Rhonda L. White is now at the University of South Carolina. Cheri L. Philip is now at the Medical University of South Carolina.
Patterns of Racial Socialization and Psychological Adjustment: Can Parental Communications About Race Reduce the Impact of Racial Discrimination?
Version of Record online: 12 AUG 2008
Copyright © 2008, Society for Research on Adolescence
Journal of Research on Adolescence
Volume 18, Issue 3, pages 477–515, September 2008
How to Cite
Neblett, E. W., White, R. L., Ford, K. R., Philip, C. L., Nguyên, H. X. and Sellers, R. M. (2008), Patterns of Racial Socialization and Psychological Adjustment: Can Parental Communications About Race Reduce the Impact of Racial Discrimination?. Journal of Research on Adolescence, 18: 477–515. doi: 10.1111/j.1532-7795.2008.00568.x
- Issue online: 12 AUG 2008
- Version of Record online: 12 AUG 2008
This study uses two waves of data to examine the relations among racial discrimination experiences, patterns of racial socialization practices, and psychological adjustment in a sample of 361 African American adolescents. Using latent class analyses, we identified four patterns of child-reported racial socialization experiences: Moderate Positive, High Positive, Low Frequency, and Moderate Negative. Experiencing racial discrimination was associated with higher levels of depressive symptoms, more perceived stress, and lower levels of well-being. On average, adolescents who experienced High Positive patterns of racial socialization reported the most positive psychological adjustment outcomes, while adolescents in the Low Frequency and Moderate Negative clusters reported the least favorable outcomes. Results suggest that High Positive racial socialization buffers the negative effects of racial discrimination on adolescents' perceived stress and problem behaviors. Together, the findings suggest that various patterns of racial socialization practices serve as risk, compensatory, and protective factors in African American adolescent psychological adjustment.
Adolescence is a period marked by significant physical, cognitive, social, and emotional changes that often result in both pain and growth. For African American adolescents, these significant changes occur within the context of a greater awareness of the role that their race plays in the way in which the broader society views them. Often this awareness comes as a result of experiencing racial discrimination (Tatum, 1997). Both observational and self-report data suggest that experiencing racial discrimination is not uncommon for a majority of African American adolescents (Fisher, Wallace, & Fenton, 2000; Gregory, 1995; Sellers, Copeland-Linder, Martin, & Lewis, 2006). A growing body of research indicates that these experiences with racial discrimination are associated with a number of deleterious psychological, educational, and social outcomes for African American adolescents (e.g., Clark, Coleman, & Novak, 2004; Neblett, Philip, Cogburn, & Sellers, 2006; Prelow, Danoff-Burg, Swenson, & Pulgiano, 2004; Sellers et al., 2006; Simons et al., 2002; Wong, Eccles, & Sameroff, 2003). As a result, experiencing racial discrimination constitutes a significant risk to the healthy development of African American youth (Cunningham, Swanson, Spencer, & Dupree, 2003; Spencer, Dupree, & Hartmann, 1997).
Although the prevalence of racial discrimination places many African American adolescents at risk for a variety of negative outcomes, African American adolescents are not without racial and cultural resources that serve as protective factors against the destructive influence of racial discrimination (Spencer et al., 1997). One set of racial protective factors are the attitudes and beliefs about the meaning of race and racism that African American adolescents often receive from their parents. The process by which these race-related messages are transmitted intergenerationally—referred to as racial socialization—has received increasing attention in the research literature as an important factor for understanding resilience in African American children (Brega & Coleman, 1999; Coard & Sellers, 2005; Fischer & Shaw, 1999; Hughes et al., 2006; Miller, 1999; Miller & MacIntosh, 1999; Stevenson, Reed, Bodison, & Bishop, 1997). Despite this increased attention, there are relatively few empirical investigations of the protective function of the parental racial socialization process in the context of African American adolescents' experiences with racial discrimination. The present study attempts to address this shortcoming with a longitudinal investigation of the associations among racial discrimination experiences, racial socialization practices, and psychological adjustment outcomes in a sample of African American adolescents.
A growing body of research on African American adults has consistently found a significant link between experiences with racial discrimination and psychological health outcomes (e.g., Broman, 1997; Jackson & Neighbors, 1996; Kessler, Mickelson, & Williams, 1999; Landrine & Klonoff, 1996; Murry, Brown, Brody, Cutrona, & Simons, 2001; Sellers, Caldwell, Schmeelk-Cone, & Zimmerman, 2003; Sellers & Shelton, 2003; Thompson, 1996). Although few studies have employed adolescent samples, recent evidence suggests that racial discrimination may be equally as damaging to the psychological health of African American adolescents. For example, Simons et al. (2002) found that racial discrimination was positively associated with childhood depressive symptoms in a sample of African American youth (aged 10–12 years). In another study, Prelow et al. (2004) found that African American adolescents' perceptions of discrimination were associated with increased delinquent behavior in the presence of economic stress and neighborhood disadvantage. Finally, in one of the few longitudinal studies of the effects of racial discrimination on African American adolescents, Wong et al. (2003) found that reports of discrimination at school were negatively associated with a variety of academic, social, psychological, and behavioral outcomes.
In understanding the impact of racial discrimination on the psychological adjustment of African American adolescents, some researchers have taken a risk and resilience approach (Sellers et al., 2006; Spencer et al., 1997; Wong et al., 2003). Such an approach recognizes the fact that, although there is increasing evidence that greater exposure to racial discrimination increases African American adolescents' risk for experiencing a variety of negative outcomes, not everyone who is exposed to the same level of racial discrimination will experience the same level of damage as a result of that exposure. In a risk and resilience framework, the presence of some racial and cultural factors may increase or decrease African American adolescents' vulnerability to racial discrimination. Within the risk and resilience literature, several models of resilience have been proposed (Garmezy, 1991; Zimmerman & Arunkumar, 1994). Two models that have been used in previous research to understand the impact of racial discrimination in African American adolescents are the compensatory model and the protective factor model (Sellers et al., 2006). The compensatory model focuses on resilience factors that are associated with more positive outcomes across all levels of risk (Zimmerman, Bingenheimer, & Notaro, 2002). Thus, the presence of the resilience factor positively compensates for the negative impact of the exposure to the risk on the outcome. The protective model focuses on resilience factors that buffer the negative effects of the risk factor on the outcome (Zimmerman et al., 2002). The relationship between the risk factor and the negative outcome is weaker for individuals who possess higher levels of the resilience factor than those who possess lower levels of the resilience factor.
African American adolescents' attitudes regarding the significance and meaning of race appear to be an important resilience factor against the impact of experiencing racial discrimination (Sellers et al., 2003, 2006; Wong et al., 2003). In a recent study, Sellers et al. (2006) found evidence that different types of racial identity attitudes served different resilience functions in the context of racial discrimination. Specifically, holding positive attitudes toward African Americans (private regard attitudes) served a compensatory function with regard to adolescents' levels of depressive symptoms, perceived stress, and psychological well-being. At the same time, believing that other groups hold less positive attitudes toward African Americans (public regard attitudes) served a protective function and thus buffered the relationship between racial discrimination and the three psychological adjustment outcomes.
