STUDENT MOBILITY IN ST. LOUIS: UNCOVERING NEW CHALLENGES THROUGH DATA

OVERVIEW

WHY THIS WORK

Though many research interests were identified, practitioners decided to focus on student mobility as a shared challenge across schools that impacts their ability to improve academic outcomes for students. St. Louis has disproportionately high rates of student mobility, as defined by the state education agency, Missouri Department of Elementary and Secondary Education (DESE), as the number of student transfers recorded in and out of a building during an academic school year, including summer transfers. The median building-level rate of student mobility was 37% for St. Louis City in 2018-2019, as reported through DESE and later examined in our first research brief.

Initially, researchers in the group cautioned that student mobility would be difficult to study and perhaps result in research that had little meaningful implications for practice. Practice-side participants in our workshop countered that student mobility is a critical barrier to academic growth in St. Louis. They surfaced many student mobility-related questions, including who’s moving, where they’re going and coming in from, and why they’re mobile, noting that if researchers could help them understand these questions, they might be able to design more effective initiatives to meet student and educator needs. The participating researchers, however, expressed concern about the locus of control of mobility-related phenomena in St. Louis schools and about the availability of high-quality data to explore these questions. They noted that homelessness and housing challenges would be identified as the culprit – which may not be within a school leader’s ability to address. Practitioners, however, pushed back and were certain that housing challenges were not the only influence on a student’s transfer from one school to another. They expressed a strong desire to examine this topic in depth. 

To underscore the RPP ethos that research questions should be directed by educators for the use of practice, SKIP brokered an agreement between researchers and practitioners to pursue this agenda, despite researcher hesitation on data availability. SKIP and educators recommended obtaining student-level data from the state department of education, increasing researcher confidence that this would address the potential roadblocks of only having access to data from participating schools. Both practitioners and researchers agreed that this would lead to a level of rigorous insight that may not be possible on their own.

WHAT THE WORK EXAMINES

Once student mobility was agreed upon as an area for research by both practitioners and researchers, the SRPC worked with both groups to further define research questions, identify and designate principal researchers and methods for each question, and develop processes to keep practitioners engaged in the following ways:

• A data committee worked to create a detailed understanding of variables and their definitions, build relationships with the state education department to access data, and develop data-sharing agreements with partnered schools. 
• A student mobility research committee, consisting of both practitioners and researchers, was created to provide oversight on the research process, interpretation of outcomes, and contextualization of implications.
• A third committee on knowledge engagement was created to develop external communications for the public and foster engagements for partnered schools to interact with the research. 

The following research questions were developed:

• What are the characteristics of students who are transferring schools? 
• When and where are students transferring?
• What is the academic impact of transferring on students? 

The student mobility descriptive report (our first research artifact listed above) was not initially outlined in the research agenda as a potential product. However, school leaders were seeking information to aid advocacy efforts and potentially influence relevant state-level education policies in real time. When data acquisition was delayed, SKIP DesignEd, acting as the RPP broker, worked to analyze publicly available building and district-level data from DESE to provide descriptive insight on patterns of student mobility in comparison to local counties and surrounding school districts. While waiting for researchers to conduct the detailed analysis of student-level data, this publicly available data helped inform school leaders’ understanding of the magnitude in which student mobility was occurring at the district- and building-level. 

FINDINGS

The first research brief shared descriptive data from publicly available records from DESE regarding building-level rates of student mobility in St. Louis City and provided a brief description of the pilot study. It introduced the goals of the RPP, explained how the pilot topic was selected, and provided a summary of the extent of student mobility in the region using analysis of publicly available state-level data. Practitioners were affirmed to find that rates were generally high on average and represented some of the highest rates in the region compared to surrounding counties and comparable cities. 

In summary, our key findings from this descriptive study were:

• In 2018-2019 (before pandemic disruptions), the median transfer rate for St. Louis City’s school population was 37%.
• This median number is the highest amongst larger cities in the state. A comparable city with school choice, Kansas City, had a student mobility rate of 31%. 
• Though these rates are high, St. Louis City’s mobility rate has declined over time, dropping from a median rate of 51% in 2010-2011 to 37% in 2019. 

