Slow growth, big disparities after 5 years of Smarter Balanced tests

Proficiency in math and English language arts rose about 1 percent in 2018-19

Students at Redwood Heights Elementary School in Oakland. The Oakland Unified School District publicly shares academic growth data for Redwood and all other schools in the district.
Credit: Alison Yin for EdSource (2017)

California’s students’ Smarter Balanced test scores rose marginally in 2018-19, the fifth year of the tests, while showing little to no progress in closing wide disparities among ethnic, racial and other student groups, the California Department of Education reported on Wednesday.

The notable exception is Hispanic students, whose 5-year proficiency rate in both English language arts and math rose faster than those of white and Asian students.

For the first time, a majority — 50.9 percent — of all students who took the English language arts test met or exceeded the standard, the top two of four testing levels, the technical definition of proficiency at grade level. In math, 39.7 percent of all students met or exceeded the standard.

The Smarter Balanced tests were designed to demonstrate students’ competency under the Common Core standards, which California adopted in 2010. Most questions are multiple choice, with a performance task requiring students to demonstrate critical thinking and problem-solving skills. The tests are given in grades 3 to 8 and 11.

Under the state’s new accountability system, test scores are just one of several measures to evaluate school improvement. The California School Dashboard, the site that rates district and school performance on all of the measures, will be updated later this year.

Overall proficiency rose only about 1 percentage point for English language arts and math, compared with 2 points in 2017-18. After five years of Smarter Balanced, students who met or exceeded standards had increased 7 percentage points in both tests. That’s an average of 1.4 percentage points annually — less after discounting the large second-year bump that reflected familiarity with a new test.

This year’s small gain, “though not as much as we want, is what I would have expected,” said Julien Lafortune, a research fellow with the Public Policy Institute of California and co-author of a recent analysis of Smarter Balanced implementation. “It’s hard to expect drastic changes from year to year” in a test taken by 3.2 million students, but “sustained improvement, even if slow, adds up over time.” In Washington and Oregon, two neighboring Smarter Balanced states, math scores declined slightly in 2018-19.

What is disappointing is that “progress in math in elementary grades is not carrying forward to middle and high school. The average student is increasingly likely not to meet standards in middle and high school,” said Neal Finkelstein, co-director of the Innovation Studies program at WestEd, the San Francisco-based research and policy organization that tracked 10 districts’ work in math over five years through the project Math in Common.

In 2018-19, 50 percent of 3rd-graders were at or above standard in math. But after 4th grade, there was a steady decline: 39 percent proficiency in 6th grade, 37 percent in 8th grade, the critical year before Algebra I, and 32 percent in 11th grade — a factor behind the California State University’s proposal to require a fourth year of high school math or quantitative reasoning.

“You can’t sugarcoat that 60 percent of students are not making standards in math,” Lafortune said.

In English language arts, the percentages of students at or above standard have increased gradually or remained constant across grades: from 49 percent of 3rd-graders to 57 percent of 11th-graders in 2018-19.

“Transitioning to Common Core was the right thing to do, and the transition has been challenging on many fronts,” said Francisco Villegas, director of school transformation focusing on K-12 math for Partnership for Los Angeles Schools, a nonprofit that manages 18 schools in Los Angeles Unified.

Not all of the news is dreary. An EdSource analysis found that increases in English language arts proficiency in districts receiving the most extra funding under the Local Control Funding Formula — those with the most low-income students, foster youth and English learners — rose three times faster than those districts receiving the least funding: an increase over five years of 9 percentage points versus 3 points,

The gap between those districts is still huge: 39 percent proficiency in English language arts for districts with the most high-needs students versus 78 percent for those with the fewest, but the difference did close 6 percentage points since 2014-15. The pattern holds, to a lesser degree, with math.

Hispanic students, who make up the majority of California’s students, made the largest 5-year gains in proficiency: 9 percentage points in English language arts, double that of whites, and 7 percentage points in math compared with 5 percentage points for whites. But 28 percent of Hispanics scored at or above standard in math, compared with 54 percent for whites and 74 percent for Asians.

The state’s 334,000 African-American students made no progress in closing the performance gap. Only 1 in 5 were at or above standard in 2018-19 in math and in English language arts.