Adolescents' racial identity attitudes such as private and public regard are often a by-product of a racial socialization process (Demo & Hughes, 1990; Sanders Thompson, 1994; Stevenson, 1995). Hughes (2003) defines racial socialization as “the transmission of parents' world views about race and ethnicity to children by way of subtle, overt, deliberate and unintended mechanisms” (p. 15). The process of racial socialization has become increasingly recognized as a primary mechanism through which youth develop a positive racial identity and become prepared to successfully cope with societal discrimination (Barnes, 1980; Bowman & Howard, 1985; Boykin & Toms, 1985; Hughes, 2003; Phinney & Chavira, 1995; Spencer, 1983; Stevenson, 1994). Evidence suggests that most African American parents engage in some form of racial socialization as they attempt to raise their children (Caughy, O'Campo, Randolph, & Nickerson, 2002; Coard, Wallace, Stevenson, & Brotman, 2004; Hughes et al., 2006; Thornton, Chatters, Taylor, & Allen, 1990). This evidence also indicates that African American parents are very heterogeneous in the ways in which they socialize their children about the meaning of race.
Adolescence is a developmental period in which parents may be more likely to engage in racial socialization activities with their children (Hughes & Chen, 1997). During adolescence, parents may feel that, for the first time, their children have the cognitive and emotional skills necessary to deal with the complex and sometimes painful truths associated with race in American society. Because parents often engage in racial socialization in response to specific situations that happen to their children, some of the normative social activities associated with adolescence may also trigger greater racial socialization behavior (Hughes & Chen, 1997). Adolescents' increasing autonomy may make them more vulnerable to exposure to racial discrimination (Fisher et al., 2000; Hughes et al., 2006). For instance, as African American adolescents begin to spend more of their free time outside of the home and in public places (e.g., malls, restaurants), they are more likely to interact with individuals and institutions that may discriminate against them. Their physical maturation may make them appear more threatening and consequently more susceptible to some of the societal stereotypes typically reserved for African American adults. Other developmentally appropriate activities such as dating may also make race salient to African American adolescents in ways that are likely to elicit conversations with their parents about race (Tatum, 1997). Parents' attempts to help their child cope with these experiences with racial discrimination are likely to result in the transmission of messages about the significance and meaning of race.
Although racial socialization is a multifaceted process (Coard & Sellers, 2005), most of the research to date has focused on the content of the racial socialization messages. As such, a number of authors have developed specific typologies representing different messages that African American parents transmit to their children about the meaning of race (for review, see Hughes et al., 2006). Several themes emerge from these various typologies. One theme that has emerged is that many African American parents convey messages that promote pride in the history and culture of their race (racial pride). Another theme focuses on teaching the child that barriers such as racial discrimination exist to impede the success and well-being of African Americans (racial barriers). A third theme centers on conveying the message that all people are equal regardless of their race (egalitarian). A fourth theme focuses on promoting feelings of individual worth within the broader context of the child's race (self-worth). Whereas racial pride messages focus on promoting pride in the race, self-worth messages attempt to build up the child's personal self-esteem. A set of messages that have received relatively less attention are messages that disparage or emphasize negative characteristics associated with African Americans (negative messages). These negative messages often reinforce societal stereotypes about African Americans. In addition to providing verbal messages, parents also often engage in racial-related activities and behaviors (e.g. buying African American literature and art) that instill racial socialization messages to their children (racial socialization behaviors). These nonverbal messages often work in concert with parents' verbal messages to communicate to the child a set of beliefs regarding what it means to be African American in this society (Bowman & Howard, 1985; Caughy et al., 2002; Coard & Sellers, 2005; Marshall, 1995; Thornton et al., 1990).
Much of the existing research has reported positive relationships between receiving various youth-reported racial socialization messages and a variety of child outcomes related to psychological adjustment (e.g., Bowman & Howard, 1985; Caughy et al., 2002; Fischer & Shaw, 1999; Knight, Cota, & Bernal, 1993; Marshall, 1995; Neblett et al., 2006; Ou & McAdoo, 1993; Stevenson, Herrero-Taylor, Cameron, & Davis, 2002). In a cross-sectional study of 377 Black adolescents and young adults (aged 14–24), Bowman and Howard (1985) found that self-development messages were associated with a greater sense of personal efficacy, while racial barrier messages were associated with higher school grades, even when the effect of personal efficacy was controlled. Similarly, Neblett et al. (2006) found that youth-reported self-worth and egalitarian messages, as well as socialization behaviors, were related to greater academic curiosity and persistence and higher grade-point-averages in a sample of African American middle and high-school students. In another study of African American adolescents, Constantine and Blackmon (2002) found evidence that racial socialization themes associated with racial pride, racial barrier messages, and egalitarian messages were associated with a variety of different domains of self-esteem. Stevenson, Herrero-Taylor et al. (2002) reported that the combination of messages that promote both racial pride and highlight the existence of racial barriers was associated with less fighting behavior for adolescent African American boys.
Fischer and Shaw (1999) conducted one of the few studies to investigate the potential of the racial socialization process as a resilience factor in the context of experiences with racism and psychological adjustment. In particular, they found that receiving more racial barrier messages attenuated the link between reports of racist events and poorer mental health for African American young adults such that individuals who reported receiving more racial barriers messages showed a significantly weaker relation between experiencing racist events and poorer mental health than those who reported receiving fewer racial barrier messages (Fischer & Shaw, 1999). In doing so, the study provides evidence that certain racial socialization messages may act as a protective factor against the influence of racial discrimination.
Two major shortcomings within the research literature limit our understanding of racial socialization as a resilience factor in the context of African American adolescents' experiences with racial discrimination and their psychological adjustment. The first shortcoming is an emphasis on the independent content of specific socialization messages, with little or no regard for the synergetic nature of the racial socialization process as a whole (Coard & Sellers, 2005; Hughes et al., 2006). Most of the existing studies have utilized analytic approaches that focus on the relationship between specific racial socialization messages (e.g., racial pride) and specific outcomes after controlling for the influence of other socialization messages. Such an approach, however, does not adequately represent the true synergetic nature of the process. Parents rarely give a single type of racial socialization message. Parents usually instill a variety of different messages. It is the confluence of different messages about the meaning of race that combine to provide children with a view of the racial philosophy that they believe that their parents are attempting to instill in them. For instance, we expect that the impact of providing a child with a number of racial barrier messages to differ as a function of what other racial socialization messages are also being provided. The meaning of receiving racial barrier messages is likely to be different for a child who also receives a number of racial pride messages as compared with a child who only receives negative messages about being African American with those racial barrier messages. The former child is likely to have a racial worldview that is both informed of danger but also equipped with feelings of possibilities. The latter child may be more likely to have a racial worldview that emphasizes a sense of inferiority as a result of being African American. While we are aware of one study to use a cluster/profile approach to examine racial socialization beliefs (Stevenson, 1997), this study focused on children's beliefs about what messages they think should be conveyed as opposed to the race socialization messages that the children actually received. To our knowledge, no studies to date have investigated the ways in which actual racial socialization messages and practices coalesce together to form meta-racial socialization messages. Thus, it is unclear how various racial socialization messages combine to form broader messages about race and how these broader messages act as resilience factors against the impact of racial discrimination.
