DEBORAH PRAISSON-STAT
Good morning. And welcome to the (UNINTELLIGIBLE) to decision
making. My name is Deborah Praisson-Stat (SP?) , I'm from the
Office Of Special Education Programs and I will be your moderator
for this session today. During this session, you will hear a
30 minute presentation by the authors of the issue paper on this
topic. Followed by nine minute presentations from the issue paper
respondents. There will be a 30 minute period for question and
answers at the end of the session. In order to provide ample
time for the presentations, please reserve your questions until
the end. This session is being video and audio taped for later
broadcast and the world wide web. It is therefore imperative
that we remain within the time limits assigned. In addition,
if you would like to address a question to the presenters during
the question and answer session, please use a microphone that
will be provided for you so that everyone can hear your questions.
Our first speakers will be doctors Don MacMillan and Gary Siperstein.
Dr. Donald MacMillan is a distinguished professor of education
at the University Of California, Riverside. And holds appointments
in the UCLA Medical School, and at the University Of Massachusetts,
Boston. He has published many articles and chapters, is a member
of the National Research Counsel Committee On Representation Of
Minority Students In Special Education. And has received numerous
awards for his contribution to the field of special education.
Dr. Gary N. Siperstein is professor, and the director of the
Center For Social Development And Education at the University
Of Massachusetts, Boston. Dr. Siperstein has published over one
hundred articles and chapters on the social relationships and
social development of children with learning problems in educational
settings. He is a past editor of the Research Monograph Series
Of AAMR, and associate editor of The American Journal Of Mental
Retardation. Welcome.
DONALD L. MACMILLAN
Good morning. Given the, uh, half hour limitation that, I guess,
I'll be doing most of the talking. Gary can write well, but he
doesn't speak. So, uh, I will be the designated speaker here.
There is a, a bit of a disconnect between the title of the paper
which we had entitled something to the effect of, uh, Learning
Disabilities As Operation To Find Better Schools, and the topic
on decision making. I think there is a connection there, and
let me try to make it as we go on through the talk. Let me just
make a few points before launching into kind of the formal presentation.
Um, what we'll be stressing here today, is that the public schools
identify a group of children as learning disabled, which differs
from the authoritative definitions in the population one might
suspect if we apply those parameters systematically to all the
children in the public schools. Secondly that the schools are
confronted with an issue of how to serve children who they perceive
need help, when some of these youngsters don't neatly fit the
criteria that are specified in a given state to be learning disabled,
or some other designation. And I would contend that the children
who do get served, do share a common characteristic. And that
characteristic is that they do deviate rather sufficiently from
their class room peers. That is, the local peers, maybe not on
some national norms that we might establish. And third, that
it's been our observation that the schools have been reluctant,
particularly in recent years, to identify children as mentally
retarded, and as behavior disorder, because it has certain consequences.
DONALD L. MACMILLAN
One, perceived stigma of the label of mental retardation. And
secondly, with children with behavior disorders, the inability
to suspend or expel once they are identified under that category.
And the LD has become a gentle designation in which we find some
children who might, in our opinion, better be described as mentally
retarded or behavior disordered. That's the first overview.
Therefore, there's one error on this overhead. Um, that the schools
in trying to comply with IDEA are not free to simply apply that
template over children and apply a psychometric criteria to who
is identified as LD. But rather are required to follow certain
prescriptions specified under IDEA. And one of the results of
that, is the children who end up in the category, in some cases
what I will call False Positive LD Children. What I mean by that,
they are called LD, but in fact they don't meet the specified
criteria in that given state. The errors in the next one, that
it includes, it excludes a group of children who probably meet
the criteria, but for reasons of never getting referred by a class
room teacher, never coming to the attention of the school personnel.
Never really get into the process, and are subsequently not identified
as LD, even though they might meet the criteria. Thirdly that
the group of children who are identified as learning disabled
nationwide, very considerably in terms of the severity of their
deficits and achievement, and also along some other behavioral
characteristics such as behavior problems that might exist co-morbidly,
uh, with the learning disabilities they present with. And that
essentially, the public school's definition of LD reflects the
perceptions of the on site educators in terms of which children,
in their opinion, are most in need of help, and secondly would
benefit from the, the services that are available in the name
of special education at that particular school site.
DONALD L. MACMILLAN
I would also like to point out that LD being a rather recent newcomer
to the disability listing, uh, came into existence after disability
categories already had kind of carved out their turf. That is,
mental retardation as already recognized as a disability category.
Serious emotion disturbance or whatever term it was going by
at that time also was on the books. And therefore in attempts
to identify and specify criteria for a group of children who needed
help but didn't quality for that help under the pre-existing categories,
LD was put in the position of one saying, these are not children
with sensory or physical impairments, they are not mentally retarded,
they are not due to cultural, or, um, economic impoverishment.
And so we have these exclusionary criteria and I want to talk
about them later, because I think that essentially they are not
applied rigorously when we consider cases, and put them into the
LD category. Lastly, with the reauthorization in 1997, a new
exclusion was really brought on the books. And that was that
children were not to be identified as learning disabled who simply
had not been exposed to adequate instruction up until that time.
So in addition to the other disability categories, and the cultural
and linguistic issues, the issue of a lack of adequate instruction
was brought into place. Again, a rather difficult thing for us
to document when we start to exclude children from the particular
category. This is the familiar definition of learning disabilities
we put on. In red what I tried to do was to highlight some of
the exclusions. And the two that I want to focus on, primarily
and the talk today, deal with the excluding children who are mentally
retarded. And excluding children due to economic or cultural
deprivation. And you'll notice at the bottom I just made the
note, that nowhere in the Federal definition is the issue of severe
discrepancy really presented.
DONALD L. MACMILLAN
Let me make a distinction. This represents one of several authoritative
definitions of mental, uh, of, uh, learning disability. What
we also have, though, what we qualify children, are the classification
criteria. That is, after the definition is put in place, then
it's kind of a toss up, you go to each and every state and ask
the state what are the criteria you are going to employ in establishing
whether or not a child is eligible, uh, as LD? Some have described
the increase in the public school identification rate of LD as
being of epidemic proportions. These figures, and I think Dan
may comment on this later, between the period of the full implementation
of the public law in 1990 to '93, we see a hundred and ninety
eight percent increase in the number of children being identified
as LD throughout the country. And the explanations for this have
varied quite a bit. Uh, Dan Hallahan in the paper, um, posited
that part of that is the fact that we've simply matured as a field.
