Latest / DiveIn: Diving into Special Education's Most Complex and Pressing Debates / The Disproportionality Series Part 2: A Conversation with Rachel Fish

Transcript

The Disproportionality Series Part 2: A Conversation with Rachel Fish · DiveIn: Diving into Special Education's Most Complex and Pressing Debates

0:08
From the Division of Research of the Council for Exceptional Children,
0:11
this is Dive In. I am your host, Federico Weitholler,
0:15
professor at the University of Illinois at Chicago.
0:20
Welcome, welcome, welcome to the 10th episode of the
0:24
third season of Dive In. Today we will continue our discussion of
0:28
a highly controversial topic in our field, the disproportional representation
0:33
of racial and linguistic minorities in special education.
0:37
The purpose of this series is to bring different lenses and perspectives into
0:41
an unsettled debate in our field. In Our
0:46
last episode, Dr. Paul Morgan and I discussed the evidence based on disproportionality
0:50
and the debates over whether students are over or underrepresented
0:54
in special education. In this episode, we'll continue
0:58
this discussion from a sociological perspective. And to that end,
1:02
I interview Rachel Fish, Associate professor at Smith College.
1:06
Rachel Fish is a sociologist of education and disability and her research
1:10
examines how inequality is produced and maintained in schools
1:14
at the intersections of disability, race and gender. She uses
1:19
multiple methods, including experimental methods, observational data analysis and
1:23
interviews to understand how disability and giftness are co
1:27
constructed with race and gender in schools, how these processes
1:31
are shaped by school context and how they relate to educational inequity.
1:35
Are you ready to continue this proportionality series? Well, let's dive in.
1:44
Hello Rachel. Thank you so much for being here with us and help us to
1:48
understand its proportionality and help us to continue to
1:52
expand in this series of this important topic. Hi Federico,
1:56
it's really nice to meet with you today and thanks for inviting me
2:00
and I'm glad that you're doing this work. Yeah, this is great. I think it
2:03
helped us to have different views on the issue. We have Kathryn Bulgaritis for those
2:07
who haven't listened to the episode, mostly talking about policy issues and how states responds
2:12
to the policy framework, but then Also we have Dr. Paul Morgan that will also
2:16
talk about different methodologies and methodologies that can
2:20
help us to better understand over or
2:24
underrepresentation and your work that also is very important.
2:28
I think you come from a sociological perspective you don't have from a healthcare.
2:32
I like because we have a little bit of everything in this series. We have
2:35
Katherine did a little bit of the legal. Dr. Paul Morgan did come
2:39
from, I would say healthcare, but a special ed perspective and you
2:42
come from sociology that I think having these interdisciplinary
2:46
views on the issue will help us to understand it better. So the first question
2:51
that I have for you, that I have asked also to the other two
2:55
colleagues, is how you define disproportionality in
2:59
your work and how if this definition
3:03
Differs from other ones, maybe. Sure. So I think I
3:07
define disproportionality pretty similar to what we see in
3:11
policy discussions and in public discourse, which is really about these disparities
3:15
that we see and the proportions of students of different racial backgrounds that are
3:20
receiving special education. And we
3:24
think about this disproportionality as important in a way that I think
3:28
probably we all agree on, which is that they reflect some kind of inequality because
3:33
we know that disabilities should be similarly distributed
3:37
across the population and should not vary by race. If you
3:41
can say, like accounting for various forms of racial
3:45
inequality. I think that where my conceptualization differs
3:49
is really in where those disparities come from and what
3:53
they mean. So my read of the literature and policy
3:57
discussions is that most people think of these disparities as
4:01
coming from two places. And this is like, I think, really germinal
4:06
to this, at least in my own academic trajectory, was reading the Donovan and Cross
4:10
piece that was looking at these patterns in special ed and
4:14
gifted. In 2000, 2002,
4:18
National Academy of Education. Yeah, right. And I think that
4:22
most of us think about there being sort of two main ways that
4:26
these inequalities are produced. And one is that there's racial inequality
4:30
in real incidents. I'm using scare quotes that I know listeners can't see,
4:35
and that these disparities are due to things like environmental racism,
4:39
like inequitable exposure to lead in drinking. Drinking water and in
4:44
homes. We can think about inequitable access to prenatal
4:49
care that might affect cognitive and behavioral development. We can think about the ways that
4:53
racism shapes the stressors that families are under that affects cognitive
4:57
outcomes and behavioral outcomes for kids. So there are
5:00
some ways that we think that children's cognitive and
5:04
behavioral development is going to vary by race because of structural racism.
5:08
And then there's this other piece that we think about ways that
5:12
schools are inappropriately either over diagnosing incorrectly
5:17
or that they're missing diagnoses that should be happening and
5:21
that kids are underrepresented. And as many
5:25
folks in disability studies and critical special education work and the sociology
5:29
of disability have argued, disability is in part socially
5:33
constructed. And so I think that in some ways I'm
5:38
trying to problematize this idea that there's incorrect diagnoses
5:43
or missed diagnoses, because a
5:47
disability diagnosis, as many folks have pointed out, has a lot of subjectivity.
5:52
It's not just like a blood test or a pregnancy
5:55
test that tells you yes or no. There are many factors that folks
6:00
are trying to figure out when we Say that a child might or
6:04
might not have a disability that are
6:08
not just like yes or no. You know, we all have skills and
6:12
abilities that fall on continua, including things like our math skills,
6:16
our attention span, our social skills. And the line
6:20
between disability and not disability is really dependent on
6:23
context, our priorities as a society, and other kinds of social factors.
6:28
And so this really complicates the idea that, like,
6:32
there's like, you know, these incorrect. These sort of false
6:36
positives or false negatives of disability categorizations.