A second limitation in the racial socialization research literature is an over-reliance on cross-sectional studies (Coard & Sellers, 2005; Hughes et al., 2006). As noted above, racial socialization is a bidirectional process in which messages are transmitted from parents to children, and children's experiences also influence the types of messages parents convey to their children. Thus, although it is possible that receiving specific racial socialization messages may result in a variety of racial discrimination and psychological adjustment outcomes, it is equally plausible that the occurrence of these outcomes may also lead parents to engage in specific racial socialization practices. For instance, providing more messages about racial barriers may lead a child to be more vigilant toward potential incidences of racial discrimination and it may also result in a child engaging in more delinquent behaviors. At the same time, greater experiences with racial discrimination and/or engaging in more fights may inspire parents to talk to their child about the racial barriers that their child may face. Unfortunately, cross-sectional studies are unable to adequately model this dynamism and thus we are limited in our ability to make any causal interpretations regarding the nature of the associations in our findings. Longitudinal investigations are needed in order to tease apart the temporal sequence among racial socialization and different outcomes (Coard & Sellers, 2005; Hughes et al., 2006).
THE PRESENT STUDY
The present study investigates the resilience function of racial socialization on four indicators of African American adolescents' psychological adjustment (depressive symptoms, perceived stress, psychological well-being, and delinquent behavior) in the context of experiences of racial discrimination. In doing so, the present study addresses the two aforementioned shortcomings in the research literature by utilizing cluster analytic techniques to capture the synergetic nature of racial socialization within a longitudinal investigation. There are four primary objectives to the study. First, we utilize cluster analytic procedures to identify different patterns of racial socialization across five different types of verbal messages as well as socialization behaviors. Such an approach will allow us to focus on the individual adolescent as opposed to the individual socialization variables as the unit of analysis. Because no other studies to date have utilized such a technique, we are unable to make any a priori predictions as to the number or nature of the different socialization patterns that our analyses will yield. Our second objective is to examine whether African American adolescents' experiences with racial discrimination are associated with more negative psychological adjustment outcomes even after we control for earlier experiences with racial discrimination and psychological adjustment. Consistent with other research on racial discrimination in African American adolescents (Wong et al., 2003), we predict that racial discrimination will be associated with poor psychological adjustment.
Our third objective is to examine whether patterns of racial socialization act as a compensatory resilience factor in adolescents' psychological adjustment. We hypothesize that there will be significant differences in psychological adjustment as a function of differences in the patterns of racial socialization that adolescents report receiving. While we are unable to make specific predictions, consistent with previous research, we do expect that those patterns of socialization that emphasize racial pride, racial barrier, and self-worth messages will be associated with more positive psychological adjustment (Bowman & Howard, 1985; Constantine & Blackmon, 2002; Stevenson, Herrero-Taylor et al., 2002). The final objective of the study is to investigate whether patterns of racial socialization play a protective role against the impact of racial discrimination on adolescents' psychological adjustment. We expect to find significant interactions between the patterns of racial socialization and racial discrimination experiences in predicting psychological adjustment. Consistent with the findings of Fischer and Shaw (1999), we predict that adolescents who report receiving patterns of racial socialization that emphasize racial barrier messages will be best insulated from the impact of experiencing racial discrimination.
The current study utilizes data from the first two waves (referred to as Time 1 and Time 2) of a longitudinal study of race-related beliefs and experiences and indicators of psychosocial adjustment in African American adolescents. Three hundred sixty-one self-identified African American adolescents in grades 7 through 11 were recruited from all 11 middle and high schools in a public school district in the Midwestern United States. The sample consisted of 146 male and 215 female adolescents who participated in the first two waves of the study. Participants were recruited from the six junior high schools and five high schools within the school district. Participants' ages ranged from 11 to 17 years with a mean age of 13.75 years (SD=1.21). Thirty-four percent of the sample was enrolled in the 7th grade, 31% were in 8th grade, 18% were in 9th grade, 16% were in the 10th grade, and 1% was in the 11th grade at Time 1. As part of the study, a primary caregiver was identified for each child. We used the primary caregiver's highest level of education completed (as reported by the child) as an indicator of educational attainment. Primary caregivers' educational attainment ranged from less than junior high to postgraduate study. The highest level of educational attainment was less than a high school diploma for a small minority of the primary caregivers (7.7%), 20% attained a high school diploma, 43.9% attended some college or received a college diploma, and 28.4% completed some graduate study or attained a graduate degree (i.e., master's, Ph.D., J.D., M.D.).
The city in which the study was conducted has a population of approximately 110,000 people. Although African Americans comprise roughly 9% of the population of the city, they comprise nearly 16% of the middle school and 12% of the high school students in the public school district. White students comprise the largest racial group within the student population with 52.4% of junior high school students and 54.8% of high school students. The racial composition of the student body, however, varied significantly across the 11 schools, with the percentage of African American students ranging from 7.2% to 64.9% of the student enrollment. The median household income for the city is above the national average at $46,299. Eighteen percent of the district's students were eligible for free or reduced lunch. Although the median family income for African Americans in the city surpasses the national average (parent-reported family median income $30,000–$39,000), there is significant variation across the distribution within the community, with parent-reported family incomes ranging from <$10,000 to >$130,000.
Initial contact information was obtained from the school district for students in grades 7 through 11 who were identified by the school system as being African American. Letters were sent to students' legal guardians requesting permission for students to participate in the study. Follow-up calls were made to parents to obtain informed consent and to check to make sure that the child identified as Black or African American. Once informed consent was obtained, an appointment was scheduled for a data collection session after school with the child. Participants who self-identified as Black or African American and agreed to participate in the study completed a survey consisting of measures of racial identity, racial socialization, discrimination, and several indices of psychological adjustment in small group administrations during the spring semester. Trained research assistants (primarily African American) administered the questionnaire. Students who were unable to participate in the study at school were able to complete the survey in a community setting. Students were informed that they would not be penalized for withdrawing from the study at any point, that they could skip any item to which they did not wish to respond, and that all responses were confidential. The questionnaire took approximately 60–90 minutes to complete. Participants received a $20 gift certificate to a local mall at Time 1. A similar procedure was used to recruit students to participate in the future waves of the study. Data were collected in small group sessions after school as well as at community settings (i.e., Public Library, mall, and community center). Participants were compensated with a $30 gift certificate for completing the second wave of the study.
