The School to Prison Pipeline: Understanding the Problem and What Research Suggests Can Get Us Back on Track By Daniel J. Losen.

The School to Prison Pipeline: Understanding the Problem and What Research Suggests Can Get Us

Back on Track By Daniel J. Losen

Outline of PresentationDescription of the school to prison pipeline with

a focus on race and disability disparitiesWhy this is an education issue, and what is at

stakeWhat the data suggests are contributing factorsThe role of povertyWhy disparate impact analysis should inform

policy decisionsWhat research suggests can be doneThe importance of addressing disparities in

remedies

Opportunities Suspended: The Disparate Impact of Disciplinary ExclusionBy Daniel Losen and Jonathan GillespieLooked at K-12Out of School and Off Track: The Overuse of

Suspensions in American Middle and High Schools, by Daniel Losen and Tia Martinez

www.schooldisciplinedata.orgResearch summary available there, too

(soon to be a book, Closing the School Discipline Gap: Research to Policy, (Teachers College Press, 2014))

My Story….

4

Son of school administratorsTaught for 10 yearsMarried to a public school

educatorDiscipline issues…

Secondary Suspension Rates: Then and Now (all students)

1972-73 2009-100

5

10

15

20

25

30

11.8

24.3

67.1

6.1

12

5.6

8.4

2.4 2.3

BlackWhiteLatinoAmerican Ind.Asian American

Suspension Rates in Secondary Schools are larger and Deeper Still For Students With Disabilities (2009-10)

BEWARE National ComparisonsThis is a tremendous national problem

with devastating impact on the lives and futures of our children.

Being at or near the average for out-of-school suspensions should be cause for alarm not complacency….

Research and Experts in Children’s DevelopmentAmerican Psychological AssociationAcademy of American PediatricsNational Association of School

PsychologistsCenter for Disease ControlNational School Boards Association all

agree:Suspension should be only utilized as a measure of last resort.

One third of all juveniles behind bars are student with disabilities

10

Photo of youth behind bars

Discipline Disparities (K-12) by Disability Category (2010-11)

Disability Category ALL

SWDsED OHI SLD MR

U.S AVG(As % of each group’s enrollment)

10-13% 29% 14% 12% 9%

Just Differences in Behavior? Or Do School Policies and Practices Matter?

Risk for Suspension in Selected Maine Districts (Secondary Schools) 2009-2010

Three Large Districts in Maine

Students WithoutDisabilities

Students With Disabilities

Male Students With Disabilities

Portland 4% 6.5% 11%

Bangor 5.3% 11% 14.3%

Lewiston 12.7% 22.2% 31.4%

Maine Sample* 4.7%* 14.5%* 18.9%*Source: Civil Rights Data Collection, U.S. Dept. of Education

*Maine sample is unpublished estimate for secondary schools.

*Not for citation

Data certified as accurate by school districts.

For your district go to:

www. schooldisciplinedata.org

If you are in a high-suspending district, you can reject the status quo….District and school level policy and

practices make a tremendous difference.Always question large disparities….

2009-10 District Distribution of K-12 Suspension Rates for Students with Disabilities in CT, MA, VT, NH, ME (Number of Districts)

0-4.99 % 5.0 to 9.99 % 10.0 to 19.99 % 20 % or above

Many small districts and charters not included

233 143 132 36 (9 over 40%)

Many School Districts Suspend Less than 3% of Their students

We assigned a risk of 3%, as “low suspending” (the approximate national average for Whites in the early 1970s). We then counted the number of districts in the sample that enrolled at least 1000 students and at least 10 students from a given subgroup.

Of 4,504 districts, for Black students it was 3% or less in 1,437 districts.

Of the 4,667, for students with disabilities, it was 3% or less in 653 districts.

Racial Differences in Behavior?Likely many factors contributing to patterns but

despite anecdotes, no research supports this as an explanation for suspensions (and it would be hard to conduct such research)

Disparate impact analysis is not trying to detect intentional discrimination but whether the policy or practice is educationally necessary….thas said….

Most extensive and robust study of Texas middle school students, controlled for other factors and found that white students were more likely than Blacks to commit a “mandatory” offense.

The Largest Racial Disparities are for less serious “Discretionary” violationsStudy of Indiana- 95% non-violent “other” Council of State Governments extensive

review in Texas, a large majority will be removed from class on disciplinary grounds at least once in 6 years.

Whites in Texas were more frequently disciplined for “non-discretionary” infractions. (Possession of weapons/drugs….)

Are most of these kids typical adolescents, or dangerous thugs?

Racial Disparities In Use of Suspension for First Time Offenders By Type of Offense

Cell Phone Dress Code Disruptive Display of Affection0

5

10

15

20

25

30

35

40

45

32.7

38.340

42.9

14.516.6

23.5

14.5

Black

White

Black first-time offenders sus-pendedat higher rates than Whites for the same minor offenses

Cell Phone Use

The Civil Rights Project

DRAFT - 20

4,838 Whites and 2,242 Blacks disciplined for first offense of Cell Phone Use.

32.7% (732) of the Blacks were suspended out of school.

14.5% (704) of the Whites were suspended out of school.

Unpacking the Poverty RationaleThere is likely more to the observed differences

than poverty can explain.Longitudinal study in Texas, adjusted for 83

variables including race, poverty, prior behavior, and found schools make a big difference.

Chicago study found that the schools serving the students from the poorest and highest crime neighborhoods could feel as safe as schools serving the wealthiest and safest neighborhoods…

We Know Poverty Matters, But…We find large differences in suspension rates by

race and disability status, often shrink, but only a little bit, after controlling for poverty.

Large race and disability and gender differences are not explained away by poverty.

But what if poverty did explain the differences?The poverty explanation, if it ever did sufficiently

explain the race and disability differences in statistical terms, would still only satisfy the concerns about unlawful different treatment driving the disparities…

…but would beg the question, why are schools suspending so many poor kids so often?

What About Poverty?What is the concept behind the question?

Is it that poor kids misbehave more often?If so, should they then spend more time at

home unsupervised as the solution?What about gender differences?What about disability differences? Can we

exclude children from educational opportunity on the basis of having a disability? [Hint, the Equal Protection Clause of the U.S. Constitution says NO!]

What about race with disability? Disability with gender?

Students With DisabilitiesSome may be misbehaving more because:

It’s a manifestation of their disabilityTheir behavioral improvement plan needs

adjusting.They are frustrated because the supports

and services they receive are inadequate.Disability bias.Other incentives to push them out.How is suspending them at double the rate of

their non-disabled peers educationally sound?

Disparate Impact ApproachDifferences in behavior are not relevant

because…the intent of the educators is not at issue.

Once a disparity is established, disparate impact analysis asks whether the policy or practice (the disciplinary response) by educators is educationally justifiable.

The Core Concept is Exclusion from Education as PunishmentIs it OK to kick out the “bad” kids out so the

“good” kids can learn?Is this the only choice before us?Is it educationally sound policy or practice for

public schools? Does sending misbehaving kids home work?

Deterrence?Behavior?Education outcomes?

Does it make economic sense?Will it make our communities safer?

Using A Disparate Impact Analysis to Address DisparitiesOne: Is a policy or practice adversely

impacting some groups of children more than others?

Two: Is the policy or practice educationally justifiable?

Three: Even if justifiable, is a less discriminatory alternative is available?

The First Question: Is there a policy or practice that has an adverse impact (unintended consequences) that harms some groups more than others?

Large disparities in many districts (race and disability and gender).

Fact: Suspensions are a leading indicator of future dropping out and incarceration

Johns Hopkins study: Associated risk of dropping out from just one out-of-school suspension (from 16% to 32%)

Non-partisan Council of State Governments longitudinal study of every middle school student in Texas showed similar associated risks, including threefold increase in risk for juvenile justice involvement. “Breaking Schools Rules.”

Question Two: Educational Justification?How about when students are a danger to

themselves or others, or exhibit other extreme misbehaviors? But unsupervised?

What about to de-escalate and/or investigate?….

Are these justifications for the frequent use of out of school suspension for less serious non-violent conduct? Suspension for truancy and tardiness? Dress code violations?

Focusing on behavior differences skirts the core question.Is out-of-school suspension justifiable, and

even if so, is it the best practice?

Three Prominent JustificationsTo get parents attentionTo deter future misbehavior…peers and

studentTo ensure a safe, orderly and effective

educational environment, “So the good kids can learn.”

What does the research say?

You need to kick out the “bad” kids so the “good” kids can learn…has no research support. Research (Indiana) shows that principals who treated discipline as

part of the educational mission, embraced clear rules and fair consequences, and regarded suspension as a measure of last resort had lower suspension rates and higher test scores than those that embraced highly punitive measures and saw misbehaving students and their parents as the source of the problem.