6:42
And so it can really help us think about the
6:46
ways that there are many pathways that can lead to a disability diagnosis,
6:51
or rather a special education classification that aren't just like,
6:55
you know, a bad teacher or an incompetent evaluator
6:59
making a bad decision, but that they reflect all of these complicated
7:03
things that are going on in the social environment for a child.
7:07
And then the other piece that I would say makes my thinking
7:11
about it different is that disability
7:15
and special education are not monolithic.
7:18
Different special education categories have different meanings and
7:22
different implications for kids. The student's race, socioeconomic status,
7:26
all of that. That accessing special education doesn't mean just one thing.
7:30
It's a complicated set of implications for kids that also vary
7:34
depending on who you are. Yeah, absolutely. So let me see if
7:38
I can clarify that for our audience and see if I understood
7:41
it. So you're separating
7:45
these two kind of exponentials. One more about issues of
7:50
experience and structural inequalities in the terms
7:54
of biological conditions like low
7:58
birth weight, lack of access to healthcare, different risks
8:03
that come associated with poverty that may result
8:06
or may contribute
8:10
to some sort of disability or cognitive
8:15
deficit or et cetera. And the other one,
8:18
it's more about subjectivity, right. Thinking about disability as
8:23
not something black and white, but something that has many grays.
8:26
And in those grades, people are making some decisions
8:30
about referral. And my question
8:33
to that is, is there a dialectic
8:38
between the two of them? Are there. Are there a point when the two of
8:42
them overlap of these two different ways
8:46
to see the issue? Well, let me see if I'm
8:50
answering that right. I mean, I think because there's the two. There's the. There's the
8:53
structural racism, and then there's sort of like what's happening in this social context.
8:58
Right. And I think that there's. There's a fuzziness
9:02
there because of what are we calling disability? And that is
9:06
also in this sort of like this real incidence of disability that we're
9:10
looking for is when we think about the ways that structural racism
9:13
can affect children's academic skills,
9:18
their behavioral and. And social skills in school. That the.
9:23
The line between, like, what really what we say
9:28
is a real disability versus, like, oh, there are other explanations for that is
9:31
very fuzzy. And it's very. I mean, I think people working with
9:36
kids, people diagnosing the teachers that I've interviewed that are making
9:40
these decisions about referrals, they're really trying to think about this in very complex and
9:45
nuanced ways about, you know, what's a real disability and what
9:49
should we really be doing here? Yeah. And do you think that's because we're talking
9:54
here disability in very broad terms? Right. Do you think that's the case with
9:58
all disabilities? Because I can hear someone push what you're saying with
10:02
the term. It's like, well, look at, you know, down syndrome. You know, it's not
10:06
a fuzzy. I mean, it's. It has a very particular definition around a chromosome.
10:10
Right. Just to give an example. Yeah, of course. There.
10:13
Go ahead, Go ahead. No, no, no. I just would like to. To see how
10:16
we respond to that. Yeah. I mean, so, you know, social construction doesn't
10:20
mean that something isn't real. Right. It has real
10:24
meaning, you know, and the. I think we
10:28
can think about the way that disability studies folks often talk about a social model
10:32
and a medical model of disability is that there
10:35
are. There are sometimes distinct lines for an impairment.
10:40
Right. Like down syndrome. We can talk about kids who
10:44
maybe need very significant supports for activities of daily living that
10:49
do need a modified curriculum because of their cognitive skills,
10:53
for instance. That I think we can say very clearly, like, here is a child
10:57
that really needs special education services. It's very clear that there's
11:01
something going on here that is both neurophysiological, but there's also the
11:05
social experience of having a disability in our society and
11:09
in our schools. Yeah,
11:12
absolutely. Yeah. That's an interesting
11:16
way to think about it. So going to.
11:22
Going to this debate, right. Why do you think,
11:26
first, it's become such a powerful topic in special education?
11:30
And why do you think it becomes so polarizing? I think there
11:34
are a few things going on here. So one is that in research,
11:38
we love a contradictory finding. We all do. Right.
11:41
Like, oh, we've turned this knowledge on its head and we found
11:45
something new that's really exciting, and it doesn't
11:48
happen a lot in education. Right. Like, most of the things. There are
11:52
some. Like, there are. There are definitely new things that we find, but most
11:57
of what we do in education research are
12:01
pretty Small findings for the most part. And often I, I tell my, I always
12:06
explain this in methods conversations with my students about.
12:09
And I, I think I saw, I remember seeing this piece that I guess was
12:13
like a joke article in a medical journal that was a randomized
12:17
control for the effectiveness of parachutes
12:21
in preventing death in skydiving.
12:25
We don't run randomized controlled trials on whether a parachute
12:29
will save your life during skydiving because it's obvious. And a
12:33
lot of the questions that we have at this point about social processes in schools
12:37
are very fine grained at this point. And trying to figure out what's
12:41
going on here is much more complicated. And the data is not always as
12:46
clear. So we love a contradictory finding. And I think that that's
12:51
part of it is that I think that there have become sort of like sides
12:55
positioned in this, even though I think lots of people are thinking about this in
12:58
much more complex ways. And some
13:03
of the divide also reflects methodological divides and what
13:08
kind of methods are privileged in academic
13:12
and policy spaces. You know, we know that there's
13:16
the old quantitative versus qualitative divide, of course, that I think
13:20
has started to dissipate, but it still rears its
13:24
head in especially in terms of, you know, I hear
13:28
sometimes like journalists reporting on research will call qualitative
13:32
work anecdotal. And I think that even,
13:36
you know, our policymakers often think of the research that way, even though
13:40
the qualitative research can give us such important information about social processes
13:44
and mechanisms. And I think that it also,
13:48
you know, that often there's an assumption that data
13:52
is just objective, that quantitative. Sorry, I'm going to back up.
