Box 1844, Department of Political Science
Brown University, Providence, RI 02912
tel: 401/863-1578 fax: 401/863-7018
EBARRETT@BROWNVM.BROWN.EDU
Paper presented at the Annual Scientific Meeting of the International Society of Political Psychology, Washington, D.C. July 1995
Abstract
As the American population grows more racially diverse, social scientists are beginning to pay close attention to the political attitudes and behaviors of minorities within the United States. So often when researchers focus on race, however, they neglect to look as well at gender. Similarly, race is frequently overlooked when researchers seek to understand the differences between men and women. This paper examines the political attitudes and behavioral intentions of American adolescents, focusing specifically on the similarities and differences across race, gender, and place of residence. Findings suggest that race and gender do not act independently of one another and that both characteristics have a different relationship to political attitudes depending on whether youths live in the city or in largely white, nonurban areas. Concluding comments suggest a possible link between political attitudes and cultural socialization.
According to the 1990 census, nearly 20 percent of the U.S. population (approximately 49 million people) considers itself non-white, up from 17 percent in 1980 and 12 percent in 1970. The population is diverse demographically, and this diversity is reflected as well in the political attitudes and behaviors of residents. For example, recent public opinion surveys have found that both African Americans and Latinos tend to be more socially liberal than whites (Page and Shapiro 1992; Gurin, Hatchett, and Jackson 1989; Welch and Sigelman 1993). Research has also highlighted the traditionally lower political participation of blacks and Latinos compared to whites (Bobo and Gilliam, 1990; Uhlaner, Cain, and Kiewiet 1989). When they do vote, blacks and Latinos -- with the exception of Cuban Americans -- are more likely than Anglos to vote for the Democratic candidate (Tate 1993; see Welch and Sigelman 1993).
Adolescence is a critical time in the development of political attitudes and intentions toward political involvement. From early childhood, individuals are developing ways to think about politics and governmental institutions, but not until the teen years, do children become capable of complex political reasoning. It is also during adolescents that children face decisions about their future roles in society -- their careers, their families, and their role in a democratic system. Young children tend to hold positive views of leaders and laws: laws are fair, the police are unprejudiced, and the president works for the people (Easton and Dennis 1969). Once individuals enter adolescence, however, they become less idealistic in their view of authority. They see unjust laws, biased police officers, and corrupt politicians. Yet despite the decreased trust among adolescents, Jennings and Niemi (1974) still found teenagers to hold more positive views than their parents of the legal system and government in general. Thus, although cynicism begins to take root during adolescence, trust continues its decay into adulthood.
Race/ethnicity is a strong predictor of adolescent values, independent of the socializing influence of the family environment (Greeley 1975), and there is some evidence that both blacks (see Abramson 1977) and Latinos (Garcia 1973) believe themselves to be less politically efficacious than Anglos believe themselves to be. Political efficacy influences both attitudes and the desire to become politically involved: individuals who believe themselves capable of having an effect are more likely to act. Thus, although there is little research to date specifically looking the impact of race/ethnicity on adolescent political beliefs, given what is known about the relationship between race/ethnicity, general values, and political efficacy, one would expect to find race/ethnic differences in adolescents similar to those observed among adults. One goal of this paper is to offer empirical evidence of the political beliefs of racially/ethnically diverse group of youths.
Minority adolescents tend to be concentrated in the more diverse urban areas rather than in relatively homogeneous white suburban neighborhoods. Beyond the influence of personal characteristics -- such as race or ethnicity -- the social context in which one lives may also act upon one's political attitudes (Huckfeldt, Plutzer, and Sprague 1993). For this reason, we might expect that race/ethnicity has a different impact on minorities living among other minorities than it has on those living as a minority among Anglos. Research on black high school students in the 1960s, found higher levels of self-esteem -- a predictor of political participation -- among students living in racially segregated neighborhoods than among black students living in predominately white districts (see Rosenberg and Simmons 1971). Chicano youths living in middle-class neighborhoods are less likely to see voting as worthwhile compared to Chicanos in lower-income, largely minority, areas (Jankowski 1986). Living among one's racial/ethnic peers seems to be an empowering experience for minorities. The rebellious black youth of culture -- enmeshed with religion and politics -- empowers its followers as is seeks to actively combat disempowering racism (Brake 1985).
