Uncovering the patterns that shape and hinder student success

When Pat Schlosser, assistant superintendent at Apple Valley Unified School District, reviewed the data from the California Dashboard for the class of 2025 this past January, he and other district administrators knew they needed to act. The data showed that while 93% of the students graduated from the district's two high schools, only 35.8% of students met A-G requirements, the set of courses seniors need to enroll at California State University or the University of California. How was it that a district with a graduation rate that was 5.5 percentage points above the state average was falling short on preparing students for four-year colleges? 

A few months prior, Trenae Nelson, superintendent of schools at Apple Valley Unified, was contacted by Stephanie Houston, assistant superintendent of innovation and impact at San Bernardino County Superintendent of Schools, to see if she’d like to join a new program to help districts improve students’ college readiness. Interested in the opportunity, Nelson soon met with Research Professor Jon Fullerton, Research Scientist Eric Larsen and Research Coordinator Marco Torres of the USC EdPolicy Hub. The research center, housed within the USC Rossier School of Education, proposed an opportunity for the district to join its Education-to-Workforce District Network. The effort, with support from the Gates Foundation and the Michael and Susan Dell Foundation, would examine thousands of data points provided by districts like Apple Valley to help them analyze complex issues and devise data-informed strategies to address them. For Apple Valley Unified, the timing couldn’t have been better.

Crunching the numbers

Schlosser and his team quickly went to work, identifying data sources and sharing them with the USC Ed Policy Hub. While Apple Valley has access to an enormous amount of data, Schlosser says, they don’t have the team or tools to synthesize and analyze it.

To analyze the data provided by Apple Valley and the other participating districts—Coachella Valley Unified School District, Downey Unified School District, Fullerton Joint Union High School District, and Long Beach Unified—the Hub is using the Education-to-Workforce (E-W) Framework. This framework was designed “to help education agencies collect and use data that advance educational and economic opportunity for all students,” Larsen explains in a recent Substack article. The framework outlines key questions, student success indicators tied to economic mobility, and ways to analyze and disaggregate data to track progress and address disparities. To establish a common language for the educational data, which will enable the researchers to combine data from a wide range of sources, the Hub is using the Ed-Fi Data Standards and associated technologies, Larsen explains. The five districts in the Education-to-Workforce Network are working on improving college preparedness for their students and/or addressing teacher workforce challenges, with the goal of “ensuring students have equitable access to high-quality teachers and courses that prepare them for success,” Larsen says.

Apple Valley, which is focusing on college preparedness, was given an early look at the Hub’s analysis of its data in January. The presentation gave them a longitudinal view of their students' progress, from middle school up until graduation. This data showed that, for Apple Valley Unified, two of the strongest predictors of A-G completion among students were 7th- or 8th-grade English Language Arts and math test scores. Students who did well on these standardized tests were more likely to complete A-G requirements. “Connecting data across systems and time allows school systems new insights into what may be driving differential outcomes for their students and, with those, the opportunity to pursue new strategies to improve results for kids,” Fullerton says. With this early delivery of data analysis, the district was learning new ways to monitor student success through the USC EdPolicy Hub’s “data-processing superpower,” Schlosser says.

The first convening

On March 27, a team from Apple Valley Unified, which included Schlosser, two high school principals and an assistant principal, attended the first convening of the Education-to-Workforce District Network. The event brought Apple Valley Unified together with three other participating districts: Long Beach Unified School District, Downey Unified School District and Fullerton Joint Union High School District. The convening gave the districts the opportunity to meet with the USC EdPolicy Hub team and their counterparts from other districts in person.

At the day-long meeting, facilitators from the USC Ed Policy Hub, including Professor of Education Morgan Polikoff, Professor of Education Patricia Burch, Research Scientist Daniel Silver and Research Analyst Dedrick McCord, began with a presentation of their data findings. The Hub aggregated all four districts' data so the group could compare and contrast outcomes. The facilitators then guided discussions as each group brainstormed strategies and interventions for the problems they decided to focus on. “The Hub really is investigatory. They are interested in the outcome only. We felt supported; the data analysis was in-depth and non-judgmental,” Schlosser says.

Another key aspect of the network is the emphasis on creating connections between the districts. Schlosser appreciated exchanging ideas with colleagues from other areas and hearing that some districts found success with strategies that had previously been dismissed, which gave them new perspectives and “widened our horizon of possible solutions and gave us a sense of hope,” Schlosser says 

What’s next

It’s only been a few weeks since the convening, but Schlosser and his colleagues at Apple Valley Unified are already taking actions based on the data analysis the USC EdPolicy Hub provided and their discussions with researchers. They discovered that there may be discrepancies in the districts’ teaching and grading practices. 

Student standardized test scores reflect that they know the content, yet student grades seem to tell a different story. The Hub’s data analysis showed that “students with similar incoming characteristics and test scores have differing probabilities of succeeding in math across schools. This could be from different grading strategies, different teaching practices, or something else,” Fullerton explains.

Schlosser says there may be overly stringent and unpredictable grading, as well as substantial differences in the number of graded assignments required for courses. Across the district, staff and educators are already meeting in small groups, comparing grading practices and seeing how they can reform policies to better support student achievement and college preparation. 

Apple Valley Unified and the other districts in the network will continue to work with the Hub at least through 2027. Right now, the Hub is helping each district develop strategies to improve student outcomes. The Hub will also monitor implementation of the selected strategies and the results of the process. “The vision,” Fullerton says, “is to be very directly helpful to our partners by supporting them in using data and rigorous analysis to make practical decisions for improvement.”