13:55
I just called it data, that quantitative data is objective and that any analyses
14:00
that we run it through are providing these facts,
14:04
these objective facts. And really all research methods
14:08
have limitations. And all research methods are
14:12
used to answer different kind. They excel at answering different kinds
14:16
of questions. So some are better for generalizability, some are better for causal
14:20
relationships, some for social processes and mechanisms.
14:24
And I think that especially because policy spaces
14:28
and popular understandings of research tend to give
14:33
more weight to large scale quantitative research studies,
14:37
I do think that it has allowed for some overstatement
14:42
of the confidence of the findings that we have in this area of research.
14:46
And there's still a lot that's uncertain. Yeah, I totally
14:50
agree. I think there's from. I'm talking about each
14:54
side of this equation and in general education, in education,
14:56
in research in general, I think we need a little
15:00
more humility as researchers to really say that
15:04
yeah, we don't. There's all things we don't know and we shouldn't so
15:08
sure about so many things that we give for granted and we
15:12
think that are so empirically grounded. And I have so many questions
15:16
and I doubt so much when someone tell me this the research proves or the
15:19
research has. You know, there is a body of research that has supported this.
15:23
You know, we're as you say, we're so fine grind still, like I think there's
15:27
so many things that are not clear. That's good to think about
15:31
one of the contributors to these different ways to see the issue.
15:35
So go a little more into the research. When you
15:40
see the research as a whole, what do you think are the most
15:44
consistent takeaways from it? So we know
15:48
that there's overrepresentation in the raw data. That's very clear.
15:51
Sorry, overrepresentation of particularly black and indigenous
15:55
children in the raw data. And we know that this reflects racial inequality
15:59
broadly. And then I think there's a lot of.
16:03
It's not just, I think there's uncertainty about to what extent
16:08
different pathways lead to that inequality, but we do
16:12
have a lot of strong evidence about various mechanisms that are
16:16
contributing to that inequality. Especially if we look beyond
16:21
just, you know, analyses that are specifically looking at say
16:24
large scale patterns of disproportionality, either, you know, nationally representative
16:29
data or district level data, but also looking at other kinds of research
16:33
on inequality in special education,
16:37
that these processes are really complicated and sometimes they operate
16:41
in opposite directions and that it's not. I think that
16:45
we need to move beyond just thinking about, you know,
16:48
the end all and be all of understanding racial inequality in special education.
16:52
It's not really going to be about overrepresentation versus underrepresentation.
16:57
From my perspective, you know, I think we
17:01
see strong evidence that teacher
17:05
referral is one of these pieces. Right. And you know, I have a few papers
17:10
showing that teachers are more likely to
17:13
notice and interpret as disability when
17:18
white boys in particular are struggling academically more than their
17:22
peers. And I do find that kind of the opposite
17:26
pattern for behavior that they're more likely to notice challenging behavior among
17:30
black children. I guess if you understand that in
17:34
your research, because I have some questions later because you have some very interesting findings.
17:37
You find like in the case of academics,
17:41
teachers are more likely to refer white students or white boys,
17:45
but in the case of behavior, teachers are more likely to refer to black students.
17:49
Right, right. And I've found this
17:53
using experimental designs where I've given teachers vignettes where
17:58
they read a vignette of a fictional student and they make
18:02
a decision about what are the ways that you would support this child? And I
18:05
change the name to be perceived as different races. And in that
18:09
research, I've found that pattern and I've also found it in my qualitative data.
18:14
Wow, that's. That's fascinating. So, and beyond that, what other
18:18
takeaways do you think people should, should know about
18:21
disproportionality?
18:25
When I, when I reviewed this question, when you sent it
18:29
ahead, I was thinking about how the takeaway, the most
18:33
consistent and well supported takeaway I have right now is that there's uncertainty and
18:37
that it's complicated. And I know that's a very unsatisfying answer.
18:41
And I don't mean to say that there's not clear evidence of a variety of
18:45
kinds of patterns, but I think that, I think that
18:49
there are many different processes
18:54
happening that are leading to these inequalities that
18:58
I don't think that it gives one clear answer about,
19:02
you know, about what we know about over or underrepresentation
19:07
of children of color receiving special ed beyond the raw data. And that's
19:10
because of major limitations with the data that we have and
19:15
just, you know, the need for more research.
19:19
Yeah. So you were mentioning one. You think one of the contributors may
19:24
be teachers. It happens at the point of teacher referral.
19:29
What do you think? We may be other contributors? And I always think about structural
19:32
inequalities, not just in the terms of resulting in
19:37
a disability because, you know, risk conditions around poverty or like low birth weight,
19:41
etc. But also about opportunities to learn. Right. Students who may go to schools
19:45
that don't have quality teachers or
19:49
language supports or resources, technology,
19:53
et cetera, may not have the
19:57
same ways, opportunities to learn or even how teachers and the school deals
20:01
with behavior. Schools will behave in very different ways. And that can also play a
20:05
role on over referring for behavioral issues. So do you think these
20:09
more structural issues may play a role? And not so much because
20:13
you were talking about individualistic level. Right. The individual level
20:17
of analysis, which is a teacher making a decision. But what do you think about
20:20
these more structural larger issues that may be affecting the dynamics of that.
20:24
Yeah. So in the recent paper that I've published with
20:28
my colleagues Ken Shores and Joao Sauto mayor,
20:31
we talk about, we kind of distinguish legitimate
20:35
from illegitimate pathways to disability status.