Superimposed on race/ethnicity and urbanicity is gender. There is ample evidence in the literature to suggest white adult men and women differ in their political attitudes and in their predisposition to participate in political activities (Page and Shapiro 1992; Sapiro 1983). The gender difference is less pronounced among minorities. Welch and Sigelman (1989, 1992) found the greatest divergence between Anglo men and women, somewhat less variance in the attitudes of Latino men and women, and virtually no gender-based disagreement among blacks. Speculating on the gender gap among adolescents leads to two possible conclusions. On the one hand, because attitudes seem to be fairly well established by early adulthood and stable throughout the adult years, we might expect the attitudes of girls and boys to be relatively reflective of those of adults (Sears 1975, 1983). Thus, we would expect to find the greatest differences among whites and smaller differences among minorities. On the other hand, as each succeeding generation of girls is being raised in an increasingly gender-equal environment and as girls today expect to be economically productive members of society (Smith 1981), we might expect to find fewer gender differences in adolescents than we see among adults. This paper will explore gender as a relevant factor in adolescent political attitudes.
Thus, the purposes of this paper are threefold. First, this paper seeks to highlight the similarities and differences in political beliefs among adolescents of differing race/ethnicites and genders. Second, the paper explores the possible interaction between race/ethnicity and gender. And finally, the paper asks to what extent race/ethnic and gender differences vary depending upon the racial milieu of the adolescents.
Research Design and Data
The Sample
During the spring of 1992, 2,662 adolescents in the state of Rhode Island completed questionnaires asking them about their political attitudes and behavioral intentions. Nearly all of the questionnaires were completed during school, most often in a social studies class. Because the principal sample of interest for the larger project was young inner-city adolescents, the study includes a sample of eighth graders from five of Providence's six middle schools, mostly ninth graders from all four of Providence's high schools, and both eighth and ninth graders from the combined junior/senior high school of Central Falls (Central Falls is one of the poorest towns in the greater Providence metropolitan area). A total of 1,154 respondents (44 percent of the sample) live in these two densely populated urban areas (areas with greater than 5,000 people per square mile).
As mentioned above, all but one of the middle and high schools in Providence and Central Falls were included. Nonurban schools were randomly sampled. Only three schools contacted refused to participate in the project: one because it is currently implementing a curriculum to make students aware of the stereotypes of welfare recipients (the larger project looks at views of welfare recipients as well) and the other two because of time constraints.
Although either the entire population of schools was included or schools were randomly selected, the students within each school were not randomly selected. Instead, participants were those in attendance the day the questionnaire was administered. Most questionnaires were administered during social studies classes, and all eighth and ninth graders in Rhode Island are required to take a such a class, thus there was no selection bias in this regard. However, it is possible that the least socially attached students were absent. In an attempt to minimize this problem, 103 youth were given the questionnaire in the context of a neighborhood club. The neighborhood club serves as a drop-in center for many youth; they come to play basketball or pool, to do their homework, or just to hang-out. Youths in the clubs were paid $5 for completing the questionnaire as an added incentive.
The sample is fairly representative of the Rhode Island youth population in urban and nonurban areas. The 34.4 percent urban white somewhat represents the urban area's non-Hispanic white population (students were asked in a single question their racial and/or ethnic background), but matches well the percent on non-Hispanic whites attending urban public schools. The 24.0 percent Latinos included in the urban sample are similar to the state's 26.0 percent overall urban Hispanic population, and the 1.6 percent nonurban sample represents the 1.8 percent present in the population. The mean white, black, and Latino students in the sample are shown in Table 1, as are the school and town means.
[Insert Table 1 about here]
The target sample of this study was eighth and ninth graders, and over 66 percent of the respondents were in either the eighth or ninth grade. Interestingly, however, though the urban sample was disproportionately ninth graders and the nonurban sample eighth graders, the students tended to be of the same age. In both sub-samples, about 44 percent of the students were 13 to 14 years old and roughly 45 percent were 15 to 16 years old, but while 47 percent of urban youth were in the eighth grade, only 23 percent of nonurban subjects were in the eighth grade. The largest group of nonurban youth were in the ninth grade (42 percent) compared to just 22 percent of urban adolescents. Although these findings demonstrate nothing conclusively, they suggest that urban youth tend to be a grade level behind their nonurban cohorts.
