Parents Involved in Community Schools v. Seattle School District No. 1 Brief for Amici Curiae

    Nos. 05-908, 05-915

In  T h e

Smpum' (Ernxrt of the Thxxteh States

Parents Involved in Community Schools,
Petitioner,

v.

Seattle School District No. 1, et al.,
Respondents.

Crystal D. Meredith, custodial parent and
NEXT FRIEND OF JOSHUA R Y A N  MCDONALD,

Petitioner,
v.

Jefferson County Board of Education, et al.,
Respondents.

ON WRITS OF CERTIORARI TO THE 
UNITED STATES COURTS OF APPEALS 
FOR THE NINTH AND SIXTH CIRCUITS

BRIEF FOR AMICI CURIAE 
THE AMERICAN PSYCHOLOGICAL ASSOCIATION 
AND THE WASHINGTON STATE PSYCHOLOGICAL 

ASSOCIATION IN SUPPORT OF RESPONDENTS

Nathalie F.P. Gilfoyle John Payton 
General Counsel Counsel o f Record

Lindsay Childress-Beatty David W. Ogden 
Deputy General Counsel LEONORA R. KRUGER 

American Psychological Wilmer Cutler Pickering 
Association Hale and Dorr llp

750 First Street, N.E. 
Washington, DC 20002 
(202) 336-5500

1875 Pennsylvania Ave., N.W. 
Washington, DC 20006 
(202) 663-6000



TABLE OF CONTENTS

Page
TABLE OF AUTHORITIES.................................................. iii
INTEREST OF AMICI CURIAE...........................................1
SUMMARY OF ARGUMENT.........................................   2

ARGUMENT..............................................................................3
I. Racial Diversity In Primary And Secon­

dary Education Challenges Racial 
Stereotypes And Promotes Harmony 
And Mutual Respect................................................. 3
A. Absent Early Intervention, Normal Cogni­

tive Processes Can Lead To Racial Stereo­
typing And Racial Bias.............................................. 5
1. The nature and consequences of racial

stereotypes........................................................... 5
2. Once formed, racial stereotypes are

difficult to abandon............................................. 8
B. Racial Diversity In K-12 Education Can

Inhibit The Formation Of Racial Stereo­
types And Racial Bias.............................................. 10
1. The importance of intergroup contact............. 10
2. Short-term effects of intergroup

contact in primary and secondary 
education............................................................. 14

3. Long-term effects of intergroup con­
tact in primary and secondary educa­
tion.......................................................................19

C. The Benefits Of Diversity In K-12 Educa­
tion Are More Likely To Accrue When 
There Is A Critical Mass Of Students Of 
Different Racial Backgrounds................................. 20



11

II. Meaningful Intergroup Contact Is 
Unlikely To Occur In Public K-12 Edu­
cation W ithout The School Districts’ 
Intervention................   22
A. Because Of Largely Unconscious, Auto­

matic Cognitive And Emotional Re­
sponses, Individuals Often Avoid Inter-
group Contact............................................................ 23

B. Because Of Intergroup Avoidance, Private
Choice Alone Is Unlikely To Produce Sub­
stantial Diversity In Public K-12 Educa­
tion............................................    25

CONCLUSION.........................................................................27

TABLE OF CONTENTS— Continued

Page



Ill

TABLE OF AUTHORITIES 

CASES

Gratz v. Bollinger, 539 U.S. 244 (2003)................................... 3
Grutter v. Bollinger, 539 U.S. 306 (2003)........................... 3, 4
Metro Broadcasting, Inc. v. FCC, 497 U.S. 547

(1990)............................................................... ...................2
Parents Involved in Community Schools v. Seattle 

School District, No. 1, 426 F.3d 1162 (9th Cir.
2005).............................................................................2,3

Regents o f University o f California v. Bakke,
438 U.S. 265 (1978)...........................................  3

BOOKS AND ARTICLES
APA Resolution on Prejudice, Stereotypes, and 

Discrimination (Feb. 2006), available at 
http://www.apa.org/pi/prejudice_discrimination
_resolution.pdf.........................................................   1

Aboud, Frances E. & Maria Amato, Developmental 
and Socialization Influences on Intergroup 
Bias, in Blackwell Handbook o f Social Psy­
chology: Intergroup Processes 65 (Rupert
Brown & Samuel L. Gaertner eds., 2001)....................8, 9

Aboud, Frances E. & Sheri R. Levy, Interventions 
to Reduce Prejudice and Discrimination in 
Children and Adolescents, in Reducing Preju­
dice and Discrimination 269 (Stuart Oskamp
ed, 2000)............................................................... ..9,16,18

Allport, Gordon, The Nature o f Prejudice (1954)........... 10,11
Baron, Andrew Scott & Mahzarin R. Banaji, The 

Development o f Implicit Attitudes: Evidence of 
Race Evaluations from. Ages 6 and 10 and
Adulthood, 17 Psych. Sci. 53 (2006)..................................9

Bettencourt, B. Ann & Nancy Dorr, Cooperative 
Interaction and Intergroup Bias: Effects of 
Numerical Representation and Cross-Cut Role 
Assignment, 24 Personality & Soc. Psychol.
Bull.' 1276 (1998)..........................................................  13

Page(s)

http://www.apa.org/pi/prejudice_discrimination


IV

TABLE OF AUTHORITIES— Continued

Page(s)
Betts, Julian R., et al., Does School Choice Work?

Effects on Student Integration and Achieve­
ment 44 (2006)...................................................................27

Bigler, Rebecca S. & Lynn S. Liben, A Cognitive- 
Developmental Approach to Racial Stereotyp­
ing and Reconstructive Memory in Euro-
American Children, 64 Child Dev. 1507 (1993)..........8, 9

Blair, Irene V., The Malleability o f Automatic 
Stereotypes and Prejudice, 6 Personality &
Soc. Psychol. Rev. 242 (2002)...................... ........... ......  10

Braddock, Jomills & James M. McPartland, Social- 
Psychological Processes That Perpetuate Ra­
cial Segregation: The Relationship Between 
School and Employment Desegregation, 19 J.
Black Stud. 267(1989).....................................................  19

Braddock, Jomills Henry, II & Tamela McNulty 
Eitle, The Effects o f School Desegregation, in 
Handbook of Research on Multicultural Edu­
cation 828 (James A. Banks & Cherry A.
McGee Banks eds., 2d ed. 2004).....................................  19

Clotfelter, Charles T., After Brown; The Rise and
Retreat o f School Desegregation (2004)........................ 26

Devine, Patricia G., Stereotypes and Prejudice:
Their Automatic and Controlled Components,
56 J. Personality & Soc. Psychol. 5 (1989)..................... 23

Dovidio, John F., et al., On the Nature o f Prejudice: 
Automatic and Controlled Processes, 33 J. Ex­
perimental Soc. Psychol. 510 (1997)........................  10,23

Dovodio, John F., et al., Reducing Contemporary 
Prejudice: Combating Explicit and Implicit 
Bias at the Individual and Intergroup Level, in 
Reducing Prejudice and Discrimination 137 
(Stuart Oskamp ed., 2000).........................................  12,14



V

TABLE OF AUTHORITIES— Continued

Page(s)
Dovidio, John F., et al., Why Can’t We Just Get

Along? Interpersonal Biases and Interracial 
Distrust, 8 Cultural Diversity & Ethnic Minor­
ity Psychol. 88 (2002)...................................................8, 24

Eberhardt, Jennifer L., et ah, Seeing Black: Race,
Crime, and Visual Processing, J. Personality
& Soc. Psychol. 876 (2004).............................................. . 7

Fiske, Susan T., Interdependence and the Reduc­
tion o f Prejudice, in Reducing Prejudice and
Discrimination 115 (Stuart Oskamp ed., 2000)............13

Fiske, Susan T., Stereotyping, Prejudice and Dis­
crimination, in 2 The Handbook of Social Psy­
chology 357 (Daniel T. Gilbert et al. eds., 4th
ed. 1998)...................................................................... 5, 6, 7

Gaertner, Samuel L. & John F. Dovidio, The Aver­
sive Form of Racism, in Prejudice, Discrimi­
nation, and Racism 61 (John F. Dovidio &
Samuel L. Gaertner eds., 1986)............................ 7, 23, 24

Gaertner, Samuel L., et al., The Contact Hypothe­
sis: The Role o f a Common Ingroup Identity 
on Reducing Intergroup Bias Among Majority 
and Minority Group Members, in What’s So­
cial About Social Cognition?: Research on So­
cially Shared Cognition in Small Groups 230
(Judith L. Nye & Aaron M. Brower eds., 1996)............13

Graham, Sandra & Jaana Juvonen, Ethnicity, Peer 
Harassment, and Adjustment in Middle 
School: An Exploratory Study, 22 J. Early
Adolescence 173 (2002)....................................................20

