Brown v. Board Doll Test: How It Worked and What the Court Made of It

The doll test in Brown v. Board of Education was a psychological experiment by Dr. Mamie Phipps Clark and Dr. Kenneth Clark showing that Black children as young as three had absorbed negative messages about their own race, and the NAACP used it to prove that school segregation inflicted real psychological harm. The Supreme Court cited the Clarks’ work in its 1954 decision striking down “separate but equal” in public education.¹

How the Experiment Worked

The Clarks used four plastic dolls, identical except for color. Two had light skin and yellow hair. Two had brown skin and black hair.² Each child, ages three through seven, sat with all four dolls and answered eight requests in sequence:³

Give me the doll you want to play with.
Give me the doll that is a nice doll.
Give me the doll that looks bad.
Give me the doll that is a nice color.
Give me the doll that looks like a white child.
Give me the doll that looks like a colored child.
Give me the doll that looks like a Negro child.
Give me the doll that looks like you.

The first four questions measured preference. The middle ones tested whether children could identify racial differences at all. The last question forced the child to place themselves in one of those categories. That final question is where the experiment became devastating, because many children who had just called the brown doll “bad” then had to identify it as the one that looked like them.

What the Results Showed

Across 239 Black children, 67 percent chose the white doll when asked which one they wanted to play with. Fifty-nine percent called the white doll “nice.” Fifty-nine percent pointed to the brown doll when asked which one looked bad. Sixty percent said the white doll had a nicer color.³

The self-identification question produced the most painful moments. Children who had spent the previous minutes labeling the brown doll as bad were then asked which doll looked like them. The Clarks observed visible emotional distress at this stage. Some children became upset or cried. Others tried to avoid answering. The Clarks described the gap between how the children felt about the brown doll and the fact that it represented them as damage to the developing personality.

One result complicated the picture. Children in integrated Northern schools showed stronger white-doll preference than children in segregated Southern schools. Seventy-two percent of Northern children preferred the white doll, compared with 62 percent in the South, and 71 percent of Northern children called the Black doll “bad,” versus 49 percent in the South.³

How the Test Reached the Supreme Court

Robert Carter, an attorney with the NAACP Legal Defense Fund, pushed the legal team to stop arguing only about unequal buildings and budgets and start arguing that segregation itself caused psychological injury. Several prominent NAACP lawyers ridiculed the idea of relying on dolls in federal court. Carter’s response, as he later recounted, was to challenge the skeptics to propose an alternative. None could.

Brown v. Board of Education was a consolidation of five cases from Kansas, South Carolina, Virginia, Delaware, and Washington, D.C.⁴ Kenneth Clark provided expert testimony in the lower court proceedings for Briggs v. Elliott out of South Carolina, along with the Davis and Delaware cases.⁵ In preparation for Briggs, the Clarks repeated the doll test with local schoolchildren and Clark presented those fresh results to the court.² Clark also co-authored a written summary of the social science testimony delivered across the trial-level cases, endorsed by 35 leading social scientists.

Thurgood Marshall argued the case before the Supreme Court on that record, contending that no amount of equal funding or identical facilities could undo the damage segregation caused to a child’s sense of self-worth.

How the Court Used the Evidence

On May 17, 1954, the Supreme Court issued a unanimous decision holding that school segregation violated the Equal Protection and Due Process Clauses of the Fourteenth Amendment.⁵ Chief Justice Earl Warren, writing for the Court, framed the harm in personal terms: “To separate them from others of similar age and qualifications solely because of their race generates a feeling of inferiority as to their status in the community, that may affect their hearts and minds in a way unlikely ever to be undone.”¹

The opinion’s Footnote 11 became one of the most discussed citations in Supreme Court history. It listed Clark’s 1950 paper presented at the Mid-century White House Conference on Children and Youth along with works by several other social scientists, including Witmer and Kotinsky, Deutscher and Chein, and Gunnar Myrdal’s An American Dilemma.¹ By anchoring part of its reasoning in social science rather than in legal precedent alone, the Court signaled that the real-world effects of a law matter when evaluating its constitutionality.

The ruling concluded: “We conclude that, in the field of public education, the doctrine of ‘separate but equal’ has no place. Separate educational facilities are inherently unequal.”¹ The decision overturned the framework established by Plessy v. Ferguson in 1896 and marked the first time the Court treated psychological well-being as a valid constitutional consideration in an equal protection case.

Criticisms of the Methodology

The doll test’s scientific rigor came under serious scrutiny in the 1960s and 1970s. The sample Clark used for the Briggs testimony was just 16 children, a number that even sympathetic observers acknowledged was too small to draw reliable conclusions from.⁶ The research also lacked a control group of white children, making it impossible to know whether the preference patterns were unique to Black children or reflected something more universal about how young children respond to differently colored dolls.

The North-versus-South data presented a deeper problem. If the doll test was meant to prove that segregation caused psychological damage, the fact that children in integrated Northern schools showed stronger white-doll preference than their Southern counterparts was hard to explain. Critics argued this result actually undercut the NAACP’s central claim. The Clarks never fully resolved the tension, and legal scholars have debated it ever since.

Other objections were more practical. Because commercially manufactured Black dolls were essentially unavailable at the time, the Clarks painted the brown dolls themselves, and critics suggested the hand-painted dolls may have looked less polished than the factory-produced white ones, potentially skewing the children’s reactions.⁶ Some also argued that because the Clarks were Black, they may have inadvertently influenced the children’s answers.

Why the Doll Test Still Matters

Researchers and filmmakers have repeated versions of the doll test in the decades since, and the results have been broadly similar. In 2005, teenager Kiri Davis conducted a version with 21 children at a Harlem school for a documentary short film. Fifteen of the 21 said the white doll was the good or pretty one, results Davis described as not far off from what the Clarks had found more than 60 years earlier.

The legal significance extends beyond the specific numbers. Before Brown, constitutional challenges to segregation focused almost entirely on whether physical facilities were truly equal. The Clarks’ work, and Carter’s insistence on using it, introduced the idea that laws can cause invisible injuries that matter as much as crumbling school buildings. Footnote 11 opened the door for social science evidence in constitutional litigation, a practice that has since reached areas from jury selection to sentencing. Whatever its methodological limitations, the doll test changed how courts think about what counts as harm.

1
Justia. Brown v. Board of Education of Topeka
2
National Park Service. Kenneth and Mamie Clark Doll – Brown v. Board of Education National Historical Park
3
The Legacy of Dr. Kenneth B. Clark. The Doll Study
4
United States Courts. History – Brown v. Board of Education Re-enactment
5
Library of Congress. Brown v. Board of Education of Topeka, Kansas – Brown v. Board at Fifty: With an Even Hand
6
Cornell Undergraduate Law and Society Review. Outside the Dollhouse and Into the Court: The Importance of the Doll Test in Brown and Beyond