The primary purpose of this paper is to examine the influence on political attitudes and behavioral intentions of gender, race or ethnicity, and place of resident. The analyses incorporates a 2 (girls, boys) x 3 (whites, blacks, Latinos) x 2 (urban, nonurban) design. Because of the demographic make-up of urban and nonurban areas, however, there are considerably more blacks and Latinos in the urban sample than in the nonurban sample. To be able to tease out the effects of race and ethnicity among nonurban youth, cases has been weighted such that nonurban blacks and Latinos are over-represented among nonurban youth. The conclusions will not be representative of the nonurban population per se, but rather representative of white, black, and Latino nonurban populations separately. [See Appendix A for the weighting protocol.] Admittedly, Latinos represent many ethnic groups. Unfortunately, because youths were given only the categories "Latino" and "Mexican-American," data analyses cannot distinguish among other ethnic groups; a brief analysis does suggest, however, little difference between those choosing the "Latino" category and those labeling themselves "Mexican-American."
Political Attitudes and Behavioral Intentions Scales
The youths answered a number of questions concerning their political attitudes and behavioral intentions. The items of interest in this article form three attitudinal factors and three behavioral factors. The first attitude factor measures trust for those adults and peers close to the individual adolescent: parents, teachers, the school principal, minister or priest, and best friend. These items of personal trust form an internally consistent factor, with a Cronbach's coefficient alpha of 0.67. (Cronbach's coefficient alpha is a measure of internal consistency and ranges from 0 [no consistency] to 1 [perfect consistency]). The second factor taps into adolescents' trust for leaders, including: Supreme Court justices, the U.S. Congress, the governor, presidents of large companies, and bank presidents (Cronbach's coefficient alpha is 0.79). The final attitude factor measures satisfaction with the current system. Included in this factor are items asking adolescents whether they believe: 1) the police make things worse, 2) the president makes things worse, 3) most laws are (not) fair, and 4) the president is not to be trusted. Cronbach's coefficient alpha for these four items is 0.60.
The survey also yielded three factors depicting adolescents' behavioral intentions within the political sphere. Adolescents first indicated how likely they felt is was that they would vote when they reached voting age and that they would talk their friends into voting. These two items have a high internal consistency (Cronbach's coefficient alpha is 0.78). The second behavioral factor asked youths about their intentions to become involved in traditional political activities, including: 1) wearing a campaign button, 2) putting a bumper sticker on their car, 3) going door-to-door for a candidate, 4) organizing people in the neighborhood, and 5) marching in Washington. Again, youths who say they intend to do any one of these activities are also those likely to say they'll do others (Cronbach's coefficient alpha is 0.79). The final behavioral factor is comprised of answers to questions about aspirations to hold elected office. Youths were asked how likely they felt it is that they would someday run for city council, state assembly, governor, U.S. Congress, or U.S. president. Cronbach's coefficient alpha for this factor is 0.92.
Results
Political Attitudes
White adolescents tend to be more trusting than minorities of people with whom they have a close relationship -- their parents, teachers, principals, friends [F(2,2382)=30.21, p<.001]. White adolescents are also more trusting than minorities of political and economic leaders [F(2,2381)=23.61, p<.001]. Finally, whites tend to be more satisfied with the American political system than either blacks or Latinos, while Latinos are more content than blacks [F(2, 2385)=45.52, P<.001]. Thus, disregarding other factors, the political attitudes of white youths tend to be more respectful of the existing system then either that of black or Latinos, and among minorities, the views of blacks lean more toward cynicism and distrust than the views of Latinos. The means are listing on Table 2.
[Insert Table 2 about here]
Female youths are more trusting than boys both of personally significant adults [F(1,2382)=9.81, p<.01] and of leaders [F(1,2381)=20.17, p<.001], but there is no gender difference in attitudes toward the political system. Nonurban youth are more trusting of their parents, teachers, principals, and friends than are urban youth [F(1,2382)=4.38, p<.04]; they are more trusting than urban adolescents of leaders [F(1,2381)=5.36, p<.05]; and they are less dissatisfied with the political system [F(1,2385)=6.96, p<.01]. These main effects of urbanicity appear even taking into consideration that the sample has been adjusted to give equal weight to whites, blacks, and Latinos, thus the difference cannot be explained by any real difference in racial composition between urban and nonurban samples.
Female adolescents seem to differ from males in their political attitudes, whites from nonwhites, and urban youth from nonurban youth. But the relationship between gender, race/ethnicity, urbanicity and political attitudes is also more complex. Although girls, regardless of place of residence, are more trusting than boys of political leaders, the difference is far greater for nonurban youths than for urban youths [F(1,2382)=7.57, p<.01]. Stated differently, urban and nonurban boys do not differ in their trust for leaders, but nonurban girls are significantly more trusting than urban girls. [See Figure 1.]
[Insert Figure 1 about here]
Main effect results suggest that girls are more trusting of significant adults than boys. In fact, this is true only of white and black girls. Latina girls are less trusting than Latino boys. [F(2,2382)=10.42, p<.001; see Figure 2a.] Furthermore, while Latino boys are more trusting than black boys, black girls are more trusting than Latina girls. White boys and girls are more trusting than blacks and Latinos of both genders. Place of residence further complicates the gender-race relationship among Latinos [F(2,2382)=9.42, p<.001]. There is little difference in trust between urban Latino boys and girls. On the other hand, nonurban Latina girls are far less trusting than nonurban Latino boys. Nonurban Latino boys are the most trusting of all Latino groups; nonurban Latina girls are the least trusting. [See Figure 2b.]
[Insert Figures 2a, 2b]
Black youths tend to be more dissatisfied with the current political institutions than either white or Latino youths. However, while nonurban Latinos express greater dissatisfaction than nonurban whites, among urban youths, there is no difference between whites and Latinos [F(2, 2385)=6.31, p<.01]. Place of residence doesn't appear to have much influence on either blacks or Latinos, but urban whites are more dissatisfied than nonurban whites (see Figure 3a). Adding gender makes little difference in the case of whites and Latinos, but shows the complexity of black suspicion. Urban black boys are slightly less dissatisfied than nonurban black boys, but urban black girls -- the most dissatisfied of all groups -- express much greater dislike for the system than nonurban black girls [F(2,2385)=7.19, p<.001, see Figure 3b].
[Insert Figures 3a, 3b]
Behavioral Intentions
Whites more likely to say they will vote than blacks, and blacks are more likely than Latinos to say they intent to vote when they come of age [F(2,2384)=51.42, p<.001]. Girls have greater intentions of voting than boys [F(1,2384)=41.82, p<.001], and girls more often than boys offer that they will participate in other political activities as well [F(1,2384)=108.18, p<.001]. Urban youth believe more often than nonurban youth that they will be involved in political activities -- other than voting -- when they become adults [F(1,2384)=7.65, p<.01], and urban adolescents have a stronger desire than nonurban youth to run for public office [F(1,2381)=28.62, p<.001]. The mean scores are presented in Table 2.
Oddly, nonurban girls are more likely than urban girls to say they will vote, but the opposite is the case for boys [F(1,2384)=6.71, p<.01, see Figure 4b]. The same is true of participation in other election activities: nonurban girls have a greater intention to participate than urban girls, but urban boys say they are more likely to act than nonurban boys [F(1,2384)=37.9, p<001, see Figure 5b]. Furthermore, in both cases, the differences between girls and boys is greater among the nonurban population -- nonurban girls are much more likely to act than nonurban boys -- than in the urban population. Although Latinos believe they are less likely to vote than either whites or blacks, urban Latinos suspect they will vote more often than nonurban Latinos believe they will vote [F(2,2384)=3.76, p<.05, see Figure 4a].
[Insert Figures 4a, 4b about here]
Gender also interacts with race/ethnicity. Girls are more likely than boys to say they will participate in election activities. But while there is little difference in the likelihood of participating in election activities between white, black, and Latino boys, Latina girls are more likely to say they will act than black and white girls. Black girls, in turn, are more likely to say they will engage in politics than white girls [F(2,2384)=4.51, p<.05, see Figure 5a]. The picture changes somewhat, though, when place of residence is considered. In fact, the race difference between girls seems to only hold true among nonurban girls [F(2,2384)=7.92, p<.001]. There is little difference in the participation expectations of white, black, and Latino urban girls. Nonurban Latino boys are slightly less likely to believe they will act than other boys, while urban Latino boys are more likely than others to think they will be engaged. In neither group, though, are the differences great (see Figure 5c). The means scores suggest a fairly unified groups of urban adolescents but a more racially and gender diverse group of nonurban youths.
