State board chooses new way of measuring school progress on tests

Credit: ideabug / iStock

After hours of discussion, the State Board of Education on Wednesday settled two much debated issues that will enable state officials to move ahead this year with the state’s new school accountability system.

One decision creates a different way to measure schools’ and student groups’ progress on standardized tests in math and English. The other, more contentious issue will determine which schools and districts will require intervention or technical help because their English learners did poorly on the math and English language arts tests.

In September, the board approved a framework for the new improvement and accountability system that will give a broader view of schools’ and districts’ performance through measures that will include students’ readiness for college and careers, school climate, parent engagement and academic performance. The board set a timeline for refining the metrics over the next year.

The decisions on Wednesday define the academic performance indicator – a primary measure that will also designate the lowest-performing schools receiving federal aid under the Every Student Succeeds Act, which will go into effect next year. It will apply to students in grades 3 to 8.

The new way of measuring performance on standardized tests responds to critics’ objections to the methodology the federal government adopted under the No Child Left Behind Act. That law measured success by the percentage of students in a school or subgroup of students who scored proficient on a test. Each year under NCLB, the federal government required a larger percentage of students to reach proficiency and penalized schools and districts that failed to do so.

One problem with this method, critics said, was that it ignored the success of low and high achievers: struggling students who improved significantly but not to the point of proficiency, as well as proficient students who reached advanced levels. Schools tended to focus on raising scores of students “on the bubble” – those performing just below proficiency.

The new system will present a more nuanced perspective of students’ progress by measuring how far they score below or above the point of proficiency and how much their scores improve or decline over time.

Under the Smarter Balanced tests in math and English language arts, proficiency is defined as the point on a scale that designates Level 3, the second highest of four achievement categories. The state board’s accountability system will combine the distance from Level 3 and how much the distance has narrowed or widened over time to designate one of five color-coded performance categories, from red to blue.

Under the new color-coded accountability system, schools and districts whose average scores fall far below Level 3, the equivalent of proficiency, on the Smarter Balanced tests, and whose scores have stagnated or declined over time will be designated red or orange – in need of assistance. The graph shows the numbers and percentages of districts that would fall in each color band based on two years of math tests.

Source: California Department of Education

Under the new color-coded accountability system, schools and districts whose average scores fall far below Level 3, the equivalent of proficiency, on the Smarter Balanced tests, and whose scores have stagnated or declined over time will be designated red or orange – in need of assistance. The graph shows the numbers and percentages of districts that would fall in each color band based on two years of math tests.

For example, any school or racial or ethnic student group with an average score falling 25 or more points below Level 3 and that declined over three years would fall in the “red” or “orange” zones – those schools most in need of help. Other schools that showed significant improvement but hadn’t yet reached Level 3 would not require outside intervention. And schools that score above proficiency and have improved would fall in the enviable “blue” or “green” zones. “Yellow” would be in the middle.

State board President Michael Kirst and State Superintendent of Public Instruction Tom Torlakson had asked the U.S. Department of Education to permit measuring academic progress by movement up and down the full range or scale of scores, as opposed to just one point – proficiency – on it. The final rules for the Every Student Succeeds Act permit that system of measuring progress.

Morgan Polikoff, an associate professor at the University of Southern California’s Rossier School of Education, wrote a letter signed by three dozen academics requesting a new approach. He praised the state board’s alternative, but noted that a system based on scale scores, not proficiency percentages, would be harder to explain. “Communicating to the public could be a challenge, given it’s slightly complicated how this all is calculated,” he wrote in an email.

Which English learners?

Measuring progress of the state’s 1.4 million English learners on standardized tests has been problematic because the population is fluid. Each year non-English speaking immigrants enroll in school, while English learners who have become adept in English exit the English learner group. As Stanford University education emeritus professors Linda Darling-Hammond and Kenji Hakuta described in an EdSource commentary this week, test scores have tended to stagnate as the least skilled in English enter the subgroup and the most skilled leave it. And, they noted, low subgroup scores created a disincentive for districts and schools to redesignate students as “fluent English proficient” – with the harmful effect of holding some of these students back from more advanced courses.

Congress, in passing the Every Student Succeeds Act, gave states a new option. For the purpose of designating which schools need assistance, they could include the test scores of former English learners in the English learner group for up to four years after they’ve been reclassified. On Wednesday, the state board adopted the policy.

For California, which already had a waiver under NCLB to fold in the scores of some former English learners, this presented a relatively small change. However,  Californians Together and a coalition of other advocacy groups initially had pressed state officials not to include scores of any reclassified English learners. They said that mixing scores of high-performing former English learners would conceal the low performance of long-term English learners – those still not fluent after six or more years. Without shining a light on their struggles, districts would continue to ignore their needs, they argued. During testimony before the board, dozens of district administrators and English learner advocates stressed that point.

State board member Feliza Ortiz-Licon agreed. Blending the two groups’ scores would produce a middling “yellow” category, she said. “If most are there, how do we show growth? We are masking two different groups with different needs.”

Darling-Hammond, Hakuta and other academics wrote that including former English learners in the group gives districts a more complete picture of successful strategies that districts have used. And data projections by the California Department of Education reinforced the case for inclusion. They found that by including only English learners, ratings of two-thirds of schools and 60 percent of school districts would immediately fall in the red or orange zones – requiring technical assistance or intervention for English learners. That massive number would potentially overwhelm the capacity of the state and county offices of education to provide effective help. With the scores of reclassified English learners included, 20 percent of districts and 28 percent of schools would be designated red or orange.

In addition, by excluding reclassified English learners, 193 school districts and 1,213 schools would not have the minimum of 30 students that the state defines as an English learner subgroup. Local officials would lack any information on how English learners were performing in those schools and districts.

In public testimony, many speakers said they agreed with combining English learners and reclassified students strictly for determining which schools need assistance but implored the board to report data for English learners, reclassified English learners and long-term English learners separately and prominently on the new California School Dashboard, a color-coded report card on school and district performance. They also want the data prominently located on the key document that districts use to determine spending priorities, the Local Control and Accountability Plan.

Districts are required to commit more money to English learners under the Local Control Funding Formula, and separating the data will enable parents and advocates to demand more resources and more effective approaches for these students, they said.

“It’s important that the LCAP process be transparent and clear to stakeholders,” said Manuel Colon, chief academic officer of the Anaheim Union High School District. “If English learners are hidden, services will be reduced, causing them to fail or drop out.”

Colleen Pagter, legislative analyst for the Los Angeles Unified School District, agreed. Not only should data be reported separately, but districts should be required to address the distinct needs of English learners and reclassified English learners in the LCAP, she said.

To a person, board members said they got the public’s message. Jenny Singh, an accountability administrator with the California Department of Education, assured the board that separate data for the English learner groups will be prominently shown.

How visible, in what form, and how many clicks it will take to find it will be decided another day – possibly at the board’s next meeting in March, when it takes a look at the latest draft of the California School Dashboard.

EdSource in your inbox!

Stay ahead of the latest developments on education in California and nationally from early childhood to college and beyond. Sign up for EdSource’s no-cost daily email.

Subscribe