Presentations of these findings were made to the local district’s School Board Legislative Committee and the St. Louis City Board of Alderman. The findings fostered deeper conversation on discrepancies between this publicly available data and the student-level data under analysis. For instance, one of the surprises that resulted from this analysis was the utilization of student mobility by the state as a percentage of the student body. In reality, the rate represented students coming and going without any specification for whether it was the same student engaging in the system versus new individuals. The potential to double-count the same student was previously not considered and therefore a challenge to school leaders who had regarded the data as unique cases. Additionally, we found that the transfer rate included summer transfers, a rate that signals perhaps less disruption when compared to a mid-year school change. This finding led to a deeper investigation of how data is collected, coded, and collated for public consumption and led to increased engagement with the state department on what was historically a non-trivial number. Though the percentage of students who transfer mid-year was found to be much lower than the rate of student mobility in schools, the administrative and educational load felt by school leaders was still noted as a challenge to be addressed. 

The second research brief examined the student characteristics of who is transferring before the end of a school year, using student-level administrative data from 2007-2022. In summary, key findings from this study were:

• In the 2018-2019 school year, St. Louis City had higher mobility than the surrounding region. In the city, 9.6% of students changed schools during the year, compared to 6.5% for the surrounding counties. 
• Students with unstable housing, especially in emergency shelters, were twice as likely to leave.
• Black students had a significantly higher chance of transferring schools than white students.
• Odds of transferring were lower in 1st through 8th grades. They increase in high school, peaking in 9th and 10th grades.

WHAT THE PARTNERSHIP LEARNED

One thing we learned is that data at the building- and district-level were under-defined. Insight into data reporting decisions when aggregating from student to district-levels influenced our understanding of true student mobility rates. In the state-reported administrative data, the building-level variable called “transfers” included a count of students transferring in and out, not only students transferring out. Therefore, building-level calculations of student mobility rates, which calculated transfers over the total enrollment, were really referring to churn. This was confusing, however, as the data dictionaries of how each variable was created and referenced were unavailable. If interventions were created based on these assumptions, they would likely be less effectively deployed. As a result of this data-deep diving, the attached data dictionary was created for partners’ mutual understanding of what exactly was being measured in each dataset. 

We also learned the importance of interpreting data alongside practitioners with knowledge of unique building-level policies. In analyzing data, we were surprised to learn that the published student mobility numbers from our state education agency may have been impacted by cutoff dates and processes around enrolling and transferring students. For example, late enrollment may have been counted as student mobility. Additionally, one student may have experienced multiple transfers in and out of the same school building over the course of a school year but be counted toward the student mobility rate. Some principals also shared that they unenroll a student who has not attended school for fourteen days or more due to school policy. These students may instead be chronically absent, which is a different challenge than being mobile or transferring at will. This affects the accuracy of student attendance accountability measures. These insights have helped us understand what the data is saying, where questions remain, and how to move forward in the future.

IMPACT AND USE OF THE WORK

Engagement with educators throughout the analysis resulted in a deeper understanding of state-reported numbers and where data may be misleading or masking other challenges. Cross-referencing from student- to building-level administrative data helped identify that some transfers may have been due to chronic absenteeism rather than true mobility. Learning about this challenge and the policies that may contribute is causing educators to consider their school practices and ways to respond.

Relatedly, the co-creation of the briefs improved literacy in data interpretation, not only for the practitioners but also for the researchers. We have observed interest across all partnership participants in tackling the next steps after learning what the data shows. What is often discussed in one paragraph of a peer-reviewed paper as “implications for practice” is the center point of discussion in the St. Louis School Research-Practice Collaborative. We’ve seen an increased interest in finding what works, for whom, and how we could approach this in our member schools.

OPEN QUESTIONS AND NEXT STEPS

One question that remains central is how to ensure that data is not used only to point to external challenges, but also to hone practices and policies within a school’s locus of control. For example, student mobility findings point to neighborhood challenges with poverty and housing policies and practices. However, the research also demonstrates some internal school-level practices that could be augmented to support students who are most likely to transfer before the end of a year. The SRPC is considering its role in supporting school leaders’ engagement with city leadership to address broader systemic challenges contributing to high student mobility, such as housing policies and family stabilization. We have also addressed the questions: Where, when, and why students move. St. Louis school leaders will determine how they would like to utilize this research – both in regard to engagement and intervention creation.

Saras Chung is the Executive Director of SKIP DesignEd and the Director of the St. Louis School Research-Practice Collaborative during its pilot phase. Deanna Childress is now Postdoctoral Research Fellow at PRiME Center at Saint Louis University and was Director of Data Analytics and Research at Community Partnership Network, St. Louis Public Schools, at the time this article was written.

Suggested citation: Chung, S. & Childress, D. (2025). Student Mobility in St. Louis: Uncovering New Challenges Through Data. NNERPP Extra, 7(2), 2-8. https://doi.org/10.25613/7DE4-FQ57