[Insert Figures 5a, 5b, 5c about here]
Boys are more likely to say they intend to run for office than girls, with the exception of Latinos [F(2,2381)=12.17, p<.001]. Latina girls have greater intentions of running for office than either white and black girls or Latino boys. Only black boys have a greater desire to serve in office than Latina girls (see Figure 6a). Breaking the effects down further by place of residence, however, shows that urban black girls are not substantially different from Latina girls and black boys. Urban black girls are much different than nonurban black girls, though: nonurban black girls are the least likely of all groups to show an interest in running for office [F(2,2381)=10.06, p<.001, see Figure 6b].
[Insert Figures 6a, 6b about here]
Summary of Findings
There are several general trends that highlight the relationship between race/ethnicity, gender, place of residence and the political attitudes and behavioral intentions of adolescents. First, and not unexpectedly, white adolescents tend to be more trusting in general and to hold more positive views about the current political system than minority adolescents. White adolescents are also more likely to say they will vote, although race does not have a singular relationship to other forms of political participation. Second, girls lean toward greater trust than boys, and they believe they will be involved in political activity more regularly than boys believe they will be. Third, the influence of place of residence is less straightforward than either race/ethnicity or gender. While nonurban youth tend to be more trusting and more supportive of the existing political institutions than urban youth, urban youth hold greater expectations of political activity than nonurban youth (with the exception of voting). Table 3 summarizes the F values and significance levels of the ANOVA results.
[Insert Table 3 about here]
These main effects are similar, for the most part, to what we would expect given adult attitudes and behaviors, and for white adolescents, the trends change little when the demographic characteristics are examined simultaneously. That is, white girls hold more positive views and are more likely to say they'll participate as citizens involved than white boys, regardless of place of residence, and urban white youth are less trusting, less satisfied, and more likely to get involved than nonurban white youth, regardless of gender.
Gender and place of residence, however, appear to have a less consistent influence on black and Latino adolescents. In general, girls are more trusting of adults than boys, but among Latinos, this holds true only for urban girls and boys. Nonurban Latina girls are less trusting than nonurban Latino boys. Urban youth are more likely to say they will participate in politics than nonurban youth, except Latina girls. Nonurban Latina girls say they will participate more than urban Latina girls, although nonurban Latina girls have less desire to run for office than urban Latina girls. Black girls, too, differ from black boys, but in ways unlike Latino boys and girls. Urban black girls are significantly less trusting than nonurban black girls -- the opposite of Latina girls -- and urban black boys are slightly less trusting than nonurban black boys -- the opposite of Latino boys. The same pattern emerges with regard to satisfaction with current political institutions: urban black girls are more negative than nonurban black girls while urban black boys are less negative than nonurban black boys. Finally, urban and nonurban black boys express a similar likelihood of running for office, but urban black girls are much more likely to intend to run for office than nonurban black girls. Urban black girls are quite similar to urban black boys in their desire to run, nonurban black girls are far less likely than nonurban black boys to plan for a political career.
Discussion
How can we make sense of these complex, sometimes contradictory findings? One possible way might be to consider the racial/ethnic, as well as economic, culture in which the adolescents live. The urban areas are much more racially/ethnically diverse than the nonurban areas. White students represent a minority 36 percent of the students in the surveyed urban public schools but 95 percent of the students in the surveyed nonurban schools. The urban cities are likewise more diverse than the nonurban areas. (Refer back to Table 1 for the exact demographic breakdown.) Black and Latino students in the nonurban schools are truly in the minority, while in the urban schools, minority students equal white students in their numbers. The results of this survey show little difference between Latino boys and girls in the urban setting. Urban Latino girls and boys share similar levels of mistrust and dissatisfaction and are generally equally likely to say they will be involved in politics at the various levels questioned. Where there are noticeable gender differences among Latinos, however, is in the nonurban areas -- areas dominated by non-Hispanic whites. In the nonurban areas, Latina girls are less trusting than nonurban Latino boys, more cynical than Latino boys, but also more likely to say they will get involved in politics both at the grassroots level and at the electoral level. Nonurban Latina girls tend to have a more negative view of politics than nonurban Latino boys, and they react to this by becoming politically energized.