Johnson, David W. & Roger T. Johnson, The Three 
Cs o f Reducing Prejudice and Discrimination, 
in Reducing Prejudice and Discrimination 239
(Stuart Oskamp ed., 2000)...............................................  13

Juvonen, Jaana, et al., Ethnic Diversity and Peer 
Perceptions o f Safety in Urban Middle 
Schools, 17 Psych. Sci. 393 (2006)..................................  16



TABLE OF AUTHORITIES— Continued

Page(s)
Killen, Melanie, et al., Morality in the Context of 

Intergroup Relations, in Handbook o f Moral 
Development 155 (Melanie Killen & Judith
Smetana eds., 2006)........................................... 6, 8, 10,16

Killen, Melanie, et al., The Social Developmental 
Benefits o f Heterogeneous School Environ­
ments, in Lessons in Integration: Realizing the 
Promise o f Racial Diversity in America’s 
Schools (Erica Frankenberg & Gary Orfield
eds., forthcoming 2006)....................................................  10

Lord, Charles G. & Delia S. Saenz, Memory Deficits 
and Memory Surfeits: Differential Cognitive 
Consequences o f Tokenism for Tokens and Ob­
servers, 49 J. Personality & Soc. Psychol. 918
(1985)................................................................................. 21

Mackie, Diane M., et al., Social Psychological 
Foundations o f Stereotype Formation, in 
Stereotypes and Stereotyping 41 (Neil C. Mac­
rae et al. eds., 1996)....................................................5, 6, 8

Margie, Nancy Geyelin, et al., Minority Children’s 
Intergroup Attitudes About Peer Relation­
ships, 23 British J. Developmental Psych. 251
(2005).................................................................................  16

McGlothlin, Heidi & Melanie Killen, Intergroup At­
titudes of European American Children At­
tending Ethnically Homogeneous Schools, 77
Child Dev. 1375 (2006)................................................ 15,16

McGlothlin, Heidi, et al., European-American 
Children’s Intergroup Attitudes About Peer 
Relationships, 23 British J. Dev. Psychol. 227
(2005)....................................... .................................... 15,16

McKown, Clark & Rhona S. Weinstein, The Devel­
opment and Consequences of Stereotype Con­
sciousness in Middle Childhood, 74 Child Dev.
498 (2003) 22



v ii

TABLE OF AUTHORITIES— Continued

Page(s)
Migdal, Michael J., et al., The Effects o f Crossed 

Categorization on Intergroup Evaluations: A 
Meta-Analysis, 37 Brit. J. Soc. Psychol. 303
(1998)................................................................................... 5

Mullen, Brian, et al., Ingroup Bias as a Function of 
Salience, Relevance, and Status: An Integra­
tion 22 Eur. J. Soc. Psychol. 103 (1992)........................... 5

Mullen, Brian & Li-tze Hu, Perceptions of Ingroup 
a?id Outgroup Variability: A  Meta-Analytic 
Integration, 10 Basic & Applied Soc. Psychol.
233 (1989)......................................................'................ ..... 6

Pettigrew, Thomas F., Intergroup Contact Theory,
49 Ann. Rev. Psychol. 657 (1998).............................  11, 13

Pettigrew, Thomas F. & Linda R. Tropp, A Meta- 
Analytic Test o f Intergroup Contact Theory,
90 J. Personality & Soc. Psychol. 751
(2006)......................................................................... passim

Pettigrew, Thomas F. & Linda R. Tropp, Does In­
tergroup Contact Reduce Prejudice? Recent 
Meta-Analytic Findings, in Reducing Preju­
dice and Discrimination 93 (Stuart Oskamp
ed., 2004).....................................................................  12, 16

Rosenthal, Robert, Meta-Analytic Procedures for
Social Science (rev. ed. 1991).......................................... 11

Schofield, Janet W., Black and White in School:
Trust, Tension, or Tolerance (1989)..............................  15

Schofield, Janet W., School Desegregation and In­
tergroup Relations: A  Review o f the Literature,
17 Rev. of Res. in Educ. 335 (1991)..........................  13,17

Schofield, Janet W. & Rebecca Eurieh-Fulcer,
When and How School Desegregation Im­
proves Intergroup Relations, in Blackwell 
Handbook of Social Psychology 475 (Rupert 
Brown & Samuel L. Gaertner eds.,
2001)..........................................................  17,18,19, 20, 21



VU1

Schofield, Janet W. & H. Andrew Sagar, Desegrega­
tion, School Practices, and Student Race Rela­
tions, in The Consequences of School Desegre­
gation 58 (Christine H. Russell & Willis D.

TABLE OF AUTHORITIES— Continued

Page(s)

Schofield, Janet W. & H. Andrew Sagar, Peer In­
teractions in an Integrated Middle School, 40
Sociometry 130 (1997)...................................................... 15

Spencer, Steven J., et al., Automatic Activation o f 
Stereotypes: The Role o f Self-Image Threat, 24
Personality & Soc. Psychol. Bull. 1139 (1998)...............23

Stangor, Charles & David McMillan, Memory for  
Expectancy-Congruent and Expectancy- 
Incongruent Information: A  Review of the So­
cial and Social Developmental Literatures, 111
Psychol. Bull. 42 (1992).................................................... 7

Steele, Claude M., A Threat in the Air: How Stereo­
types Shape Intellectual Identity and Per­
formance, 52 Am. Psych. 613 (1997)................................ 22

Stephan, Walter G. & Cookie White Stephan, Inter­
group Relations in Multicultural Education 
Programs, in Handbook o f Research on Multi­
cultural Education 782 (James Banks & Carrie
McGee-Banks eds., 2004)................................................. 18

Stephan, Walter G. & Cookie White Stephan, Im­
proving Intergroup Relations 47 (2001)......................... 18

Stephan, Walter G. & Cookie White Stephan, Inter­
group Anxiety, 41 J. Soc. Issues 157
(1985)....................................................................... 8, 24, 25

Stephan, Walter G. & Cookie White Stephan, Inter­
group Relations (1996)....................................................  19

Taylor, Shelley E., et al, Categorical and Contex­
tual Bases of Person Memory and Stereotyp­
ing, 36 J. Personality & Soc. Psychol. 778 (1978)..........21



TABLE OF AUTHORITIES— Continued

Page(s)
Tropp, Linda R. & Mary Prenovost, The Role o f In­

tergroup Contact in Predicting Inter-Ethnic 
Attitudes: Evidence from Meta-Analytic and 
Field Studies, in Intergroup Relations: An In­
tegrative Developmental and Social Psycho­
logical Perspective (Sheri Levy & Melanie Kil-
len eds., forthcoming 2006).............................................  14

Venezia, Andrea, et al., Betraying the College 
Dream (Stanford University Bridge Project
2006)..................................................................................4

Wells, Amy Stuart & Robert L. Crain, Perpetua­
tion Theory arid the Long-Term Effects of 
School Desegregation, 64 Rev. Ed. Res. 531
(1994)...........................................................................  19, 20

Wells, Amy Stuart, et al., How Desegregation 
Changed Us: The Effects o f Racially Mixed 
Schools on Students and Society, available at 
http://cms.tc.columbia.edU/i/a/782_ASWells0415
04.pdf (last visited Oct. 9, 2006)..................................... 20

Wilder, David A., Perceiving Persons as a Group: 
Categorization and Intergroup Relations, in 
Cognitive Processes in Stereotyping and Inter­
group Behavior 213 (David L. Hamilton ed.,
1981).......................................................................... 6, 7, 13

Wright, Stephen C., et al., The Extended Contact 
Effect: Knowledge of Cross-Group Friendships 
and Prejudice, 73 J. Personality & Soc. Psy­
chol. 73 (1997)......................................................'.............16

Yee, Mia D. & Rupert Brown, Self-Evaluations 
and Intergroup Attitudes in Children Aged 
Three to Nine, 63 Child Dev. 619 (1992)...................... 5, 8

http://cms.tc.columbia.edU/i/a/782_ASWells0415


INTEREST OF AMICI CURIAE

The American Psychological Association (APA)1 is a 
voluntary, nonprofit, scientific and professional organization 
founded in 1892. It is the major association of psychologists 
in the United States, with more than 145,000 members and 
affiliates. The APA has 54 divisions representing the full 
array of areas of emphasis within the field of psychology. 
The objects of the APA include advancing psychology as a 
means of promoting human welfare, diffusing psychological 
knowledge, and encouraging the application of research find­
ings to the promotion of health and public welfare. The APA 
places a high priority on the amelioration of stereotypes, 
prejudice, and discrimination among individuals and institu­
tions. See APA Resolution on Prejudice, Stereotypes, and 
Discrimination (Feb. 2006), available at http://www.apa.org/ 
pi/prejudice_discrimination_resolution.pdf.

Members of APA research psychological aspects of im­
portant social issues such as the causes and consequences of 
racial and ethnic prejudice and stereotypes and the devel­
opment of such prejudice and stereotypes in children. A 
body of social scientific knowledge provides a substantive 
scientific context for considering these cases. Pertinent 
studies are scientifically reliable and peer-reviewed, and, 
taken together, bear directly on the empirical claims at the 
heart of this matter.

The Washington State Psychological Association 
(WSPA) is a statewide nonprofit scientific and professional 
organization. Founded in 1947, WSPA has approximately 
700 members and affiliates. WSPA’s mission is to support 
psychologists and psychologists-in-training and to promote 
the practice of psychology in order to maintain the vitality of 
the profession in the public interest. WSPA promotes reli­

1 The parties have filed with the Court letters consenting to the fil­
ing of all amicus curiae briefs in these cases. No counsel for any party 
had any role in authoring this brief, and no one other than amici curiae 
provided any monetary contribution to its preparation or submission.

http://www.apa.org/


2

ance on scientific evidence in crafting policies that enhance 
the mental and behavioral health of Washington citizens. 
WSPA joins the APA in this brief to present compelling evi­
dence supporting the principle that local school districts 
should be permitted to create school assignment plans to en­
sure diversity throughout public education.2

SUMMARY OF ARGUMENT
I. Extensive psychological research shows that, under 

certain conditions, interaction among persons of different 
races can diminish racial stereotypes and promote cross- 
racial understanding, empathy, and mutual respect. These 
findings apply with particular force in the context of K-12 
education. Adults may find it difficult to abandon racial 
stereotypes already formed, but children who interact regu­
larly with persons of other races are less likely to fall into 
patterns of stereotypical thinking about other racial groups. 
Primary and secondary schools provide an excellent setting 
for this type of sustained interaction. Children and adoles­
cents in racially diverse schools—particularly schools de­
signed to facilitate cross-racial interaction among students— 
are more likely to learn to regard others as individuals, 
rather than simply “the product of their race,” Metro Broad., 
Inc. v. FCC, 497 U.S. 547, 604 (1990) (O’Connor, J., dissent­
ing).

It is noteworthy that the courts below broadly acknowl­
edged, in the words of one judge, “the importance of teach­
ing children, during their formative years, how to deal re­
spectfully and collegially with peers of different races.” 
Parents Involved in Community Sell. v. Seattle Sch. Dist., 
No. 1, 426 F.3d 1162, 1194 (9th Cir. 2005) (Kozinski, J., con­
curring). “The idea that children will gain social, civic, and

2 APA and WSPA gratefully acknowledge the assistance of 
Mary Margaret Brabeck, Ph.D., John F. Dovidio, Ph.D., Susan T. Fiske, 
Ph.D., Sandra Graham, Ph.D., Melanie A. Killen, Ph.D., 
Clark Atwater McKown, Ph.D., Janet Ward Schofield, Ph.D., and Linda 
R. Tropp, Ph.D., in the preparation of this brief.



3

perhaps educational skills by attending schools with a pro­
portion of students of other ethnicities and races, which pro­
portion reflects the world in which they will move, is a no­
tion grounded in common sense.” Id. at 1196 (Bea, J., dis­
senting). As this brief explains, a substantial body of psy­
chological research confirms this understanding.

II. The diversity necessary to generate these substan­
tial benefits in public K-12 education is unlikely to result 
purely as a matter of private parental choice. Fear of the 
unknown and unfamiliar often leads individuals to shy away 
from substantial interaction with persons of other races, and 
instead to gravitate toward homogeneous neighborhoods, 
communities, and schools. Even those who consciously dis­
claim racial prejudice often unconsciously harbor, and act in 
accordance with, racial stereotypes and racial bias. These 
phenomena help to explain why, all things being equal, many 
parents are unlikely to choose to send their children to 
schools predominantly populated by children of other races. 
Thus, without school district involvement, children are far 
less likely to reap the benefits of learning, at an early age, to 
resist the racial stereotypes that so often result in division 
and discrimination.

ARGUMENT
I. Racial Diversity In Primary And Secondary Educa­

tion Challenges Racial Stereotypes And Promotes 
Harmony And Mutual Respect

This Court has long recognized that student body diver­
sity is a compelling governmental interest that “legitimately 
may be served” by the consideration of race in admissions to 
selective public universities. Regents o f Univ. o f Cal. v. 
Bakke, 438 U.S. 265, 320 (1978); accord Grutter v. Bollinger, 
539 U.S. 306, 325 (2003); Gratz v. Bollinger, 539 U.S. 244, 268 
(2003). The Court has explained that the benefits of racial 
diversity in higher education are both “substantial” and 
“real”: Assembling a racially diverse class “promotes cross- 
racial understanding, helps to break down racial stereo­
types, . . . enables students to better understand persons of 
different races, promotes learning outcomes, [and] better



4

prepares students for an increasingly diverse workforce and 
society[.]” Grutter, 539 U.S. at 330 (internal quotation 
marks and citations omitted).

Peer-reviewed professional literature shows that racial 
diversity in K-12 education promotes precisely the same 
goals, and therefore represents an equally compelling inter­
est. Indeed, racial diversity in primary and secondary edu­
cation is likely to have an even greater effect on inter-race 
relations than diversity in selective institutions of higher 
education. As a practical matter, because the majority of 
children in the United States do not go on to attend selective 
four-year colleges, the benefits of K-12 racial diversity are 
likely to have a broader reach.3 And for many children who 
do not go on to attend college, primary and secondary school 
may be the only educational settings in which they have 
meaningful interaction with persons of different racial back­
grounds.

Equally—if not more—important, early interaction with 
individuals of different racial backgrounds can have pro­
found and lasting effects. Interaction between children and 
adolescents of different races helps not only “to break down 
racial stereotypes,” but to prevent the development of 
stereotypical thinking. Children begin to classify people by 
race and develop the capacity for stereotypical thinking in 
their early years. Childhood and adolescence are important 
periods for the formation of biases that may be expressed 
either consciously or unconsciously throughout a person’s 
lifetime. Interaction with others from different racial back­
grounds also enables children to develop notions of racial 
equality and fairness. Early intervention thus can signifi­
cantly lessen racial prejudices among children and, ulti­
mately, the likelihood that they will engage in discrimina­
tory behavior.

3 Cf Andrea Venezia et at, Betraying the College Dream 14 (Stan­
ford University Bridge Project 2006) (approximately 20% of first-year 
postsecondary students attend selective 4-year institutions, while 80% 
attend non-selective or minimally selective 2- and 4-year institutions).



5

A. Absent Early Intervention, Normal Cognitive 
Processes Can Lead To Racial Stereotyping And 
Racial Bias
1. The nature and consequences o f racial 

stereotypes
As young children become aware of the world around 

them, they begin to divide individual objects, situations, and 
people into categories. This categorization is a normal cogni­
tive process that permits children and adults alike to sim­
plify a complex world. Confronted with an overload of indi­
vidual stimuli, people conserve scarce cognitive energy by 
identifying the similarities between different stimuli and 
grouping them together on that basis. When the stimuli are 
people, this process leads people to group individuals into 
social categories. See generally Diane M. Mackie et al., So­
cial Psychological Foundations o f Stereotype Formation, in 
Stereotypes and Stereotyping 41, 44-45 (Neil C. Macrae et al. 
eds., 1996); Susan T. Fiske, Stereotyping, Prejudice and Dis­
crimination, in 2 The Handbook o f Social Psychology 357, 
362, 375 (Daniel T. Gilbert et al. eds., 4th ed. 1998).

This simple act of categorization can have dramatic ef­
fects on one’s attitude toward group members. Research 
indicates that individuals tend to assign value to differences 
between others and the group to which they belong, devel­
oping more favorable attitudes toward “ingroup” members. 
See Michael J. Migdal et al., The Effects o f Crossed Categori­
zation on Intergroup Evaluations: A Meta-Analysis, 37 
Brit. J. Soc. Psychol. 303 (1998) (presenting evidence con­
firming the existence of “ingroup bias”); Brian Mullen et al., 
Ingroup Bias as a Function of Salience, Relevance, and 
Status: An Integration, 22 Eur. J. Soc. Psychol. 103 (1992); 
Mia D. Yee & Rupert Brown, Self-Evaluations and Inter­
group Attitudes in Children Aged Three to Nine, 63 Child 
Dev. 619 (1992) (finding in-group favoritism among young 
children). Indeed, “the mere existence of different social 
groups is sufficient to foster biased behavior.” See David A. 
Wilder, Perceiving Persons as a Group: Categorization and 
Intergroup Relations, in Cognitive Processes in Stereotyp­



6

ing and Intergroup Behavior 213, 228 (David L. Hamilton 
ed., 1981) (citing Henri Tajfel, Cognitive Aspects o f Preju­
dice, 25 J. Soc. Issues 79 (1969)).

