The disruption to education caused by COVID-19 is well-documented, but less is known about the implications facing the special education system. A new study from scholars at the Michigan State University College of Education explores how less K-5 students with disabilities were identified for needing support, especially among Black, Asian and economically disadvantaged students.
Declines in special education identification during remote learning
Published October 2024 in Educational Evaluation and Policy Analysis, the study examined special education identification trends from a sample size of over 2.9 million Michigan students from 2012 to 2023.
During the 2019-20 and 2020-21 school years, identifications fell by a staggering 19% and 12% respectively. Specific learning disabilities — brain-based disorders that affect a person’s ability to understand or use language, read, write or do math — were the most impacted. Economically disadvantaged students and students from minority backgrounds, particularly Black and Asian students, experienced disproportionately larger declines in special education services.
The study attributes the disruption in special education identification to the challenges in providing pre-referral intervention and other required evaluation procedures during remote learning across most of state’s public-school districts.
Identification rates returned to pre-pandemic trends during the 2021-22 school year and saw continued growth into 2022-23.
However, the sharp decline in identifications from 2019-21 could have long-lasting implications, according to the researchers.
“There is a lot of evidence suggesting that the earlier a student is identified and receives support, the better the long-term outcomes are,” said Matthew Guzman, the study’s second author and fourth-year Ph.D. student in the Education Policy program. “Missing that window could have lasting negative effects, which is why it is so important to address these gaps now.”
Why in-person learning matters
The research shows that districts with fully remote learning had the steepest declines in special education identification. This trend appeared less severe in hybrid and fully in-person settings, highlighting the importance of physical classroom interactions in identifying and supporting students with disabilities.
The scholars also address the complex process schools must follow to identify specific learning disabilities, which involves weeks of targeted interventions, detailed data collection, and strict criteria to ensure accurate diagnoses.
“Even before the pandemic, our evaluation processes for specific learning disabilities were difficult for schools with strained resources to implement and were not sensitive enough to detect specific learning disabilities before grades 3 through 5,” said Associate Professor Adrea Truckenmiller. “These data from the pandemic show how events can exacerbate already vulnerable systems.”
Truckenmiller also notes that providing early intervention in grades K-3 can help close the achievement gap between students with and without disabilities. It also reduces the need for intensive instruction outside the general classroom, allowing more students with disabilities to stay fully included in general education.
Spartan scholars on the study include:
- Bryant Hopkins – Former postdoctoral researcher with the college’s Education Policy Innovation Collaborative (EPIC)
- Katharine Strunk – Former director of EPIC
- Ph.D. student Matthew Guzman – Education Policy program
- Professor Scott A. Imberman – Department of Economics and Education Policy Ph.D. Program
- Associate Professor Adrea Truckenmiller – Department of Counseling, Educational Psychology and Special Education
- Associate Professor Marisa Fisher – Department of Counseling, Educational Psychology and Special Education
The scholars hope the study will inform future policy on improving early identification processes, addressing disparities in special education services and ensuring equitable support for students with disabilities, especially during times when resources are stretched thin.
More: