California’s public schools continued to show gains on the Academic Performance Index (API), a measure of how well students do on the California Standards Tests and, in high school, on the exit exam. For the first time since the testing program began in 1999, a majority of schools reached or exceeded the state’s target of 800 on the index.

Results of the spring 2012 exams, released yesterday by the State Department of Education, showed that 53 percent of schools met the mark, 4 percentage points above last year.

State Superintendent of Public Instruction Tom Torlakson said the growth is especially noteworthy given the years of budget cuts that schools have endured. “The incredible efforts of teachers, administrators, school employees, parents, and students should serve as an inspiration to us all. While there’s still more work to do, California’s schools have earned a vote of confidence,” said Torlakson in a written statement accompanying the announcement.

Percentage of schools meeting or exceeding 800 API by grade levels. Source: California Dept. of Education. (Click to enlarge)

A little disaggregation, however, calls to mind former President Bill Clinton’s rallying cry at the recent Democratic National Convention: “It’s arithmetic!” Break down the numbers and it’s basically elementary schools that are responsible for pushing the average over the halfway point. Fifty-nine percent of elementary schools met or exceeded an 800 API, followed by 49 percent of middle schools. High schools lagged considerably, with just 30 percent meeting or going beyond the target.

There’s no clear-cut answer for why this is so. Some studies have found that high schools are failing to engage students, others say students are bored. Sherry Griffith, a legislative advocate for the Association of California School Administrators (ACSA), suggests there’s no incentive to do well because the tests have no bearing on a student’s grades, graduation, or college applications. “We believe high schools are challenged by the fact that students don’t care about the California Standards Tests or the results,” said Griffith. “It counts for nothing for them.”

That’s why ACSA sponsored AB 2001 in the just concluded legislative session. The bill, by Democratic Assemblymember Susan Bonilla of Concord, would have

Gaps in achieving API target of 800 by racial and ethnic subgroups. Source: California Dept. of Education. (Click to enlarge)

required the State Department of Education to expand California’s assessment program through grade 12 (currently students in second through eleventh grades take the exams) and to develop a plan for making the tests more meaningful to students. The bill died in the Senate appropriations committee, but ACSA may reintroduce it.

Stanley Rabinowitz, the director of Assessment & Standards Development Services at WestEd, said how much effort high school students put into the California Standards Test is often up to the values of school leaders. “I’ve seen schools work with their students to say, ‘we’re all in this together, the content matters for you individually and the results matter for the school. You need to take pride in what you’re doing and you need to take pride in your school,’” said Rabinowitz. “It works very well, but only if it’s a joint decision between the administration working with the teachers and the students.”

The buzz saw of No Child Left Behind

Notwithstanding the positive news on API scores, a record seven out of 10 California schools receiving federal Title I aid now fall under sanctions under the No Child Left Behind Law.

Parents wouldn’t know that from looking at the School Quality Snapshot, a snazzy two-page graphic summary of every school’s performance that the Department of Education released yesterday. It includes standardized test results, API scores, records on student suspensions and expulsions, even physical fitness test results (go here to search for your favorite school’s report). What’s not included is whether the school is in Program Improvement, the formal name for the much disliked NCLB penalties.

Torlakson made it clear the state is downplaying Program Improvement while it waits to hear from the U.S. Department of Education on whether the state’s request for a waiver from NCLB has been approved.

“California’s request for a waiver from the requirements of NCLB is still pending,” Torlakson said in a press release. “While we’re waiting for the flexibility we need, we’re not going to allow a flawed system to distract us from the work we’re doing to help schools improve.”

Thirty-three states have received waivers from NCLB releasing them from the restrictions of  Program Improvement, which, at a minimum, requires schools that don’t make academic targets two consecutive years to commit 20 percent of their Title I dollars to offer after-school tutoring and to transport students to a higher-scoring school if their parents choose to transfer. Schools that have continued to miss their targets and remain in Program Improvement for four or more years are required to submit a plan for significant school improvement, although there is no longer any federal funding for it and not a lot of enforcement to see if the plans really make a difference. About a third – 2,079 of 6,209 Title I schools in California – have now been in Program Improvement for four or more years, and most have little chance of exiting it. That’s because the requirements for meeting academic targets have been getting harder every year; most educators agree the requirements are implausible.