A cross-sequential design was employed in which a new cohort of participants was recruited to participate in the study during each spring from 2002 through 2004. As a result, we collected 3 years of data from our first cohort, 2 years of data from our second cohort, and 1 year of data from our third cohort. The data were collected in the spring and summer across the 3 years of the study and spacing between data collections points was relatively stable across and within cohorts. The present study focuses on the first two cohorts and the first two waves of data. During the 3 years of the study, the households of 742 children were contacted to participate in the study. A total of 546 adolescents participated in one of the three cohorts for an overall participation rate of 74%. Of the 465 students in cohorts 1 and 2 who completed wave 1, 361 students also completed wave 2 for a retention rate of 78%. Of these 361 students, 258 (71%) were in cohort 1 and 103 (29%) were in cohort 2. There were no appreciable differences in the retention rates across the two cohorts (83%–76%). Adolescents who participated in both waves 1 and 2 did not differ in gender composition, age, or parent educational attainment from those who dropped out after wave 1.
Racial socialization messages. Participants' scores on the Racial Socialization Questionnaire-teen (RSQ-t) at Time 2 were used as the primary measure of racial socialization in the study (Lesane-Brown, Scottham, Nguyên, & Sellers, 2006). The RSQ-t consists of 26 items that were developed to measure five types of racial socialization messages that have been identified in the research literature. In addition, the scale includes a measure of the different types of racial socialization behaviors in which African American parents often engage. Participants were asked to respond to each item using a three-point Likert-type rating scale (0=never to 2=more than twice) indicating how often their parents had communicated each message or behavior to them. The scale consists of six subscales. The Racial Pride subscale (Time 2: α=.71) consists of four items measuring the extent to which primary caregivers emphasize Black unity, teachings about heritage, and instilling positive feelings toward Blacks (e.g., “Told you that you should be proud to be Black”). The Racial Barriers subscale (Time 2: α=.76) consists of four items measuring the extent to which an awareness of racial inequities and coping strategies is emphasized (e.g., “Told you that some people try to keep Black people from being successful”). The Egalitarian subscale (Time 2: α=.64) consists of four items measuring the extent to which messages regarding interracial equality and coexistence are emphasized (e.g., “Told you that Blacks and Whites should try to understand each other so they can get along”). The Self-Worth subscale (Time 2: α=.70) consists of four items measuring the extent to which messages emphasizing positive messages about the self are conveyed (e.g., “Told you that you are somebody special, no matter what anyone says”). The Negative subscale (Time 2: α=.70) consists of five items measuring the extent to which messages are conveyed regarding disparaging or negative messages about Blacks (e.g., “Told you that learning about Black history is not that important”). The Socialization Behaviors subscale (Time 2: α=.77) consists of five items measuring the frequency of various socialization activities or behaviors related to Black culture (e.g., “Bought you books about Black people”). Subscales were created by averaging across each of the items within the subscale such that higher scores indicated a greater frequency of the particular message, socialization activity, or behavior.
Lesane-Brown et al. (2006) utilized the first-wave of data from all three cohorts in the present study to investigate the factor structure of the scale with a confirmatory factor analysis. A confirmatory factor analysis was employed in which each of the subscales were purported to be separate oblique factors. The results indicate a satisfactory fit to the data for the entire sample as well as for a split-half replication using the different cohorts of the study. The subscales of the RSQ-t were also predictive of subsequent racial identity attitudes and intergroup relations in ways that are suggestive of the predictive validity of the measure with this sample.
Racial discrimination. Adolescents' experiences with discrimination during the past year were measured at Time 1 and Time 2 using The Racism and Life Experience Scales (RaLES). The RaLES (Time 1: α=.92, Time 2: α=.93) consists of 18 items measuring how often adolescents were discriminated against because of their race (Harrell, 1997). A sample item includes: “In the past year, how often have you been ignored, overlooked, or not given service (in a restaurant, store, etc.) because of your race?” Participants were asked to respond on a six-point scale regarding the frequency with which they experienced the event (0=never to 5=once a week or more). Higher scores correspond to higher levels of experiences with discrimination. The RaLES has proven to be a reliable and valid measure of perceived discrimination (Harrell, 1997). Both Time 1 and Time 2 measures were used in the present analyses.
Sociodemographic variables. Participants' self-reported age, gender, and the educational attainment of their primary caregiver were included in the study as sociodemographic variables. At Time 1, participants provided their age in years as well as their gender (male coded as 1 and female coded as 2). They also provided information regarding the highest level of education that their primary caregiver had obtained using an eight-point scale: 1=junior high or less; 2=some high school; 3=high school graduate; 4=some college; 5=college graduate; 6=some graduate school; 7=master's degree; and 8=doctoral degree.
Psychological Adjustment Outcome Measures
Depressive symptomatology. The Center for Epidemiological Studies—Depression Scale (CES-D) assesses the presence and frequency of symptoms associated with depression (Radloff, 1977) at Time 1 and Time 2. The 20-item inventory (Time 1: α=.76, Time 2: α=.87) allows participants to rate the frequency of specific feelings such as loneliness in the past week, ranging from 1=rarely or none of the time (less than 1 day) to 4=most or all of the time (5–7 days). Items were coded such that higher scores on the composite scale indicate higher occurrences of depressive symptomatology. Sample items include “I was bothered by things that do not usually bother me” and “I felt depressed.” Scores were averaged across the 20 items such that higher scores on the scale denote more depressive symptoms.
Perceived stress scale. Participants' levels of stress were measured at Time 1 and Time 2 using the Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983). The Perceived Stress Scale (Time 1: α=.63, Time 2: α=.71) consists of 14 items that ask how often in the last month the participants experienced symptoms of stress. Sample items include feeling nervous, feeling like difficulties were piling up, and being upset because of something that happened unexpectedly. Participants used a five-point response scale ranging from 0=never to 4=very often. Scores were averaged across the 14 items such that higher scores on the scale denote more stress.
Problem behaviors. Adolescents' engagement in behaviors associated with externalizing or acting out behavior problems were assessed at Time 1 and Time 2 with items adapted from work by Elliott, Huizinga, and Menard (1989). The scale (Time 1: α=.72; Time 2: α=.64) consists of seven items which allow participants to rate the frequency of participation in the specified activities during the last year, ranging from 0=never to 5=more than 20 times. The items asked about each of the following: (1) damaging property for fun, (2) skipping classes without a valid excuse, (3) getting into fights inside or (4) outside of school, (5) being sent to the principal's office, (6) getting into trouble with the police, and (7) cheating on tests. Scores were averaged across each of the seven items such that higher scores indicate greater involvement in problem behaviors.