Categorization does not, however, inevitably lead to 
stereotyping. Mackie et al., supra, at 47. Rather, stereotyp­
ing occurs only when a person ascribes certain traits to indi­
viduals solely on the basis of their group membership. Id. at 
44, 47; Melanie Killen et al., Morality in the Context o f In­
tergroup Relations, in Handbook o f Moral Development 155, 
163 (Melanie Killen & Judith Smetana eds., 2006). Viewing 
social groups through the lens of stereotypic assumptions 
tends to produce at least four kinds of cognitive responses.

First, people perceive greater homogeneity and less dif­
ferentiation within an outgroup than an ingroup. That is, 
people tend to assume that members of their own group will 
possess a diversity of attitudes and beliefs, while others will 
conform to stereotypic expectations. See Wilder, supra, at 
226 (“ [Ijngroup members [a]re assumed to have a wider 
range of beliefs than outgroup members. Relative to the in­
group, the outgroup [is] thought to be more homogene­
ous . . . . ”); see also Brian Mullen & Li-tze Hu, Perceptions of 
Ingroup and Outgroup Variability: A  Meta-Analytic Inte­
gration, 10 Basic & Applied Soc. Psychol. 233-52 (1989) (pre­
senting evidence confirming the outgroup homogeneity ef­
fect). People thus tend to see ingroup members as individu­
als, and members of other groups as “all the same.”

Second, people explain the causes of the actions of mem­
bers of other groups and their own group in different ways. 
Negative behaviors of members of other groups are attrib­
uted to stable, internal factors (e.g., their personality or 
character) and positive behaviors are dismissed as situa- 
tionally caused. See Fiske, Stereotyping, Prejudice, and 
Discrimination, supra, at 370 (citing studies). In contrast, 
people tend to discount negative actions of their own group 
and attribute positive behaviors to internal causes. See id. 
In communication, they also talk about negative behaviors of 
members of other groups and positive behaviors of members 
of their own group in deeper, more abstract ways. See id.



7

Thus, these processes contribute not only to the formation of 
personal stereotypes but also to the way members of groups 
are thought about by others.

Third, stereotypic assumptions affect perception: Peo­
ple often resort to stereotypes to fill in the gaps in short, un­
focused, or partial sense impressions. See id. at 368. Stereo­
typic associations between social groups and concepts can 
guide how people process visual stimuli in their environ­
ment. See Jennifer L. Eberhardt et ah, Seeing Black: Race, 
Crime, and, Visual Processing, J. Personality & Soc. Psy­
chol. 876, 889 (2004) (discussing stereotypic associations be­
tween African-Americans and crime). We are also quicker 
to identify and process information that is consistent with a 
stereotype. See Fiske, Stereotyping, Prejudice, and Dis­
crimination, supra, at 368 (citing studies).

Fourth, stereotypic assumptions distort memory. Par­
ticularly when people are overloaded with information—as is 
often the case in everyday life—it is easier to recall informa­
tion that is consistent with a stereotype than information 
inconsistent with that stereotype. See Wilder, supra, at 226 
(“ [T]he categorization of persons into an ingroup and an out­
group is sufficient to bias subjects’ recall of information 
about the groups.”). See generally Charles Stangor & David 
McMillan, Memory for Expectancy-Congruent and Expec­
tancy -Incongruent Information: A  Review o f the Social and 
Social Developmental Literatures, 111 Psychol. Bull. 42 
(1992).

Stereotypical thinking often has more tangible negative 
ramifications. Such thinking is one source of prejudice—that 
is, negative feelings about other groups. Negative thoughts 
about other racial groups often contribute unconsciously to 
prejudiced attitudes. This type of implicit prejudice, in turn, 
often manifests itself in discriminatory behavior, anxiety 
when dealing with members of other groups, and in avoid­
ance of substantial interaction with members of other 
groups. See generally Samuel L. Gaertner & John F. 
Dovidio, The Aversive Form of Racism, in Prejudice, Dis­
crimination, and Racism 61-89 (John F. Dovidio & Samuel



L. Gaertner eds., 1986); John F. Dovidio et al., Why Can’t We 
Just Get Along? Interpersonal Biases and Interracial Dis­
trust, 8 Cultural Diversity & Ethnic Minority Psychol. 88, 92 
(2002); Walter G. Stephan & Cookie White Stephan, Inter­
group Anxiety, 41 J. Soc. Issues 157-75 (1985). In short, 
stereotypical thinking poses real and substantial obstacles to 
harmonious relations among members of different racial 
groups.

2. Once formed, racial stereotypes are difficult 
to abandon

The cognitive processes that lead to the formation of 
stereotypes begin early in life. Preschool-aged children have 
been shown to categorize people along such varied dimen­
sions as gender, ethnicity, occupation, and age. See Mackie 
et ah, supra, at 46-47. Racial categories become salient at a 
young age. See Rebecca S. Bigler & Lynn S. Liben, A Cog­
nitive-Developmental Approach to Racial Stereotyping and 
Reconstructive Memory in Euro-American Children, 64 
Child Dev. 1507,1516 (1993).

This process of categorization has predictable effects. 
As early as ages 5 and 6, children’s “perception of individual 
differences declines in favor of ethnic differences,” a mani­
festation of the so-called homogeneity effect. Frances E. 
Aboud & Maria Amato, Developmental and Socialization 
Influences on Intergroup Bias, in Blackwell Handbook o f 
Social Psychology: Intergroup Processes 65, 71 (Rupert 
Brown & Samuel L. Gaertner eds., 2001). At ages 3 through 
9, children display demonstrably more favorable attitudes 
toward members of the group to which they belong than 
they do to members of other groups. See, e.g., Yee & Brown, 
supra. Researchers have found that, between ages 6 and 9, 
children attribute characteristics based on race. Killen et ah, 
Morality in the Context o f Intergroup Relationships, supra, 
at 164-65.

During roughly the same period, children become aware 
of racial stereotypes and become susceptible to memory ef­
fects that lead them to reinforce those stereotypes. These



9

effects were dramatically illustrated in an experiment exam­
ining the role of cognitive skill and racial stereotyping in a 
sample of white children, ages 4 to 9, who had relatively lit­
tle contact with African-American children. The children 
were asked to recall stories involving either African- 
American or white characters associated with one of three 
negative traits—meanness, dirtiness, or laziness. See Bigler 
& Liben, supra, at 1507-18. The data showed that children’s 
memory for stories that were consistent with racial stereo­
types was better than their memory for stories involving 
counter-stereotypic themes and behaviors. See id. at 1515. 
That is, white children found it easier to remember stories 
with negative depictions of African-Americans than stories 
in which whites were portrayed negatively.

Critically, parents are generally not the principal influ­
ences on their children’s attitudes about race. Children do 
not learn stereotypes solely from their parents, nor do chil­
dren necessarily imitate their parents’ attitudes toward dif­
ferent racial groups. See Aboud & Amato, supra, at 74. In­
deed, most studies have found only a small correlation be­
tween the racial attitudes of children and their parents. See 
Frances E. Aboud & Sheri R. Levy, Interventions to Reduce 
Prejudice and Discrimination in Children and Adolescents, 
in Reducing Prejudice and Discrimination 269, 278 (Stuart 
Oskamp ed., 2000). Children’s racial attitudes largely stem, 
instead, both from their cognitive development and their 
own social experience. See Aboud & Amato, supra, at 74-76.

As children grow older, their thinking about racial cate­
gories tends to become more flexible. Around age 10, pref­
erence for ingroup members decreases. Andrew Scott 
Baron & Mahzarin R. Banaji, The Development o f Implicit 
Attitudes: Evidence o f Race Evaluations from  Ages 6 and 10 
and Adulthood, 17 Psych. Sci. 53, 56 (2006). Their earlier 
perception of homogeneity within racial groups gradually 
becomes “more elaborated in structure, . . .  allowing children 
to differentiate within ethnic groups and to use multiple 
cross-cutting categories” to classify others. Aboud & 
Amato, supra, at 72 (citations omitted). Older children be­



10

come better able to recognize similarities between different 
racial groups, as well as to distinguish between societal 
stereotypes and their own personal beliefs. Killen et al., 
Morality in the Context o f Intergroup Relationships, supra, 
at 163-64. This is a time when, in the right circumstances, 
stereotypical thinking can fall away.