And the diagnostic process has become improved, and therefore
we're now able to detect cases that heretofore have been undetected.
Uh, he also pointed to the increase in what is referred to as
the new morbidity. A term Al Valmeister (SP?) had introduced,
looking at some of it (STAMMERS) effected development of children,
in it's most extreme forms may result in severe forms of retardation,
but possibly milder forms result in a little less, um, pronounced
manifestation and it might present as a learning disability.
Reed Lion (SP?) was quoted in the L.A. Times as characterizing
LD as basically not being a disability, but rather being a failure
of general education. In which he characterized it as a sociological
sponge that soaks up the failures of general education. And some
of the work that we have done at UC Riverside, um, would suggest
that children are knowingly classified as LD who the public schools
know full well probably better fit the definition of mental retardation
or emotional disturbance. Let me go to the process that is prescribed
under IDEA, which the schools are compelled to follow, even though
we as researchers sometimes do not have to. And that is, when
we see children in the public schools who experience severe under
achievement, we typically try to provide them with some rather
less intense interventions that would be available in special
education.
DONALD L. MACMILLAN
So if the youngster is in a school where Chapter One funds are
available, we might provide some support under that. Or in the
name of pre-referral intervention, try certain interventions that
can be implemented by a regular class teacher, at least within
the regular class context. If youngsters respond to those treatments,
they obviously remain in the regular tract as a normal student,
so to speak. If they fail, however, frequently they are then,
at that point, referred for formal assessment to find out what
does this child qualify for in terms of special (STAMMERS) education
services, if anything? If the child is found ineligible, presuming
they go back into the regular class, however, if in fact they
are found eligible as mentally retarded, as learning disabled,
as emotion disturbed, or one of the other designations, they are
now able to go for special education. Our experience would indicate,
in the data that we have reported in a number of different publications,
when children are found to be in in need of help, the are almost
invariably classified as learning disabled. Uh, we have found
youngsters with IQ's as low as 53 still being called learning
disabled. When I came into this field, that, I saw down syndrome
children inevitably classified and put in programs for trainable
mentally retarded. As my career proceeded, they came into programs
for educable mentally retarded. Several years ago I'm involved
in a court case in Connecticut and going out to the schools, to
different school sites, every LD, or every down syndrome child
I saw was in an LD class room, or an LD program. So the field
has changed in terms of who is in what programs, and LD, it's
going to be my contention, is basically the being operationalized
in the school as absolute low achievement, not discrepant low
achievement, and frequently that absolute low achievement is accompanied
by expected under achievement because of a low, uh, IQ score acing
that. We've tried to simplify this in some of the research we've
done in trying to characterize what is the school doing? Is it,
in fact, lawful?
DONALD L. MACMILLAN
What we've tried to do is to break it into three stages. One
is, a youngster, no matter what IQ, achievement discrepancy they
might present with, will never get into the educational system
special ed programs unless they are referred by their teacher
or by their parent. Referral is a key element in this. (WORD?)
years ago, said it's the most important element with these judgmental
categories. (WORD?) Sigmond has commented similarly on the importance
of teacher referral, and in some cases it can be by a parent as
well. Frequently that is more apparent in upper and middle class
families. Once referred, the assessment process comes into place.
And it has been hammered heavily in the mental retardation literature
over the years in terms of IQ testing, and eligibility criteria.
But essentially the child is (STAMMERS) secondly assessed to
find out if in fact they're eligible. And then thirdly this goes
to a committee as prescribed by IDEA, where it's to be a team
decision, not an individual decision. Not based on simply psychometric
profiles. But the team considers whether or not this youngster
should be served in special education. At each of these three
stages, it's been our contention that different norms apply.
When a teacher makes a determination to refer, that teacher typically
uses local norms. That is, comparison to regular class peers,
is this youngster's achievement so deficient that they are not
keeping up with the other kids in the class? Or some absolute
norms that the teacher might have. This is where a second grader
ought to be at this time, if they are not there, and they sufficiently
deviate, the decision refer may take place. The second stage,
national norms come into play.
DONALD L. MACMILLAN
In which the psychologist administering the test compares that
child's performance, on both intelligence and in achievement,
to national norms of how children are doing. And establishes
whether they meet the criteria specified in the state, which usually
requires across most states a severe discrepancy. And thirdly,
when it goes to the committee, the local norms that the teacher
considered, the national norms the psychologist considered, and
socio-cultural factors enter into the decision. Unfortunately,
some of our special ed teachers are not that good. And when teachers
see a kid saying, Suzy really qualifies, we think they would benefit
in the best of all worlds. But you know what? At this school
it's not the best of all worlds. Our special ed teacher is weak.
A teacher might decide to keep the youngster in general ed, feeling
that they will be better served there, and I don't think that's
an unwise kind of decision. But at each of these stages what
we see is a degree of subjectivity enters into the process. The
teacher's decision to refer, whether or not a divorce accounts
for the child being upset, and therefore we shouldn't place him
or her, fits into the process as well, and some of the suspected
issues of gender and race might enter into it as well. Okay,
and finally, we've characterized these stages in a paper that
Kathy Boshun (SP?) is the senior author of, um, saying at stage
one it's an issue of relativity. How does this perform relative
to classmates? The second stage, one of acceptability. Is this
level of performance acceptable in terms of is it congruent with
what we would base, expect based on their performance in mathematics,
let's say, or an IQ test. And finally at the last stage, that
of profitability. If we put the youngster into special education,
what is the likelihood they will profit from it compared to what
they're doing at the, in the regular class. With both Gary and
my interest in mental retardation, one of the things that has
jumped out at us over the years is trying to find out where these
youngsters who used to be classified as mentally retarded turned
up.
DONALD L. MACMILLAN
And low and behold, one thing we're finding is many of them are
cropping up in LD programs. We have been engaged in a project
where we have taken children referred to student study teams,
doing psychometric work ups on the children, applying research
criteria to it, that is reflecting the California criteria for
learning disabilities and mental retardation. And then looking
at what the school does with the child. So if we look across
the top, the research diagnostic categories, we use an IQ of 75
or lower to put the youngster into the category of mental retardation,
having very low achievement in getting referred. The borderline
group is really a, a throw back to the old borderline mental retardation
group, which used to go up to IQ 85. So we have that group although
it's truncated 75 to 85. Then children with LD, which in California
requires a standard, uh, discrepancy of one and a half standard
deviations. And then we have kids that don't meet any of the
criteria. What you can see here in the red are what we consider
errors. That is, if you take the research diagnostic criteria,
and say those are right, then look at what the school does with
it. What we have our a number of children who are mentally retarded,
and borderline IQ children who are non-discrepant in their achievement.
What you see here are the red cases that have either been left
in general education, with no special ed services, or have been
classified as learning disabled. And a substantial number. What
you'll notice, the one blue thing is the children with IQs under
76, 81 in number, only six were classified as mentally retarded
by the public schools. The rest were either LD, put into one
of the other categories, or unserved. We'll also point out that
54 percent of the children in our sample, which is a well represented
sample of the kids referred to student study teams. 54 percent
had IQs of 85 or lower. Suggesting that in what are fairly urban
school districts, that the children who are most difficult to
teach in the perception of teachers, in terms of who they refer,
many of them are low cognitive children. Whether there are a
number of discrepant low achievers who don't get referred, we
can't say from our project.
DONALD L. MACMILLAN
I think that is a Beta set which is absent to special education
literature. I don't know anybody who has really looked at the
magnitude of an un-referred group of children who meet the criteria
for LD or for mild mental retardation. We looked at just the
cases who had IQs below 75. In our first run, look at the first
hundred and fifty out of three hundred and twenty eight cases.
This is how they came out. You will notice, again, our classic
six cases, um, that that class that is mentally retarded. But
a majority of those cases getting put into the LD category. And
let me comment a little bit here. Why is it that the schools
ignore the criteria specified in the California Education Code
from our data, and the exclusionary criteria? Jay Gottlieb has
commented on the New York City situation, of saying we're simply
looking at scant resources that could be made available for kids
who are struggling. And the schools are tending to overlook some
of the criteria in order to get those scant resources that kids
need in help. Lori Sheppard commenting on the Colorado situation
made the observation that teachers and the school's personnel
would be more adamant in implementing some of these criteria,
if they didn't feel that denying eligibility was denying the needed
help that a child has. And we were told by the teachers we worked
with then, they said, you guys at the university are more concerned
with what kind of kid it is, we're really more concerned with
what to do with them. And therefore we don't agonize as much
over which pigeonhole to put them in, we're more concerned with
how are we going to teach them reading? How are we going to teach
them math? How are we going to teach them life skills? One of
the things that seems to me to have happened, is that the importance
of differential diagnosis, of is the kid really LD, are they MR?
Uh, has changed over the years. When I was back to the 1960s,
when I entered this field, we agonized over is the kid LD or MR?
And one of the reasons was, that at that time, the eligibility
criteria had profound implications for where the kid would be
taught, and what they would be taught. If you were called MR,
if you look at the Robinson And Robinson checks published in 1965,
they said mild MR kids are educated in special day classes. End
of debate. So if you were called mildly retarded, you were typically
put into a special day class, and you got a functional curriculum.
DONALD L. MACMILLAN
Conversely, if you were called LD toward the end of that decade,
you typically were taught in a resource kind of setting, and remediation
of skills was emphasized. When we go to the public schools today,
and ask the situation around the turn of the century, they say
it's irrelevant. It doesn't matter what we called them, because
placement is going to be determined in terms of LRE. So we can
take a kid who's IQ is however low you want to look at it, call
them LD, we can still put them into a special day class, and conversely
in terms of the treatment, we can give them a functional curriculum
in the IEP negotiation. So that the importance of the classification
category for the child is not as important today as it was back
at an earlier point in time. Nevertheless it results in considerable
variability in terms of the number of kids that are LD and the
nature of it. Taking out, and this is, this is twenty second
annual report, not the twentieth, they'll look at extreme states.
In Georgia we're looking at three point two eight percent of
their school children being called LD. Conversely in Rhode Island
where over a nine and a half percent of their kids being called
LD. This kind of variability reflects probably variations in
criteria. Some states use a standard (WORD?) discrepancy of one
versus one and a half standard evasion. Some are regress discrepancy.
If you look at the Cecil Mertze's (SP?) work, um, and the surveys
that he has done of the state criteria, you see considerable variability
across states in the LD criteria, um, implemented. We also see
variability in terms of gender, with a split of approximately
60 to 40. Yet when we look at the Shaywitze's (SP?) data from
Connecticut, there is no gender difference. True in the MR category,
where we've always had it, the gender difference in favor of males,
being in favor means that you're more likely to be identified.
Uh, work that Dan Reschly did in Arizona showed again, when you
apply a psychometric profile, that gender difference disappears.
So there are other contextual factors operating in terms of who
gets into the system, and who gets classified into what, uh, category.
Let me turn my attention to say, when we see this confound of
mental retardation and learning disability, it presents several
problems to us.
DONALD L. MACMILLAN
The work that Jay Gottlieb has done in New York City, in comparing
LD kids in the inner city to LD kids in suburban school districts.
When we looked at the mental retardation, um, situation, mental
retardation is being reserved for only the most patently disabled
children. And we're seeing the same thing in California, we saw
the same thing in Connecticut. If a youngster was classified
as mentally retarded, I think that most of us would agree that
youngster does have severe problems. In LD we see a slightly
different case. Also the fact that the condition of mild mental
retardation is a phenomenon almost exclusively of low social class.