Well-being. Adolescents' level of overall psychological well-being was measured at Time 1 and Time 2 using a shortened 24-item version of the Psychological Well-being Scale (Ryff, 1989). The Psychological Well-being scale (Time 1: α=.83; Time 2: α=.85) assesses individuals' level of well-being along six dimensions: self-acceptance, positive relations with others, autonomy, environmental mastery, purpose in life, and personal growth. Participants indicated their agreement with a series of items using a five-point Likert-type scale ranging from 1=strongly disagree to 5=strongly agree. Sample items include, “In general, I feel I am in charge of my life” and “In general I felt confident and positive about myself.” A single composite Well-being scale score was derived by averaging the scores from all 24 items. Higher scores on the composite scale indicate higher levels of psychological well-being.
Data were analyzed for adolescents with complete data for both waves of the study (N=361). We used Analysis of Variance to determine whether participants who completed only one wave of the study differed from those who completed both waves of the study on any of the discrimination, racial socialization, and psychological adjustment variables at Time 1. The analyses indicated no significant differences (p<.05) between the two groups on any of these study variables.
Preliminary Analyses: Racial Socialization Variables, Racial Discrimination Experiences, and Psychological Adjustment Outcomes
Preliminary analyses focused on descriptive statistics among racial socialization variables, experiences with racial discrimination, and psychological adjustment variables at Time 2. Participants reported receiving racial barrier messages (M=1.28; SD=.59), egalitarian messages (M=1.25; SD=.55), and behavioral messages (M=1.13; SD=.56) from their parents approximately once or twice. They reported receiving slightly more racial pride and self-worth messages (M=1.43; SD=.54; M=1.66; SD=.45, respectively). Adolescents in the sample reported receiving very few negative race messages (M=.19; SD=.34). Examination of the psychological adjustment variables indicate low levels of depressive symptoms (M=1.76; SD=.50), low levels of involvement in problem behaviors (M=1.50; SD=.50), and high levels of well-being (M=3.58; SD=.54). Reports of perceived stress were moderate (M=2.91; SD=.52).
In general, adolescents reported experiencing racial discrimination at least once in the past year (M=1.53; SD=.99). The modal number of discrimination encounters reported by participants was 18. Specifically, 54 participants indicated that they had experienced each of the 18 discrimination encounters at least once in the past year. Seventeen adolescents indicated that they had not encountered any discrimination experiences within this time frame. Data also indicate that some discrimination experiences were more common among participants than others. Participants reported being treated rudely or disrespectfully (74.7%), being mistaken for someone of their same race (74.7%), and being accused of something or treated suspiciously (72.2%) most frequently. On the other hand, the fewest number of participants reported being disciplined unfairly because of their race (44.7%), other people avoiding them (52.2%), and being laughed at, made fun of, or taunted (53.8%).
The associations among racial socialization, racial discrimination, and psychological adjustment at wave 2 were examined using zero-order correlations (Table 1). In general, adolescent reports on the six racial socialization subscales were positively related to each other. Racial pride messages were strongly related to racial barrier messages (r=.54; p<.01), egalitarian messages (r=.52; p<.01), self-worth messages (r=.53; p<.01) and racial socialization behaviors (r=.66; p<.01). Strong positive relationships were also present between egalitarian messages and self-worth messages (r=.53; p<.01). Receiving negative messages was inversely related to receiving racial pride (r=−.14; p<.01) and self-worth messages (r=−.26; p<.01). With the exception of racial pride and self-worth messages, racial discrimination was positively associated with each of the racial socialization subscales.
|1. Racial pride||—|
|2. Racial barriers||.54**||—|
|7. Racial discrimination||.04||.22**||.11*||−.01||.26**||.14**||—|
|8. Depressive symptoms||−.15**||.04||−.05||−.09||.25**||−.11*||.35**||—|
|10. Problem behaviors||−.09||.05||−.13*||−.16**||.18**||−.04||.24**||−.42**||−.51**||—|
There were also significant bivariate relationships between the racial socialization variables and the psychological adjustment outcomes. First, racial pride messages and racial socialization behaviors were associated with fewer depressive symptoms. Next, racial pride and self-worth messages were negatively associated with perceived stress. With respect to problem behaviors, individuals who experienced more egalitarian and self-worth messages were less likely to have engaged in problem behaviors. Bivariate correlations also revealed that egalitarian, racial pride, and self-worth messages, as well as racial socialization behaviors, were associated with greater well-being. Finally, a consistent pattern of relationships emerged for negative race messages such that adolescents who received more negative race messages reported more negative psychological adjustment outcomes.
Racial discrimination experiences were also significantly associated with psychological adjustment outcomes. Discrimination encounters were positively correlated with depressive symptoms, such that more experiences with racial discrimination were associated with higher levels of depressives symptoms. Similarly, discrimination was positively related to levels of perceived stress for adolescents in this sample. Experiencing more racial discrimination was related to more engagement in problem behaviors and lower levels of well-being. By and large, these results indicate that reports of more racial discrimination encounters are associated with decreased psychological adjustment.
Racial Socialization Profiles
Latent class analysis (LCA) implemented by the Latent Gold program (Vermunt & Magidson, 2005) was used to determine racial socialization clusters from the adolescent sample. LCA is a model-based cluster analysis that provides statistical criteria for selecting a plausible cluster solution among alternatives (Magidson & Vermunt, 2004). Selecting the most appropriate cluster solution is accomplished in several ways. First, model fits and comparisons are assessed using the likelihood ratio chi-squared statistic (L2). For the baseline (one-class) model, L2 can be considered as the total association among the indicators. When the baseline model is compared with an alternative model from the same data with a higher number of classes, the percent reduction in L2 reflects the total association explained. Typically, L2 is evaluated against the chi-squared distribution. However, when the number of indicators or the number of categories of these indicators is large, L2 is not well approximated and the alternative bootstrap p-value is recommended (Langeheine, Pannekoek, & Van de Pol, 1996). Next, the percent reduction in L2 is used in conjunction with the Bayesian information criterion (BIC), an index of model fit and parsimony. In general, a model with the largest association explained (i.e., the greatest reduction in L2) and the lowest BIC value is preferred. Finally, recent advances in LCA allow the specification of models that account for associations between indicators. The Latent Gold program provides a diagnostic statistic, the bivariate residual (BVR), to assess the bivariate relationships among indicators (Magidson & Vermunt, 2004). By allowing local dependence among indicators, more parsimonious models can be estimated from data when the association between indicators is not adequately explained solely by the latent classes.