By the time children reach adulthood, however, re­
search shows that it may be difficult for them to abandon the 
stereotypes they maintain. See, e.g., Irene V. Blair, The 
Malleability o f Automatic Stereotypes and Prejudice, 6 Per­
sonality & Soc. Psychol. Rev. 242, 255-58 (2002) (although 
stereotyping and prejudice can be moderated, they can also 
operate automatically and outside an individual’s conscious 
intent). After years of formulating racial attitudes, adults do 
not easily change their perceptions of racial difference. 
Even adults who consciously desire to change their own 
stereotypes, and who are generally able to keep from openly 
expressing these stereotypes, often nevertheless, spontane­
ously and unintentionally display implicit biases. See John 
F. Dovidio et al., On the Nature of Prejudice: Automatic and 
Controlled Processes, 33 J. Experimental Soc. Psychol. 510, 
534 (1997); cf. Melanie Killen et al., The Social Developmen­
tal Benefits o f Heterogeneous School Environments, in Les­
sons in Integration: Realizing the Promise o f Racial Diver­
sity in America’s Schools (Erica Frankenberg & Gary Or- 
field eds., forthcoming 2006). These findings underscore the 
importance of early intervention to reduce stereotyping and 
prejudice among children and adolescents, and better to 
prepare them to live productively in a multiracial society.

B. Racial Diversity In K-12 Education Can Inhibit 
The Formation Of Racial Stereotypes And Racial 
Bias

1. The importance o f intergroup contact
For the past half-century, psychological research has fo­

cused on intergroup contact as a touchstone for strategies to 
reduce racial bias and conflict. Intergroup contact theory, 
described in Dr. Gordon Allport’s seminal book The Nature



11

of Prejudice (1954), holds that interaction with members of 
other groups can disarm stereotypes, while promoting un­
derstanding and mutual respect. More specifically, inter­
group contact theory posits that, where four key conditions 
are present—equal status between groups, common goals, 
intergroup cooperation, and the support of authorities, law, 
or custom—interaction among members of different groups 
can be expected to reduce intergroup prejudice. See Thomas 
F. Pettigrew, Intergroup Contact Theory, 49 Ann. Rev. Psy­
chol. 65, 66-67 (1998) (describing the basic features of All­
port’s hypothesis).

Intergroup contact theory is supported by voluminous 
social science research. Field studies, laboratory experi­
ments, surveys, and meta-analytic reviews all confirm that 
positive contact between people of different races typically 
reduces racial bias and promotes positive race relations. 
See, e.g., Thomas F. Pettigrew & Linda R. Tropp, A Meta- 
Analytic Test o f Intergroup Contact Theory, 90 J. Personal­
ity & Soc. Psychol. 751, 752-53 (2006). A recent meta­
analysis—that is, a statistical analysis that pools the results 
of a number of studies to determine their size and consis­
tency4— examined more than 500 individual intergroup con­
tact studies involving more than 250,000 people from 38 
countries. More than 90% of these studies showed that 
greater intergroup contact resulted in lower incidence of 
prejudiced attitudes among both majority and minority 
group members. See Pettigrew & Tropp, A Meta-Analytic 
Test o f Intergroup Contact Theory, supra, at 757; cf. Thomas 
F. Pettigrew & Linda R. Tropp, Does Intergroup Contact 
Reduce Prejudice? Recent Meta-Analytic Findings, in Re­
ducing Prejudice and Discrimination 93 (Stuart Oskamp 
ed., 2000).

Moreover, the research showed that intergroup contact 
changes perceptions of the entire outgroup—not just those

4 See generally Robert Rosenthal, Meta-Analytic Procedures for So­
cial Science (rev. ed. 1991).



12

outgroup members directly involved in the contact. See Pet­
tigrew & Tropp, A Meta-Analytic Test o f Intergroup Con­
tact Theory, supra, at 759. Scholars have thus concluded 
that “ [a]ctual intergroup contact, under specified conditions, 
can be a powerful way of reducing intergroup biases.” John 
F. Dovidio et al., Reducing Contemporary Prejudice: Com­
bating Explicit. and Implicit Bias at the Individual and 
Intergroup Level, in Reducing Prejudice and 
Discrimination 137,147 (Stuart Oskamp ed., 2000).

Some commentators had speculated that the findings of 
intergroup contact studies might be affected by a potential 
participant bias, such that intergroup contact would be ef­
fective in reducing prejudice only if people had willingly cho­
sen to engage in the contact. Specifically, because preju­
diced people are more likely to avoid intergroup contact than 
tolerant people, it was thought that reduced prejudice may 
simply be correlated with, rather than the result of, interac­
tion with members of outgroups. Studies have demon­
strated, however, that “optimal contact reduces prejudice 
over time, even when researchers have eliminated the pos­
sibility of participant selection.” Pettigrew & Tropp, A  
Meta-Analytic Test o f Intergroup Contact Theory, supra, at 
753 (citations omitted). In other words, the positive effects 
of intergroup contact are consistent and significant, whether 
or not participants had a choice about whether to engage in 
the contact. Id.

Intergroup contact is effective in part because it pro­
vides an opportunity for members of different groups to get 
to know one another and to develop affective bonds— 
feelings of personal closeness and common connection that 
transcend race. See, e.g., Thomas F. Pettigrew & Linda R. 
Tropp, Does Intergroup Contact Reduce Prejudice? Recent 
Meta-Analytic Findings, in Reducing Prejudice and Dis­
crimination 93, 108 (Stuart Oskamp ed., 2004) (finding that 
cross-group friendships are strongly associated with reduced 
prejudice); see also Pettigrew & Tropp, A  Meta-Analytic 
Test o f Intergroup Contact Theory, supra, at 766-67 (posit­
ing that “the process underlying contact’s ability to reduce



13

prejudice involves the tendency for familiarity to breed lik­
ing”). But intergroup contact has effects on cognitive, as 
well as affective, processes. Many scholars have posited that 
intergroup contact works because interaction with members 
of other racial groups helps us to deconstruct racial stereo­
types and create new social categories that cut across race. 
As one study explains:

Negative stereotypes tend to lose their primary po­
tency and to be reduced when interactions reveal 
enough detail that group members are seen as indi­
viduals rather than as members of an ethnic group.
All collaborators become “one of us.” In other 
words, cooperation widens the sense of who is in 
the group, and “they” become “we.”

David W. Johnson & Roger T. Johnson, The Three Cs of Re- 
ducing Prejudice and Discrimination, in Reducing Preju­
dice and Discrimination 239, 247 (Stuart Oskamp ed., 2000); 
see also Samuel L. Gaertner et al., The Contact Hypothesis: 
The Role o f a Common Ingroup Identity on Reducing Inter­
group Bias Among Majority and Minority Group Me?nbers, 
in What’s Social About Social Cognition?: Research on So­
cially Shared Cognition in Small Groups 230, 232 (Judith L. 
Nye & Aaron M. Brower eds., 1996); Pettigrew, Intergroup 
Contact Theory, supra, at 75; Wilder, supra, at 245. This 
phenomenon is particularly well-documented in dozens of 
studies of cooperative learning and work environments. See, 
e.g., B. Ann Bettencourt & Nancy Dorr, Cooperative Inter­
action and Intergroup Bias: Effects o f Numerical Represen­
tation and Cross-Cut Role Assignment, 24 Personality & 
Soc. Psychol. Bull. 1276, 1288 (1998); Janet Ward Schofield, 
School Desegregation and Intergroup Relations: A  Review 
of the Literature, 17 Rev. of Res. in Educ. 335, 360 (1991) 
(citing, inter alia, research by Eliot Aronson et al. and 
Robert E. Slavin); see generally Susan T. Fiske, Interde­
pendence and the Reduction o f Prejudice, in Reducing 
Prejudice and Discrimination 115 (Stuart Oskamp, ed., 2000) 
(describing how cooperative environments create the condi­



14

tions for seeing outgroup members as “individuals” and 
“all[ies]” ).

These processes of individuation and recategorization 
have extremely significant cognitive and behavioral conse­
quences. To begin with, they temper the harmful perceptual 
distortions triggered by racial categorization. But even 
more important, once individuals of different racial back­
grounds are categorized together on some non-racial basis, 
powerful ingroup biases cause group members to view one 
another more favorably. As leading proponents of the re­
categorization perspective have explained, “ [transforming 
members’ representations of the groups to recognize a com­
mon ingroup identity can harness the psychological forces 
that contribute to intergroup bias and redirect them, thus 
improving attitudes towards people who would otherwise be 
recognized only as outgroup members.” Dovidio et al., Re­
ducing Contemporary Prejudice, supra, at 158.

