SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

The 2008 presidential election presented a unique opportunity to examine children's attention to racial issues in politics. We conducted interviews with 6- to 11-year-old children (70 boys, 60 girls; 29 African Americans, 58 European Americans, 43 Latinos) within 3 weeks prior to and after the election. Interview questions concerned knowledge, preferences, and perceptions of others’ attitudes concerning the election, views of the implications of the election for race relations, and personal aspirations to become president. Results indicated that children were highly knowledgeable about Obama's status as the first African American president. Most children felt positively about the presence of an African American candidate for president, although a few children showed clear racial prejudice. Overall, children expected others to show racial ingroup preferences but simultaneously endorsed the optimistic view that Obama's race was a slight asset in his bid for the presidency. Older children were somewhat more likely to view Obama's race as negatively impacting his chances of being elected than younger children. African American and Latino children were more interested in becoming president than European American children; aspiration rates did not change from pre- to post-election.

Barack Obama's election to the presidency of the United States in 2008 was an event of major historic importance. Many social commentators suggested that the election would have meaningful, far-reaching consequences for Americans’ views of race, perhaps most powerfully among children. In an interview with the Washington Post shortly after the election (Fletcher, 2009), President Obama himself stated that “[t]here is an entire generation that will grow up taking for granted that the highest office in the land is filled by an African American.…I mean, that's a radical thing. It changes how Black children look at themselves. It also changes how White children look at Black children.” Although such claims are certainly plausible, scant research has examined children's attention to national elections or their perceptions of racial discrimination within politics. The primary goal of the present study was to examine children's knowledge and views of the role of race in the 2008 presidential election, a single highly visible and widely discussed national event.

To examine children's views of the role of race in the presidential election, we interviewed children between the ages of 6 and 11 within the 3 weeks prior to and following the election. Interview questions addressed children's knowledge and views of racial issues related to the election, as well as personal political aspirations. Although we interviewed children before and after the election, our primary focus was not on change over time but instead on fully documenting children's attention to, knowledge of, and views of the role of race in the election. We recruited African American, European American, and Latino participants with the goal of testing hypotheses about racial group variations in children's political knowledge and attitudes, as assessed by both quantitative and qualitative measures. Given Obama's racial group membership, and the extensive research on Black–White attitudes in developmental psychology, we readily generated hypotheses concerning African American and European American participants. However, far less is known about Latino children's views of race, including whether Latino youth perceive themselves and other Latinos as sharing stigmatized status with African Americans, or, in contrast, perceive themselves and other Latinos as relatively more similar to European Americans (Mindiola, Nieman, & Rodriguez, 2002). Thus, some of our analyses of Latino children's responses are exploratory in nature; nonetheless, we expected these analyses to provide important information about the fastest growing segment of American youth. Finally, a note about terminology: we use the terms “race” and “racial groups” to be inclusive of ethnicity, although we recognize that one might legitimately do the reverse (see Buriel & Parke, 2007).

Practical and Theoretical Motivations for Examining Youths’ Views of the 2008 Election

Although clearly an event of historic importance, why might the election be of interest to developmental psychologists? First, political commentators have long been concerned about the low levels of political participation in the United States, particularly among people of color (Sherrod, 2003). Despite the need to better understand the factors that shape political engagement, the topic has been a neglected area of developmental inquiry (Rozin, 2006). This is especially concerning given that political interest and engagement emerge well before adulthood and are relatively stable across the life course (Sears & Levy, 2003). Salient political events, such as elections, may play an especially important role in the formation of political attitudes (Sears & Valentino, 1997). The dearth of empirical work on the topic of political development conducted with preadolescent youth means that we know little about when, why, and how children come to attend to political events, engage in political discourse, or form political values. We expected this study to contribute to the development of a more complete understanding of the development of political interest and engagement among racially diverse children.

The second reason to be interested in the election concerns children's perceptions of issues of racial diversity and discrimination within the political sphere. Most developmental research on perceptions of discrimination has examined children's perceptions of interpersonal discrimination, typically within familiar contexts such as schools or playgrounds, rather than societal-level discrimination (e.g., Brown, Alabi, Huynh, & Masten, 2011; Killen, 2007). This distinction is important because perceptions of the two forms of discrimination are likely to affect youth in distinct ways. For example, instances of interpersonal discrimination are likely to be perceived as reflecting the attitudes of the particular individuals involved and thus to be circumscribed in their influence. Instances of institutional or societal-level discrimination, in contrast, are likely to be perceived as reflecting the views of large segments of society and thus to have a wider scope of influence. National elections reflect the attitudes of Americans as a whole, and thus we expected data on youths’ perceptions of the role of race in the 2008 election to make a valuable contribution to the understudied topic of perceptions of societal-level racial bias and discrimination.

In formulating hypotheses concerning children's views of the role of race in politics, we drew on Brown and Bigler's (2005) developmental model of children's perceptions of discrimination, which outlined group, situational, and individual factors that affect children's perceptions of discrimination. Brown and Bigler argued that (1) children's attributions to discrimination should increase with age as a result of their increasingly sophisticated cognitive abilities (e.g., perspective-taking) and (2) African American and Latino children should be more likely to make attributions to discrimination than their European American peers as a result of their greater personal experience with discrimination and higher levels of parental racial socialization. As Brown and Bigler noted, however, the tenets of their model were based almost exclusively on the empirical study of children's perceptions of interpersonal discrimination, and thus the present study was among the first attempts to generalize the model to children's perceptions of societal-level discrimination.

