La la la la, la la la la, Elmo’s world. La la la la, la la la la, Elmo’s world!
After listening to Dr. Claire Duquennois, it’s come to my attention that we might actually be living in Elmo’s world. On February 29, Duquennois, an assistant professor in the Department of Economics at the University of Pittsburgh, spoke at the Sanford School of Public Policy about her research on the impact of “Sesame Street” on voter turnout and behavior. As the first of a series of papers on child media representation, Duquennois and her co-author Jiangnan Zeng examined the impact of the highly popular television show on voters born in the 1960s.
For those who didn’t have “Sesame Street” as a cornerstone of their childhood, the show first aired in November 1969, and quickly attracted a large audience of young children from 2-5 years old. The show was unique in its academic and socio-emotional curriculum, as well as in its diverse and integrated cast. Duquennois described the show as having two intents: the first was to create academic curriculum for preschool age children. But the second, more implicit goal was to improve children’s self image, increase their racial tolerance, and highlight the importance of different perspectives, cooperation, and fairness. This is exhibited by the amount of documentation from the creation of the show, as well as the consultation of psychiatrists like Dr. Chester Pierce, who was an expert in the consequences of racism and television’s impact on the portrayal of minorities.
Whereas other shows like “Mister Rogers’ Neighborhood” featured a more white and suburban cast and setting, “Sesame Street” aimed to relate to kids in more urban or low income areas. For example, both the adult and adolescent cast featured numerous people of color, and the show’s set was reminiscent of Harlem brownstones. The show also brought on numerous diverse guest stars, many of which were important figures in the Civil Rights movement. For many children living in white-dominated suburbs at the time, “Sesame Street” was their first introduction to people of different cultural backgrounds. This “hidden agenda” did not go unnoticed by more conservative governments. For example, the Mississippi commission for education TV vetoed the airing of “Sesame Street” due to the messaging of integration and diversity, although this decision was later overturned due to popular support for the show. Duquennois and Zeng wanted to know: Can child media reduce prejudice in the long-run, impacting voter preferences and behaviors in adulthood?
There had already been a lot of research on mass media in terms of short-term voting outcomes, Duquennois said. In particular, she spoke about research on the news and mass media creating a negative impact on racial and ethnic tensions. However, there was a lack of research on both child media and its impact on later life voting, as well as media’s ability to reduce biases in the majority group. In particular, Duquennois frequently referenced a paper by Melissa S. Kearney and Phillip B. Levine titled “Early Childhood Education by Television: Lessons from Sesame Street.” Duquennois also spoke on the previous research done on contact theory, which has proven that interactions with other groups can help to reduce biases. For example, research done on random college roommates has found that introducing college-age students to people from differing cultural backgrounds has a positive impact on reducing prejudice.
To any readers still waiting to hear the connection, here it is. Duquennois used a difference in difference model with four different segments (really, a scale of low to high coverage, but she simplified for our sake). The treatment group is identified as children younger than six (“Sesame Street’s” target audience, as well as kids who would be home the majority of the day instead of at school) and with high coverage. This methodology is primarily based on Kearney and Levine’s 2019 study mentioned earlier. Since it’s impossible to tell which children were actually watching “Sesame Street,” Kearney and Levine relied on the statistic that nearly 50% of children were watching the show if it was available to them. They also controlled for general patterns in a particular cohort in that particular state like migration and attenuation bias.
In terms of getting voting reports, the study used election year responses from 2006-2020 on the Cooperative Congressional Election Study (CCES) as well as Implicit Association Test (IAT) scores. Specifically, the report used data in major party ballots for US House elections.
In terms of results, Duquennois broke elections into various different demographic compositions. This included elections between two white men, a Republican white man and Democrat woman of color, vice versa and et cetera.
The results were quite interesting. In the case of electoral participation, an increase in coverage by one standard deviation (20 ppts) increased the treated cohort’s voter turnout by 2.8 ppts (4.4%). Voter registration increased by 1.8 ppts (2.4%), and treated cohorts were more likely to know whether they were registered or not.
Additionally, those with more television coverage in their childhood later on expressed increased political knowledge, including more interest in public affairs, better recognition of elected officials’ names, and increased engagement for marginal voters. There was also increased identification with a party and political ideology. However, there were null effects on more costly forms of political engagement like protesting or primary turnout rates.
The most interesting part to me, however, is the impact on voter preferences. Duquennois found that former watchers of “Sesame Street” are more likely to vote for minority and female candidates, regardless of political party.
Even more interesting, the decreased race and gender bias in voting patterns does not translate to policy views. There’s evidence that “Sesame Street” viewers both support gay marriage and restrictive immigration policies, which are often seen as opposing political views. That said, what is consistent is that those in the treated cohort were more likely to have an opinion, regardless of what the opinion actually is.
Moreover, it appears that the hidden messaging of “Sesame Street” was effective in decreasing bias. According to the IAT score results, one standard deviation increase in television coverage reduced the race IAT scores of white subjects by 0.013 standard deviations. However, it had null effects on non-white respondents. There was no evidence of selection bias of taking the race IAT in treatment versus non-treatment groups. As for the gender-career IAT test scores, there was no clear change on bias results, but there was evidence of a selection bias with the treated cohort more likely to take the gender-career IAT.
Duquennois concluded her presentation with a few final takeaways: “Preschool age exposure to child media portraying an inclusive, egalitarian and diverse America reduced prejudice in the long run, with consequential implications for voter preferences.”