2. Short-term effects o f intergroup contact in 
primary and secondary education

Dozens of studies have focused specifically on the appli­
cation of the intergroup contact theory to K-12 education. 
Taken as a whole, these studies show that intergroup con­
tact among school-aged children is a vital condition for re­
ducing stereotyping, bias, and prejudice during this impor­
tant developmental period. A recent meta-analytic review 
of 198 studies involving intergroup contact among children 
and adolescents—57% of which examined the effects of con­
tact in school settings—shows that school contact between 
youth from different racial groups corresponds with more 
positive intergroup attitudes, and such positive outcomes 
become even stronger when Allport’s optimal conditions are 
established in the school environment. Linda R. Tropp & 
Mary Prenovost, The Role o f Intergroup Contact in Predict­
ing Inter-Ethnic Attitudes: Evidence from  Meta-Analytic 
and Field Studies, in Intergroup Relations: An Integrative 
Developmental and Social Psychological Perspective (Sheri 
Levy & Melanie Killen eds., forthcoming 2006).



15

Individual field studies provide concrete illustrations. 
One early study, for example, analyzed student behavior in a 
newly opened, integrated middle school whose student body 
was drawn largely from segregated schools. See Janet W. 
Schofield & H. Andrew Sagar, Peer Interactions in an Inte­
grated Middle School, 40 Sociometry 130-38 (1977). 
Schofield and Sagar analyzed voluntary seating patterns in 
the cafeteria during the first year of the school’s existence. 
Their analysis revealed that seating became more and more 
integrated over time—that is, members of different racial 
groups sat together more as intergroup contact persisted. 
See id. at 135 (taking note of “increasing interracial interac­
tion” over time). The authors of this study reported, there­
fore, that “highly integrated interracial schooling. . .  does 
foster increased voluntary association between Blacks and 
Whites.” Id. at 137. Researchers also found that contact 
among school-age children reduces intergroup anxiety and 
helps them to learn to work together more effectively. See 
Janet W. Schofield, Black and White in School: Trust, Ten­
sion, or Tolerance 162 (1989).

In more recent studies, researchers have compared the 
racial attitudes of children who attend non-diverse schools 
against those who attend racially heterogeneous schools. 
The results of the comparison have been striking. Presented 
with an ambiguous situation involving characters of differ­
ent races, white children who attended racially homogeneous 
schools displayed implicit racial bias in their interpretation 
of the characters’ behavior, rating African-American charac­
ters more negatively than white characters. See Heidi 
McGlothlin & Melanie Killen, Intergroup Attitudes o f Euro­
pean American Children Attending Ethnically Homogene­
ous Schools, 77 Child Dev. 1375, 1377, 1382-84 (2006) (exam­
ining implicit racial bias among first- and fourth-grade white 
children who attend schools with white student populations 
of 91.2% and 86.1%, respectively). By contrast, students 
who attended racially diverse schools displayed no bias or 
minimal bias. See Heidi McGlothlin et al., European- 
American Children’s Intergroup Attitudes About Peer Rela­
tionships, 23 British J. Dev. Psychol. 227, 236 (2005) (exam­



16

ining bias among first- and fourth-grade white children at­
tending racially heterogeneous schools); Nancy Geyelin 
Margie et al., Minority Children’s Intergroup Attitudes 
About Peer Relationships, 23 British J. Dev. Psychol. 251, 
264 (2005) (examining bias among first- and fourth-grade 
minority children).

In a study of middle-school children of a variety of racial 
backgrounds, the majority of whom were Latino and Afri­
can-American, children in more diverse schools reported 
greater feelings of safety and fewer feelings of loneliness 
than their peers in less diverse schools. Moreover, children 
in diverse classrooms reported greater feelings of self-worth 
than children in less diverse classrooms. See Jaana Juvonen 
et ah, Ethnic Diversity and Peer Perceptions of Safety in 
Urban Middle Schools, 17 Psych. Sci. 393 (2006).

Studies have also found that, although the relationship 
between racial diversity and views concerning cross-race 
friendships are complex, children in non-diverse schools are 
more likely to assume that they cannot form friendships 
with children of other races. Compare McGlothlin & Killen, 
supra, at 1383-84, with McGlothlin et al., supra, at 242-43, 
245, and Margie et al., supra, at 264. Meta-analyses of re­
search studies focusing largely on adults have shown that 
cross-race friendships are strongly correlated with reduc­
tions in prejudice. Pettigrew & Tropp, Does Intergroup 
Contact Reduce Prejudice?, supra. The same is true among 
children. See, e.g., Aboud & Levy, supra, at 272. Signifi­
cantly, when people become friends with individuals of other 
races, they are far more likely “to see that there are similari­
ties between people of different ethnicities and that people 
of other ethnicities are not all the same.” Killen et al., Mo­
rality in the Context o f Intergroup Relations, supra, at 167. 
Cross-group friendships can also have positive effects on the 
intergroup attitudes of other in-group members who become 
aware of the existence of others’ cross-group friendships. 
See Stephen C. Wright et al., The Extended Contact Effect: 
Knowledge o f Cross-Group Friendships and Prejudice, 73 J. 
Personality & Soc. Psychol. 73 (1997). Children in integrated



17

schools are far more likely to reap these benefits than chil­
dren in homogeneous schools.

To be sure, some of the studies on the impact of school 
desegregation on intergroup relations have yielded inconclu­
sive or inconsistent results. See, e.g., Janet Ward Schofield, 
School Desegregation and Intergroup Relations, supra. 
Many of these studies were undertaken in the 1970s and 
early 1980s, preceding significant changes in intergroup atti­
tudes and behavior, and many focused on schools’ experi­
ences during the first few years of desegregation. See id. at 
343. Although intervening demographic and societal 
changes do not render the earlier research invalid, they may 
“limit its usefulness for drawing conclusions about the pre­
sent.” Janet W. Schofield & Rebecca Eurich-Fulcer, When 
and How School Desegregation Improves Intergroup Rela­
tions, in Blackwell Handbook o f Social Psychology 475, 477 
(Rupert Brown & Samuel L. Gaertner eds., 2001).5

Perhaps more important, the variation in the results of 
the earlier studies may well reflect the fact that “desegrega­
tion can be implemented in very different ways, and . . . 
these differences may well affect their outcomes.” Schofield 
& Eurich-Fulcer, supra, at 477-78. Few of the early studies 
were conducted in environments providing particularly fa­
vorable conditions for intergroup contact. See Schofield, 
School Desegregation and Intergroup Relations, supra, at 
360, 381. Notably, those few early studies that did “tak[e] 
contact theory seriously” yielded “generally promising re­
sults.” Id. at 360. Studies focused on school conditions found 
markedly more positive intergroup relations where Allport’s

5 Pettigrew and Tropp explored the question of timing in their meta- 
analytic review of intergroup contact studies, noting that more recent 
studies have tended to find larger effects in terms of prejudice reduction 
than studies conducted before 1980. Pettigrew and Tropp concluded that 
much of the differences in findings can be explained by differences in re­
search rigor, but that there is in fact a statistically significant difference 
between the results of pre-1980 and post-1980 studies. Pettigrew & 
Tropp, A Meta-Analytic Test o f Intergroup Contact Theory, supra, at 765.



18

conditions were met relative to school environments in 
which Allport’s conditions were not met. See Janet Ward 
Schofield & H. Andrew Sagar, Desegregation, School Prac­
tices, and Student Race Relations, in The Consequences of 
School Desegregation 58, 67-68 (Christine Rossell & Willis 
Hawley eds., 1983); see also supra page 11 (describing All­
port’s conditions). Some scholars have concluded that inter­
group contact alone typically reduces intergroup prejudice. 
See Pettigrew & Tropp, A Meta-Analytic Test o f Intergroup 
Contact Theory, supra, at 753, 766. But it is widely recog­
nized that positive effects are more likely to be realized 
when the proper conditions exist for productive contact 
among members of different racial groups. See, e.g., 
Schofield & Eurich-Fulcer, supra. Scholars are thus gener­
ally in agreement that intergroup contact is a necessary— 
even if not wholly sufficient—condition for producing re­
spectful and positive relations between students of all races. 
See, e.g., Schofield & Eurich-Fulcer, supra, at 478.6

6 The literature indicates that there are no equally effective substi­
tutes for intergroup contact. Research on the effects of educational pro­
grams designed to reduce prejudice among children in the absence of in­
tergroup contact has been for the most part inconclusive. Some studies 
have shown positive effects from interventions designed to focus chil­
dren’s attention on individuating information about outgroup members, 
but have also noted that curricular approaches may also have no effect, or, 
rarely, a negative effect. See Walter G. Stephan & Cookie White Stephan, 
Intergroup Relations in Multicultural Education Programs, in Hand­
book of Research on Multicultural Education 782, 793-94 (James Banks 
and Carrie McGee-Banks eds., 2004); see also Schofield & Sagar, 
Desegregation, supra, at 87.