And let me just go to the next thing. This is a, a table that
was presented, um, in the case of Marshal Versus Georgia in which
Dan Reschly was one of the expert witnesses testifying. I want
to point out to you here, the blue part of the bar indicates a
form of mental retardation where the IQ is below 50, and where
there is central nervous system involved and apparent. The red
bar is where IQ is above 50, and central nervous system is evident
in the case. Such as Cerebral Palsy with low functioning. And
the green form, the one of interest to us here, is a form of mental
retardation where the IQ is higher than 50, and there's no apparent
evidence of central nervous system damage. The social case categories
to up from top to bottom. The top being the highest, down to
the bottom being the lowest. And what is apparent here, as you
move down and get to the lower social classes, is green form.
Low I, or IQ above 50, but with no central nervous system damaged
becomes more and more apparent, and more dominant.
DONALD L. MACMILLAN
Okay, this study was done in Aberdeen Scotland with all White
subjects. So we don't have the confound of race that we have
in many of the studies done in this country. We've long known
that this form of mental retardation is linked to low social class,
and heaven knows what causal factors that are associated with
low social class. You don't see it in the suburban schools.
And it's primarily a function of the more urban schools. You
can go to the next thing. And if Gottlieb looked at the classification
of the (STAMMERS) urban children, the most here, the mean IQ in
the suburb and urban schools are about the same, 54 and 55. Much
lower than what it was in the '60s where we'd had a much higher
mean IQ for our MR category. In a survey done in 1992, the mean
IQ in the urban setting for the LB children now (UNINTELLIGIBLE)
to 81. But in the suburban school district, the mean IQ for the
LD's was over a hundred and two. So in terms of intellectual
functioning, an urban LD and a suburban LD look like quite different
children in terms of the aptitude level. On a survey done in
1984, 41 percent of the children classified as LD in the New York
Public schools had IQ scores between 70 and 85. That old borderline
category of mental retardation. An additional seven point five
percent had IQs below 70. And still were classified as LD. And
here again, just further confirmation of it, if you look at 16
percent in another survey had IQs of less than 70. And it's not
just in terms of aptitude that the urban LD kids differ. When
we look further at his data, in a school district with high degrees
of poverty, you had 80 percent of the students in the districts
living in poverty. 90 percent of the LD students in the New York
Public schools are on some form of public assistance for their
families.
DONALD L. MACMILLAN
But 95 percent of the LD kids were minority group children. Let
me just say for perspective, 93 percent of the district students
are minorities, so this is not a statistically significant difference.
But 19 percent were from foreign born, were foreign born, and
44 percent of them came from families where English was not the
primary language spoken. Now just submit to you, it does not
appear that the exclusionary criteria for LD are being rigorously
applied in New York City schools. If cultural and economic properties
do exclude you from eligibility, if mental retardation is to exclude
you from eligibility for LD, it's not being rigorously opposed.
Conversely, I would argue, the New York public schools are saying
we got a group of kids who need help, and we're gonna help them.
And what we have found in our research is that once they make
that commitment, they say, okay, you guys with IDEA are going
to make us go through the eligibility criteria by applying these,
these tests. We'll do it, but you know what? We can do creative
financing with them. We're going to find the child eligible if
we think that kid needs help, to the point where we have found
one discrepancy that qualified a kid for eligible was between
an achievement test score and a Peers Harris. (SP?) And the
schools are not that dumb. They know what they're doing. And
they're saying, we know this kid is mentally retarded, there's
no upside to classifying him or her as mentally retarded.
DONALD L. MACMILLAN
We can take care of that in the IEP, and an LRE, and LD is a much
more acceptable definition in classification for children. Yet
that presents problems, because we do see this disconnect between
what the research criteria are supposed to be for LD, and the
characteristics of the children currently being served in the
public school as LD. There are several things with this that
strike me as problematic with what we are doing now. One is with
the one time only assessment that we do to qualify a child. When
we get a low achievement test score, interpreting that becomes
difficult. The new exclusionary criteria where it should not
be due to inadequate instruction comes up here. Call your attention
to some work that Shep Callum (SP?) has done in the Baltimore
Public Schools. Now his interest was more in psychopathology,
but nevertheless, they were actually able to randomly assign first
grade children to teachers within the Baltimore Public Schools
in his research. And what they found was, after a short period
of time, about six weeks, the distribution of behavior problems
in those first grade classes was bimodal. Either all hell broke
loose, or they were pretty well under control. The interesting
thing, when they filed them longitudinally, the kids that were
in the classes where all hell broke loose, carried that with them
into middle school. They were still misbehaving in middle school.
And he characterized these first grade classes where all hell
broke loose as chaotic classrooms. Now as I say, his interest
was primarily psychopathology. John Reed from the University
Or Oregon really, recently reanalyzed some of their data that
they had, and found the same thing held for reading. When a teacher
can't control the class, they probably don't provide a real stimulating
reading program either.
DONALD L. MACMILLAN
And these kids were significantly lower in reading in these chaotic
classrooms. Now, as we look at the urban settings, more unqualified,
or marginally qualified teachers are functioning there with kids
who are more dependant on the instructional skills of their teacher.
Recently was out at a school in San Bernadino, California. Out
of 18 elementary teachers in this, in this school, 16 were uncertified.
That doesn't give these kids a fair chance. And when we just
go in with an achievement test, an intelligence test, ruling out
that as a contributing factor seems to be the extremely difficult,
and it gets a little touchy when you start saying, you know what?
It was Siperstein that, there was a teacher last year, we know
he can't teach. Therefore we'll put that exclusionary criteria
in. When you say in concluding, some of the things that seem
in place, one we've characterized what's gone on at the schools
as kind of a titration of intensities. Where we try pre-referral
intervention, chapter one services. If that doesn't result in
some improvement in performing, we then go to assessments and
put them into special education. Try resource rooms, if that
doesn't work, special day class. And we keep titrating until
we get a response to our treatment approach. Since the IQ issue
is central to some of the issues that ought to be discussed in
this conference and elsewhere. I think the biggest condemnation
of it is that the intelligence test, as used now, has no curricular
validity. It doesn't give us any direction as to what to do with
the child. It may tell us which pigeonhole to put them in, but
then it doesn't add any information to it, in terms of how to
instruct the child. So for classification purposes, one might
defend it. For curricular purposes it seems very weak.