Children's Knowledge of Race and Politics

Did children understand the historic importance of the 2008 election? To grasp the importance of the election, children would have needed to be aware that Obama was African American and that no African American had ever been elected president. Several literatures led us to expect elementary-school-aged children to be aware of these facts. First, children routinely attend to race because it is perceptually discriminable and serves as a basis for the sorting of individuals in the culture (e.g., via de facto segregation; Bigler & Liben, 2006). Second, children are aware of race as a social category and can understand and apply racial labels in early childhood (Aboud, 1988; Susskind, 2007). Thus, we expected our elementary-school-aged participants to identify correctly the racial group memberships of both Obama and McCain. We were, however, unsure whether Obama would be labeled as African American or biracial, given the sparse research on children's perceptions of biracial individuals.

In addition to noticing race, children become aware of racial differences in occupational roles, including the presidency, during the elementary school years. For example, in a study conducted several years prior to the 2008 election, Bigler, Arthur, Hughes, and Patterson (2008) found that 69% of 5- to 10-year-old children were aware of the absence of African Americans from the U.S. presidency. Thus, we expected the majority of our participants to be aware that Obama, if elected, would be the first African American U.S. president. We also, however, expected variations in knowledge across participant age and race. From a theoretical perspective, older children and African American children should be more aware of the lack of African American presidents than their peers because of their greater exposure to information on the topic via school and parental socialization (Hughes et al., 2006). Consistent with this notion, Bigler et al. (2008) reported that older and African American children showed greater knowledge of the lack of African American presidents than younger and European American children.

Race, Racial Attitudes, and Voter Preferences

Political scientists have long argued that racial attitudes affect voter attitudes and behavior. Research from the 2008 election indicated that adults’ racial group membership and racial attitudes were related to their candidate preference (Dwyer, Stevens, Sullivan, & Allen, 2009; Greenwald, Smith, Sriam, Bar-Anan, & Nosek, 2009). African Americans and Latinos were more likely to vote for Obama than European Americans (CNN, 2008). In addition, individuals who held more positive views of African Americans were more likely to support Obama than individuals with less positive attitudes, even after accounting for other relevant factors, such as race and political party affiliation (Dwyer et al., 2009; Greenwald et al., 2009).

Given Obama's lead over McCain in the weeks prior to the election (Gallup, 2008), we expected that a majority of children in our sample would support Obama's candidacy. Nonetheless, children, like adults, show robust ingroup biases related to race (e.g., Aboud, 1988). Because European American children typically demonstrate much stronger anti-African American affect and stereotyping than African American children (Aboud, 1988; Hughes, Bigler, & Levy, 2007), we expected European American children to show less support for Obama than African American children.

In addition to their candidate preferences, we asked children about their own affective reactions to learning about the presence of an African American candidate. This question allowed us to assess children's reactions to Obama's race per se (as distinct from other characteristics). We expected children to state that they felt positively when they learned that an African American was running for office, consistent with past work indicating that children typically reject overt race-based exclusion as morally wrong (Killen, 2007). Nonetheless, we also expected children's personal reactions to show evidence of ingroup bias. That is, we expected African American children to report feeling more positively about Obama's candidacy than European American and Latino children. We did not predict age effects in children's reports of their affective reactions, given that racial bias does not show a consistent pattern of change across the elementary school years (Doyle & Aboud, 1995).

Perceptions of Racial Bias and Discrimination

Children's awareness of racial bias and discrimination within interpersonal contexts increases over the course of childhood (McKown, 2004). Far less is known, however, about children's perceptions of societal-level discrimination. To address this gap, we collected data on a range of dependent variables, including children's: (1) perceptions of others’ affective reactions to the presence of an African American candidate, (2) views of African American and European American adults’ voting behavior, and (3) assessment of whether and how Obama's race would affect the election. We employed multiple, newly developed measures because of the absence of extant measures of perceptions of societal racial biases for use with children and our desire to tap both perceptions of prejudice (i.e., affective reactions) and discrimination (i.e., differential treatment; Dovidio & Gaertner, 1986).

It seemed possible that children would perceive Obama's race as an asset in the election because they expected Americans to be eager to end the exclusion of African Americans from the presidency (Killen, 2007). This view might be especially prevalent among young children, who have greater difficulty understanding the perspectives of others and therefore respond egocentrically on some social tasks (Selman, 1980). However, the opposite outcome—the perception that Obama's race would negatively affect his presidential bid—also seemed possible. Bigler et al. (2008) reported that approximately one third of elementary-school-aged children believed racial discrimination to have played a role in producing the lack of African American U.S. presidents. Furthermore, children typically assume that ingroup bias is a normative process (Abrams & Rutland, 2008; Dunham, Baron, & Carey, 2011); as a consequence, they might assume Obama's race would undercut his popularity with European American and Latino voters.

Although it was not clear whether children—at the group level—would perceive racial bias to play a significant role in the election, we expected children's perceptions of discrimination to vary across racial groups. There are well-documented racial differences in the perception of racial equality and discrimination among adults and adolescents (Eibach & Ehrlinger, 2006; Hughes & Bigler, 2011) and children of color show greater awareness of racial discrimination than their European American peers (Brown et al., 2011; Hughes & Bigler, 2011; McKown, 2004). In addition, African American and Latino children receive more parental racial socialization than European American children (Hughes et al., 2006). Based on these empirical data, as well as Brown and Bigler's (2005) model, we expected that, across measures, African American and Latino children would perceive racial bias as playing a more important role in the election than would European American children.

