Do Early Warning Systems Work? Measuring Impact to Improve Student Attendance in Pasco County

“On their own, early warning systems only give us a certain type of information. Data can show where there is a problem with attendance, but it doesn’t really tell us how to solve it.” - Yusuf Canbolat 

Chronic absenteeism was a stubborn challenge before the Covid-19 pandemic, with about 15 percent of all U.S. students missing 15 or more days of school each year. That number doubled during pandemic lockdowns and has remained high, at 26 percent in 2022–23. 

Many districts have data dashboards and early warning systems, which alert teachers, school social workers, and other staff when a student’s attendance record puts them at risk of becoming chronically absent. These alerts are intended to prompt interventions improve student attendance. But do early warning systems actually reduce absenteeism? If so, how? And for which students? 

SDP Fellow Yusuf Canbolat, a research analyst at Pasco County Schools in suburban Tampa, Florida, sought to answer these questions in his district, which implemented an early warning system in 2018–19. He first analyzed attendance and early-warning data to measure the impact of warnings on student absenteeism. Then, he conducted group interviews with educators and parents to identify common attendance barriers and school-based responses to early warnings. 

Canbolat’s findings, which were published in Educational Evaluation and Policy Analysis, Education Week, and the Hechinger Report, revealed differences in the impacts of early warnings based on students’ socioeconomic status and sparked a broad conversation about how educators and social workers could more readily reach and help families who are struggling to get students to school. 

“The analysis showed that the early warning system is effective in reducing absenteeism for socioeconomically advantaged kids, but not their disadvantaged peers,” Canbolat said. “In our focus group meetings, we learned that disadvantaged students face more structural barriers to attendance, and that those barriers are not easy to dismantle through our current school-based efforts.” 

Assessing Quantitative Evidence 

When a student is absent, they miss out on teaching and learning. And when they miss more than 15 days of school and are “chronically absent,” those short-term impacts compound and can have serious consequences for grades, high-school graduation, and health and economic security in adulthood. An analysis by the Council on Economic Advisors found that absenteeism accounts for up to 45 percent of post-pandemic declines in student test scores in 4th and 8th grade reading and math. In May 2024, the White House hosted the Every Day Counts Summit, where leaders urged states and districts to establish “cultures of attendance” and highlighted resources and best practices to promote consistent attendance at school. 

In 2022–23, chronic absenteeism at Pasco County Schools was in line with national trends: about one in four students and one in three low-income students were chronically absent. On paper, the district had already implemented first-line responses to combat absenteeism: a robust data dashboard built in SQL that links students’ attendance, academic, and behavior records, and an early warning system that alerts teachers and social workers when a student crossed two thresholds of risk. After seven absences, students are flagged as “at risk.” After 18 missed school days, they reach “off track” status. At both the moderate and chronic thresholds, teachers and social workers are cued to reach out to families and stress the importance of coming to school. 

Canbolat measured the impact of the district’s early warning system through a regression discontinuity analysis that examined trends and impacts in two years of student attendance data. During that time, 28 percent of students were “at risk” and 16 percent were “off track.” 

Pasco County’s early warning system had no impact on attendance for students at either threshold, Canbolat found. However, in comparing impacts based on student socioeconomic status, he found benefits for one group: socioeconomically advantaged students considered “off track.” An alert reduced chronic absenteeism for this group by about a third, from 5.9 percent to 4.0 percent. 

Gathering Qualitative Insights 

The project’s second phase focused on exploring why students were chronically absent, including the economic, social, and environmental factors that contribute to poor attendance and hampered the effectiveness of early warnings. Focus groups of teachers and social workers discussed common attendance barriers for low-income students, including housing instability, family health challenges, and responsibilities to care for younger siblings at home.  

“In many cases, students did not have easy access to available resources,” like housing support, Canbolat said. “We decided to help families and students access the resources they need to dismantle those barriers so students can attend schools at the same rates of their more advantaged peers.” 

In addition to reconsidering supports for low-income students, the district also is planning a comprehensive attendance awareness campaign and will establish school attendance teams. 

These conversations revealed another important impact of attendance data: it highlighted the scope of the problem, which was larger than many teachers realized. Overall, attendance at Pasco County schools is high, at 90 percent or more on an average school day. 

“When kids miss school, they can become invisible,” said Canbolat. “These data tools can remind teachers to follow up on absent students and interact with social workers to intervene.” 

Lessons Learned 

One major takeaway from Canbolat’s work is the sheer complexity of the challenge. Students are chronically absent for many reasons, and those factors typically influence and amplify one another. School-based efforts to combat absenteeism, starting with data collection and an early warning system, are necessary but not sufficient to reduce absenteeism, particularly for low-income students. 

Analysts tackling this issue should consider the context of the data and look for trends by student group. When attendance differs by student group, researchers and leaders should examine environmental factors and investigate how policy and practice can help dismantle those barriers. 

In addition, students attendance patterns emerge very early in a school year. Tracking student attendance in the first few weeks and investigating attendance challenges before students become chronically absent seems a promising approach. Data review and intervention should be ongoing, so schools can help prevent chronic absenteeism before it occurs. 

In looking ahead, Canbolat sees the potential for machine learning to use attendance data to predict student absenteeism.  

“Most tools and existing data follow up on student absenteeism after it happens. In the future, we hope to create more early warning indicators and new incentives to support and motivate kids to attend school, such as individualize texts and messages to students and families,” he said. “That’s the goal in the next few years.”