As a recent survey of techniques for reducing intergroup preju­
dice among school-age children explains, “there is too little research avail­
able to inform educators about which anti-racist programs successfully 
reduce prejudice and stereotyping.” Aboud & Levy, supra, at 282; see 
also Walter G. Stephan & Cookie White Stephan, Improving Intergroup 
Relations 47, 62-63 (2001).



19

3. Long-term effects o f intergroup contact in 
primary and secondary education

The benefits of early exposure to members of other 
races reach well into adulthood. As a number of studies 
have found, school integration has long-term positive effects 
on relationships between individuals from different back­
grounds. See Schofield & Eurich-Fulcer, supra, at 476; Wal­
ter G. Stephan & Cookie White Stephan, Intergroup Rela­
tions 79-82 (1996).

A recent review of studies on the long-term effects of 
school desegregation demonstrates that attending a diverse 
school typically leads to increased interaction with members 
of other racial groups in adulthood. Jomills Henry Braddock 
II & Tamela McNulty Eitle, The Effects o f School Desegre­
gation, in Handbook o f Research on Multicultural Educa­
tion 828, 834-35 (James A. Banks & Cherry A. McGee Banks 
eds., 2d ed. 2004); see also Amy Stuart Wells & Robert L. 
Crain, Perpetuation Theory and the Long-Term Effects of 
School Desegregation, 64 Rev. Ed. Res. 531, 552 (1994). Re­
search shows that students who attend integrated schools 
are more likely than other students to work and live in inte­
grated environments as adults. Jomills Braddock & James 
M. McPartland, Social-Psychological Processes That Per­
petuate Racial Segregation: The Relationship Between 
School and Employment Desegregation, 19 J. Black Stud. 
267 (1989) (presenting evidence that school desegregation 
promotes desegregation in work environments).

In one study, white adults who attended racially diverse 
schools reported their “decreased fear of people of color.” 
Amy Stuart Wells et al., How Desegregation Changed Us: 
The Effects o f Racially Mixed Schools on Students and So­
ciety 16, available at http://cms.tc.columbia.edU/i/a/ 
782_ASWells041504.pdf (last visited Oct. 9, 2006). And Afri­
can-American adults who attended racially diverse schools 
cited their greater preparedness “to function in predomi­
nantly white environments.” Id.

Having reviewed the long-term effects of school deseg­
regation, scholars have concluded that racially diverse

http://cms.tc.columbia.edU/i/a/


20

schools can effectively “break[] the cycle of segregation.” 
Wells & Crain, supra, at 531. Their work shows that the 
benefits of K-12 diversity extend well beyond graduation: 
Racially integrated schools have made demonstrable differ­
ences in the lives of adults who spent their youth learning 
alongside children of different races.

C. The Benefits O f Diversity In K-12 Education Are 
More Likely To Accrue When There Is A Critical 
Mass O f Students Of Different Racial Back­
grounds

Reduction in stereotypical thinking and prejudice de­
pends, at least in part, on the proportion of children of dif­
ferent races in the classroom. To begin with, meaningful in­
tergroup contact is possible only when a significant number 
of children of different racial backgrounds are present: 
“When a group is proportionately very small, outgroup 
members have very little opportunity to interact with mem­
bers of [that group] even if they are inclined to do so.” 
Schofield & Sagar, Desegregation, supra, at 71.

There is evidence, moreover, that where a minority ra­
cial group lacks critical mass, there are substantial implica­
tions for members of that group. Children whose ethnic 
group is a numerical minority in a school setting may be at 
greater risk for peer harassment. See, e.g., Sandra Graham 
& Jaana Juvonen, Ethnicity, Peer Harassment, and Ad­
justment in Middle School: An Exploratory Study, 22 J. 
Early Adolescence 173, 191 (2002). Children are also more 
likely to self-segregate by race in order to maintain group 
identity. See Schofield & Sagar, Desegregation, supra, at 71; 
see also Schofield & Eurich-Fulcer, supra, at 483 (“Consis­
tent with the contact hypothesis, when minorities are a very 
small proportion of the total student body self-segregation 
seems to be heightened.” ). Researchers have found, for ex­
ample, that when African-Americans represent less than 
15% of a student body, they are significantly more likely to 
choose friends on the basis of racial group membership. 
Schofield & Sagar, Desegregation, supra, at 71. By its very 
nature, self-segregation will significantly reduce intergroup



21

contact and thereby decrease the likelihood of cooperation. 
Where students have self-segregated, all students are less 
likely to learn to challenge social stereotypes or to focus on 
the similarities, rather than the differences, between them­
selves and members of other racial groups.

Some scholars have theorized that intergroup contact is 
ineffective at dispelling prejudice absent critical mass be­
cause without critical mass, the “equal status” factor identi­
fied by Allport is lacking. “ [I]f only a small number of stu­
dents of a given background are present, they are unlikely to 
enjoy equal status since they form such a small group they 
will be unlikely to exert much influence in the institution.” 
Schofield & Eurich-Fulcer, supra, at 482-83; see also 
Schofield & Sagar, Desegregation, supra, at 72 (when there 
are small numbers of minority students, “traditional status 
relations may be maintained because the . . . students lack 
the sheer numbers to become an influential force in the life 
of the school”). This lack of influence means, among other 
things, that teachers and administrators may not recognize 
the significance of diversity among their students and may 
not see the need to organize classes in such a way as to pro­
mote cooperation and understanding among members of dif­
ferent racial groups.

Other scholars emphasize that where members of a ra­
cial or ethnic group are reduced to token status, they may be 
evaluated unfairly. See Shelley E. Taylor et al, Categorical 
and Contextual Bases of Person Memory and Stereotyping, 
36 J. Personality & Soc. Psychol. 778, 792 (1978) (“ [A] situa­
tion of token integration . . . would . . .  be especially condu­
cive to extreme evaluations and stereotyping of the minority 
group member.”). Minority group representation in token 
numbers can also negatively affect the performance of mi­
nority students even in the absence of active discrimination. 
In particular, minority students with token status feel that 
others view them primarily in terms of their race and feel 
that they are constantly being evaluated on this basis. This 
feeling of being “under the microscope” may cause emotional 
stress and cognitive and behavioral deficits. See Charles G.



22

Lord & Delia S. Saenz, Memory Deficits and Memory Sur­
feits: Differential Cognitive Consequences o f Tokenism for  
Tokens and Observers, 49 J. Personality & Soc. Psychol. 918, 
918, 923-25 (1985).

Minority students present in only token numbers may 
also face “stereotype threat” : that is, the threat that others 
will reduce them to a negative stereotype about the racial 
group to which they belong. In the short run, the possibility 
of confirming a negative stereotype can trigger an emotional 
response in minority students that interferes with academic 
performance. In the long run, it can affect minority stu­
dents’ very sense of identity, leading them to disidentify 
with domains—such as scholastic achievement—in which 
they face the threat of negative stereotyping. See generally 
Claude M. Steele, A Threat in the Air: How Stereotypes 
Shape Intellectual Identity and Performance, 52 Am. Psych. 
613 (1997); Clark McKown & Rhona S. Weinstein, The De­
velopment and Consequences of Stereotype Consciousness 
in Middle Childhood, 74 Child Dev. 498 (2003) (reporting 
stereotype threat effects among elementary-school stu­
dents).

Thus, much as in the higher education context, in K-12 
schools, a “critical mass” of students from different racial 
groups is necessary to secure the benefits of intergroup con­
tact. For schools to lay the foundation for productive cross- 
racial interaction among their students, numbers matter.

II. Meaningful Intergroup Contact Is Unlikely To Oc­
cur In Public K-12 Education Without The School
Districts’ Intervention

Research shows that many individuals tend to avoid 
situations in which they will have substantial interaction 
with persons they perceive to be significantly different from 
themselves. Crucially for purposes of the present litigation, 
this tendency to avoid intergroup contact means that par­
ents often will not make the kinds of choices that will afford 
their children substantial opportunity to interact with chil­
dren of other races.



23

For that reason, to secure the benefits of racial diversity 
in K-12 schools, there is no practical substitute for consid­
eration of race in school assignments. In particular, alterna­
tives that rely primarily on parental choice to achieve racial 
diversity are likely to be far less effective than a system in 
which school districts manage parental choice with the goal 
of reducing racial concentration and providing opportunities 
for meaningful intergroup contact in local schools.

A. Because Of Largely Unconscious, Automatic Cog­
nitive And Emotional Responses, Individuals 
Often Avoid Intergroup Contact

Over the past several years, many have documented a 
substantial decrease in overt expressions of racial prejudice. 
See Gaertner & Dovidio, supra, at 61 (citing studies). Nev­
ertheless, numerous studies show that prejudiced attitudes, 
like racial stereotyping, remain widespread. As described 
above in Section I.A.l, ordinary cognitive processes cause us 
to place people in social categories, and to favor our own 
category over others. When people are exposed to societal 
stereotypes about race, even those who maintain a conscious 
commitment to racial equality often develop, if uncon­
sciously, a variety of negative feelings and thoughts (i.e., 
prejudices and stereotypes) about other racial and ethnic 
groups. See generally Gaertner & Dovidio, supra.