20:40
And we can think about legitimate effects of racism,
20:44
of structural racism on children's cognitive behavioral development like we talked about. Right.
20:48
And that doesn't make Them. Okay, but it makes these legitimate reasons for a
20:52
special education qualification that if, as you said, like, as.
20:56
As we've talked about, if a child has experienced lead
21:00
exposure and has an intellectual disability that's
21:04
resulting from that, we can say, like, this is. This is legitimate reason to
21:09
provide special education and why we might expect more black children
21:13
to need special education than white children, for instance.
21:17
But then there are these illegitimate pathways that we talk about,
21:21
and this includes things like teacher referral, which I think is just one small piece
21:25
of it. We think about, you know,
21:28
there could be bias in the evaluation procedures. It's an
21:32
older study, but the book by Hugh Meehan and colleagues where
21:36
they. They followed the. The psychologists running tests on students, and for
21:41
some of the students, they ran up to 16 different. I might have the number
21:45
slightly wrong, but they ran many, many tests until they found
21:49
something that had a significant effect. And,
21:53
you know, I think that these folks in schools that are making
21:57
the determinations for children are, I think,
22:01
in general, wanting to provide help and trying to find ways to provide help.
22:06
So I'm not saying that these are necessarily evidence
22:09
of, like, bad racist actions by educators, but that's
22:13
one piece. There's also differences in parent advocacy,
22:17
inequitable access to, say, private evaluations, or resistance
22:22
to certain disability labels and advocacy for other labels.
22:25
And then we can think about the other ways that structural
22:30
racism can shape also illegitimate
22:34
pathways to special education. So thinking about resource
22:38
differences in schools and what kind of academic supports kids have gotten
22:42
before they get to a point of being tested,
22:45
when we think about, you know, is the school a space where
22:49
children feel welcomed? Like, where, you know, that black children might feel
22:53
less belonging in our schools because of racism in our schools, for instance,
22:58
where we can. We can think about all of these different ways,
23:02
population differences in schools that can affect whether a student
23:06
stands out, whether they're seen as being typical,
23:11
all of those things can really, you know, shape whether a child
23:14
might get diagnosed with a disability that leads to a special ed classification.
23:19
And I mean, in all this, I mean, what is very important to me is
23:23
the issue of methods, right. Or how we go about examining all these different
23:27
possible explanations, or maybe they're all. But you
23:31
mentioned that sometimes we're very limited in methods
23:35
and sources. Right. Data sources as well, that all have their limitations.
23:39
So I'm thinking, like,
23:43
one of the conversations we had with Dr. Paul Morgan in our last show was
23:47
that, I mean, the quality of that evidence and what kind of evidence do
23:51
we need to make statements that there is, for example, systemic Bias or
23:55
structural bias for you,
23:59
mythologically speaking, what kind of evidence will make a strong case for
24:04
either discriminatory practices or structural issues? I mean,
24:08
what, what kind of research do we need to make a strong case one
24:12
way or the other? Well, I think that continued
24:17
experimental designs that can look at race
24:21
are, at racism are one piece of this. Right.
24:24
I think that we also need to,
24:28
I guess before I can talk about what I think we need, I think if
24:31
I can talk a little bit about what I think are some of the limitations
24:35
of the existing research would be helpful. That'd be great.
24:38
So, you know, the researchers,
24:43
including in my own work, when we use large scale data analyses,
24:47
it's been a huge leap forward for folks to be able to use
24:51
multiple linear regression or logistic regression or whatever method you're using that
24:55
includes a variety of control variables. Right. This is a
24:59
big change from just looking at raw differences in
25:03
the population level data to be able to say,
25:06
let's compare similar students that are different races and
25:10
see whether they are receiving special education.
25:14
But we need to think really carefully about what we
25:19
mean by clinical need. Like when we introduce
25:23
controls that are supposed to tell us that say, I'm going to stick.
25:27
Again, I don't mean to neglect everything outside of white black differences, but just to
25:31
be able to stick to one example. When we try to create
25:35
comparable black and white students in terms of clinical need
25:39
and we control for something like academic achievement,
25:43
we are saying the implication is that
25:47
a black student and a white student with the same level
25:51
of achievement has the same meaning about
25:55
their underlying risk of disability. And can
25:59
you expand on that? Yeah. So it's collapsing. All of the things
26:03
that we know that lead to black and white children having different levels
26:08
of academic achievement in schools that are related to structural racism, that are
26:12
related to the kinds of schools that they have access to, their experiences in
26:15
school, the structural racism that affects their experiences outside of school,
26:19
all of those things come together and they do they have real effects
26:23
on children's academic achievement. Do we really want to be in
26:27
the business of saying that all of those things are,
26:30
are evidence of disability when a white child who
26:35
had, let's say, I'm going to take, I'm going to take an example from my
26:38
qualitative data because I think teachers are going through this same thought process about referring.
26:43
When a teacher in my study has a white middle class child
26:47
who has had all of the, you know, has had
26:52
a low stress environment, has not been exposed to high levels of racism in
26:56
his life, has had Consistent access to teachers who create
27:01
a sense of belonging for him at school has, you know,
27:04
parents who are not undergoing extreme stress from poverty and racism
27:08
versus, say, a black child who is also low
27:12
socioeconomic status who might be performing at the. And they're performing
27:17
at the exact same level when that white middle class child is
27:21
struggling. I think teachers, and also what we're essentially doing
27:25
in research is saying, like, okay, like all of these things are explained. There's no
27:29
reason for this child to be struggling other than disability.
