California has a long history of attempting to reform Algebra I in the name of equity. From the misguided “Algebra for all” push in the ‘90s to San Francisco’s disastrous “Algebra for none” initiative, the Golden State has repeatedly embraced well-meaning but ill-conceived math policies. Both of these particular initiatives failed for the same reason: They treated all students the same, ignoring the reality that students come to the classroom with different levels of preparation.
Another Bay Area district, however, has tried a slightly different approach with its Algebra I (A1) Initiative accelerating below-grade-level ninth graders into Algebra I to learn alongside their on-grade-level peers. This effort by the Sequoia Union High School district excluded high-achieving students who had already completed Algebra in middle school, allowing them to continue on advanced tracks. At the same time, it offered extensive teacher support to foster differentiation for the students who remained in the program. A recent random-assignment study, made possible through a partnership between the district and Stanford University, examined how this approach is playing out.
Sequoia Union is a diverse community just north of Palo Alto, where 40 percent of students are socioeconomically disadvantaged, 14 percent are English learners, and 43 percent identify as Hispanic. Before the A1 Initiative, students were placed into ninth grade math courses using placement charts based on transcripts and test scores. The new program shifted students entering ninth grade at or below grade level into an accelerated Algebra I course and equipped teachers with a comprehensive set of supports, including extra planning time, additional instructional training (including “math language routines”), and professional learning communities.
The study followed a cohort of over 1,000 students entering ninth grade in the fall of 2019, who were randomly assigned to either the treatment group (A1 Initiative classrooms) or the control group (business-as-usual classrooms were divided into three tracks based on the district’s traditional placement charts).
The researchers evaluated the A1 Initiative by examining a range of outcomes, including attendance rates, chronic absenteeism, student retention within the district, course progression, and standardized math test scores.
The non-academic outcomes were promising for students below grade level at baseline. Their absence rates were between 2 and 7 percent lower than the control group throughout each year of high school, with chronic absenteeism dropping even more significantly, by 5 to 9 percent. Meanwhile, attendance patterns for on-grade-level students remained largely unaffected.
The initiative also appeared to improve student retention, particularly for those at the highest and lowest ends of prior achievement. Below-grade-level students’ retention was 13 percentage points higher than their peers’ in the control group, while on-grade-level student retention was about 6 percentage points higher. Researchers speculated that these gains could reflect stronger feelings of belonging and satisfaction.
Academic outcomes, however, paint a more complex picture. Below-grade-level students assigned to the initiative faced higher course failure rates in ninth grade than those placed in remedial classes, likely due to the increased rigor of Algebra I compared to Algebra Readiness. Roughly half of these students needed to retake Algebra I or enroll in a pre-geometry summer course. Still, the long-term benefits were notable. By their senior year, these students were more than twice as likely to have passed Algebra II compared to their remedial-track peers. They also earned over 25 percent more math credits, on average, though they were no more likely to complete a post-Algebra II course.
Test scores followed a similar pattern. Below-grade-level students did not show significant improvement on the tenth-grade standardized math test, but by junior year, their scores were 0.19 standard deviations higher—about 7 percentile points—than their control group counterparts. Researchers attributed these gains to improved attendance and persistence.
For other baseline achievement groups, the results were less compelling. Students who were at grade level entering ninth grade math showed similar performance in both the treatment and control groups. Meanwhile, “nearly-at-grade-level” A1 students were less likely to complete two semesters of geometry by the end of tenth grade. This may have been because their control group peers benefited from “double dose” math under the traditional placement system. Additionally, standardized test score improvements for these students were not statistically significant.
While the study’s use of random assignment lends itself to strong internal validity, several factors warrant caution. The pandemic disrupted the cohort’s freshman year, introducing potential confounders like changes in motivation and parental involvement during remote learning. The relatively small sample size further limits generalizability, and the fact that teachers volunteered to participate in the initiative adds another layer of complexity, as teacher assignment was not random.
Even when taken at face value, the A1 Initiative raises important questions about both feasibility and scalability. The district made thoughtful adjustments compared to previous Algebra reforms, such as exempting high achievers and focusing on robust teacher supports, but these efforts required significant resources and partnerships. For below-grade-level students, the A1 Initiative clearly delivered benefits, both academically and non-academically, while leaving on-grade-level students unaffected. However, the approach’s resource-intensive nature—relying on extensive professional development and a partnership with Stanford University—may not be realistic for most districts.
Ultimately, the A1 Initiative represents a well-intentioned and innovative effort to address math achievement gaps, but its broader applicability is limited. Sure, for districts with the budget and partnerships to provide the necessary teacher training and support, this model could be worth considering. For the rest, it’s back to the drawing board.
SOURCE: Thomas Dee and Elizabeth Huffaker, “Accelerating Opportunity: The Effects of Instructionally Supported Detracking,” Annenberg Institute at Brown University (2024).