DONALD L. MACMILLAN
I've indicated before, in differential diagnosis, we don't seem
to be using it as an important fact. We made the call of LD,
saying that we can take care of that later. So distinguishing
between mental retardation and learning disabilities, IQ doesn't
seem to be of central importance to the public's (WORD?) And finally
in using it to apply exclusionary criteria again, we don't see
any evidence it is being used in a systematic way, and therefore
I find little purpose, given the expense of administering intelligence
tests that we currently have, and it's lack of relevancy in instructional
program, to continue and use it as a criterion for LD. I think
also we must recognize comorbidity. We see many children who
not only have a reading problem, but they have a behavior problem.
The current system in special ed, and under IDE puts a kid into
one hole and one hole only. But we find retarded kids who are
behavior disordered. We find kids with reading problems who also
act out. If there is a linkage of treatments to labels, teachers
naively assume that this kid, because they're LD, only presents
with a reading problem. And frequently they don't, they present
with a multitude of problems that have to be addressed in the
IEP and in the instructional program. And I guess recognizing
the fact that IDEA plays out very differently in urban from suburban
school districts is essential. Or we have the naive assumption
that we can read a study done by Gottlieb in New York City, and
generalize to an affluent middle class community in terms of what
LD kids are like. And then I guess, finally, before sitting down
and let others take the heat, um, I think the assumption of a
neurological basis for the learning problem is irrelevant to the
educational program we have, and ought to be deleted from the
definition.
DONALD L. MACMILLAN
And I guess the abolishment of the severe discrepancy. We don't
see that being a criterion. Absolute low achievement is what
we're seeing put in the public schools and classified as LD.
Thank you very much, and I think I made it under half hour.
DEBORAH PRAISSON-STAT
Thank you Dr. MacMillan. We will now hear from our respondents.
And please remember to save your questions until the end of the
session. Our first respondent is Dr. Michael Gerber who is a
full professor and director of the Center For Advanced Studies
Of Individual Differences at the University Of California, Santa
Barbara. Dr. Gerber has conducted research on early acquisition
of basic skills by student with learning disabilities, school
implementation of special education policy. And applications
of computer technology to remediation, teacher training, and assessments.
MICHAEL GERBER
Thank you, good morning. (CLEARS THROAT) I was so anxious yesterday
about staying on time that I forgot to say that I really enjoyed
the paper by, uh, Dr. MacMillan and Dr. Siperstein. I congratulate
them on a great effort. Um, we're going to use individualized,
uh, slides today, which means that you can conjure up any image
of a slide that is relevant to what I'm saying, in your mind.
It would be axiomatic, uh, should be axiomatic that decision
making in schools is different from the judgement processes that
we associate with psychological measurement, testing and research,
pertaining to learning disability. It should be axiomatic, but
it seems that we are not able to get past this fact. If you're
going to make progress not merely in identifying, but also in
better educating students with learning disability, we will need
to invest more intelligently on research on the schools that do
the identifying, and hopefully the educating. I mean, that knowing
more about learning disability in the absence of more detailed
knowledge about the structure of decision making in schools will
not yield a sufficient basis for improved policy. Is it possible
that we are so distracted by the fact that different teachers
in different schools identify students with different characteristics
as learning disabled, that we fail to recognize a lawfulness that
meets the variability? Another way to ask this question is this,
why is it so difficult to reliably scale what we believe are valid
classification and intervention practices to the level of whole
schools, districts and states. I was going to say everyone has
a theory about this. But the fact is, that everyone has an opinion
about this. We really have no theory whatsoever.
MICHAEL GERBER
Typically we assume that variability of identification rates across
large units of analysis represent error variance. However, error
variance in this kind of formulation rarely means idiosyncratic
or random. It simply means that there's a portion of observed
variability for which our models cannot account. A portion of
variability that remains unexplained, but in principal not unexplainable.
What is it that a good theory and the research it would guide
should include? First, a good theory of identification of learning
disability would account for decisions by teachers. There is
very little imperial data to support the argument that simple
and easily correctable variations in local staff, or more specifically,
variations in their training, ethics, or competence can account
for much or most of the observed variations in incidents. Specifically
we need to understand not decisions to formerly refer students
as possible learning disabled, but rather the formative decision
making that leads from instructional choices, to modifications,
to perplexity, to expressed concern. The following, the, the
thinking that follows this general root, and the forces that act
on it, are not well understood at all. We believe that some of
it may be stimulated by objective characteristics of students.
But surely there are other factors was well, factors related
to teachers knowledge of subject matter perhaps. Or knowledge
of pedagogy. Or ability to form mental models of students as
a basis for planning and revising instruction. Or the ability
to reformulate models of students based on instructional experiences.
Second, a good theory of identification would account for decisions
that may be quite distill from the classroom. These factors are
organizational in nature.
MICHAEL GERBER
They concern not only policies and mechanisms of administration,
but also the optimizing and (SOUNDS LIKE) satisficing strategies
or solutions that undergird policy and administrative practice.
Moreover, these organizational factors create the structure within
which teacher decision making occurs. They act as critical constraints
on teachers decision making. Such as when, uh, externally imposed
reforms involving curriculum standards, uh, high stakes testing,
or performer basic countability are imposed. Organizational factors
also include algorithms, some only implicit, for allocating resources.
Especially and most importantly teacher attention, time, and
effort. Three year agos, three years ago in California, some
colleagues and I studied regional variability in incidents of
disabilities for the office of (STAMMERS) a legislative analyst,
the Department Of Education and the Department Of Finance. The
Department Of Finance in this research project, specifically insisted
that we not study what they perceive to be idiosyncratically variable
disability categories. Particularly learning disability. Instead
they prescribe that we study only severe, medically identified,
or other high cost and low incidents disabilities. In other words,
they wanted us to study disabilities that were, to them, not only
indisputable, but also in some sense distributed in the population
by chance. Why, what they failed to recognize however, was that
(WORD?) are not distributed across the landscape of California
by chance alone. Nor are our quality of schools. Preschools,
or school staffs, or sources of tax revenue, or job opportunities,
or housing, or availability of medical services. So it was no
surprise to us when we found huge regional variations and incidents
of low incidence disabilities.