Scores for English language learners, a group that changes as students become proficient in English and are reclassified, remain dismal, with 13 percent at or above standard in both math and English language arts. About one-fifth of the state’s students are classified as English learners.

“At the rate we’re going, my five-year-old-son will be old enough to be a grandparent before California achieves educational justice for low-income students and underrepresented students of color. That’s simply not good enough. We have to do much better, much faster,” said Elisha Smith Arrillaga, executive director of the advocacy organization Education Trust-West.

Islands of progress

Discouraging state statistics, however, don’t reveal a deeper, more hopeful story, Finkelstein said. Look locally, where differences in effectively implementing the standards reveal big variations among districts and schools.

“It’s all the more important to find champions in this work. The exceptions can be highlighted and learned from,” Finkelstein said. He cited Garden Grove, a district often praised for its stable leadership and collaboration among teachers, where 3 out of 4 students are low-income and English learners. In the latest results, 61 percent of students were at or above standards in English language arts and 52 percent in math — twice that of Santa Ana, its Orange County neighbor.

Observers have suggested multiple reasons for the low proficiency in math and the drop in scores in middle school. Arun Ramanathan, CEO of Pivot Learning, a nonprofit organization that works with schools on improvement strategies, said many districts were too quick to adopt subpar instructional materials when they began implementing the Common Core and haven’t switched to better textbooks. Villegas said that some districts signed long-term contracts early in the implementation process, putting them in a bind.

Then there’s the nature of middle school, where many students develop the I’m-no- good-at-math mindset, and the nature of math itself. More so than with English language arts, math builds on prior knowledge. Middle school students who are weak in fractions, a building block for algebra introduced in the early grades, will struggle later on, experts said.

There’s often no time during the year to work with students who are falling behind, said Lisa Andrew, CEO of the Silicon Valley Education Foundation. Last summer, 4,000 students from grades 3 to 10 in 32 Bay Area districts did a 19-day math intervention program. Elevate [Math] targets students who tested one level below standard, not the furthest behind. It addresses their weaknesses and introduces them to content they will see in the fall.

“Extended time is not optional any more to close the achievement gap,” Andrew said. “The research is clear: Students need to be engaged in high-quality learning during the summer, during breaks.”

Common Core demands that math be taught differently, with an emphasis on students’ conceptual understanding. That can be a big lift for teachers who learned to teach with mnemonic devices and formulas. Finkelstein, Ramanathan and Andrew agree that teachers aren’t getting enough in-the-classroom coaching.

Elevate [Math] puts the 178 credentialed teachers in the program through 60 hours of professional development, starting in the spring. Rocketship Public Schools, a charter school organization with 13 K-5 schools in the Bay Area, has had success by differentiating between English/social studies and STEM teachers, who teach only math and science.

“We give new teachers the choice of STEM or humanities,” said CEO Preston Smith, “but we encourage science majors to try to teach STEM.”

All Rocketship teachers get 400 hours of training over the course of a year, where they can concentrate on math and science without having to feel they must be the master of all subjects, he said.

In 2018-19, Rocketship reported 61 percent of students at or above standard in math — way above the statewide average; 80 percent of its students are low-income students or English learners.

Under local control, each district is responsible for its own improvement strategies. Some have formed data collaboratives and networks. Michael Kirst, the former State Board of Education president and an architect of local control under former Gov. Jerry Brown, acknowledges that, with scores stagnant in many districts, the state should play a larger role. He’s not sure what the priority should be: fund more specialists for county offices of education; re-establish math academies, as Gov. Gray Davis did for algebra; create more summer programs like Elevate or encourage teacher specialists.

“We’re doing better in English language arts than I predicted and worse in math,” he said. “The problem is serious.”

Morgan Polikoff, an associate professor of education policy at the USC Rossier School of Education, who recently published a brief challenging the state’s method of measuring test scores, said, “If the state is concerned about the magnitude of performance gaps, they probably should exercise more of their constitutional authority over education and take a stronger hand in telling districts to do things that will help improve outcomes.”

“Local control never really boosted equity anywhere,” he added.

Villegas characterized the issue differently. “Letting each district decide is fine, as along as there is guidance on what effective practices and effective implementation look like,”  he said.

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