27:32
And when a black child is struggling and has the exact same
27:36
level of achievement, we're collapsing all
27:40
of those things that can create low academic achievement in a
27:44
black child into evidence of disability and saying
27:48
it's the same risk of disability as the white child with
27:52
that level of achievement. And so the pushback.
27:55
Go ahead, go ahead. No, no, no, you go. No, the pushback of that is
27:59
like, well, I was talking to Dr. Morgan in our last show.
28:03
This is what the main indicator there of sending to
28:06
special education will be academics. So if we control for that,
28:12
we make sure that teachers who face the same
28:16
academic performance of the students are
28:20
making maybe a bias or not a biased decision because we're controlling that and we're
28:25
exposing to students who are academically similar.
28:28
So I agree,
28:32
and this is where we're stuck in many ways as researchers
28:36
using these kinds of analyses is that if you don't include
28:40
academic achievement, you're getting these kids in your study that are
28:44
very different groups of students. Yeah. But if you do control for
28:49
it, you're controlling away the same racism
28:53
that's creating those academic achievement differences.
28:57
And so are you left with something that might be
29:01
able to tell you the level of like just what's happening
29:05
in a dia, you know, diagnosis versus not maybe,
29:09
but you're not able to see all of these other ways that
29:13
structural racism is shaping differential academic achievement
29:17
among. In those children. Yeah.
29:20
And we see the same thing with a lot of these studies will use a
29:24
control variable of ratings of behavior. Yeah,
29:28
I was going to ask you. Yeah. And we know that these are. We.
29:31
There's so much evidence. Both. There's some from special education, but there's
29:35
evidence, you know, in education research more broadly, we know
29:40
that, you know, teachers have, and other evaluators have
29:44
bias in their evaluations of children's behavior.
29:48
Behavior is very racialized in our society. And like,
29:52
whether we see it as a problem or not, we can see there's
29:56
a, a paper by Michaela Musto showing that
30:00
when white boys interrupt and do other Kinds of things that are misbehavior,
30:05
they're understood as. As evidence
30:09
of their brilliance, whereas their misbehavior for other kids.
30:13
We can see experimental designs like the, you know,
30:17
my paper where I show that children of color are more likely to be identified
30:20
as having disabilities when they have behavioral difficulties.
30:24
There's a video experiment that Jayanti Owens did where
30:29
she had teachers rating children's behavior.
30:33
It was rating their behavior in school. And these were child
30:37
actors that were engaging in the same.
30:40
They were. They were designed to engage in the same behaviors, and the people rating
30:44
them saw them differently by race. Yeah. And so,
30:47
you know, we can't. On the one hand, these are indicators
30:52
of the kinds of things that schools are looking for when
30:56
they think they might want to classify a child with a disability. And so it
31:00
makes sense. And at the same time, we're just collapsing a
31:04
lot of things into what we think are real evidence of clinical need.
31:08
Yeah. Or collapsing a lot of things in a measure
31:12
that tells two students are identically situated or similar.
31:16
Situated in the terms of the behavior.
31:20
Would your argument change thinking that some of those. It's not just the teacher
31:24
who fill out the behavioral ratings, but also the parents. Right. Some of the
31:28
studies also have behavioral rating scale from Aren't. Would that change a
31:31
little bit your answer? I mean, it changes it some. Right. I think that there
31:36
are a few things here. One is that it's the same problem as with the
31:39
academic achievement. Right. Is that we're then saying that the
31:43
same behaviors by a white child as
31:48
a black child are indicative of
31:53
the same risk of disability, even if they're being measured by a parent or an
31:57
outside observer. That we're still seeing
32:01
that racism shapes these
32:05
behaviors in the first place and the experiences
32:09
that kids are having in school. And then we can think about how, you know,
32:12
are parents understanding their children's behavior also because of the
32:16
ways that teachers are talking to them about their children's behavior. Like,
32:20
these are. These are all messy. And I mean, it helps some,
32:24
but it still has the same problems as the academic achievement variable.
32:28
Yeah. It collapses a lot of different factors and variables into
32:33
one single indicator. Right. Something in those terms.
32:37
So do you think we have, I mean,
32:41
for what I'm hearing, saying and the conversation we're having, you would say,
32:44
as if, I guess, that there is no strong evidence either for over
32:49
or under. That we still have a lot of research to do and that we
32:54
are still trying to figure it out.
32:58
What are the underlying causes of
33:02
either? Both, I would say, I Would.
33:05
I would qualify that by saying that I think that we really do have strong
33:10
evidence from the raw data. Right. That there is
33:13
overrepresentation of children of color and that tells us that there's
33:17
racial inequality. I think that figuring out,
33:21
as you said, like, figuring out the extent to which we can explain it through
33:25
these different pathways is less clear. We have evidence of
33:29
a lot of different mechanisms, but I think that determining,
33:33
you know, does that mean. Does that really mean that more black children
33:37
should receive special education than currently are?
33:41
Like, I. I don't think we have strong evidence to be able to
33:45
say that. And do we have strong evidence
33:49
that are being overrepresented?
33:53
I think that the raw data tells us that. That they are
33:57
overrepresented. It's. The question is like, does that mean that it's appropriate
34:03
overrepresentation because of things like exposure to lead?
34:06
Or does it. Or is it.
34:09
And it's pro. I. We have plenty of research showing us that
34:13
it can be all of these things at once, that all of these things are
34:16
happening. And at the end of the day, it means that
34:20
we have this inequality that we need to address. That I think
34:24
we can move forward and do. While we try to get better answers.
34:27
Yeah. So it seems that descriptively. Let's see if I can reference it descriptively.
34:32
There are different patterns of overrepresentation,
34:36
sometimes underrepresentation in some cases, I will argue.