MICHAEL GERBER
Regardless of how we modeled low incidence disabilities. Regardless
of how we modeled the definition or the composition of severe
disabilities. These results are consistent with earlier findings
from samples of states, districts in the same state or metropolitan
areas. However contrary to earlier findings from national or
state data, uh, we found, uh, that measures of poverty or ethnicity
provided no explanation for observed variability in the incidence
of low incidents disability across California regions. However,
a small portion of variability was, of severe disability was explained
by incidents of students with limited English proficiency. Possibly
indexing the relative poverty and lower social status of California's
large immigrant population and their lack of access to health
care. When severity was modeled in terms of higher than normal
costs, 48 percent of that variability was explained, as expected,
by percentage of low incidence disabilities, uh, enrolled. But
also by district's resource capacities, and the degree to which
regions were urbanized. These results are only a start, of course,
but I suggest that they strongly indicate that there are complex
organizational features of schooling that systematically influence
identification rates and service levels, over and above any confusion
we have about definition, identification models, or psychometric
instruments.
DEBORAH PRAISSON-STAT
Thank you Dr. Gerber. Our next respondent is Dr. Dan Reschly
who is a professor of education and a of psychology at Peabody
College, Vanderbilt University. There he has chaired the Department
Of Special Education. DR. Reschly has published on topics of
school psychology, professional practices, systems reform, assessments
of disability with minority children and of youth, and legal issues
in special education. Welcome.
DANIEL J. RESCHLY
Thank you it's an honor to be part of this summit, and particularly
the comment on the paper by MacMillan and Siperstein, a really
fine piece of work. Minority over representation in special education
is a significant but often unrecognized contributor to 25 years
of increasing LD prevalence, and of the current diagnostic confusion
over what LD is. It does work. I will not elaborate on the changes
in prevalence over the last 25 years. MacMillan did a very fine
job with that. I will point out, however, that the changes in
mild mental retardation are probably greater than those shown
by the official OSEP figures. The reason for that is, that after
1976, '77, students with moderate, severe, and profound levels
mental retardation in many, many places throughout the United
States gained access to the public schools for the first time.
Thus the decline in mental retardation is even greater for the
mild level of mental retardation. Now before we go too far with
discussions of state prevalence, I'm sorry, of national prevalence,
we need to recognized the enormous variations among the states.
States use different names of disabilities, different conceptual
definitions, and different classification criteria. That turns
into enormous variations in prevalence of high incidence disabilities
across the states. MR prevalence varies by a factor of nine,
that is the highest state versus the lower state is different
by a factor of nine. LD by a factor of three. ED by a factor
of 33. The highest state, highest state prevalence in ED is Minnesota,
which carries about a two percent prevalence. The lowest, I believe,
is in Mississippi, which has about a half a percent prevalence.
I guess it shows that the persistent severe cold temperatures
have a substantial impact. So analysis of disability prevalence
has to take a state by state as well as a national perspective.
Now the reasons for the declining mild mental retardation, mild
MR, uh, prevalence have been discussed elsewhere. I made a living
doing that in the 1980s. And if you're interested in that, I'd
be happy to send you a reprint, in fact I'd be flattered if anybody
asked. Um, over representation litigation had direct and indirect
influences on that decline. In fact, I think over representation
litigation was the greatest, uh, influence. Massive declassification
generally followed consent decrees.
DANIEL J. RESCHLY
For example, the (SP?) Deeana Consent Decree in California, Guadeloupe
and Arizona, (UNINTELLIGIBLE) the holiday descent decree in Mississippi,
producing markedly lower MR prevalence and contributing significantly
to L, increased LD prevalence. One current effect of the over
representation litigation is the reluctance to use the mild mental
retardation category, which MacMillan has described very well.
The kind of declassification experience in California, Mississippi,
Arizona, and other places experienced in the 1970's and 80's is
less frequent today. It is occurring as we speak, in the state
of Alabama, which is under a consent decree called (SOUNDS LIKE)
Lee V. Macon, which is, establishes considerably more stringent
and mild mental retardation, uh, classification criteria. You
could expect, by the December first report in 2002, to see increasing
incidents in LD in that state, and markedly lower MR. Declassification
pressures also come from OCR investigations of districts. Now
LD is generally not implicated in current discussions of national
over representation, and I present data here on the OCR rates,
um, uh, for different disabilities. I urge you to distinguish
between a composition index, and a risk index, and I'll be happy
to send this table for explanation. Uh, if, if you're interested.
Only a small proportion of minority students, and this is the
point I wish to emphasize, only a small proportion of minority
students are in special education. And the difference in the
proportion of minority students in special ed is relatively small.
The highest incidents of minority students in special education
have the greatest disproportionately, or about two point six three
percent of the highest incidence group is in, uh, that, that category.
DANIEL J. RESCHLY
Scholars associated with the NICHD Dyslexia studies program concluded
that IQ testing and ability achievement discrepancy determination
contribute little to identifying appropriate groups for reading
interventions. A position that I strongly agree with. Moreover,
use of the discrepancy model to identify LD produces a way to
fail kind of phenomenon that I think everybody is now familiar
with. The NICHD recommendations suggest universal early screening
with early intervention. One likely effect of early screening
is the identification of a relatively high proportion of students
as needing early intervention. They suggest as many as 25 percent.
And many more females as having low reading achievement being
identified receiving early interventions, perhaps eliminating
the current LD, uh, ratio, male to female ration of about two
to one. (CLEARS THROAT) Now, and anticipated and unanticipated
effect of early screening not discussed today, is the virtually
inevitable outcome of identifying a markedly disproportionate
number of students as needing early reading interventions. If
the early interventions prevent reading disabilities, with a high
proportion of students, and are differentially even more effective
with minority students, then current patterns of over representation
could diminish. However, if the early interventions have the
effect of raising (STAMMERS) reading achievement generally, but
continuing to preserve approximately the same distribution, the
same relative levels of achievement, then the universal early
screening could very easily have the effect of greater minority
over representation, and moving that minority over representation
to the field of learning disabilities. The reason this would
occur is that early, universal early screening is more efficient
in identifying children with low achievement as opposed to the
very imperfect procedures of teacher referral today.
