Context for decline in special education

Category 2005 2008 2011
LD 4.14 3.77 3.43
ED 0.72 0.62 0.54
ID 0.81 0.71 0.63
Data from

Percentages of students 6-21 yrs
identified in categories of LD, ED, or ID
for US schools

Ever wonder why the number of students identified as having learning disabilities (LD), emotional or behavior disorders (ED), or intellectual disabilities (ID) in the US is declining? The numbers are not declining, you say? Hmmmm. Well, as the table here shows, they are, indeed, declining. I drew these data from the child counts mandated under the Individuals with Disabilities Education Act. I used the data for percentage of students ages 6 through 21 served under IDEA, Part B, as a percentage of population, by disability, and state. (I used 6-21, because those are reported consistently across the years; therefore these percentages are lower than one might expect for school-age students.) As indicated, one can check my work by referring to the data tables for US Office of Special Education Programs’ state-reported data. And, as the data show, it’s not just LD that’s decreasing.

Some of our colleagues will argue that the reason for the decline is improvement in preventative services, especially as represented by response to intervention or instruction methods that have been touted extensively in the time represented in the table here. I’ve discussed the question of whether RTI is reducing identification of LD on LD Blog more than once (i.e., does RtI reduce numbers of children in special education?), so I shan’t return to it now.

Early intervening services may well be doing good things for students with reading problems and even some behavior problems. But they probably have had little to do with the decline in identification of students with ID. Instead, let’s examine two different possible contributing factors. One is diagnostic substitution and the other is (for lack of a better term at the moment) the sped=bad perspective.

Diagnostic Substitution

Some observers have noted that declines in the numbers of one category are associated with increases in the numbers of another category. Shattuck (2006; Why has autism prevalence increased?) provided pretty solid evidence that this, along with other factors such as changes in the criteria used for identification, accounts in some part for the rise in the prevalence of autism (Matson & Kozlowski, 2011), for example.

Perhaps this phenomenon is reflected in the child count data. If the declines in LD, ED, and ID were a consequence of increases in diagnoses of autism, then we ought to see something like a 1% increase in the number of students identified as receiving services for autism in the OSEP child find data. That’s about the amount that ID, ED, and LD declined (1.07%) over the period from 2005-2011. (Remember that these percentages are for 6-21, not school ages, so they’re already muted.) Well, of course! The number of students receiving special education under the autism category has increased.

Category 2005 2008 2011
AU 0.29 0.44 0.59
OHI 0.85 0.97 1.07
Data from

Percentages of students 6-21 yrs
identified in categories of Autism or
OHI for US schools

However, as the data show, that number has only risen 0.3%. Not enough to account for the 1% decline. Another possible substitute diagnosis, other health impaired (where students with ADHD or ADD might receive services), has also increased. It’s grown by 0.22% over the time period.

In aggregate, then, autism and OHI have grown by a bit more than 0.5%. So, maybe diagnostic substitution accounts for about half of the decline in ED, LD, and ID.

Special Ed in Decline

Category 2005 2008 2011
SE 9.15 8.80 8.43
Data from

Percentages of students 6-21 yrs
identified as needing special
education in US schools

There is another row of data to add to this array. The percentage of students ages 6-21 receiving special education has decreased almost 3/4ths of a percent. So, overall, there’s a decline in percentage of students served, and it’s led by a greater than 1% decline among LD, ED, and ID. Meanwhile, there’s an increase in those receiving special education for autisma nd OHI.

Again, some educators will see this as a beneficial result of improved services in general education, better pre-referal work, and so forth. This may be the case. There may be cheers, celebrations, and smiles. Children have been served well and, therefore avoided special education.

An alternative case builds on that very thinking. The case is that special education has become something to be avoided. That it’s bad. One doesn’t want students “put in special ed.” This is a point made by my colleague Jim Kauffman on more than one occasion (he can add the appropriate references in a comment!). It’s a point that’s hard to prove, of course, but I think Jim had some examples.

There are some intriguing trends that run consistent with this perspective. They also loop back to the decline of LD and ID.

When people search for things on the Internet, what terms do they use? Google has a substantial record of those searches that cover a time span similar to the one I’ve been considering here. The accompanying figure shows the relative frequency of searches on Google for “learning disability” between 2004-2013. The picture that emerges for LD is that people are searching less and less frequently for “learning disability.” Now, this includes other use of the term (e.g., searchers from the United Kingdom might be using it to search for what people in the US can ID or used to call mental retardation; I ran the analysis with only US data, and it was pretty similar), but the overall picture is clearly one of declining interest.