Studies have shown that stereotypes operate automati­
cally, often independent of conscious attitudes, beliefs and 
perceptions. See, e.g., Steven J. Spencer et al., Automatic 
Activation o f Stereotypes: The Role o f Self-Image Threat, 24 
Personality & Soc. Psychol. Bull. 1139 (1998). The effect of 
unconscious stereotyping on cognition has been demon­
strated through a number of “reaction-time” experiments 
measuring the speed with which two concepts are associ­
ated. The association between racial groups and racial 
stereotypes is outside the conscious control of the individual. 
See Patricia G. Devine, Stereotypes and Prejudice: Their 
Automatic and Controlled Components, 56 J. Personality & 
Soc. Psychol. 5 (1989); see also John F. Dovidio et al., On the 
Nature o f Prejudice: Automatic and Controlled Processes,



24

33 J. Experimental Soc. Psychol. 510 (1997). As explained 
above, such stereotypes have significant cognitive effects. 
Among other things, stereotypes cause people to minimize 
the differences among individual members of another racial 
group and to dismiss or forget information inconsistent with 
their stereotypes. See supra Section I.A.l.

Research also shows that prejudice, like stereotypical 
thinking, operates implicitly. Studies demonstrate that even 
those who firmly maintain and articulate explicit attitudes of 
racial equality and acceptance nevertheless implicitly harbor 
a variety of negative feelings about members of other racial 
and ethnic groups. See generally Gaertner & Dovidio, su­
pra. Studies focusing specifically on “aversive racism”—that 
is, racist prejudice harbored by those who would find it 
aversive to acknowledge their racial biases—has demon­
strated that these subconscious prejudices can trigger dis­
criminatory behavior. See id. It can also trigger avoidance: 
that is, people who harbor prejudice—even implicit preju­
dice—will often shy away from contact with persons of other 
races. See, e.g., Pettigrew & Tropp, A Meta-Analytic Test of 
Intergroup Contact Theory, supra, at 753.

Some researchers emphasize that people often experi­
ence anxiety about interacting with members of other 
groups. See generally Stephan & Stephan, Intergroup 
Anxiety, supra. The level of anxiety varies with a number 
of factors. Among other things, lack of prior contact and the 
presence of negative stereotypes lead to increased levels of 
anxiety. Id. at 161, 163. Individuals are “unlikely to antici­
pate positive interaction” with members of groups about 
whom they harbor negative stereotypes. Id. at 163. And as 
other researchers have noted, those whose implicit prejudice 
is averse to their conscious beliefs are particularly likely to 
feel a sense of anxiety and unease, rather than open hostility 
or clear dislike, about interacting with persons of other ra­
cial groups. See Dovidio et al., Why Can’t We Just Get 
Along?, supra, at 90.

Perhaps unsurprisingly, the “dominant response” to in­
tergroup anxiety is avoidance. Stephan & Stephan, Inter­



group Anxiety, supra, at 165. Individuals forced to interact 
with one another may cope with their discomfort through 
various other strategies, including resort to formal and 
superficial interactions, withdrawal, hesitance, confusion, or 
even aggression. See id. at 165-67. But the easiest way to 
reduce anxiety is simply to avoid its source. Id. at 165. Par­
ticularly for those with less experience dealing with mem­
bers of other racial groups and those who harbor implicit 
racial stereotypes and prejudice, that means avoiding inter­
actions with members of other racial groups when possible.

B. Because Of Intergroup Avoidance, Private Choice 
Alone Is Unlikely To Produce Substantial Diver­
sity In Public K-12 Education

The phenomena described above have important impli­
cations for the question of how to achieve K-12 diversity. 
Petitioners and their amici suggest that one race-neutral 
alternative to race-conscious school assignments would be to 
increase the quality of educational offerings at predomi­
nantly minority schools to give the parents of majority white 
children an incentive to send their children to these schools. 
See, e.g., Br. for the United States as Amicus Curiae Sup­
porting Petitioner, No. 05-908, at 23-24, 25-26 (suggesting 
magnet schools and “additional investment in racially con­
centrated schools” as alternatives to race-conscious assign­
ment plans). The effectiveness of suggested alternatives 
that depend on additional incentives to attract parents to 
schools must be assessed in light of the distinct disincentives 
to integration that are associated with implicit prejudice and 
intergroup anxiety.

To be sure, incentives in the form of magnet programs 
are often designed to ameliorate racial concentration in pri­
mary and secondary education, and they have led to in­
creased diversity in a number of schools. But magnet 
schools generally serve a limited number of students. For 
most students in a given school district, investment in pre­
dominantly minority schools to make them “equal” to other 
schools in terms of resources and educational quality is 
unlikely to provide their parents a sufficient incentive. Re­



26

search on the cognitive and emotional processes associated 
with intergroup interaction predicts that, given the choice 
between two schools of equal quality, parents may not per­
ceive the schools as equal. See supra pp. 6-7, 23-25. They 
are therefore likely to choose the school whose student body 
appears more familiar to them, and likely to decide against 
the school where their children will encounter significant 
numbers of children of other races.

School choice patterns appear to bear out that predic­
tion. Studies of school districts across the country have 
shown that white parents tend to choose schools with larger 
percentage populations of white students, suggesting a pat­
tern of “white avoidance of racially mixed schools.” Charles 
T. Clotfelter, After Brown: The Rise and Retreat o f School 
Desegregation 92-93 (2004). One study documented a similar 
phenomenon among parents of minority children in his study 
of the school transfer choices made by parents in the Mont­
gomery County, Maryland school system in 1985. Minority 
parents tended to choose schools with larger percentages of 
minority students, leading the researcher to conclude: 
“ ‘Both whites and minorities seem to direct their choices to­
ward schools in which their children will be less likely to be 
racially or socioeconomically isolated. Cultural familiarity is 
a strong point of attraction.’” Id. at 92, 94 (quoting Jeffrey 
R. Henig, The Local Dynamics of Choice: Ethnic Prefer­
ences and Institutional Responses, in Who Chooses? Who 
Loses?: Culture, Institutions, and the Effects o f School 
Choice 95,105 (Bruce Fuller et al. eds., 1996)).

Some studies have shown that some minority parents, 
perhaps conscious of the practical problems associated with 
racial isolation, also choose schools with predominantly 
white student bodies. See Clotfelter, supra, at 93. Even in 
these studies, however, parents’ choices have proved less 
effective than other strategies in producing student body 
diversity. One such study, a recent examination of San 
Diego schools, compared a pure-choice open-enrollment pro­
gram against other school choice programs employed in the 
school district. The study found that the open-enrollment



27

plan “increases the exposure of whites to Asians but de­
creases the exposure of whites to blacks and Hispanics.” 
Julian R. Betts et al., Does School Choice Work? Effects on 
Student Integration and Achievement 44 (2006). By con­
trast, other school-choice plans specifically designed to pro­
mote racial diversity “unambiguously increase the exposure 
of whites to nonwhites, and vice versa.” Id. at xvii. These 
findings led the authors to conclude that “built-in mecha­
nisms aimed at promoting integration” may well be neces­
sary to achieve racial diversity. See id. at xvii; see also id. at 
47.

The principle of intergroup avoidance described in the 
psychological research on implicit prejudice and intergroup 
anxiety helps to explain these and other similar findings re­
garding the effects of school choice plans. These studies in­
dicate that private choice alone is unlikely to produce schools 
in which children of different races have the opportunity to 
engage in meaningful intergroup interaction. Experience 
thus suggests that school districts have an important role to 
play in providing schools that help students overcome the 
implicit biases and prejudices that have historically resulted 
in de facto segregation.

CONCLUSION
For the foregoing reasons, amici urge the Court to af­

firm the decisions below.



Respectfully submitted.

N a t h a l i e  F.P. G i l f o y l e  
General Counsel 

L i n d s a y  C h i l d r e s s -B e a t t y  
Deputy General Counsel 

A m e r i c a n  P s y c h o l o g i c a l  
A s s o c i a t i o n  

750 First Street, N.E. 
Washington, DC 20002 
(202) 336-5500

J o h n  P a y t o n  
Counsel o f Record 

D a v i d  W . O g d e n  
L e o n o r a  R . K r u g e r  
W i l m e r  C u t l e r  P i c k e r i n g  

H a l e  a n d  D o r r  l l p  
1875 Pennsylvania Ave., N.W. 
Washington, DC 20006 
(202) 663-6000

O c t o b e r  2006