Although Hardy-Fanta (1993) found Latina adult women in Boston to be quite politically active, noting that Latina women have been more likely to run for office in Boston than white women and just as likely as Latino men, she also observed that the urban Latinas in her study tended to take a more grassroots approach to politics. The women sought to build connections between community and leadership. Hardy-Fanta speculates that the differences between men and women may be, to some extent, reflective of both sexism and cultural conditioning. Hispanic cultures have a tradition supporting men in public leadership positions but not women. Women's power remains hidden -- respected but not publicly acknowledged. Women are the unseen strength behind the family, morality, and religious tradition. It is possible that the nonurban Latina girls, living among Anglos, feel freer from sexist cultural constraints and, thus, are driven toward traditionally male positions, such as elected officials.
The urban-nonurban dimension appears to have a different influence on blacks than on Latinos. Urban black girls are similar to urban black boys in their level of trust and in their desire to run for elected office, and though urban black girls are more cynical about the nature of political institutions than urban black boys, they are also more likely to say they will get involved. In other words, the pattern between urban black boys and girls is similar as that between nonurban Latino boys and girls. Nonurban black girls, however, haven't the same spark toward involvement nor the same striking dissatisfaction than either urban black or nonurban Latino girls hold. Thus, the nonurban environment -- that is, living largely among whites -- seems to have a more discouraging influence on black girls than on black boys. Brake (1985) noted in his study of youth cultures that the ghetto "offers a supportive culture which makes a dent in hegemony" and that black popular culture creates for youth "a 'space' which enables them to resist" (p. 126). At the same time, other researchers have noted reduced public -- but not personal -- self-esteem among black youths (Rosenberg 1981, Turner and Turner 1982; Dukes and Martinez 1994), suggesting that racism has an affect on public efficacy but not personal efficacy. The results of this survey seem to confirm both these previous findings, but more so perhaps for black girls than for black boys.
Minority women tend to be better represented among minority politicians than white women are among white politicians (Barrett in press; Hardy-Fanta 1993). Whether they live largely among other whites, as in nonurban areas, or live alongside other minorities in urban areas, the white female adolescents in the survey were generally the least likely group to believe they will ever run for public office. Neither were they among those most likely to say they will be involved in other political activities, with the exception of voting. Thus, white girls tend to hold the most positive views of leaders and politics, and do not feel strongly charged to become involved. Despite the dissatisfaction and distrust expressed by blacks and Latinos, it is perhaps the white girls that present the most discouraging picture. Without the anger or frustration felt by minorities, and without the political expectation taught white boys, white girls may continue to remain relative outsiders in the political system. They will be good citizens -- voting regularly -- but many lack the drive to seek power from the inside.
Robert Dahl (1956) wrote that "a central guiding thread of American constitutional development has been the evolution of a political system in which all the active and legitimate groups in the population can make themselves heard at some crucial stage in the process of the decision" (p.137). Women, blacks, and other minorities within the U.S. have remained relative outsiders in the political realm. The results from this study suggest that in the coming years the color and gender of politics is headed for change.
References
Abramson, Paul R. 1977. The Political Socialization of Black Americans: A Critical Evaluation of Research on Efficacy and Trust. New York: The Free Press.
Barrett, Edith J. In press. "Black women in state legislatures: the relationship of race and gender to the legislative experience." In Susan Carroll (ed.), Gender and Policymaking: Studies of Women in Office. Bloomington, IN: Indiana University Press.
Bobo, Lawrence, and Franklin D. Gilliam, Jr. 1990. "Race, sociopolitical participation, and black empowerment." American Political Science Review 84: 377-393.
Brake, Michael. 1985. Comparative Youth Culture: The Sociology of Youth Cultures and Youth Subcultures in America, Britain, and Canada. New York: Routledge & Kegan Paul.
Dahl, Robert A. 1956. A Preface to Democratic Theory. Chicago: University of Chicago Press.
Dukes, Richard L. and Ruben Martinez. 1994. "The impact of ethgender on self-esteem among adolescents." Adolescence 29: 105-115.
Easton, David and Jack Dennis. 1969. Children in the Political System: Origins of Political Legitimacy. New York: McGraw-Hill.
Garcia, F. Chris. 1973. Political Socialization of Chicano Children: A Comparative Study with Anglos in California Schools. New York: Praeger
Greeley, Andrew M. 1975. "A model for ethnic political socialization." American Journal of Political Science 19: 187-206.