34:39
But those descriptive statistics really tell us about the underlying
34:43
issues of this systemic disparities that we're saying
34:47
that we're still trying to figure it out. Right. And there are other.
34:51
There are different factors like you mentioned that are play that we're
34:55
figuring out, like teacher bias, maybe some more structural aspects.
35:00
And. Yeah. And we have.
35:03
We still have a long way to understand. It seems like.
35:06
And I know it's really unsung. This is proportionality serious.
35:09
Nobody's going to listen that to tell us that we don't know anything.
35:13
I know it's a very unsatisfying kind of. Especially when it
35:17
comes to, you know, what do we do in terms of policy. Right.
35:20
But we, you know, we do need to keep tracking this. We do need to
35:24
figure out what's going on. And in the meantime, there are things
35:28
we can do to reduce the effects that this has on kids.
35:32
Yeah. Yeah. You know, I want to get a little more into your
35:37
latest articles. I think they were very interesting. One of the great
35:41
parts about. On this podcast, I get to read people's work that I'm going to
35:45
interview. So it's very fun and rewarding.
35:48
So in your research on
35:52
teachers perceptions during the referral process and you mentioned a little bit of this.
35:57
What were. The main patterns you observed? You mentioned one,
36:01
but there were other stuff in your articles I think that were worth mentioning.
36:05
I would love to expand on that. Sure. So I've
36:08
looked at this a few different ways. So I've used some experiments as
36:14
I described earlier, where I give the teachers a vignette of a fictional student
36:18
and I ask them, how would you support this child? And they have an option
36:22
of a variety of the kinds of responses that schools have, like call
36:27
parents and you know, spend time during after school
36:30
tutoring or you know, those kinds of things. And one of the options is
36:35
begin the referral process for special education education
36:39
testing or MTSS or rti.
36:42
And, and just in case folks don't know, those are various.
36:47
They incorporate a lot of,
36:50
a lot of different processes. But one of their, one of
36:54
the things they're doing is identifying students with disabilities.
36:58
And in those studies I find that
37:03
teachers are more likely to perceive behavioral difficulties as disability
37:07
for black and Latino boys and
37:12
more likely to. And also black girls in another study.
37:16
And that they're more likely to perceive academic difficulties as
37:20
disability when they're exhibited by white boys.
37:24
Yeah, that's what's mentioned. Erfin. So that will reflect,
37:28
for example, that white boys will be more likely to be identified,
37:32
for example, for a specific learning disability. Exactly. And black boys and
37:36
black girls or Latinx boys and girls for a behavioral disorder.
37:40
Right, right. And what do you think are the implications of that?
37:46
Why is that happening? And it's like why some students
37:49
seen us having an academic needs and other ones
37:54
a behavioral needs. Well, these reflect the racialized
37:57
status beliefs that we have in our society where intelligence
38:02
is associated with whiteness and with boys,
38:07
where we. And where misbehavior
38:13
is understood as more intentional and more problematic
38:17
when exhibited, particularly by black children.
38:21
And so what kinds of problems are teachers noticing? Is part of
38:26
it. I find similar patterns in my qualitative data.
38:30
And there's where I can nuance this
38:34
a bit and not, you know, this is not about pointing fingers at teachers and
38:38
saying like bad basis teachers. Right.
38:42
Most of the teachers that I interview, they really
38:45
want, they, they all, you know, they work with kids because they want
38:49
to help kids. And many of them, while there
38:54
are some teachers that have some pretty explicit racialized
38:59
framings that, you know, go along with our
39:03
broader cultural understandings of behavior and intelligence
39:08
and race in our society, there are a lot of teachers that are
39:11
thinking about this in very complicated ways. So, for instance,
39:17
when a child of color is struggling in school,
39:21
I do have some teacher. I still recall, you know, one of
39:25
the teachers saying, like, you know, and this was an indigenous girl and
39:30
making all of these implications about her parents being bad
39:34
parents and not caring and like, no evidence, but just saying that
39:38
they, you know, they were using drugs. And these are very
39:42
like racialized deficit perspectives of this child's parents.
39:46
And, you know, or talking about a
39:51
black child or a Latina child as, you know, like,
39:55
oh, this is a choice. This isn't. This isn't because they can't do it.
40:00
But I have other. A lot of teachers are looking
40:03
at, you know, there's this girl who was
40:08
Latina and was having all of these academic difficulties,
40:11
and the teacher was like, you know, her father got deported
40:16
last year. She's dealing with all of these difficulties. Of course she's
40:20
struggling in school. I don't want to call that a disability. Yeah. And that's
40:24
what I saw for the most part, is teachers are wrestling with the same
40:28
things that I think we're trying to do with these statistical analyses.
40:32
We're trying to distinguish what we
40:36
think is real disability from all of these other ways that we
40:40
know that racism, xenophobia, socioeconomic inequality can
40:45
shape the academic and behavioral
40:50
and social skills that kids have in schools. Yeah.
40:53
Seems like we're putting teachers in a very difficult position in
40:58
the kind of framework we have to provide services. Right,
41:01
Absolutely. They gotta make the decision, you got him,
41:05
you don't got him, you got him, you don't got him.
41:08
And it's, as you say in particular, thinking about that sometimes disability,
41:12
it's a gray area that there is not.
