“What About the Children?” The Psychological and Social Well-Being of Multiracial Adolescents

Abstract

We used the National Longitudinal Study of Adolescent Health (Add Health) to examine the social and psychological well-being of multiracial adolescents. Using two different measures of multiracial identity, we investigated the ways in which these adolescents compare to their monoracial counterparts on five outcomes: depression, seriously considering suicide, feeling socially accepted, feeling close to others at school, and participating in extracurricular activities. We found that multiracial adolescents as a group experience some negative outcomes compared to white adolescents, but that this finding is driven by negative outcomes for those with American Indian and white heritage. We found no consistent evidence, however, that multiracial adolescents as a group face more difficulty in adolescence than members of other racial and ethnic minority groups. The results were similar, whether the multiracial population is defined by self-identification or by their parents' racial identifications.

ACKNOWLEDGMENTS

We wish to thank Peter Kivisto, Lincoln Quillian, Kerry Ann Rockquemore, Gary Sandefur, the members of the Social Psychology Brownbag at the University of Wisconsin–Madison, and three anonymous reviewers for their helpful comments. This research uses data from the Add Health project, a program project designed by J. Richard Udry (PI) and Peter Bearman, and funded by grant P01-HD31921 from the National Institute of Child Health and Human Development to the Carolina Population Center, University of North Carolina, Chapel Hill, NC, with cooperative funding from 17 other agencies. Persons interested in obtaining data files from the National Longitudinal Study of Adolescent Health should contact Add Health, Carolina Population Center, 123 West Franklin Street, Chapel Hill, NC 27516-2524 (http://www.cpc.unc.edu/addhealth).

NOTES

Notes

¹ Of course, Drake and Cayton's research was done almost 20 years after Park's and was restricted to Chicago. Therefore, it may be the case that the empirical reality of biracial experiences had changed.

² Specific self-concept was measured in the domains of social acceptance, physical attractiveness, and romantic appeal.

³ The exact question was: “What is your race? You may give more than one answer.” We use the home rather than the school survey (which has a larger sample) because some of the variables we use are only available for adolescents and their parents who are interviewed at home.

⁴ For all of the descriptive statistics and models, we exclude those who identify themselves as Hispanic, since Add Health's use of two separate questions to assess Hispanic and racial identity make it unclear whether or not individuals who choose a Hispanic and a single racial identity are thinking of themselves as mixed race. Responses on the Hispanic and racial identity questions sometimes reflect mixed family heritage, but can also reflect national origins or simply pressure to answer both questions (Rodriguez 2000). As described in Harris and Sim (2002), issues of Hispanic multiracial identities can only be adequately addressed when a combined race and Hispanic ethnicity question is included in the survey.

⁵ For these models, we can include only students who live with both biological parents, since Add Health collected information on the racial identities of both parents only when they were both living with the student.

⁶ The CES-D is a 20-item scale with an alpha of >85. The scale we have created includes 18 of those items, as well as a question not asked on the CES-D (how often “you felt life was not worth living?”). The score for each question ranges from 0 to 3, with a high score indicating a high level of depression. These items were summed and logged (after 1 was added). For more information, see Radloff (1977).

⁷ Both school heterogeneity and family income have a significant number of missing cases. Missing data for both variables were imputed, with a dummy variable indicating missing data on the variable.

⁸ However, the distributions of single-race adolescents are different under the two different measurements. When defined by parental race, a greater proportion of the adolescents are identified as white, a smaller proportion as black, and the American Indian group became so small that we were forced to combine it with the “other race” group.

⁹ The means and proportions shown in Figures 2 to 6 were calculated using Stata's commands for survey data. They are therefore weighted estimates that control for the complex survey design of Add Health (but do not control for any other variables).

¹⁰ Since few significant differences in club participation were found, a comparable model with the log of the number of clubs participated in was also estimated (using both self- and parent-identification of race). Few significant racial differences were found for that outcome as well, and the relationships of the control variables to the outcome were similar.

¹¹ Add Health might not be the ideal data set with which to do this, however, since the measure of racial identity in wave 3 is different than the measure in wave 1.