In the fast march toward 100 percent student proficiency two years from now, 78 percent of middle and elementary school students were supposed to be proficient in English language arts and 79 percent proficient in math to meet this year's target called Adequate Yearly Progress. Source: State Adequate Yearly Progress Report Information Guide (click to enlarge).

In the fast march toward 100 percent student proficiency two years from now, 78 percent of middle and elementary school students were supposed to be proficient in English language arts and 79 percent proficient in math to meet this year’s target called Adequate Yearly Progress. Source: State Adequate Yearly Progress Report Information Guide (click to enlarge).

Passed by Congress a decade ago, NCLB requires 100 percent proficiency by 2014, for every subgroup of students in every school, including students with disabilities, English learners, low-income students, and significant racial and ethnic groups. The bar has been raised every year; for the current year, between 78 and 80 percent of students must be proficient on CSTs in math and English language arts. Only 26 percent of all 9,905 schools and 18 percent of Title I schools made those targets. Most of those were wealthier schools with fewer subgroups.

“As the Annual Measurable Objectives (yearly targets) got higher and higher,” said Rabinowitz, “the system was rigged against diverse schools and large schools, and California has a lot of those. Moving forward we need to think about what’s the next generation of accountability systems that are fairer and more appropriate for the conditions we’re dealing with.”

The state and federal school accountability systems are very different, which is why the state Department of Education can boast that 59 percent of elementary schools met their API goal of 800, while, at the same time, only 27 percent met their annual target under NCLB. The state measures growth in achievement, while the feds measure the percentage of students who have reached proficiency, without crediting progress toward that goal. (The state’s 800 goal for API is just a round number; 875 is actually the score for reaching schoolwide proficiency.)

There are virtues in both systems, Rabinowitz said, but NCLB, with its aspirational goal of proficiency for all, ran into problems. “While it made sense from a non-technical perspective, implementing it had a lot of challenges. And that’s what we’ve been dealing with in California and other states,” he said.

Recognizing that states were about to run into a wall, yet unable to reach agreement with Republicans in Congress over how to fix or replace NCLB, President Obama and Secretary of Education Arne Duncan offered states the opportunity for a waiver. Gov. Jerry Brown, Torlakson, and the State Board of Education didn’t like the demands that came with it, which included a commitment to adopt a statewide teacher and principal evaluation system that incorporated test scores. So they ignored Duncan’s waiver form and submitted their own version through another provision in NCLB.

Arun Ramanathan, executive director of the advocacy group Education Trust-West, predicts that the feds will reject the waiver soon after next month’s presidential election. He said California missed an opportunity.

“It’s an application in name only,” he said. If California had submitted a real waiver, it could now be in the position of other states that have created one accountability system – their own – that identified the lowest-performing schools most in need of improvement as well as those schools that have closed the achievement gap and deserve recognition. California’s system does neither, Ramanathan said.

Until the state does get a waiver, California must live with NCLB’s regulations. Districts will not have flexibility to use Title I money, they must notify parents of students in a Program Improvement school that they can transfer, and the Department of Education must keep its compliance bureaucracy intact.

Christine Swenson, director of the state Education Department’s Improvement and Accountability Division, said that monitoring of districts will continue. Districts with schools in Year Five of Program Improvement that fail to implement their school turnaround plan could lose Title I money, although she acknowledged that this has yet to happen.

A technical note: Program Improvement is often portrayed as a Venus Flytrap, easy to fall into, impossible to escape. That’s not quite true. NCLB includes an easier, alternative way of determining whether a school has made its annual academic targets.

Seventy-nine schools exited Program Improvement this year using the safe harbor provision under NCLB. Source: State Department of Education (click to enlarge).

Seventy-nine schools exited Program Improvement this year using the safe harbor provision under NCLB. Source: State Department of Education (click to enlarge).

Called a safe harbor provision, its chief requirement is that a school decrease the number of non-proficient students in every subgroup by 10 percent each year. So if only 40 percent of a subgroup such as students with disabilities scored proficient, next year, an additional 6 percent of students must be proficient (10 percent of 60) or 46 percent. That’s a lot less than 80 percent proficiency, the overall target for this year.