Using the data from the six subscales of the RSQ-t measured at Time 2, six latent class models (ranging from 1 to 6 clusters) were estimated. Summary statistics for these six models are displayed in Table 2. Of the six models estimated, the four-cluster model appeared to be the most appropriate solution. It had the lowest BIC (4,816.8), a nonsignificant bootstrap p-value (.44), and a substantial reduction in L2 (32.6%) over the baseline. Although the five-cluster and six-cluster models showed a slightly further reduction in L2 (33.9% and 35.2%, respectively) than the four-cluster model, these models also had a larger BIC (4,837.8 and 4,857.6, respectively) suggesting that they are not as parsimonious as the four-cluster model. Next, the BVRs for each variable pair of the four-cluster model were examined for local dependence. The self-worth/egalitarian message pair had a substantially large BVR (6.75). Consequently, a four-cluster model with the direct effect between self-worth and egalitarian messages was then estimated. The direct effect accounted for the residual correlation between the two indicators and provided a more parsimonious model with a better fit. This modified four-cluster model had a smaller BIC (4,802.0), a larger reduction in L2 (34.0%), and acceptable BVRs. Thus, we adopted this model as our final cluster solution.
|Model||BIC(LL)||L2||df||Bootstrap p-Value||% Reduction in L2||Maximum BVR|
|With no direct effects|
|With direct effects|
|Four-class with direct effect between Egalitarian and self-worth||4,802.0||1,021.6||289||.47||34.0||.91|
Next, the raw and standardized means of each racial socialization variable were used to describe and label the clusters (see Table 3). The largest cluster was labeled Moderate Positive (n=164, 45% of sample). This cluster was characterized by scores near the sample mean on all six of the racial socialization subscales. With respect to the raw means, the Moderate Positive cluster has relative high scores on the racial pride and self-worth subscales, moderate scores on the racial barrier, egalitarian, and socialization behavior subscales, and low scores on the negative subscale. The second largest cluster was labeled High Positive (n=95, 26%). This cluster was characterized by high scores relative to the rest of the sample on all of the racial socialization subscales with the exception of the negative message subscale (which was near the sample mean). The third largest cluster (n=67, 19%), Low Frequency, had low scores on all of the subscales except self-worth (approximately 1 SD below the mean) and negative messages (approximately .3 SD below the mean). Finally, the smallest cluster (n=35, 10%) was labeled Moderate Negative. This cluster was characterized by high scores relative to the rest of the sample on the negative message subscale, scores near the mean on the racial barrier, egalitarian, and socialization behavior subscales, as well as low scores on the racial pride and self-worth scales. Figure 1 presents a graphic summary of the four racial socialization clusters using standardized means of the racial socialization variables.
|Racial Socialization Variable||Moderate Positive||High Positive||Low Frequency||Moderate Negative|
|Racial pride||1.53 (.28)||1.96 (.09)||.65 (.33)||.98 (.39)|
|Racial barriers||1.31 (.49)||1.71 (.40)||.6 (.49)||1.22 (.39)|
|Egalitarian||1.3 (.49)||1.6 (.37)||.63 (.49)||1.22 (.36)|
|Self-worth||1.79 (.29)||1.91 (.20)||1.29 (.49)||1.11 (.31)|
|Negative||.12 (.25)||.14 (.25)||.08 (.13)||.9 (.34)|
|Behaviors||1.08 (.44)||1.72 (.24)||.48 (.38)||1.07 (.34)|
Cluster Group Differences in Demographic and Racial Discrimination Variables
Analyses were conducted to assess whether cluster groups differed by age, gender, or primary caregiver's educational attainment. The results indicate no significant cluster differences in age (F[3, 357]=.72, ns) or gender (χ2 [3, N=361]=5.90, ns). Although a significant omnibus F was found for parent educational attainment (F[3, 357]=2.72, p<.05), post hoc analyses reveal that adolescents in the Moderate Negative cluster (M=4.20) had parents who were only marginally (p=.06) less educated than the parents of adolescents in the High Positive cluster (M=5.08). No other cluster differences were found.
An analysis of covariance (ANCOVA) was employed to assess cluster differences in adolescents' experiences of racial discrimination at Time 2. Covariates including gender, age, parent education, and discrimination experiences at Time 1 were included in the model. A significant main effect was found for cluster membership, (F[3, 357]=9.18, p<.01). Post hoc comparisons revealed that adolescents in the Moderate Negative cluster (M=2.21; SD=.70) reported experiencing significantly more racial discrimination at Time 2 than adolescents in the Moderate Positive (M=1.48; SD=.99), High Positive (M=1.60; SD=1.04), and Low Frequency (M=1.18; SD=.87) clusters. Post hoc tests also indicated that participants in the High Positive cluster reported significantly more encounters with racial discrimination at Time 2 compared with adolescents in the Low Frequency cluster. Gender, age, and parent educational attainment were unrelated to reports of discrimination across the four clusters; however, discrimination encounters at Time 1 were associated with racial discrimination at Time 2 (b=.37, p<.01).
Relations Among Discrimination Experiences, Racial Socialization, and Psychological Adjustment
To investigate the role of racial discrimination as a risk factor and patterns of racial socialization as a resilience factor in African American adolescents' psychological adjustment, a series of four general linear model (GLM) analyses of variance (ANOVA) was estimated with each of the four psychological adjustment outcomes as dependent variables (see Tables 4–7). Age, gender, parent's highest level of education completed, and discrimination experiences at Time 1 were included as covariates. The Time 1 measure of the respective psychological adjustment measure was also entered as a covariate in each model. Discrimination experiences at Time 2 and cluster group membership were included in the model as main effects. Finally, interaction terms were created between the discrimination experiences at Time 2 and the cluster group membership variables. All variables were mean-centered and the interaction term was the cross-product term of the centered discrimination variable at Time 2 and the cluster-group membership variable.
|Source||df||B(SE)||Type III Sum of Squares||Partial Eta Squared||F||p|
|Discrimination (T1)||1||−.02 (.02)||.09||.00||.51||.47|
|Depressive Symptoms (T1)||1||.38 (.05)||11.01||.16||65.67||.00|
|Discrimination (T2)||1||.13 (.03)||3.01||.05||17.98||.00|
|Cluster group (C)||3||—||2.26||.04||4.49||.00|
|C × Discrimination (T2)||3||—||.56||.01||1.12||.34|
|Source||df||B(SE)||Type III Sum of Squares||Partial Eta Squared||F||p|
|Discrimination (T1)||1||−.02 (.02)||.14||.00||.74||.39|
|Perceived stress (T1)||1||.47 (.05)||15.22||.19||83.32||.00|
|Discrimination (T2)||1||.16 (.04)||.88||.01||4.80||.03|
|Cluster group (C)||3||—||3.09||.05||5.63||.00|
|C × Discrimination (T2)||3||—||1.74||.03||3.17||.02|
|Source||df||B(SE)||Type III Sum of Squares||Partial Eta Squared||F||p|
|Discrimination (T1)||1||.02 (.02)||.17||.00||1.02||.31|
|Problem behavior (T1)||1||.48 (.04)||21.36||.27||125.21||.00|
|Discrimination (T2)||1||.00 (.04)||1.38||.02||8.10||.94|
|Cluster group (C)||3||—||.29||.02||.56||.64|
|C × Discrimination (T2)||3||—||1.13||.02||2.21||.09|
|Source||df||B(SE)||Type III Sum of Squares||Partial Eta Squared||F||p|
|Discrimination (T1)||1||.01 (.02)||.01||.00||.04||.84|
|Well-being (T1)||1||.41 (.05)||14.79||.18||76.24||.00|
|Discrimination (T2)||1||−.08 (.04)||1.22||.02||6.26||.03|
|Cluster group (C)||3||—||43.60||.06||7.49||.00|
|C × Discrimination (T2)||3||—||.64||.01||1.10||.35|
Consistent with Wong et al. (2003), an “instantaneous effect model” (see also Long, 1983) was employed, and, thus, we viewed the discrimination experiences at Time 2 variable as our measure of the risk of experiencing discrimination. The main effect term for the cluster group membership variables represented our measure of compensatory resilience of racial socialization. We interpreted a significant main effect for the cluster group membership variable as indicative of the fact that certain patterns of racial socialization may act to compensate for the risk of discrimination. We utilized findings from our post hoc analyses of these significant main effects to identify what patterns of racial socialization were most compensatory. Finally, we interpreted a significant effect for the interaction term as evidence that particular patterns of racial socialization serve a protective function in providing resilience in the face of experiencing racial discrimination. We also used post hoc analyses of the significant interactions (Aiken & West, 1991) to illuminate which patterns of racial socialization were the most protective.
Covariates of the Psychological Adjustment Outcomes
The first GLM ANOVA explained 32% of the variance in depressive symptoms (Table 4). There were two significant covariates in the model. The level of depressive symptoms at Time 1 was strongly related to the level of depressive symptoms at Time 2 (b=.38; p<.01). There was also a significant gender effect such that girls reported significantly higher levels of depressive symptoms at Time 2 (b=−.13; p<.01). With respect to perceived stress, the overall model explained 32% of the variance (Table 5). Three of the covariates were significantly related to perceived stress at Time 2. There was a strong relationship between perceived stress at Time 1 and perceived stress at Time 2 (b=.47; p<.01). Gender (b=−.17; p<.01) and parents' level of education (b=.03; p<.05) were related to perceived stress at Time 2 such that girls and adolescents whose parents completed more formal education reported higher levels of stress. The GLM explained 33% of the variance in problem behaviors at Time 2 (Table 6). The only significant covariate coefficient in the model was problem behavior at Time 1 (b=.48; p<.01). No other significant coefficients were found. Finally, the GLM explained 32% of the variance in well-being at Time 2 (Table 7). The only significant coefficient was the well-being variable at Time 1 (b=.41; p<.01).
Racial Discrimination as a Risk Factor for Psychological Adjustment Outcomes
Significant main effect terms for racial discrimination experiences were found for three of the four models. After controlling for the covariates, the results from the analyses indicate that individuals who experienced greater discrimination in the past year experienced more depressive symptoms (b=.13; p<.01), higher levels of perceived stress (b=.16; p<.01), and lower levels of well-being (b=−.08; p<.05) at Time 2. The main effect for racial discrimination experiences (b=.00; ns) was not significant for problem behaviors.
Patterns of Racial Socialization as Compensatory Resilience Factors
In the GLM examining depressive symptoms, the main effect for cluster group membership was significant (F[3, 348]=4.49, p<.01: Table 4), indicating that certain patterns of racial socialization served as compensatory resilience factors. The lowest level of depressive symptoms was found in the High Positive cluster (M=1.66). Individuals in the Moderate Negative (M=1.90) and Low Frequency (M=1.89) clusters reported significantly higher levels of depressive symptoms relative to the High Positive cluster. The group mean for the Moderate Positive group (M=1.73) was not significantly different from the mean for the High Positive cluster group.
The GLM examining perceived stress also yielded a significant main effect for cluster group membership (F[3, 348]=5.63, p=.001: Table 5). The High Positive group had the lowest levels of perceived stress at Time 2 (M=2.79) of all the cluster groups. The Moderate Negative (M=3.18) and Low Frequency cluster groups (M=3.01) reported significantly higher levels of perceived stress at Time 2. The difference between the Moderate Positive (M=2.89) and the High Positive group approached significance (p<.10). There was also a main effect for cluster membership (F[3, 348]=7.49, p<.001: Table 7) for well-being. Post hoc analyses indicate that adolescents in the High Positive cluster group (M=3.73) reported significantly higher levels of well-being than those in the Moderate Positive cluster (M=3.59), the Low Frequency cluster (M=3.47), and the Moderate Negative cluster (M=3.24).
Finally, there were no significant differences in problem behaviors by cluster membership (see Table 6).
Patterns of Racial Socialization as Protective Resilience Factors
An examination of the discrimination × socialization cluster interaction term in the four models found a significant interaction for perceived stress (F[3, 348]=3.17, p<.05), while the interaction term for problem behaviors (F[3, 348]=2.21, p=.09) approached significance. Figure 2 plots the relationship between racial discrimination at Time 2 and perceived stress separately by cluster membership. Racial discrimination had a significant positive relationship with perceived stress for both adolescents in the Moderate Positive (b=.16; p<.001) and Low Frequency clusters (b=.17; p<.01). Racial discrimination was unrelated to perceived stress for individuals in the High Positive (b=.03; ns) and Moderate Negative clusters (b=−.05; ns). Figure 2 also indicates that of the individuals who experienced higher levels of racial discrimination those adolescents who were in the High Positive group reported lower levels of perceived stress than individuals in any of the other three groups. This pattern of findings is consistent with High Positive racial socialization acting as a buffer against the influence of racial discrimination.
Although the discrimination × socialization cluster interaction term only approached significance for problem behaviors, post hoc analyses indicated significant cluster differences in the relationship between racial discrimination and problem behaviors (see Figure 3). Racial discrimination was significantly related to engaging in problem behavior for individuals in the Moderate Negative (b=.27; p<.01) clusters, but was unrelated for individuals in the Moderate Positive (b=.00; ns), Low Frequency (b=.06; ns), and High Positive clusters (b=.05; ns). Of those who report experiencing higher levels of racial discrimination, adolescents in the High Positive cluster engage in fewer problem behaviors than adolescents in the Low Frequency and Moderate Negative cluster groups, suggesting that receiving High Positive racial socialization messages may buffer the negative influence of racial discrimination on problem behaviors.