In addition to racial group variations, we expected age-related variations in children's perceptions of the role of racial bias in the election. As noted earlier, Brown and Bigler (2005) proposed that children's perceptions of discrimination increase with age as a result of greater stereotype awareness and perspective-taking ability. Importantly, the role of racial bias in political preferences is typically ambiguous; individuals who opposed Obama's candidacy might have done so for various reasons unrelated to his race. Under ambiguous conditions, younger and older children's reasoning about racial exclusion is likely to differ (Brown & Bigler, 2005). Young children tend to reason about fairness using the principle of equality (i.e., everyone should receive the same treatment), whereas older children tend to use the principle of equity (i.e., treatment should depend on factors such as effort or intentions; Damon, 1975). Thus, young children might consider the absence of non-European American presidents as sufficient evidence of racial discrimination in the political process, whereas older children might consider intraindividual factors as responsible for the absence of African American presidents (e.g., African Americans’ lack of motivation to pursue, or qualifications for, the role; Bigler et al., 2008).

Political Engagement

Observers have long been concerned about the low levels of political engagement demonstrated by American youth, particularly among African Americans and Latinos (Sherrod, 2003; Torney-Purta, Barber, & Wilkenfield, 2007). Importantly, the presence of candidates who are racial ingroup members may increase youths’ interest in politics (Barreto, 2007; Tate, 1991). To assess children's interest in politics, we asked participants to report whether they were interested in becoming president and whether they viewed the presidency as an attainable goal, consistent with measures used by Bigler et al. (2008). Given Obama's victory, we expected aspirations to the presidency, and perceived attainability of the role, to increase from pre- to post-election among African American, but not European American, children.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

Participants

Participants were 130 children (70 boys, 60 girls; ages 6 to 11) recruited from after school care programs in the Midwestern and Southwestern United States as part of a larger study of children's political socialization. Participants in both regions resided in communities that supported Obama (i.e., Obama received over 60% of the vote; New York Times, 2008). The sample included 29 African Americans or biracial African American-European Americans, 58 European Americans, and 43 Latinos or biracial Latino-European Americans (see Table 1). Although we did not assess participants’ cultural ancestry, Mexican Americans comprise the majority of Latino students in the regions in which we collected data.

Table 1. Participant Demographic Characteristics
 BoysGirlsAfrican AmericanEuropean AmericanLatinoCombined
6-year-olds127311519
7-year-olds159413724
8-year-olds1613713929
9-year-olds15859923
10-year-olds91269621
11-year-olds31144614
Combined7060295942130

Overview of Procedure

Children were interviewed individually on two separate occasions at their after school care sites. Pre-election interviews were completed within the three weeks prior to the 2008 United States presidential election. Post-election interviews were completed within the three weeks following the election. The majority of children (86%) participated in both pre- and post-election interviews. Children were interviewed by a same-race interviewer when possible; 83% of African American, 98% of European American, and 67% of Latino participants had a same-race interviewer.

Pre-election Measures

Knowledge of candidates

Participants were shown individual color photographs of the two major-party presidential candidates (Barack Obama and John McCain) and asked to identify each individual and explain why he was in the news. The candidates’ names and roles were provided for children who stated that they did not know them or who answered incorrectly.

Candidate preference

Children were asked whom they wanted to win the election.

Knowledge of candidates’ race

Participants were shown individual color photographs of Barack Obama and John McCain and asked to identify the race of each candidate. Response options were Black, White, and Half Black, Half White.

Knowledge of past presidents’ race

Participants were asked whether an African American had ever served as president of the United States.

Affective responses to an African American candidate

To assess children's own emotional reactions to an African American's candidacy, they were asked, “How did you feel when you found out that a Black man was going to run for president?” To assess children's perceptions of others’ emotional reactions, they were asked, “How do you think Black kids [other Black kids] felt when they found out that a Black man was going to run for president?” and “How do you think White kids [other White kids] felt when they found out that a Black man was going to run for president?” Participants’ open-ended answers were recorded.

Perceptions of the Influence of Race and Racial Bias on the Election

Predictions concerning voter behavior

Children were asked to predict the proportions of African American and European American adults that would vote for Obama using response options that ranged from almost none (1) to almost all (5). Responses were converted into difference scores by subtracting ratings for European Americans from ratings of African Americans and thus scores ranged from –4 (i.e., more European Americans than African Americans) to +4 (i.e., more African Americans than European Americans).

Perceptions of effect of Obama's race on election

Children were asked how Obama's race might affect his ticket's chances of being elected, with response options ranging from hurt his chances a lot (1) to help his chances a lot (5).

Endorsement of experimenter-provided attributions for Obama loss

Children were asked (yes or no) whether each of three experimenter-provided reasons might account for a hypothetical Obama loss: lack of leadership ability (“He doesn't have good enough leadership skills”), poor policies (“He doesn't have good policies and plans for making the country better”), and racial discrimination (“He is Black and people don't want to vote for a Black man”).

Aspirations

Children were asked, “Would you like to be president of the United States when you grow up?” and “Do you think that, if you wanted to, you really could be president of the United States when you grow up?”

Post-election Measures

Perceptions of effect of Obama's race on election

Children were asked how Obama's race had affected his chance of being elected, with response options ranging from hurt his chances a lot (1) to helped his chances a lot (5).

Endorsement of experimenter-provided attributions for Obama victory

Children were asked (yes or no) whether each of three experimenter-provided reasons contributed to Obama's victory: his leadership ability (“He had the best leadership skills”), his policies (“He had the best policies and plans”), and voters’ desire for an African American president (“People wanted a Black person to be president”).

Significance of election for race relations

Children completed a measure of perceptions of race relations in light of Obama's election. Questions assessed children's beliefs in two areas. Three items concerned post-election racial equality: Obama's victory (1) “proves that White and Black people have the same chances to succeed in the United States,” (2) “proves that everyone who works hard in the United States has an equal chance to become president,” and (3) “proves that White people like Black people just as much as they like White people.”

Four additional items concerned post-election racial bias: (1) “Even though Obama won, it is still easier for White people to be successful in the United States than for Black people,” (2) “Black people have to work harder than White people in order to become president,” (3) “Even though Obama won, people still sometimes treat other people badly or unfairly just because of their skin color,” and (4) “Even though Obama won, White people are still sometimes mean to Black people just because they are Black.” For all items, response options ranged from not at all true (1) to really true (4).

Aspirations

As at pretest, children were asked to report if they were interested in becoming president and if they viewed the role of president as attainable.

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

Data Analysis Plan

For questions with binary outcome variables, binary logistic regression models were used to examine variations in responses across participant race and age. For questions with continuous outcome variables, linear regression models were used. For all regression models, two sets of dummy codes were created to investigate differences across racial groups. First, European American was set as the reference group, allowing European American children's responses to be compared to those of African American and Latino children. Second, Latino was set as the reference group, allowing Latino children's responses to be compared to those of African American children. All regression models were run twice: once using European Americans and once using Latinos as the reference group. To examine variations over time, McNemar tests were used. The McNemar test is a nonparametric method used on categorical data in which responses on two items come from the same participants. We tested both a priori hypotheses and exploratory hypotheses, using one and two-tailed tests of significance, respectively. Bonferroni t-tests were used to probe significant F-tests; only comparisons significant at p < .05 are reported.

Pre-election Responses

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

Knowledge of Candidates

Participants were generally knowledgeable about the identities of the two presidential candidates. Most participants were able to identify the candidates by name (82% for Obama; 69% for McCain) and state that they were running for president of the United States (66% for Obama; 64% for McCain).

To examine children's knowledge of the presidential candidates’ names and roles, composite knowledge scores were created (possible range: 0 to 4). A regression analysis using participant race and age (in years) as predictors indicated a significant effect of age (β = .45, p < .001). Older children were more knowledgeable than their younger peers. European American children (M = 3.24, SD = 1.35) were more knowledgeable than their African American (M = 2.71, SD = 1.46; β = .22, p = .01) and Latino (M = 2.23, SD = 1.46, β = .37, p < .001) peers, who did not differ from each other.

Candidate Preference

The majority of children (84%) expressed a preference for Obama over McCain. Binary logistic regressions indicated that participant race and age were unrelated to candidate preference. Of the 20 children who preferred McCain, 13 were European American, 2 were African American, and 5 were Latino.

Knowledge of Candidates’ Race

As expected, participants were highly knowledgeable about the race of the two candidates. Nearly all children identified Obama as either biracial (69%) or African American (29%). The majority of children (84%) identified McCain as European American. To examine variations in identifying Obama as biracial across participant race and age, binomial logistic regression analyses were conducted. African American (82%) and Latino (84%) children were more likely than European American children (52%) to identify Obama as biracial (Wald statistic = 6.73, p = .009, OR = 4.33 and Wald statistic = 10.06, p = .002, OR = 5.20, respectively).

Knowledge of Past Presidents’ Race

As expected, the majority (77%) of participants knew that an African American had never been president of the United States. A binary logistic regression using participant race and age as predictors indicated a significant effect of race; more African American (86%) and European American (81%) than Latino (64%) children knew that no African American had held the post (Wald statistics = 3.58, p = .06, OR = 3.39 and 4.28, p = .04, OR = 2.78, respectively). Age was also positively related to knowledge (Wald statistic = 3.77, p = .05, OR = 1.35).

Personal Affective Response to an African American Candidate

Children's responses to the three open-ended questions regarding affective responses to the presence of an African American candidate were coded as reflecting positive, neutral, or negative affect. Children could, and occasionally did, give responses that fell into more than one affective category. Responses that reflected more than one category of affect (e.g., “surprised and excited”) were counted within each of the relevant categories (e.g., neutral and positive) and thus response categories were independent. Interrater reliability was κ = .91.

When reporting their personal affective reactions, children largely reported positive responses (see Table 2 for category exemplars and Figure 1 for data). We examined whether responding within each category varied by participant race and age.

Table 2. Personal Affective Reactions to an African American Candidate
PositiveNeutralNegative
Note
  1. AA = African American; EA = European American; LA = Latino.

“Happy because I like Black people” (EA girl, age 6)“I just go on with life” (EA boy, age 7)“Sad because all of the presidents are White. He's the only one” (LA girl, age 6)
“For me it was really good. Because all presidents have been White and now we can learn about other people of color” (LA girl, age 8)“I didn't really feel anything” (EA boy, age 9)“I felt not good because I don't like Black people being president. I only like White people being president” (AA girl, age 7)
“Think it was cool because no Black man has even been president” (EA boy, age 9)“I didn't feel too bad about it” (EA boy, age 9)“Felt mad because all the other White presidents would be mad that a Black man was running for president” (AA boy, age 7)
“Happy because I never thought a Black person would be president” (EA girl, age 9)“Surprised cause there has never been a Black president” (LA girl, age 9)“Upset because I don't really like Black men because they're weird and crazy and there's pretty much only White men involved in everything” (EA girl, age 8)
“Excited because my family wants a Black president” (LA boy, age 9)“Interesting because a lot of people will vote for McCain and African Americans will vote for Obama” (LA girl, age 9)“I was scared” (EA boy, age 9)
“Me and my mom felt happy” (AA girl, age 11)“I felt surprised because I didn't know there would be Black presidents. Only White presidents” (LA boy, age 9) 
“I felt happy because it would be a historical change or event” (LA girl, age 11)“I felt like he was any other candidate” (EA boy, age 10) 
“Good because we finally got a Black man running” (AA girl, age 11)“Weird because we only have White presidents that run” (AA girl, age 10) 
image

Figure 1. Personal and perceived affective reactions to an African American candidate by participant race.

Note. In some instances, children's responses reflected more than one type of affect. Such responses received multiple codes and thus percentages may not add to 100%.

Download figure to PowerPoint

Positive responses

Children typically show ingroup favoritism based on race (see Aboud, 1988) and thus we expected a higher percentage of African American than European American or Latino children to report positive responses to Obama's candidacy. A binary logistic regression using participant race and age as predictor variables and positive (vs. all other) responses as the outcome variable supported our hypothesis. African American children (75%) were more likely than European American (54%) and Latino (51%) children to report positive reactions to the presence of an African American candidate (Wald statistics = 3.25, p = .07, OR = 2.54 and 3.31, p = .07, OR = 2.69, respectively).

Neutral responses

A binary logistic regression using participant race and age as predictor variables and neutral (vs. all other) responses as the outcome variable revealed that European American (30%) and Latino (32%) children were more likely than African American children (11%) to report neutral emotions (Wald statistics = 3.45, p = .06, OR = 3.57 and 3.03, p = .08, OR = 3.44, respectively).

Negative responses

The very small number of negative responses prohibited analysis of this category of responding.

Perceptions of African Americans' Response to an African American Candidate

When asked to report their perceptions of African Americans' responses to an African American candidate, children largely reported positive responses (see Table 3 for category exemplars and Figure 1 for data). We examined whether responding within each category differed by participant race and age.

Table 3. Perceptions of Others' Affective Reactions to an African American Candidate
African AmericansEuropean Americans
Note
  1. AA = African American; EA = European American; LA = Latino.

PositivePositive
“Happy because he's Black and they're Black” (LA girl, age 6)“Kind of happy” (LA boy, age 6)
“A lot happy because they're poor and he helps poor people”(LA girl, age 9)“A little excited, because they've never experienced one” (EA boy, age 8)
“Excited, thrilled, finally recognized” (EA boy, age 9)“Good” (LA boy, age 8)
“Excited because the president would be their kind” (LA boy, age 9)“Some thought it was cool” (EA boy, age 9)
“That it was cool” (EA boy, age 10)“Glad and surprised” (AA girl, age 10)
“Really excited, like ‘OMG finally! It's about time'” (EA boy, age 10)Neutral
“Glad too and they might want to be like him” (AA girl, age 10)“No special feeling” (EA girl, age 7)
“Good because they noticed he would be the first Black president” (EA boy, age 10)“I don't think it really affected them because race isn't an issue anymore” (EA boy, age 10)
“Excited because it will show Black people can do just as good as White people if he's elected” (EA boy, age 10)Negative
Neutral“They want them all to be White and not Black” (LA girl, age 7)
“Felt strange” (EA boy, age 6)“I guess they would feel disturbed that one was running but not at all ashamed” (EA boy, age 7)
“Felt the same way as if he wasn't Black” (AA boy, age 8)“Mad, they'll say ‘boo'” (AA boy, age 7)
Negative“Sad, because they might not want a Black man president” (LA girl, age 8)
“Sad” (EA boy, age 6)“Disappointed” (AA girl, age 8)
“Sad because they like the president before him” (AA boy, age 7)“They would feel like they would hate him. They really really wanted McCain to win.” (AA boy, age 8)
“Probably suspicious and stunned” (AA girl, age 10)“Sad because White people don't play with Black people much” (AA girl, age 8)
 “Some of them might have felt pretty bad because [they] don't like Black people” (EA boy, age 9)
 “They were really, really scared” (EA boy, age 9)
Positive responses

A binary logistic regression using participant race and age as predictor variables and positive (vs. all other) responses as the outcome variable indicated a significant effect of age (Wald statistic = 3.78, p = .05, OR = 1.36). Older children were more likely than their younger peers to report that African Americans felt positively about an African American candidate.

Neutral and negative responses

The very small numbers of neutral and negative responses prohibited analyses of these categories of responding.

Perceptions of European Americans' Response to an African American Candidate

When asked to report their perceptions of European Americans' responses to an African American candidate, children most frequently reported negative responses (see Table 3 for category exemplars and Figure 1 for data). We examined whether responding within each category differed by participant race and age.

Positive responses

A binary logistic regression using participant race and age as predictor variables and positive (vs. all other) responses as the outcome variable revealed no significant effects.

Neutral responses

A binary logistic regression using participant race and age as predictor variables and neutral (vs. all other) responses as the outcome variable revealed that European American children (30%) were more likely than Latino children (14%) to report that European Americans felt neutral emotions about an African American candidate (Wald statistic = 4.67, p = .03, OR = 2.95). In comparison, 22% of African American children reported that European Americans felt neutral emotions, which did not differ from either group. There was also a significant effect of participant age (Wald statistic = 3.90, p = .06, OR = 1.30). Reports than European Americans felt neutral emotions about an African American candidate increased with age.

Negative responses

A binary logistic regression using participant race and age as predictor variables and negative (vs. all other) responses as the outcome variable indicated a significant effect of participant race. Latino children (56%) were more likely than European American children (33%) to state that European Americans had negative reactions to an African American candidate (Wald statistic = 8.18, p = .004, OR = 4.22). The percentage of African American children who believed European Americans experienced negative emotions (41%) did not differ from either group. Reports that European Americans felt negatively about the presence of an African American candidate decreased with participant age (Wald statistic = 4.12, p = .04, OR = 0.74).

Perceptions of the Influence of Race and Racial Bias on the Election

Predictions concerning voter behavior

Overall, participants believed that more African Americans than European Americans would vote for Obama, M = 0.37, SD = 1.55; this score was significantly different from zero, t(121) = 2.66, p = .009. Results of a linear regression using participant race and age as predictors indicated a significant effect of age, β = .20, p = .03, with the perceived racial gap in voter behavior growing with increasing participant age.

Perceptions of effect of Obama's race on election

Overall, children reported that Obama's race would help his chances of being elected slightly (M = 3.54, SD = 1.33). A linear regression indicated that neither participant race, participant age, nor their interaction predicted children's responding.

Endorsement of experimenter-provided attributions for Obama loss

Children were asked to indicate whether (yes or no) they thought each of three factors might play a role in a hypothetical Obama loss: lack of leadership ability (endorsed by 22%), poor policies (endorsed by 31%), and racial discrimination (endorsed by 33%). Binary logistic regression analyses indicated that the likelihood of endorsing the discrimination attribution increased with age (Wald statistic = 4.59, p = .03, OR = 1.32). Participant race was unrelated to attributions to discrimination; 21% of African American, 36% of European American, and 35% of Latino children endorsed this attribution. Neither participant race nor age predicted the endorsement of leadership or policies as causes of a hypothetical Obama loss.

Aspirations

When asked whether they would like to be president of the United States, half of the sample (50%) answered affirmatively. A binary logistic regression was run with participant race and age as predictor variables and interest (vs. disinterest) in the presidency as the outcome variable. African American (57%) and Latino (68%) children were more likely than European American (34%) children to report wanting to be president (Wald statistics = 4.99, p = .03, OR = 3.00 and 11.51, p = .001, OR = 4.88, respectively). In addition, more girls (61%) than boys (40%) reported wanting to be president, Wald statistic = 6.57, p = .01, OR = 0.37.

When asked whether they really could be president if they so desired, 64% of children answered affirmatively. A binary logistic regression using participant race and age as predictors of the belief that it was possible (vs. impossible) to become president indicated a significant effect of participant race. African American children were more likely than European American children to report that they could be president (82% and 59%, respectively; Wald statistic = 4.74, p = .03, OR = 3.47). Latino children's responding (58%) did not differ significantly from that of African Americans (Wald statistic = 3.69, p = .06, OR = 3.15) or European Americans (Wald statistic = .05, p = .82, OR = 0.91). Boys and girls were equally likely to report that they could become president if they so desired.

Post-election Responses

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

Attributions for the Election Outcome

Perceptions of effect of Obama's race on election

Overall, children reported that Obama's race helped his chances of being elected slightly (M = 3.64, SD = 1.21). A linear regression using participant race and age as predictors indicated a significant effect of age, β = –.22, p = .04. With increasing age, children viewed Obama's race as having been more detrimental to his chances of being elected.

To examine change in participants' perceptions from pre- to post-election, we created a difference score. On average, perceptions of the effect of Obama's race did not change (M = .04, SD = 1.73); the mean difference score was not significantly different from zero, t(94) = .21, p = .84. A regression using participant race and age as predictors of difference scores indicated no significant effects.

Endorsement of experimenter-provided attributions for Obama victory

Children generally agreed that Obama's leadership ability (79%), his policy positions (77%), and voters' desire for an African American president (50%) contributed to his election victory. Exploratory binary logistic regressions were conducted using participant race and age as predictors of the likelihood of endorsing (or not) each attribution. Participant race was a significant predictor of the endorsement of two of the attributions. European American children (65%) were less likely than African American (95%) and Latino (92%) children to attribute Obama's victory to his leadership ability (Wald statistics = 4.61, p = .03, OR = 10.01 and 4.30, p = .04, OR = 5.31, respectively). European American children (61%) were more likely than Latino children (35%) to attribute Obama's victory to voters' desire for an African American president (Wald statistic = 4.32, p = .04, OR = 2.94). African American children's responding (43%) did not differ from either group.