Texas reports compared similar districts and after controlling for over 80 variables and there were no benefits to higher rates of suspension in terms of test scores.

Qualitative research has demonstrated that students often behave differently in different classrooms.

Research by Pedro Noguera suggested that when teachers were allowed to remove the disruptive kids, new disruptive kids soon emerged.

Kicking out the “bad” kids does not improve the learning environment as measured by achievement, after controlling for race and poverty.

Myth “BUSTED”

Will deter future misbehavior…Truancy? Mendez study: suspension in grade 6 highly

correlated with repeated suspensions in the future.

Data demonstrate increasing numbers of students suspended between grades 6 and 9.

No research behind this common assumption.

Pinellas County’s Cohort’s Suspension Rate As Students Advanced in Grade

Grade

4 5 6 7 8 9 10 11 12

% OSS

4 5 11 13 15 19 20 18 19

The Civil Rights Project

DRAFT - 34

Percentage of cohort’s enrollment suspended at least once: (Mendez 2003)

Get Parental Attention

Increased parental attention is a sound goal.American Academy of Pediatrics conclusions

Adds stressIncreases risk factors for child and familyWon’t help anyone if seriously dysfunctional

parenting is the root cause

Many other ways to improve parental involvement

Relieves Teacher Stress?In the short-term, perhaps, but….Studies suggest that frequent suspensions

break down teacher-parent and teacher-student trust and harms these relationships.

And we know it adds to student disengagement.

Most teachers will admit that students do not come back better behaved.

Frequent suspensions can undermine a teacher’s authority in the classroom.

Question Three: Are There Less Discriminatory Alternatives?

Major Remedies IncludeLess punitive responses (school code and

change in theory)Tiered intervention strategies (PBIS; RtI;

Threat Assessment Protocol)Strategies targeting social and emotional

learningRestorative practices

Common Threads: What WorksImproving teacher/student engagementTeacher/parent engagement“Buy in” by leaders and teachersPositive rather than punitive approachesReflection on the dataRejecting the status quo and problem

solvingAcknowledging how the adults and the the

school system contributes to the problems

Frequently suspending adolescents for minor infractions does not work.Eliminate the unsound practice, starting with the

most obvious ones: Baltimore forbids the use of suspension for truancy

and several other minor offenses McKinley High School in Boston reversed its closed

door policy that constructively suspended all tardy students.

Connecticut limits out-of-school suspensions. Reduce discretion around out-of-school suspension.

Research indicates that PBIS is effective reducing disciplinary referrals, but should be aligned with school code.

Need to Move Beyond ComplianceLearn from the experience of other school

districts: http://www.nbcbayarea.com/news/local/Change-in-Suspension-Policy-H

elps-San-Jose-High-School-223997471.html

http://www.npr.org/2013/06/02/188125079/why-some-schools-want-to-expel-suspensions

Beyond Changes to Discipline Policy: AlternativesHigh quality pre-school (associated with fewer

anti-social behaviors in school).Mandatory classroom and behavior

management training for certification as well as professional development for new teachers.

The Seattle Social Development ProjectCoaching for teachers to improve instruction andDevelopmentally sequenced parenting workshopsSocial and citizenship skills training for children to

learn and practice these skills

Social and Emotional HealthImprove procedural protections for

students with disabilitiesReduce misidentification in special

educationWrap-around servicesMulit-systemic therapyTrauma awareness and support

Social and Emotional LearningTeaches students directly about how to deal

with frustration and angerProblem solving techniquesGroup work skillsConflict resolution skills….

Restorative PracticesOffender accountability is centralAs is the value of each child to the

community as a whole.Seeks to make the victim whole again.Seeks to get at the root cause of the

offender’s behavior.Emphasis on prevention.

Tiered Intervention StrategiesSchool-wide Positive Behavioral

Intervention SystemsRtI: Response to intervention approach has

behavioral componentVirginia’s Threat Assessment Protocol

Tiers of a Process (not a new set of labels)

PBS/RTI Three-Tiered Model

Will these alternatives reduce the disparities?They don’t alwaysHow change is measured and evaluated

can lead to different conclusions

What about unconscious bias?Disparities remain a problem, even if there

is no clear evidence of intentional discrimination based on race or disability status.

Systemic failure to address the behavioral challenges experienced by students with disabilities is unlawful.

Implicit bias against students with disabilities and students of color can influence both the perceptions of misbehavior and the chosen responses.