And it’s not just declining interest in LD. Here is the comparable analysis of the searches for “mental retardation” (I used that in place of ID, given that the terms had changed during the period under consideration and evidence that ID had not taken hold in the general public). In fact, another figure shows that ID and LD declined in a highly correlated way.

Meanwhile, the interest, as expressed by Google searches, in ADHD and autism is consistently high. Indeed it is remarkably higher than the interest in LD and ID.

Yet another indicator of the decline: Membership in professional organizations. The Council for Children with Behavior Disorders (CCBD) and the Division for Learning Disabilities (DLD) have approximately half the number of members that they had in 2004. Overall, the Council for Exceptional Children (CEC) is about 60% of the size it was in 2004. Only one group has grown appreciably in membership since 2004; the Division for Children with Developmental Disabilities which rechristened itself as Division on Autism and Developmental Disabilities (DADD) has almost three times the membership it used to have. (I doubt many people who left DLD and CCBD moved to DADD; that would be a very different form of diagnostic substitution!)

Most people who leave CCBD and DLD also leave CEC. So, it’s an overall decline in affiliation with special education that probably accounts for the losses. Although I don’t have the data, I suspect that sibling professional organizations (e.g., Council for Learning Disabilities) have seen declines in membership during the same time period.

There’s nothing inherently bad about a shift from one categorical group to another. I’m not talking about turf, not pitting LD and ED against autism. I’m noting that the numbers of students getting services have decreased, the number of people joining organizations aimed at helping those students has declined, and the public inquiry about those topics has waned.

Well, this is not a fancy and sophisticated argument. It’s interesting that during the time when special education numbers appear to be in decline, lead by the high-incidence categories, the interest in them (as reflected by Google searches and membership in professional organizations) is also in steady decline. As we race to promote prevention and to improve services in general education, I hope we don’t run away from securing services for students who need special education.

I’m tempted to make comments about safety nets….


Matson, J. L., & Kozlowski, A. M. (2011). The increasing prevalence of autism spectrum disorders. Research in Autism Spectrum Disorders, 5, 418-425. doi:/10.1016/j.rasd.2010.06.004

Shattuck, P. T. (2006). The contribution of diagnostic substitution to the growing administrative prevalence of autism in US special education. Pediatrics, 117, 1028-1037 doi: 10.1542/peds.2005-1516.

Published by

John Lloyd

John Lloyd--founder and lead editor for

One thought on “Context for decline in special education”

  1. My two cents on the decline looks at two items, the economy and referrals,
    and then a possible bit of good news, if it’s good that our numbers
    increase. I think the decline parallels the economic decline of the
    country. While I am far from knowledgeable about how special education is
    funded throughout the country (here in NY, our 700+ school districts get
    state money and then, outside of NY City, ask the taxpayers each spring to
    vote on a local school budget which adds a significant amount to the state
    dollars), I assume most states were under financial pressures to reduce or
    level fund education. With the unwillingness of local taxpayers to support
    significant increases in their local share (we now have an unworkable 2%
    cap on local school budget increases here in NY ), schools districts have had little
    recourse but to trim the budgets or at least hold them to minimal
    increases. With the poor economy keeping higher salaried veterans teachers
    from retiring and normal inflation, local schools have been under pressure
    to reduce their expenditures and I read of staff cuts in many districts. My usual review of
    the NY Times shows very, very few new special education jobs being

    Now, I think I’m correct in that the most assured way to get a child
    classified is to refer that kid. I also assume that most referrals come
    from teachers, although knowledgeable parents will always get their kids
    what they need. Thus, I might guess that referrals are down across the
    country. With financial pressures on schools, it would not surprise me if
    teachers were dissuaded from making those referrals. While I think this is
    illegal, I have heard a few stories of principals telling teachers not to
    make referrals. Of course RTI and other newer strategies may have some
    impact, but fewer referrals mean fewer classified kids.

    As for the good news, well here in NY we have started testing kids on the
    Common Core using new tests, even while we await for PARCC. The reports
    are that these new tests are really tough on all kids and that we should
    expect scores to drop by at least 30% or more, and that is for all the
    kids taking the tests. If there is such a drop, I’d guess that children
    with some learning/behavior issues will do far worse and the schools may
    look to increase their special education referrals out of fear. (We are
    also a state that has implemented an teacher and principal evaluation
    system based in part on these scores.) Thus, I might see some increases as
    the Common Core tests start showing the need.

Comments are closed.