To exit from Program Improvement, a school must meet its academic targets two consecutive years. As the accompanying chart shows, this year 79 school escaped Program Improvement, probably through Safe Harbor. An additional 323 schools in Program Improvement made their target the first of two years.

 

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 (8)

Leave a Reply to Manuel

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. CarolineSF 11 years ago11 years ago

    I do know that in other states (some? most? all?), students’ standardized test scores do have impact on them personally in various ways. So that would have to be taken into account in any comparison. I guess it would take a comprehensive study to find out how students with lackluster standardized test scores did on the SAT and ACT, which are voluntary and DO personally affect them.

  2. Doug McRae 11 years ago11 years ago

    CarolineSF: When one looks at APIs for EL, MS, and HS from back in 1999 when APIs were first initiated, the numbers were pretty comparable. So, over the past 13 years, using the API as a rough measureing stick, one can say that elem schools have had larger achievement gains than middle schools, and middle schools have had larger gains than high schools. I don't know of any data confirming the same pattern in … Read More

    CarolineSF: When one looks at APIs for EL, MS, and HS from back in 1999 when APIs were first initiated, the numbers were pretty comparable. So, over the past 13 years, using the API as a rough measureing stick, one can say that elem schools have had larger achievement gains than middle schools, and middle schools have had larger gains than high schools. I don’t know of any data confirming the same pattern in other states, but I suspect that many educators would say it is more of a universal pattern than a unique California situation. I’ve always remembered a Q & A between a well respected urban Supt from the LA area and a CA State Board member back in 2000 or 2001. The Q from the board member was “How is the implementation of the new academic content standards going in your district?” The new content standards were the ones adopted by the board in 1997/98. The A from the Supt was “Well, we see a lot of progress at the elem grade levels, some signs of light at the middle grade levels, but at the HS level it’s like trying to move a cemetery.” I suspect most parents would agree that trying to modify behavior [and learning is indeed a behavior] is harder for adolescent kids than it is for younger kids — I know it was for my four kids who are now adults. Whether that has something to do with differential gain scores for EL, MS, and HS is probably a pretty decent speculation.

    Replies

    • navigio 11 years ago11 years ago

      That's an interesting observation Doug. Anecdotally, I would say the very existence of the API has offered people an additional reason to choose against certain public schools or even public school in general. In my area, it is fairly common to leave the public system after elementary. This of course is a self-feeding dynamic, and clearly is not based only on API, but I do think it has something to do with it. Even if … Read More

      That’s an interesting observation Doug. Anecdotally, I would say the very existence of the API has offered people an additional reason to choose against certain public schools or even public school in general. In my area, it is fairly common to leave the public system after elementary. This of course is a self-feeding dynamic, and clearly is not based only on API, but I do think it has something to do with it. Even if it only has a minor impact, I would not be surprised to see it magnified over a decade or more.
      Again, anecdotally, our average elementary API was about 50 points higher than our high school API and about 10 points higher than our middle school API in 1999. Elementary is about 100 points higher this year than both middle and high averages, which are about the same now.

  3. CarolineSF 11 years ago11 years ago

    Does the pattern of higher API for elementary schoolers and lower for middle and high schoolers hold true in other states? Did it hold true in the past? I think this reporting needs to inform us about whether this is a common or universal pattern or some unique California situation.

  4. Manuel 11 years ago11 years ago

    Well, it is arithmetic. The API has been until now nothing else but a weighted average of a schools' CST scores with a few wrinkles thrown in for students with disabilities plus the CAHSEE for high schools (I may be forgetting a few details, but this constitutes the bulk of the API's calculations; here's the link to the spreadsheet used by the CDE: http://www.cde.ca.gov/ta/ac/ap/documents/calc11b12g.xls). I have somewhere in my files an LAUSD powerpoint presentation explaining how to … Read More

    Well, it is arithmetic.

    The API has been until now nothing else but a weighted average of a schools’ CST scores with a few wrinkles thrown in for students with disabilities plus the CAHSEE for high schools (I may be forgetting a few details, but this constitutes the bulk of the API’s calculations; here’s the link to the spreadsheet used by the CDE: http://www.cde.ca.gov/ta/ac/ap/documents/calc11b12g.xls).