Gurin, Patricia, Shirley Hatchett, and James S. Jackson. 1989. Hope and Independence: Blacks' Response to Electoral and Party Politics. New York: Russell Sage.
Hardy-Fanta, Carol. 1993. Latina Politics, Latino Politics: Gender, Culture, and Political Participation in Boston. Philadelphia: Temple University Press.
Jankowski, Martin Sanchez. 1986. City Bound: Urban Life and Political Attitudes Among Chicano Youth. Albuquerque: University of New Mexico Press
Jennings, M. Kent and Richard G. Niemi. 1974. The Political Character of Adolescence: The Influence of Families and Schools. Princeton, NJ: Princeton University Press.
Page, Benjamin I. and Shapiro, Robert Y. 1992. The Rational Public: Fifty Years of Trends in American's Policy Preferences. Chicago: University of Chicago Press.
Rosenberg, M. 1981. "The self-concept: social product and social force." In M. Rosenberg and R. H. Turner (eds.), Social Psychology: Sociological Perspectives. (pp. 593-624). NY: Basic.
Rosenberg, Morris, and Roberta G. Simmons. 1971. Black and White Self-Esteem: The Urban School Child. Washington, D.C.: American Sociological Association.
Sapiro, Virginia. 1983. The Political Integration of Women: Roles, Socialization, and Politics. Urbana, IL: University of Illinois Press.
Sears, D. O. 1975. Political socialization. In F.I. Greenstein & N.W. Polsby (Eds.), Handbook of political science: Vol. 2 (pp.93-153). Reading, MA: Addison Wesley.
Sears, D. O. 1983. The persistence of early political predispositions: the roles of attitude object and life stage. In L. Wheeler and P. Shaver (Eds.), Review of Personality and Social Psychology: Vol. 4 (pp. 79-116). Beverly Hills, CA: Sage.
Smith, E. S. 1981. "The career development of young black females." Youth & Society 12: 277-312.
Tate, Katherine. 1993. From Protest to Politics: The New Black Voters in American Elections. Cambridge, MA: Harvard University Press.
Turner, C.B. and B. T. Turner. 1982. "Gender, race, social class, and self evaluations among college students." The Sociological Quarterly 23: 491-507.
Uhlaner, Carole J., Bruce E. Cain, and D. Roderick Kiewiet. 1989. "Political participation of ethnic minorities in the 1980s." Political Behavior 11: 195-231.
Welch, Susan and Lee Sigelman. 1993. "The politics of Hispanic Americans: insights from national surveys, 1980-1988." Social Science Quarterly 74: 76-94.
Welch, Susan and Lee Sigelman. 1992. "The gender gap among Hispanic?: a comparison with blacks and Anglos." Western Political Quarterly 45: 181-199.
Welch, Susan and Lee Sigelman. 1989. "A black gender gap?" Social Science Quarterly 70: 120-33.
Appendix A
Weighting Protocol to Equalize Cell Size in Urban and Nonurban Samples
Urban Nonurban
N Weight N Weight
White Males 138 1.45 566 0.35
White Females 192 1.04 697 0.29
Black Males 149 1.34 33 6.06
Black Females 157 1.27 23 8.70
Latino Males 127 1.57 13 15.38
Latino Females 156 1.28 10 20.00
Table 1
Mean Sample, School, and Town Demographics, by Place of Residencea
______________________________________________________________________________
Percent Percent Percent Percent
Anglo Black Latinob Povertyc
______________________________________________________________________________
Nonurban
Sample 90.0 ( 3.0) 4.1 ( 2.0) 1.6 ( 1.2) 10.6 ( 3.1)
School 95.1 ( 4.9) 2.4 ( 4.2) 1.1 ( 1.0) 10.7 ( 8.2)
Town 96.2 ( 4.5) 2.1 ( 3.1) 1.8 ( 1.0) 5.5 ( 2.1)
Urban
Sample 34.4 ( 4.7) 30.4 ( 4.6) 24.0 ( 4.3) 35.5 ( 4.8)
School 35.6 (17.4) 22.8 (11.4) 31.5 (12.1) 49.6 (25.7)
Town 54.6 ( 4.8) 20.9 ( 5.0) 26.0 ( 4.2) 21.3 ( 0.2)
______________________________________________________________________________
a Unweighted samples, standard deviations in parentheses.
b Latino is not an exclusive category. An individual may identify as Latino as well as with a race.
c Poverty is defined differently in at the different levels. For the sample, poverty defined as self admission of self or immediate family member receiving AFDC benefits. For school: percent of students qualifying for free or reduced lunch program. For town, percent of population falling at or below official poverty line, as calculated by U.S. Bureau of the Census.