41:16
It may be socially constructed or maybe
41:20
difficult to tease out. It's very difficult
41:24
position to put on teachers. Absolutely. And our schools offer few
41:28
alternatives. Yeah, yeah, yeah. And so in many
41:32
ways, and. And this is where I think sometimes
41:37
I understand this argument of, you know, these kids
41:41
who are struggling in school should be receiving special education at higher rates
41:45
for children of color, because in the absence of other kinds
41:50
of services, you know, they're like,
41:54
what. What are people supposed to do? Yeah, yeah,
41:57
yeah. Is that. Or what else we do? I mean, we keep doing what we're
42:01
doing before we're failing the student. And if we identify with disability
42:05
that's had maybe with some benefits, but also have. With that other kind of like
42:09
limitations and risk and segregation and stigma and maybe in
42:14
some cases, maybe a lower down curriculum that is not A general education.
42:17
It comes. Yeah. Yes, I definitely, while I
42:22
understand the inclination. Well, I'll just say that I understand the inclination to say it
42:25
to. That kids should receive special ed, but because of all of these ways that
42:29
special ed can both help and harm kids and that we
42:33
don't want to be in the business of calling racial inequality and xenophobia
42:37
and socioeconomic inequality disability.
42:40
Yeah, yeah, yeah. Like we, like, if we're going to start,
42:44
this is bringing us back to, you know, eugenics of like
42:48
saying that. That what we're seeing by race is reflective
42:52
of, of something that we want to call disability when it's not.
42:57
Yeah, fascinating. You know,
43:01
in one of other of your articles, you also
43:04
suggest that we need to be
43:08
more careful about findings around underrepresentation.
43:12
And you pinpointed some mythological
43:16
issues about, around covariance around and
43:21
around databases that we were using. Can you expand a
43:25
little bit on that? Yeah. So this is where we
43:28
talk about how when we control for academic achievement that
43:32
we're really collapsing. We're over controlling is what
43:37
we call it in the quantitative methods.
43:41
But what I think about this intuitively is that this is really collapsing
43:46
both legitimate pathways to disability and
43:50
illegitimate pathways to disability. Both the
43:55
things that really do affect cognitive development that are due to racial inequality
43:59
like exposure to lead, and also the things that can create
44:03
low academic achievement, that when we are using all
44:08
of these control variables, we are saying that these control variables are
44:13
equivalent risk of disability when they're not, they are evidence
44:17
of inequality. So we're essentially controlling
44:21
away the racism and then saying that there's no racism.
44:25
Yeah, yeah, I think that you were mentioning that before.
44:29
Yeah. And then the other piece is that a lot of these
44:33
studies have used the same sets of national data
44:37
that, you know, that's, it's, this is where when we try to
44:41
have generalizability in our studies, using these nationally representative data
44:46
sets is a great resource. But when we keep using the
44:50
ECLSK and the NAEP data, part of the problem
44:54
is that these data sets,
44:58
if you look at the unconditional estimates,
45:02
when they're just looking at just the relation between race and disability
45:05
status, these studies are actually finding that
45:10
the raw data is showing underrepresentation of children of color too.
45:14
And this does not map on to what we know from
45:19
the federal full population data. So the, you know,
45:23
the ECLSK and the NAEP are samples, and it
45:26
looks like there's something wrong with these samples that they're not actually appropriately
45:32
representing. They're not representing the same Patterns that we
45:36
see in the full population. And so the raw data does not match
45:40
the raw data from other sources that we've been using as well. Exactly. Not just
45:43
the methodology before that. The descriptive statistics
45:48
around whether they're over. Under just with very descriptive
45:52
terms are not matching what we know without larger data sets. Okay,
45:57
would you consider. I mean, one of the things that I struggle with these
46:01
larger data sets is that because they're larger and they
46:06
aggregate all the data from the state, it may completely
46:10
mask processes or inequalities
46:14
happen at the state level. Do you think that's a valid concern?
46:18
Absolutely. I think these processes are very contextualized.
46:23
I think that we see very different patterns of
46:27
racial inequality across the US in different spaces. And that's at state levels,
46:31
and that's by urbanicity. That's by a lot of different factors.
46:34
Factors. And,
46:38
you know, so I have a paper.
46:42
And then this is. We've also seen similar findings in a number of other studies
46:46
using different data sets and different methods that particularly
46:51
the racial composition of the school matters quite a bit for
46:55
racial inequality in special education. And so
46:59
what we're seeing across these studies is that
47:02
when a child stands out racially, so like a
47:07
white child in a school with more black children or a black child in a
47:11
school with more white children, that their risk of disability status changes
47:15
depending on that, which really, you know,
47:19
to me, provides some pretty good.
47:24
It suggests that there's something really going on here in terms of the way
47:29
that disability is racialized and. And what that means
47:33
to be. To stand out racially in that space.
47:38
Do you think these cases that you mentioned
47:42
are outliers? Because when we think about how schools are segregated,
47:48
it's less likely that it's more likely that black students go to school with
47:52
black students. Right, sure. But we also have plenty of diverse schools.
47:56
I mean, we know segregation is a big problem in our society.
48:00
And I think, you know, more than, like, what does
48:04
it mean, you know, at. At the extreme cases of,
48:08
say, like, a white child in a school that is truly like a very segregated
48:11
black school that has just, like, a couple of white kids in it. Are those
48:15
outliers? Sure. Like, in terms of our. The. The schools and the populations,
48:20
but it. It reflects that something is going on
48:24
in the interaction between the racialized organization of the
48:28
school and the way that disability is racialized.
48:32
And so it tells us something about what
48:36
disability means. So, like, in my study,
48:39
I found that when white children are in
48:43
schools with more children of color, and it's not Just highly segregated.
48:47
It's a relationship that goes across the continuum of like.
