago

      Let me try again without confusing you more. Don. All the statements and signals from the board and its advisers point toward downplaying standardized test scores as the sole or even primary element in a new accountability system. I don't think anyone knows -- including board members -- what the new system will look like, but board members will have plenty of metrics from which to choose and many aren't related to testing: A-G course … Read More

      Let me try again without confusing you more. Don.

      All the statements and signals from the board and its advisers point toward downplaying standardized test scores as the sole or even primary element in a new accountability system. I don’t think anyone knows — including board members — what the new system will look like, but board members will have plenty of metrics from which to choose and many aren’t related to testing: A-G course completion, suspension rates, grad rates, English learner reclassification rates, maybe even measurements of perseverance and determination, if the metrics even get sophisticated enough. There are many ways to measure whether students leave school ready for whatever they choose to do in life. And they can’t credibly take the form of a single grade — even if some states insist that it can.

      The Public Schools Accountability Act advisory committee tried to include some non-test measures, like grad rates, into the API, but ended up punting and recommending instead the switch to some other system than a 3-digit number that sums everything up.

      That is not to say there will not continue to be some form of an API — an index that weights results of various standardized tests scores, including history, science and a new test for English learners, and spits out one three-digit number. But if it continues to exist — that’s anyone’s guess — it will be just one piece of a larger system. And that system won’t be called API.

      Meanwhile, parents and schools will get scores on the Smarter Balanced results this spring, and they should hold their districts for making progress on the specific goals that they set in their yearly LCAPs. That’s accountability, though, for now, it’s locally based.

      Hope that’s clearer.

      • Manuel 9 years ago9 years ago

        John, since LAUSD asked the SBoE not to count the scores from this Spring’s SBAC, I suspect that there will be no meeting the LCAP in our neck of the woods.

        Then what?

  9. navigio 9 years ago9 years ago

    It is arguable that our supposed 'objective' goals are not so much that anyway. And since it is only alignment with those goals where the 'new' numbers don't provide enough information, they are more than sufficient for our traditional 'stack and rank' uses. Waiting is not going to repair any flaws in test design and making the API more complex is not going to make it more simple. My guess is the SBE is simply waiting … Read More

    It is arguable that our supposed ‘objective’ goals are not so much that anyway. And since it is only alignment with those goals where the ‘new’ numbers don’t provide enough information, they are more than sufficient for our traditional ‘stack and rank’ uses.
    Waiting is not going to repair any flaws in test design and making the API more complex is not going to make it more simple.
    My guess is the SBE is simply waiting because they have no clue what is going to happen with the tests (in terms of logistics, modifications or reactions). Knowing more will allow them to ‘manage expectations’, or put another way, manipulate results.
    This seems to be made clear by some SBE members lauding the previous API as a helpful and valid gauge for parents, while others claiming it needed to be done away with.
    And not all the numbers used in the former API were publicly available. Perhaps the biggest question right now should be whether the intent is to continue to hide some of the data.

    Replies

    • Paul Muench 9 years ago9 years ago

      What aspects were not publicly available?

      • navigio 9 years ago9 years ago

        School-level CST results for subgroups are not reported at a performance band granularity, rather only proficiency rates. That is not enough information to calculate API values for a school's subgroups or other potentially interesting cross-sections of students. This means gap analysis based on API isn't possible using public data at a school level (it still is for gaps based on proficiency rates at a school level tho). Some districts make that information available but it … Read More

        School-level CST results for subgroups are not reported at a performance band granularity, rather only proficiency rates. That is not enough information to calculate API values for a school’s subgroups or other potentially interesting cross-sections of students. This means gap analysis based on API isn’t possible using public data at a school level (it still is for gaps based on proficiency rates at a school level tho). Some districts make that information available but it is not part of what is published by the state.
        I expect as the number of measures is increased, the more scenarios will arise where concern about publishing data will exist.
        This is not to say that not publishing literally everything is wrong, rather we shouldn’t always assume we can just figure it out on our own.

        • Don 9 years ago9 years ago

          You haven't explained why in your opinion the CST mean scale score cannot be used to accurately calculate an API. I understand you believe the overall quality of test information gleaned is inadequate, (though I doubt there would ever be consensus on adequacy of testing), but relative scores are still accurate, if you agree with the way API is adjusted. Otherwise one can use STAR results at grade level. In … Read More

          You haven’t explained why in your opinion the CST mean scale score cannot be used to accurately calculate an API. I understand you believe the overall quality of test information gleaned is inadequate, (though I doubt there would ever be consensus on adequacy of testing), but relative scores are still accurate, if you agree with the way API is adjusted. Otherwise one can use STAR results at grade level. In any case, what kind of further breakdown or granularity would you like reported? – Info at the content level/instructional level like a 3rd grader’s understanding of multiplication?

          Though only partially related, it should be noted that tens of thousands or even perhaps hundreds of thousands of subgroup scores are not reported when they don’t meet the minimum number of 10 for the purpose of confidentiality, but they were calculated nevertheless. OTOH, non-participation rates certainly did impede overall accuracy and that’s rarely broached and is not part of your argument against the validity of achievement gap analysis.

  10. Paul Muench 9 years ago9 years ago

    "Several board members acknowledged that highlighting the achievement gap must not be lost in whatever system replaces the API." Exactly! The federal accountability system still requires that goal be met. At least to begin with these changes are all about managing public impressions of schools. Best I can tell all the data that might be used is already available, it's just that the school system and public have a history of focusing on … Read More

    “Several board members acknowledged that highlighting the achievement gap must not be lost in whatever system replaces the API.” Exactly! The federal accountability system still requires that goal be met.

    At least to begin with these changes are all about managing public impressions of schools. Best I can tell all the data that might be used is already available, it’s just that the school system and public have a history of focusing on one number that is a proxy for test scores. But once numbers exist they can be used to rally political support and make public decisions. We’ll find out what that leads to.