Table 2
Mean Attitude and Behavioral Intention Scores,
Broken Down by Race/Ethnicity, Gender, and Urbanicitya
______________________________________________________________________________
Attitudes _ Behaviors_________
Trust in Trust in Satisfaction Vote Campaign Run for
Adults Leaders with system Activities Office__
Race/Ethnic
White 6.29*** 4.34*** 2.88*** 2.82*** 1.16 0.87
Black 5.85 3.86 3.27 2.41 1.19 0.97
Latino 5.89 3.99 3.05 2.28 1.21 0.96
Gender
Male 5.93** 3.93*** 3.05 2.36*** 1.00*** 0.96
Female 6.09 4.20 3.08 2.65 1.36 0.90
Urbanicity
Nonurban 6.06* 4.13* 3.02** 2.52 1.14** 0.83***
Urban 5.96 3.99 3.11 2.49 1.23 1.03
______________________________________________________________________________
a Weighted sample.
* p<.05; **p<.01; ***p<.001
Table 3
Summary of F values from ANOVA Testsa
_____________________________________________________________________________________________
Attitudes ____ Behaviors___________
Trust in Trust in Satisfaction Vote Campaign Run for
Adults Leaders with system Activities Office__
Main Effects
Race/Ethnicity 30.21*** 23.61*** 45.52*** 51.42*** 0.62 2.71
Gender 9.81** 20.17** 0.60 41.82*** 108.18*** 2.47
Urban 4.38* 2.36* 6.96** 0.40 7.65** 28.62***
2-Way Interactions
Race x Gender 10.42*** 0.58*** 0.51 2.00 4.51* 12.17***
Race x Urban 2.88 2.57 6.31** 3.76* 0.27 1.19
Gender x Urban 0.03 7.57** 1.12 6.71** 37.90*** 1.14
3-Way Interactions
Race x Gender x Urban 9.42*** 2.27 7.19*** 0.91 7.92*** 10.06***
Total degrees of freedom 2393 2392 2396 2395 2395 2392
______________________________________________________________________________________________
a Weighted sample.
* p<.05; **p<.01; ***p<.001
Figure 1
Mean Trust in Leaders Scores, by Gender and Place of Residence
girls
boys
Nonurban Urban
Figure 2a
Mean Trust in Significant Adults Scores, by Gender and Race
whites
blacks
Latinos
Boys Girls
Figure 2b
Mean Trust in Significant Adults Scores, by Gender, Race, and Place of Residence
white girls
white boys
Latina girls
Latino boys
black girls
black boys
Nonurban Urban
Figure 3a
Mean Dissatisfaction in Political Institutions Scores, by Race and Place of Residence
black
white/Latino
Nonurban Urban
Figure 3b
Mean Dissatisfaction Political Institutions Scores, by Gender, Race, and Place of Residence
black girls
black boys
Latina girls/
white boys
latino boys
white girls
nonurban urban
Figure 4a
Voting Intentions Scores, by Race and Place of Residence
white
black
Latino
Nonurban Urban
Figure 4b
Mean Voting Intention Scores, by Gender and Place of Residence
girls
boys
Nonurban Urban
Figure 5a
Mean Campaign Participation Scores, by Gender and Race
Latino
black
white
Boys Girls
Figure 5b
Mean Campaign Participation Scores, by Gender and Place of Residence
girls
boys
Nonurban Urban
Figure 5c
Mean Campaign Participation Scores, by Gender, Race, and Place of Residence
black girls
lat/white girls
latino boys
black boys
white boys
nonurban urban
Figure 6a
Mean Likelihood of Running for Office Scores, by Gender and Race
Latino
white/black
Boys Girls
Figure 6b
Mean Likelihood of Running for Office Scores, by Gender, Race, and Place of Residence
boys
Latina girls
black girls/boys
latino boys
white boys
white girls
girls
nonurban urban