Entangled Contributors

Discrimination based on gender?Think about gender bias.Do we still accept deteriorating performance by girls

in math?

The MIT Beer Experiment (Ariely)

In a blind-fold test, tasters overwhelmingly preferred vinegar laced MIT beer over a commercial brand lite beer.

When the test was repeated, but participants were told that the MIT beer had a trace of vinegar added, participants crinkled their noses as they tasted the MIT beer and overwhelmingly chose the commercial lite beer.

Expectations altered their experience of the beer.

Bias can alter our senses….Bias can alter what we experience, what we

pay attention to and what we ignore.Biased expectations can affect our

evaluation of what we hear and see… Auditions behind blinds

What to do about implicit bias?

Expect bias, as part of being an irrational human being.

Think about bias in other areas, such as economics: The Hidden Forces That Shape Our Decisions (Predictably Irrational by Ariely)

Test yourself: www.implicit.harvard.eduMonitor and discuss the racial and disability

disparities as part of the work of the school and district.

Avoid the comfortable explanations…

www.implicit.harvard.edu

55

You Can Test Yourself

Cognitive Dissonance

Perceptions of Behavior in the Classroom and Subsequent ResponsesConnections to student and parentsMutual respectConfrontation or problem solvingDe-escalationPreparation and engagementPredictions of outcomes

Challenges:Reluctance to own the problemBlame povertyBlame bigger problems outside the

district’s control

We all likely harbor some biases that we are unaware of…

59

http://www.youtube.com/watch?v=Tt9d8CKsypsAlan Alda and test makers take the test and discuss their

own biased results.

Reflect on the DataRace, disability, and gender data disparities

are often shocking.Often districts do not routinely review or

use data that breaks out discipline disparities by race, disability or gender.

One way to push back on bias is to use the disaggregated data on a regular basis and reject the status quo of disparities if these are revealed.

Lack of access to the data

Advocacy for better data collection…

• And use the data once it is collected…• The “Pledge of ____” ends with:

The Civil Rights Project

DRAFT - 62

The Difference Training Can Make

• My inexperience and frequent referrals.• Typical mistakes easily corrected:

– Focusing on wrongs rather than rights. – Assuming bad motives and taking things personally.– Poor preparation.– Deferring to higher authorities.– Ignoring special needs.– Group punishments.– Parental outreach reluctance– Classroom confrontations….

The Civil Rights Project

DRAFT - 63

Do Schools Meet Their Legal Obligations to Students With Disabilities?

• Two districts in Delaware: • About 50% of the principals knew that students

with disabilities had additional due-process rights• Manifestation determination: – Behavior caused by disability, or– Resulted from failure to properly implement the IEPAdministrators often complain that they have too few

resources to meet their special education obligations….

The Civil Rights Project

Police in Our Schools• How well trained?• Cost benefit analysis compared with support for

students and teachers?• How evaluated? Arrests or lack thereof.• Increase in police presence, without education

protocol resulted in dramatic rise in misdemeanor offenses and minor rise in felonies.

• Costs associated with over-use by educators…• Cops or Counselors?

DRAFT - 65

Clayton County Georgia Referrals to Juvenile Court (Source: Judge Steven Teske)

The Civil Rights Project

DRAFT - 66

A Positive Outlook for ChangeNational initiative and growing consensusNew understanding of the harmsSome simple solutions (i.e. Baltimore)Promising long-term remedies based on

researchUnacceptable economic implications of the

status quoUnions reconsidering their positionsSchools proving that they can make a

difference!

How to Measure Success?

• Improving sense of safety and school climate for teachers as well as students

• Improving achievement scores• Improving attendance• Improving graduation rates• Decreasing involvement in the juvenile

justice system• Cost savings from more effective schools

69

The EndNew web tool: www.schooldisciplinedata.orgNew Policy Brief: Daniel Losen, Discipline Policies, Successful

Schools, and Racial Justice. National Education Policy Center Release on October 5, 2011 National Press Club, D.C.

http://nepc.colorado.eduNew Reports: Out of School and Off Track; and Opportunities

Suspendedhttp://www.civilrightsproject.ucla.edu/National Council of State Governments Justice Center Report:

Breaking Schools’ Rules: A Statewide Study on How School Discipline Relates to Students’ Success and Juvenile Justice Involvement

http://justicecenter.csg.org/resources/juvenilesContact for Daniel J. [email protected]