Special Education II

How integrated are special education and general education students

The last post dealt with the incidence of special education across the state i.e. what fraction of a district’s students have a disability. Recall that three disability categories account for 75% of all special education students:

Separate or Integrated?

The only data published about how special education students are served is the percentage of their time that they spend in a regular general education classroom setting i.e. one “which includes at least 50 percent nondisabled students.”

Nearly three-fifths (58.2%) of Special Education students spend at least 80% of their time in general education classes with majority non-disabled classmates. These are the students with the least severe disabilities. A sixth (15.9% to be precise) of disabled students attend regular classes 40-79% of the time and another sixth (17.4%) are in regular schools but spend less than 40% of their time in general education classes with non disabled students. Students in Special Day Classes fall into this category.

How much time a student spends in general education varies by disability. 67% of those with Specific Learning Disabilities, 78% of those with Speech or Language Impairments but only 31% of those with Autism spend at least 80% of their time in general education classes. Conversely, only 8% of those with Specific Learning Disabilities, only 5% of those with Speech and Language Impairments, but 38% of those with Autism, are in Special Day Classes or similar programs where they spend less than 40% of their time in general education.

The horizontal bars above don’t extend to 100% because some special education students are not in standard public schools. 6% of SpEd students are served in a preschool setting (defined as “a home, regular early childhood program, separate class, or at a service provider location”) and another 3% are in separate schools or other settings (“receive special education programs and/or related services in a separate school, residential facility, homebound/hospital, correctional facility, or are parentally placed in a private school”).

Preschool and Special School Settings

Students with a Speech or Language Impairment or Autism account for 90% of the students in preschool special education settings. Presumably, these are cases that are easy to diagnose early. At the opposite end of the spectrum, even though over one-third of all special education students have a “specific learning disability”, only 0.01% of preschool special education students have this diagnosis. Clearly, “specific learning disabilities” don’t manifest at an early age.

Over half of all special education students are diagnosed with a Specific Learning Disability or Speech/Language Impairment but these tend to be less severe disabilities. Only 9% of students who receive their special education services in a separate school or other setting have either of these disability classifications. Instead, as figure X shows, the Special School settings contain a lot of students with Autism, Intellectual Disability, Emotional Disturbance, or Multiple Disabilities

(Not) Explaining the Variation

In the last post, we saw that the share of White and Asian students in a district is a significant predictor of the percentage of Black and Latino children who are diagnosed with a disability. This persuaded me that diagnosis is not based on objective medical or educational criteria but on comparisons with district norms. A Black student surrounded by high-achieving White and Asian students is more likely to be placed in special education than the same student would be if surrounded by other Black students.

We also saw in the last post that the percentage of children with disabilities varied significantly from district to district. Some districts simply have a greater propensity to place students in special education.

I formulated a couple of hypotheses:

1. Districts with high numbers of special education students would not have elevated levels of the most severe disabilities (i.e. those requiring separate schools or more isolated classroom settings). Instead, those districts would simply have more students who spend 80% or more of their time in general education.

2. Districts with high numbers of special education students would not have proportionately high numbers of all disabilities but would instead have high numbers of some particular disabilities that might have squishier definitions e.g. “Specific Learning Disability” or “Speech/Language Impairment”. This would indicate that those districts were using those diagnoses as dumping grounds for struggling students.

These hypotheses both turned out to be untrue.

1. As the share of special education students in a district increases, the fraction that are in general education at least 80% of the time does not increase. It actually trends down but this is not significant and the R-squared is only 0.04.

2. The correlation between the overall special education rate in a district and the share of students suffering from particular disabilities was low for each of the most common disabilities (-0.03 for Specific Learning Disability; -0.01 for Specific Learning Disability; 0.12 for Autism).

That was a bit of an anticlimax, wasn’t it. Clearly, I have no conclusions of any value to draw! I have no idea why some districts have special education rates so much higher than others.