48:51
As the proportion of children of color increases, that white children are
48:56
more likely to be identified with what I have categorized,
49:00
drawing on other people's work, too, as higher status disabilities
49:05
that they're more likely to be diagnosed with autism,
49:08
other health impairment, which is most frequently ADHD and
49:13
speech language impairment. And that when black children are
49:18
in schools with more children of color and
49:22
indigenous children are in schools with. Sorry,
49:25
I'm getting my. Getting my words mixed up. When children of color are
49:29
in schools with more white children, that they're more likely to be diagnosed with
49:34
what I classify as these lower status disability categories
49:38
of emotional disturbance and intellectual disability.
49:42
And so it raises this question of,
49:46
like, how is special education operating for students by race?
49:50
When we look at white children being more likely to get services through
49:55
a specific learning disability, through for adhd,
50:00
for speech language impairment, we're largely looking at relatively
50:05
low stigma services that children can receive
50:09
generally in inclusive environments. They still get access to the
50:13
general education curriculum, they're getting accommodations,
50:17
they're getting supports, versus what we see happening
50:22
for kids that are getting diagnosed with emotional disturbance or
50:26
intellectual disability that often result in a removal from the classroom.
50:30
Yeah, yeah, yeah, yeah. Interesting. Yeah.
50:34
Wow. You know, Rachel, I have a couple of questions just
50:38
to wrap up the interview. One is, it's twofold,
50:42
the first one, and it's about things
50:46
that we're having and things that we're not. And that is what conversations we're
50:50
not having in the disproportionality debate
50:55
that we should have and what assumptions are we having
50:58
that maybe they're faulty? So, I mean,
51:02
these are related, right? Yeah, yeah, yeah. They just have two sides of the
51:06
same coin. Yeah. I think that.
51:11
I mean, one is that we've talked about, right. That we need to stop assuming
51:15
that low academic achievement is a good indicator
51:19
of clinical need for special education services.
51:22
Right.
51:25
And you know that I think where
51:30
we need to be going, I think that we need to
51:34
think about this beyond just disproportionality. We need to
51:38
think about the variety of processes that are created
51:43
by racism and ableism in our society and in our educational system.
51:48
That the way that we need to. That our schools categorize
51:52
students, the way that we. The way
51:56
that we allocate services, that we
52:00
think about these in very black and white
52:04
ways,
52:07
and they are shaped. Our schools
52:11
are centered around this idea that we can sort children into these different categories and
52:15
that they have meanings and they serve important purposes.
52:19
Right. Like, I would never say that we shouldn't classify kids with disabilities because from
52:24
a civil rights perspective, we want to make sure that disabled children are getting what
52:27
they need. And at the same time,
52:33
we're dividing kids into disabled or not, we're dividing
52:37
it into like, is it over representation or underrepresentation when there
52:41
are so many different processes happening in schools that reflect the
52:45
ways that these are ableist spaces and racist spaces that are not serving so many
52:49
of our children? Well, yeah, it seems like the over under is like the tip
52:53
of the iceberg and which is looking at the iceberg from above
52:57
the water without paying too much attention what's going on underneath.
53:00
Exactly. Especially because, you know,
53:04
the. So if you identify overrepresentation
53:08
of, say, black children, does that mean,
53:12
do you identify that as a good thing that they're getting services, or do you
53:15
identify it as a bad thing that they're, you know, that they might be excluded
53:19
and stigmatized and it's all of those things at once.
53:24
Yeah, yeah, yeah, yeah. It's not one or the other. Right. Those things happen at
53:27
the same time and. And it has, you know, different implications depending on
53:31
your race. Right. Like white children's experiences in special ed are, on the whole,
53:35
better than the experiences of black children.
53:38
Yeah. So my last question is similar, but it's looking at
53:42
the future. Like, if you look at 10 years ahead, what would you like to
53:45
see change in research or policy around this issue?
53:49
So as much as I as a researcher want
53:53
to keep figuring out what's going on here, I think that what we
53:57
really need to be doing to achieve real equity at the intersection
54:01
of race and disability is to think about, okay,
54:05
we have uncertainty here about what the nature of the disproportionality
54:10
is and what it means, but what can we do
54:13
to minimize the negative effects? And what I keep
54:18
coming back to is that we need to improve
54:22
the kinds of educational experiences that kids
54:25
are having before they get to a point where they get identified with a disability.
54:30
And we also. So that we can make sure that kids who
54:35
aren't officially identified are still getting the supports they need. And on the
54:39
other hand, among the kids that are receiving special edges, we need to make sure
54:42
that services are as inclusive as possible. Yeah. Because that
54:47
way we can reduce some of the harms of segregation. So if
54:51
we could create truly inclusive and well
54:56
differentiated classrooms with universal design for learning,
55:00
we could make sure that we are really reducing the impacts
55:04
of these inequitable selection effects.
55:07
Because everybody could be getting what they need regardless of
55:11
their label. And I. That would be great. And it's
55:15
like, it's what, you know, multi tiered systems of support is supposed to be doing
55:19
in many ways. Yeah. How. How to provide the
55:23
supports they need with how. How to rely on making those kinds of decision of
55:27
year in, year out with those services. Right, exactly.
55:30
Exactly. And we need, you know, I would never just say, hey, teachers,
55:34
do more things. We need to provide resources to make that happen.
55:38
We need more teachers in schools, we need support staff to support teachers,
55:42
we need interventionists, and we need inclusive education
55:46
support folks to make sure that schools have
55:50
the resources they need to implement that plan. Yeah.
55:53
Well, we'll leave with those great ideas and thoughts. Food for thought.
55:57
Thank you so much, Rachel, for being with us. It's been a pleasure to have
56:00
you on the show. Thank you so much, Federico. It's really nice to talk with
56:03
you. Thank you for
56:07
listening to Diver. I hope you learned from this episode as
56:11
much as I did. Please help us to spread the word about the show.
56:15
See you next time.