    I have somewhere in my files an LAUSD powerpoint presentation explaining how to increase the API by improving the lower bands rather than trying to push more students into the “advanced” band. This is easier to do in elementaries because students 1) have a single teacher that is after them to do well in the test while middle and high school students have many to answer to, 2) all elementary students take the test at the classroom of their single teacher while middle and high schoolers take it where they happen to be when the test is given, and 3) elementary students have not found out that nothing will happen to them if they don’t score well.

    A few years ago, LAUSD’s Superintendent Cortines threatened to close magnet middle and high schools who did not increase their CST scores on the premise that if their students had scored high in elementary why aren’t they scoring high as they move up? Magnets were in a panic, but that threat went nowhere.

    Lastly, these increases won’t last as the system will eventually regress to the mean. Why? Because a standardized test does not allow for “growth”. If it does, then it is not a standardized test. Don’t believe me? Take a look at the tables listing the score distributions in the Technical reports. Have fun!

  5. el 11 years ago11 years ago

    I think the right way to engage the high school kids with the STAR exam is to make it meaningful for them if they do well on the tests, but unmeaningful if they do poorly. Several districts in this area have experimented with a proficient STAR result being worth a letter grade of extra credit in the appropriate class at the high school level. There's no penalty for a bad score. This makes the test … Read More

    I think the right way to engage the high school kids with the STAR exam is to make it meaningful for them if they do well on the tests, but unmeaningful if they do poorly. Several districts in this area have experimented with a proficient STAR result being worth a letter grade of extra credit in the appropriate class at the high school level. There’s no penalty for a bad score. This makes the test seem like a better use of everyone’s time and it makes the STAR testing time a bit more festive, and to my surprise, it got kids actually choosing to study for it.

    Replies

    • navigio 11 years ago11 years ago

      Sometimes I wonder how much difference the lack of seriousness has on the CST. If you look at proficiency rates by PEL (parent education level), there is a difference in results that pretty much stays constant over all grades. If, as is posited, elementary students dont yet know that they cannot take it seriously, one would expect those differences to rise as the grade level increases. That seems to imply that, although clearly there are … Read More

      Sometimes I wonder how much difference the lack of seriousness has on the CST. If you look at proficiency rates by PEL (parent education level), there is a difference in results that pretty much stays constant over all grades. If, as is posited, elementary students dont yet know that they cannot take it seriously, one would expect those differences to rise as the grade level increases. That seems to imply that, although clearly there are some who dont take it seriously, whether its enough to make a huge difference in results, is not entirely clear. Perhaps that makes a huge difference in some schools, but none in others?
      One concern I have with trying to make CSTs meaningful by rewarding those who score well is highlighted by this same proficiency difference based on PEL. Statistically, more affluent kids will get better grades, due largely to ‘circumstance’. If we could show that circumstance has a disproportionate impact on not caring over and above the other impacts of circumstance, then it would make more sense. Otherwise, I think it would be problematic.
      I also think that since these differences exist starting right away (ie in early elementary), the impact of how we segregate ability means that, in large part, we are probably accumulating any negative impacts of the education and societal systems disproportionately on kids who are scoring the lowest. It does not seem right that we can then expect them to simply be able to counter years of that history because they all of a sudden care about the test.
      That said, I want to be wrong. It is really troubling that there is this unknown factor in the reliability of our test data, but that we simply tend to ignore that. It might be one thing if we could quantify it, but I’m not clear we can even do that.

  6. navigio 11 years ago11 years ago

    "... and the Department of Education must keep its compliance bureaucracy intact." So we're spending money to do something we know is a waste of time. Nice. I would add that school districts also employ people to do these things. Btw, thank you for using the term 'disaggregation' and actually doing some of it. Finally! ;-) Read More

    “… and the Department of Education must keep its compliance bureaucracy intact.”

    So we’re spending money to do something we know is a waste of time. Nice. I would add that school districts also employ people to do these things.

    Btw, thank you for using the term ‘disaggregation’ and actually doing some of it. Finally! 😉