DANIEL J. RESCHLY
The higher minority identification rates are likely to occur with
the universal early screening because the current system identifies
fewer minority students as disabled, than those who would be eligible
if universal screening measures were applied. Now the greater
(WORD?) early screening is likely to increase over representation
will also depend on whether national or local norms are used.
If national norms are used, the early screening proportions are
going to be even more disproportionate. Here's the critical point,
the acceptability of great disproportionality will depend entirely
on outcomes. If there are good outcomes from disproportionately
identified numbers of persons in early intervention, it will not
be a problem. If, however, there are mediocre or poor outcomes,
and it simply serves as the way of identifying more kids who end
up being, quote, disabled, then there will be severe difficulties
associated with that. Special education outcomes were the principal
issue in the over representation litigation, although most commentators
focused on IQ tests and other aspects of assessment. The point
was made clear by the (WORD?) trial judge in a later commentary
on the case, when he said, the 1979 decision was largely concerned
with, quote, the harm to African America children resulting from
improper placement in dead end educational programs. The greatest
challenge to special education is to show that programs are effective
and that they make a difference in improving academic skills and
opportunities. There are, I forgot about this, there are system
reforms underway today that emphasize noncategorical eligibility,
elimination of IQ and IQ achievement discrepancies, functional
assessment of formative evaluation, and student performances with
basis for special education decisions. There are significant
parallels as well as differences in recent history of mild MR
and LD.
DANIEL J. RESCHLY
Both are high incidence disability categories that are rarely
identified before school entrance. LD is the current dominant
category in special education. A position held by mild MR until
the late 1970s. Over representation litigation based on concerns
about (WORD?) and program effectiveness lead to a sharp decline
in mild MR. Current LD practices and identification trends have
similar problems to the mild MR programs in the 1960s and 1970s,
specifically problems in demonstrating positive outcomes, and
concerns about stigma. Reforms to address these trends, especially
improving outcomes, will determine whether LD follows the mild
MR path. Thank you.
DEBORAH PRAISSON-STAT
Thank you Dr. Reschly. Our final respondent is Dr. Martha Brooks
who is director of the Exceptional Children And Early Childhood
Group at the Delaware Department Of Education. She is a parent
of three sons, one of whom has severe disabilities. Dr. Brooks
is a member of the board of directors of the National Association
Of State Directors Of Special Education. Welcome Dr. Brooks.
MARTHA BROOKS
Thank you and good morning. I am honored to have the opportunity
to be a respondent to this particular paper, but I've also had
the, the pleasure of being part of this activity from it's inception
a year ago, and I can't tell you how exciting it is to actually
be here and listen to all of these papers being presented. To
meet all of the authors that I have read over the past, uh, several
months. It's been a very exciting couple of days for me. But
I want to start my various brief, um, comments this morning by
issuing you a challenge. 'Cause I think it's not so much what
we're learning over this two days that's as important as what
we're going to do with it, uh, in the future. That is, is really
the heart of this issue. And I would challenge us all that we
need to continue this discussion 'til we find a solution where
there are no losers, and all children have the opportunity to
learn. A solution where as to, to quote Bob, our new leader,
uh, no child will be left behind. But I think he was quoting
some other new leader with that particular, um, quote. But I
think it really is an important one, and I think it's one we need
to keep in mind. Whatever we do to, to change, to refine, to,
to work with, uh, a definition of learning disabilities and how
we implement special education services for our children. We
cannot forget that the goal is to leave no child behind, and to
make sure that all children have a real opportunity for an education.
This was one of my favorite papers out of the nine, because it,
it clarifies the reality of the world in which I live. It talks
about research identified students with learning disabilities,
and it talks about school identified children with learning disabilities.
MARTHA BROOKS
And as you can tell from the discussions that are going on around
here, they are somewhat different. And I would ask, I'm going
to read one sentence out of their paper that I think kind of sums
it up from a, a practitioners perspective that you take away with
you. We urge recognition of one reality driving the public school's
focus on planning for services. That is, they are going to continue
to serve those students they perceive to be most in need of help.
That is not going to change. And I think, um, you heard a very
good description by Dr. MacMillan of how that definition of who
are the children most in need varies by school, and sometimes
by class room, and certainly by district and by state. I want
to very briefly comment on what I think are some of the significant
points that need to be kept in the forefront of further discussion.
One is that critical role of teacher referral. Children are
referred because teachers do not feel children are meeting the
expectations of the norm of their class and or their school.
Teachers are concerned that they themselves are not meeting the
needs of the child, and the further the child is from the norm
of the class, the more likely that referral will be. That is
a reality that sends us some incredibly important professional
development messages. We need teachers who have more skills,
both in terms of correctly diagnosing and helping children in
their classes, but having the skills to know how to help. Get
them the kinds of programs that they need that will be effective.
Another point that I found very interesting that I want to keep
in the forefront of my head, because it periodically surfaces
in, in cases that I deal with in, in the states, and that's the
false negative concept. Uh, a group, the false negative group
is a concern to researchers but it's also a concern to advocates
and educators and parents.
MARTHA BROOKS
In the sense that these are children that do have a disability,
and yet because they are able to keep up they are not identified
for special education services. This group is becoming more critical
as the high stakes nature of accountability in our public schools
moves up. We're seeing more and more of these children with,
uh, a count five oh four accommodation plan, and in some cases
parents pushing to get them into special education services in,
in part because of the benefits that they will get in terms of
accommodations on testing programs, to be very honest. It's not
a group we want to ignore, and one that we really want to pay
attention to. Uh, the eligibility of the sessions have been very
informative and, and, uh, I will just comment that I have yet
to find a psychologist who cannot find a discrepancy somewhere
if they really need to. And for the most part, the recommendations
for eligibility center that a team makes center on whether or
not the student will benefit from services of eligibility will
make available for them. And as long as that is the reality,
let's deal with that, and not worry, um, quite so much about how
we get to, to that point. The early childhood focus I was pleased
to hear a little more mention today, actually, than I heard yesterday,
and maybe I was just listening, uh, the third time I go through
this presentation I'll pick up something else, yeah. Um, but
I, I really think that we need to really pull back and focus more
on, on early childhood and, because that's part of my responsibility
in the department, I spend a lot of time with our three, four,
and, uh, five year old head start programs, we have a state head
start program that I also oversee. And I think getting the screening
and the services into those programs is equally important.