Significance of Election for Race Relations

Children completed scales measuring belief in: (1) post-election racial equality and (2) post-election racial bias. Children's responses were analyzed using a 3 (participant race: African American, European American, Latino) X 2 (belief type: racial equality, racial bias) repeated measures ANOVA, with the last variable as a within-subjects factor and age as a covariate. Results indicated a significant interaction of belief type and participant race, F(2, 95) = 3.90, p = .02. African American and European American children were more likely to endorse post-election racial equality than racial bias, whereas Latino children showed equally strong endorsement of both constructs [Equality Ms (SDs): AA = 2.97 (0.84), EA = 3.16 (0.68), LA = 2.69 (0.85); Bias Ms (SDs): AA = 2.29 (0.71), EA = 2.16 (0.68), LA = 2.26 (0.73)].

Aspirations

Following the election, 44% of participants stated that they wanted to be president. As was true prior to the election, African American (61%) and Latino (56%) children were more likely than European American (28%) children to report a desire to be president (Wald statistics = 7.16, p = .007, OR = 4.18 and 7.50, p = .006, OR = 3.47, respectively). In contrast to pre-election findings, girls (53%) were not significantly more likely than boys (37%) to report a desire to become president. McNemar tests indicated that the proportion of children who wanted to be president did not differ across time (i.e., pre- vs. post-election) among any racial or gender group.

When asked whether they really could be president, 66% of children answered affirmatively. Binary logistic regressions indicated that participant race, gender, and age were unrelated to the belief that one could become president. McNemar tests indicated that the proportion of children who thought they could be president did not differ across time (i.e., pre- vs. post-election) among any racial or gender group.

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

The 2008 election is widely viewed as both reflecting and influencing Americans' racial attitudes and views of the role of race in society. Indeed, some individuals have claimed that the election will have lasting effects on children's views of both race and politics. We sought to provide empirical evidence that elementary-school-aged children were attentive to issues of race in the election and to test hypotheses concerning children's knowledge and views.

Nearly all children in our sample knew that an African American candidate was running for president. Interestingly, African American and Latino participants were more likely than European American participants to label Obama as biracial. This may reflect underlying differences in African Americans', Latinos', and European Americans' views of race; in previous research, adults with more categorical views of race were less likely to view Obama as biracial than adults with less categorical views of race (Malahy, Sedlins, Plaks, & Shoda, 2010). Furthermore, more than three-quarters of our sample knew that, if Obama were victorious, he would be the first African American president of the United States. Such high rates of attention to issues of race lay the groundwork for the possibility that the election will have long-term consequences for those youth who experienced it.

Children generally expressed positive reactions to the historic presence of an African American candidate, although positive reactions were more common among African American than European American or Latino children. Despite children's generally positive views of Obama, there is some evidence that racial biases may have influenced children's perceptions. For example, European American children were less likely to attribute Obama's victory to his personal competence (i.e., leadership ability) than were African American and Latino children. Explicit negative reactions to Obama's candidacy were rare, but those that did occur were quite interesting. In one instance, an 8-year-old European American girl stated that her reaction to learning that an African American was running for president was “upset, because I don't really like Black men because they're weird and crazy and there's pretty much only White men involved in everything.” However, statements that seemed to indicate racial bias or the endorsement of racial stereotypes were not exclusively limited to European American children: a 7-year-old African American girl stated that she felt “not good because I don't like Black people being president. I only like White people being president.”

To examine children's views of the possible influence of race and racial bias in the election, we employed multiple dependent measures. The broad pattern of results indicated that children—at the group level—expected others to show racial ingroup preference while, simultaneously, endorsing the optimistic view that Obama's race was a small asset in his bid for the presidency. What might account for these conflicting patterns of responses?

Previous work has indicated that children perceive racial exclusion as morally unacceptable (Killen, 2007). If children believe that their own views on exclusion are culturally shared (i.e., inclusivity is a cultural norm), they should then report that most Americans (regardless of race) were pleased that an African American had the opportunity to end the racial exclusivity of the presidential office (Nesdale, 2004). Instead, children's responses suggested a perception of ingroup bias, with African Americans reacting positively and European Americans reacting negatively. For example, an 8-year-old Latino boy said that European American children would feel “bad because [Obama] is not going to be fair with rules—he will choose Black people before White people.” Similarly, approximately one third of children endorsed discrimination as an explanation for a hypothetical Obama loss. These findings suggest that awareness of racial discrimination in American society is relatively common among even young children. Furthermore, African American, European American, and Latino children were equally likely to attribute a possible Obama loss to race, indicating that a general perception of ingroup bias (rather than race-specific socialization) may underlie these beliefs. The antecedents and consequences of such knowledge should be studied more extensively.

When asked specifically about the voting behavior of African Americans and European Americans, children were less likely to endorse the view of others as racially biased. For example, children—at the group level—indicated that only slightly more African Americans than European Americans would vote for Obama. Although children were correct in perceiving that African Americans supported Obama at a higher rate than European Americans, they seriously underestimated the magnitude of the actual difference in voting behaviors—African Americans were more than twice as likely to vote for Obama as European Americans (CNN, 2008).

Furthermore, when explicitly asked about the effect of Obama's race on the election, children reported that Obama's race would slightly help (rather than hinder) his chances of winning the election. Children's optimism may reflect their generally favorable views about contemporary race relations. Both African American and European American children believed that Obama's election was indicative of post-election racial equality (rather than a continuation of racial bias). Interestingly, Latino children were conflicted about the implications of Obama's victory; they were equally likely to endorse views of post-election equality and continuation of racial bias. Latino children's greater skepticism may reflect the continued lack of Latino major-party presidential candidates. Future research should examine whether African American children's optimism about the effects of Obama's election on race relations has continued, or whether post-election optimism has decreased over time (consistent with research on African American adults' views; Agiesta & Cohen, 2010).

