Science, Civil Rights, and the Doll Test

Two black scientists explored the effects of segregation on children. Their “Doll Test” played a key role in the Civil Rights movement.


A great article from @kristel. Thank you for this!

1 Like

Thanks for the article? This caught my attention:

“Most importantly, and unexplained at the time, the study showed that children from the integrated (mixed) schools were more likely to evaluate the brown dolls negatively.”

What’s going on here? Did it have to do with the prevalence of white toys even in integrated contexts? Also, I don’t think the author ever followed up on this point:

“As we will soon see, the order of the questions matters too.”

Very curious! Best wishes!

1 Like

Agreed. The article seemed to stop rather abruptly before these questions were clarified. Do we have to go hunt down those results ourselves now?

On another note, that part where the children are asked what doll they look like themselves is gut-wrenching, and some suspicious part of me can’t help wonder if the whole test was designed by some sadist and set up to elicit such a response in the first place.
As if it was made to have children go through a particular chain of reasoning like “I want you to tell me what is good and what is bad, and then to realize you’re the bad one”.

1 Like

So this was an editing error that hints at some content that was dropped when we simplified the article. Good catch.

@Kristel, would you be willing to share the text relevant to this point?

Have you looked at the video? It is heartbreaking.

I think it is better to ask who set up and tolerated the segregated world, and if they were sadists. Surfacing issues caused by racism isn’t sadistic, but it certainly does force us to confront the pain we’ve gratuitously inflicted on one another.

Hi Joel_B,

Thanks for the comment. As Joshua said, we had to cut down the article and in that process have missed some explanations for the paragraph you referred to. Let me address it here:

1. Children from the integrated schools were more likely to evaluate the brown dolls negatively:

One part of the Clarks’ study compared the responses of black students who go to a mixed school (those who interacted frequently with white children) vs. segregated school (only interacted with black children). The result: 71% of black students from mixed school (Northern US) thinks that the colored dolls look bad, compared to 49% of back students from the segregated schools (South). There wasn’t a clear explanation of “why” this is the case. But it was important, as it turned out, that this is the main piece of data that would support the argument that segregating school has a negative impact towards the children. A conclusion that influenced the supreme court ruling.

2. The order of the questions matters.

About 40 years after the publication of the Clarks’ Doll Test, Walter C. Farrell, Jr. and James L. Olson revisited the study. They recruited 151 five year-old students from both racially segregated and non-segregated schools. Instead of using brown and white dolls, they use eight photograph cutouts of dark-skinned black, light-skinned black, dark-skinned white, and light-skinned white female and male. They asked different sets of questions to that of Clarks’ in the original study. The boys and girls were asked to identify the following about the doll:

  1. Most like you
  2. Other children like best
  3. The teacher likes best
  4. Who likes school
  5. Other children don’t like
  6. The teacher doesn’t like
  7. Who doesn’t like school

Farrell and Olson’s questions were more nuanced. Instead of asking the students to self-identify last, they placed the identification question first. That ordering likely matters. The three questions that followed (questions 2 - 4) were meant to gauge the children’s positive preference towards the doll, while the last three questions (questions 5 - 7), their negative preference. Look closely at how these questions were framed. Instead of asking the questions which ones they think “looks nice” or “looks bad” as in the Clarks’ study, Farrell and Olson wanted the children to answer on behalf of someone else’s preference.

At the end of the study, the results were compared to that of Clarks’. As previously found, these children did not have any issues identifying the doll that is most like them. What was different, however, was that in this study, racial identification (based on skin color) did not show any statistically significant association with the children’s racial preference.

The choice for white dolls made up for 47% of the responses, compared to the previous 65% in Clarks’ study. When it comes to negative preference (that is, those who think that the colored doll 'looks bad’, or in this case, the doll that other children don’t like), Clarks’ study reported a lofty proportion of 77% negative responses, while Farrell-Olson only discovered 36%. Farrell-Olson’s subjects also had a more balanced frequency of choosing the black figure. This study demonstrated that when it comes to understanding racial perception, the way the questions were framed is important.

You can read the study here:

I hope that helps clarify things.

Happy to answer any other questions that you may have.



Thanks @Kristel - this is the question I had after reading your article last week: Did anyone repeat the test with self-identification first? I assumed that children might have more positive view of the dolls that looked like themselves after thinking about how they looked like first.

On a different note: As I’ve read baby / young children’s books to my kids, I’ve wondered whether there is a real racial bias in the way diversity is portrayed. In the books we have that show diverse children rather than just white children, it seems like the white children more often are happy or energetic, and the children who are not white are more often sleeping, crying, or annoyed. I would like to see if someone has done a survey of children’s books to know whether it’s just books that I have, or this is a real problem.

Hi Valerie,

Hopefully the additional info on Farrel-Olson’s study helped.

With regards to baby/young children’s books. I have come across this discussion too while researching the topic.

Someone wrote on this issue on The Conversation last month, which you might be interested in checking out: Rooting out racism in children's books

This one is on some children’s books you might be interested in looking into: 23 Children’s Books on Race (Because It’s Never Too Early to Learn) | SELF



more examples: Children’s Books That Tackle Race and Ethnicity - The New York Times

1 Like

Thanks! I have put a few for younger kids on hold at my library.

If I can make any suggestions for anyone else who has kids:

I really liked some articles Shai Linne wrote on Gospel Coalition, so I have his kids book on my kids’ wishlist. I don’t have it yet, but it’s got five-star reviews and I’m excited to get a book from a Christian perspective:

I also read the book The House on Mango Street by Sandra Cisneros in college. It’s a book for teens, but it’s pure poetry. One of my favorites. Just incredible writing and powerful.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.