MARTHA BROOKS
And though a lot of the discussion that you're hearing starts
with kindergarten, I think there is some very valid research that
talks about what kinds of things can be happening for children
three and four years of age. And I think they will have a very
critical impact if we can put some of those things in place, in
preventing more severe levels of disability for children as they
enter the school system. The mental retardation thing has been
very well discussed, and I'm not going to spend any more time
on it other than to say, I learned a lot in, in reading several
of these articles in helping me to understand how that should,
because I kind of lived through it in the trenches. And to see
it all pulled together and explained made a lot of sense to me
as to why we're where we are today. I just think it's something
that we need to, to, uh, keep, again, in the forefront of our
thinking. The poverty in the urban schools, and the disproportionality,
uh, Dad did a wonderful job of explaining. And I think those
are realities in my state. We have a population that is in our
public schools across the entire state, 30 percent minority.
And so these issues are very, very real in every single school
in our state. And they're, they're ones that we wrestle with
all of the time. But I want to kind of focus on what I think
is one that has been touched on, but I, I really want to emphasize
from a, a practitioners perspective. LD is an okay label. In
our society today it is okay to be identified as having a learning
disability. It is a label that parents, teachers, and society
at large do not object to, and it opens doors to services including
accommodations and modifications that have increased importance
in high stakes world. Eligibility determination requires identification
under one of the IDEA categories, that's all it requires. But
the (WORD?) has very little meaning in terms of what's in a child's
IET, and the programs and services that child is provided.
MARTHA BROOKS
It is a way to open the door. So putting on a categorical label,
if we have to do that, that is comfortable and that people like.
Uh, or are more comfortable with. I'm not sure that people like
it, but they are more comfortable with it, it's more acceptable,
is it just makes everyone's life a lot easier. If you go back
to the underlying theme in this article, that schools are already
identifying students who need the help the most under the LD category,
you can begin to see that unless we develop ways to ensure supports
and services for all children who need it in order to be effective
learners, we will continue to serve those most in need within
special education. With a continued focus on high standards and
accountability for all students, the number of children referred
to special ed will continue to grow. A couple of comments on
the proposed models that the authors came up with. The only place
I really disagreed with them here is that they, they push for
more definitive labels. I'm kind of at a point in my life where
I'd like to see the labels go away and talk about building programs
based on the needs of children. IPT members including parents
are not going to be any more accepting of concepts or terms such
as low general intelligence, or environmental disadvantage, um,
than they are currently of mental retardation and emotional disturbance.
Those are just not, uh, they're more negative sounding terms,
and that's how they are interpreted. We need to ask ourselves
if we can meet the needs of children in public schools, and those
of researchers wishing to study specific processing to the cultures
with the same eligibility process. And what I challenge you to
think about, do we need to? Do we even want to? Although we
understand and appreciate the value of research in terms of improving
education for children with disabilities, we may be better served
by adding some additional research dollars to provide assessment
at the (STAMMERS) discrimination level researchers need, and allow
assessments for access to services (STAMMERS) in public schools
to be driven by the needs of the child and the instructional process.
MARTHA BROOKS
We do have other forms of accountability, but they're not, the
IDEA and within state's accountability legislation, which will
help to keep, um, some control on numbers and things like that.
Focusing our efforts on improved results for all children may
be a better investment of our assessment energies. The, the model
that they came up with, though, is one that I, I believe in very,
very strongly. They called it the Titration Of Intensity Of Treatments
In Decision Making. I call it the Instructional Support Team,
or the problem solving model. Which is, I probably best demonstrated
in states like Iowa, Pennsylvania and Maryland. And Delaware
is, is trying very hard to really beef up our ability to, to have,
um, really functional problem solving teams at the school level.
But the formative assessment process that is the basis of those,
um, models, and which was discussed, uh, earlier, is designed
to rely on an instructional problem solving team at the building
level that would be looking at every child who is not performing
at grade level. Interventions are planned on, by the child and
the teacher, and are based on research based best practices.
Plans are implemented with integrity, and data are gathered on
an ongoing basis to measure the effectiveness of the intervention
strategies. This curriculum based assessment, and intervention
process, continues until the right combination of services and
support (WORD?) for the individual child. If the level required
for the child to be successful reaches the level of special education,
and remember that's defined by what special ed within that school
and that district, um, which may not be the same in all places,
then the child would be moved into a formal referral process and
be found eligible for special education services. If not, the
child would still have a learning plan that would be able to support
them as they moved on up through the grades. There are a number
of issues, and I've run out of time so I've got to be very quick
on these. And I'm going to concentrate on the one that I think
is the most important, and I haven't heard brought a lot.
MARTHA BROOKS
Uh, in any of the discussions I've heard so far. And that is
the role of parent involvement in this process. If we make changes
in how we move children into the system, or we involve children
in the system, we cannot forget the incredible partnership that
special education has developed with, with parents. Many parents
and advocates view this prolonged pre-referral process, or this
formative assessment process as a delaying tactic. However, if
they are part of the problem solving process right from the beginning,
and they recognize that the needs of their child are being addressed,
and the process will not stop until the right supports and services
are identified, I think they can, we can develop the trust level
that we need to make some of these changes. If a model such as
this is implemented, children who meet the research identified
definition of learning disabilities will be found. Their learning
needs will be addressed, and those needing a level of supports
and services that may include accommodations and or special education
will be identified. However, all of the other students who need
help in order to access their education and be effective learners
will also receive it. Thank you.
[END OF TAPE]