The Massachusetts Analysis of Dropout Data: 2021-22 Dropouts

Introduction

The purpose of this tool is to analyze Massachusetts public high school dropouts in grades 9–12 in order to determine the size of particular groups of dropouts for strategic interventions and planning.

Specifically, this analysis includes characteristics of students in grades 9–12 who dropped out of school in the 2021-22 school year (between July 1, 2021, and June 30, 2022). These students were examined to determine dropouts' patterns of student success prior to dropping out of school (e.g., attendance, behavior and course performance) as well as an approximation of the core course accumulation at the time of dropping out.

Along with our Early Warning Indicator System (EWIS), this analysis is another important type of data for schools and districts to examine. While EWIS looks at students who are currently enrolled in Massachusetts public schools, this tool looks at data on students who dropped out of high school the previous year and their behavior in the years leading up to dropping out to: a) plan appropriate re-engagement programming for specific groups of students and b) plan prevention programming for students currently enrolled based on the profiles of students who have dropped out.

Prior Research

This study is built on prior research conducted on the early warning indicators, specifically for dropout outcomes. Most of this analysis is based on the work of Robert Balfanz & Vaughan Byrnes at Johns Hopkins University and initial segmentation studies with five school districts in the Commonwealth.

Studies have consistently determined that students' in-school behaviors (e.g., attendance, behavior, and credit accumulation/course marks) were more predictive than demographic factors, and in particular that students who fell below attendance rates of 90%, failed courses, or received one or more suspensions, were the most likely to drop out.

This analysis uses the most recent data on dropouts grouping them according to past attendance, behavior, and course taking patterns. It uses descriptive statistics to summarize students' academic behaviors over the four-year period immediately prior to their dropping out, to identify the proportions of dropouts who fell below key indicator levels, and to create "flags" based on that analysis.

While the focus remained on those key indicators established by prior research, the demographic background and prior achievement levels (the MCAS and ACCESS state tests) of students were also examined. In addition, dropouts were segmented into groups based upon their age, grade, and the rough number of courses that students were short of obtaining projected graduation requirements (note: local requirements for graduation vary district by district). The analysis around course accumulation is intended to show the "academic distance" to meeting local graduation requirements (e.g., those within only a couple of credits of graduation vs. overage/undercredited who may need a different type of high school completion programming).

Using the Analysis

Data for 2021-22 dropouts were analyzed for districts that had 20 or more dropouts. Users may also filter the results and see data for subgroups (Low Income, Students with Disabilities, English Learners, First Language Not English, and High Needs) that had 10 or more students who dropped out.

For these students, we examined longitudinal data readily available at the state level for the year they dropped out and the three years prior to their dropping out. All data analyzed came from the Massachusetts' Student Information Management System (SIMS), Student Course Schedule (SCS), School Safety Discipline Report (SSDR), and MCAS and ACCESS test results.

This analysis enables district and school leaders to identify the student subgroups that are unlikely to graduate without an intervention. The results of this type of analysis can provide guidance to district and school leaders concerning what types of interventions and programming different populations need. It can enable districts to use resources strategically to redesign existing options or to create new ones tailored to the specific needs of each group, based on age and distance from high school graduation.

The analysis includes research-based "flags" that may indicate student disengagement from school and serve as a contributing factor to their dropping out. These flags (i.e. Attendance issues: attendance below 90%, Behavior: one or more suspensions, and Course failures: failing one or more courses) can serve as a signal a student was going to drop out. Studies done by Balfanz and Byrnes indicated that 90% of dropouts showed at least one of these flags in the year previous to dropping out and the vast majority exhibited them as far as three years before dropping out.

Combined with EWIS data (that details information about currently enrolled students) and locally held data and information, especially around students' complete course taking and credit accumulation tied to local graduation requirements, this analysis can provide additional clues concerning programming and interventions in districts, schools and communities.

Last Updated: March 24, 2023