Both before and after the election, a greater percentage of African American and Latino children were interested in becoming president than were European American children. Intuitively, one might hypothesize that the higher rates of personal interest among African Americans than European Americans were the result of Obama's candidacy. This notion is not, however, supported by extant data; Bigler et al. (2008) found that African American children were more likely than European American children to aspire to the presidency in data collected several years before Obama's presidential candidacy. It seems possible that children may be motivated to seek political office out of a desire to change the status quo of perceived racial bias. Knowledge of the civil rights movement may be especially influential in children's views of political office as a mechanism by which to fight racial bias and inequality.

The major theoretical foundation for this study came from Brown and Bigler's (2005) model of perceptions of discrimination. Brown and Bigler argued that perceptions of discrimination should be more likely among older than younger children and among African American and Latino children compared to their European American peers. This study is one of the first attempts to generalize Brown and Bigler's model to perceptions of societal-level (rather than interpersonal) discrimination. In terms of age effects, our data partially supported Brown and Bigler's model. Compared to their younger peers, older children were more likely to agree that discrimination was an explanation for a hypothetical Obama loss and perceived greater racial group disparities in voting behavior.

In terms of racial group differences, Brown and Bigler's model was generally not supported. There were no observed racial group differences in attributing a hypothetical Obama loss to discrimination, perceptions of the influence of Obama's race on the election, or perceptions of voter behavior. (There were, however, racial group differences on other measures, such as perceptions of European Americans' affective reactions to an African American candidate and the significance of the election for race relations, indicating that race-specific socialization experiences may influence some aspects of children's attitudes.) One possible explanation for similarity in children' responding across race is that pervasive media coverage of the election increased the salience of racial issues in politics for all children (Payne, 2010).

One major way in which this study extends our knowledge of children's political understanding is the inclusion of Latino participants. These data are particularly important because of the growing Latino population in the United States and relative lack of emphasis on Latinos in research on racial attitudes. Overall, African American and Latino children were similar to each other (and different from European Americans) in their knowledge of Obama's racial background and their aspirations to become president, perhaps reflecting similar racial socialization experiences. However, Latino children expressed less positive personal reactions to the presence of an African American candidate than did African American children, consistent with a view of Obama as an outgroup member. Future research should examine more closely the role of culture, including the diverse cultures that make up the U.S. Latino population, in political socialization and engagement.

Although we believe these data make important contributions, it is important to acknowledge the limitations of our work. Data collection occurred in only two locations, both of which were predominantly Democratic regions. The experiences and attitudes of children in predominately Republican regions may differ, and this will be an important area for future research. In addition, because we were interested in children's views of race within the election, our interview questions often explicitly addressed or labeled race. It is possible that children's answers to our questions were distorted by our explicit focus on race. Future research should examine the effects of explicitly labeling race by comparing children's responding under implicit and explicit testing conditions (as in Rutland, Cameron, Milne, & McGeorge, 2005). Finally, fewer Latino than European American or African American participants were interviewed by a same-race experimenter, and thus we cannot rule out the possibility that their responses were affected by social desirability concerns. Two findings argue against the presence of strong social desirability concerns, however. First, young children tend to be less affected by such concerns than older children (Apfelbaum, Pauker, Ambady, Sommers, & Norton, 2008; Katz, Sohn, & Zalk, 1975) and second, it is especially difficult for children to identify Latinos (relative to the identification of African Americans and European Americans; Quintana, 1998) and thus there was likely to be some element of uncertainty in Latino participants' judgments of their interviewers' ethnicity.

Overall, we believe these data advance our understanding of children's political views, racial attitudes, and perceptions of discrimination in significant ways. It is clear that most American elementary school students attended to the election and were aware of its importance. Psychologists and political scientists undoubtedly will study the long-term effects of such knowledge on this generation's political and racial attitudes and behavior. These data also extend our knowledge of children's views of societal-level discrimination, highlighting the nuanced views that children develop about the impact of ingroup biases on their own and others' preferences. We hope that these findings spark additional theoretical and applied research aimed at maximizing the extent to which all children are motivated to participate in civic debate and work to prevent racial exclusion.

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies

Biographies

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Pre-election Responses
  6. Post-election Responses
  7. Discussion
  8. References
  9. Biographies
  • MEAGAN M. PATTERSON is an assistant professor in the department of Psychology & Research in Education at the University of Kansas. Her primary research interests concern the formation and integration of personal identities, social group identities, and intergroup attitudes among children. In recent research, she has examined the relations between children's self-perceived gender typicality and their endorsement of gender stereotypes.

  • ERIN PAHLKE is an assistant professor of psychology at Whitman College in Walla Walla, Washington. Her research interests include children's and adolescents' understanding of discrimination, stereotyping, and current gender and racial inequalities across contexts. In recent research, she has examined the relations between mothers' racial socialization practices and their children's racial attitudes.

  • REBECCA S. BIGLER is a professor of psychology and women's and gender studies at the University of Texas at Austin. Her work examines the causes of social stereotyping and prejudice among children. She is also interested in children's perceptions of discrimination and the design of interventions aimed at reducing gender and racial bias.