Patterns of Racial Socialization Experiences
Our first objective in the present study was to identify different patterns of racial socialization experiences. Using latent class analyses, we identified four patterns of child-reported racial socialization experiences within our sample of African American adolescents. These four profiles provide a more detailed picture of the various messages that parents convey to children regarding the meaning of race. Consistent with other studies, the majority of adolescents in the present study reported that they had received a wide variety of racial socialization messages from their parents (Coard & Sellers, 2005; Hughes et al., 2006). Our two largest clusters emphasized positive messages such as racial pride and self-worth along with messages that emphasize the barriers that African Americans are likely to face. The relative proportion of different messages and behaviors looks strikingly similar between the High Positive and Moderate Positive clusters with the difference in the two groups lying in the overall frequency of messages. This emphasis on racial pride messages and racial barriers is consistent with the idea that a major goal of racial socialization is to protect African American children against the stigma that they face in this society because of their race (Bowman & Howard, 1985; Hughes & Chen, 1999; Marshall, 1995; Miller & MacIntosh, 1999; Stevenson, Cameron, Herrero-Taylor, & Davis, 2002; Thornton et al., 1990). Interestingly, while adolescents in both the High Positive and Moderate Positive clusters report receiving messages about both racial pride and racial barriers, they also report receiving egalitarian messages that encourage them not to use race in determining how they will interact with other people. Such a combination of messages may buffer them from developing bitterness towards individual members of other races (especially whites), while also instilling a healthy vigilance against racism and pride in their race. Parents of children in these two clusters also seem to be reinforcing these verbal messages with racial socialization activities and behaviors.
Our third cluster consisted of those adolescents who reported receiving the fewest messages and engaging in the least amount of racial socialization activities. Roughly one-fifth of our sample fell into the Low Frequency cluster. This finding is consistent with other research which has reported that a significant minority of African American parents engage in little or no explicit racial socialization (Bowman & Howard, 1985; Hughes & DuMont, 1993; Marshall, 1995; Parham & Williams, 1993; Peters, 1985; Spencer, 1983). While adolescents in the Low Frequency cluster reported receiving almost no messages related to race, they did report receiving more self-worth messages than any other type of messages. This suggests that some parents may have consciously decided not to discuss race with their children as an explicit racial socialization strategy (Hughes & Chen, 1999). Some researchers have conceptualized the absence of racial socialization messages as a possible form of egalitarian messages (see Hughes & Chen, 1999). Interestingly, however, adolescents in the Low Frequency cluster also reported receiving few egalitarian messages. While it is possible that some of the parents of children in the Low Frequency cluster have made a conscious effort not to discuss race with their children, it is also possible that the dearth of race-related messages may be indicative of more general problems with the parent-child relationship. In a study of African American mothers, Frabutt, Walker, and MacKinnon-Lewis (2002) found that parental positivity, warmth, involvement, and monitoring were generally associated with racial socialization. Unfortunately, there were no such measures in the present study. Future research is needed to examine racial socialization within the context of general parenting (Coard & Sellers, 2005; Hughes et al., 2006).
Although most of the racial socialization literature has focused on positive messages, negative stereotypes are also internalized by some African Americans and transmitted across generations (Taylor, 1990; Thornton et al., 1990). As a whole, adolescents in the present study reported receiving few negative messages about African Americans from their parents. Roughly 10% of the sample, however, reported a pattern of racial socialization in which they more frequently received negative messages about African Americans from their parents. Not only did these adolescents report receiving more negative messages than all of the other adolescents, but the adolescents in the Moderate Negative cluster also reported receiving fewer positive messages such as racial pride and self-worth than all of the other adolescents except those in the Low Frequency cluster. Such a combination of messages seems to have dire consequences for the way in which African American adolescents' function psychologically. Because so little research has focused on this type of racial socialization, more research is needed that illuminates the circumstances in which parents convey these messages. It is also important for future research to determine to what extent adolescents are internalizing these negative messages.
Although some research has found that sociodemographic factors often shape the types of racial socialization messages that parents convey to their children (Bowman & Howard, 1985; Hughes & Chen, 1997; Thornton et al., 1990), we found no age or gender differences in the composition of the racial socialization clusters. We did find, however, a marginally significant difference in primary caregivers' educational attainment between the Moderate Negative and High Positive cluster groups such that children in the High Positive cluster, on average, had primary caregivers with higher levels of educational attainment. This trend is consistent with several studies that have reported links between parental socioeconomic background and racial socialization (Caughy et al., 2002; Hughes & Chen 1997). Indeed, Hughes et al. (2006) suggest that parents with different socioeconomic backgrounds may have different ideas about race and ethnicity, which, in turn, shape racial socialization. In light of evidence that caregivers who attain higher levels of education perceive more prejudice and discrimination than their less educated counterparts (Williams, 1999), it may be that the frequency and kind of messages parents convey about race are a function of their own experiences with racial discrimination. African American children who come from more affluent families are often more likely to have social interactions with children of other racial backgrounds than children of less affluent backgrounds. As a result of these more frequent interracial interactions caregivers may be more likely to spend time discussing issues of race with their children. The dearth of age differences in cluster group membership may be due to the selective nature of the sample. Although there was significant age variation within the sample, our sample does not include children at other developmental stages. Unlike other studies, we found no gender differences in our cluster membership. Bowman and Howard (1985) found that African American girls reported receiving more racial pride messages than boys, while African American boys reported receiving more racial barrier messages from their parents than girls. It is possible that the gender differences that Bowman and Howard reported were an artifact of focusing on each type of message in isolation from other socialization messages. When we used a similar approach to examine gender differences in our own sample, we found that boys reported receiving more racial barrier messages than girls. Further research that examines patterns of racial socialization messages will be necessary before any definitive conclusions can be made regarding demographic differences and even developmental shifts in racial socialization processes (see Fisher et al., 2000; Hughes & Chen, 1999).
We did find cluster differences in adolescents' experiences with racial discrimination. Adolescents in the Moderate Negative cluster reported experiencing racial discrimination significantly more frequently than all of the other adolescents. Because our analyses examined racial discrimination and racial socialization within the same time period, we are unable to determine the direction of causation. It is possible that experiencing racial discrimination more frequently may make certain racial socialization messages (i.e. negative ones) more salient to adolescents, and thus the adolescents are more likely to report receiving those specific types of messages. It is also possible that receiving more negative messages and fewer positive messages about African Americans may lead adolescents to be more racially vigilant to ambiguous situations that may be appraised as racist (see Mendoza-Denton, Downey, Purdie, Davis, & Pietrzak, 2002). We plan to conduct future research using prospective research designs looking at parents' reports of the messages that they convey to further address this issue. As noted above, some parents choose not to talk to their children about race because they are afraid that such discussions may lead their children to become hypervigilant toward racial discrimination. Our results do provide evidence that talking to children about race does not necessarily make them overly sensitive to racial discrimination. The adolescents in the Low Frequency cluster did not differ significantly in their experiences of discrimination from those in the High Positive and Moderate Positive cluster. Our findings suggest that it is not whether or not parents talk to their children about race, but rather, the content of the racial socialization messages that is associated with adolescents' experiences with racial discrimination.
Racial Discrimination and Psychological Adjustment
Our second objective was to investigate the relationship between racial discrimination and psychological adjustment outcomes. Our results add to a growing body of research indicating that experiencing racial discrimination is associated with a wide variety of negative outcomes for African American adolescents (Fischer & Shaw, 1999; Neblett et al., 2006; Prelow et al., 2004; Sellers et al., 2006; Wong et al., 2003). The present findings indicate that experiencing racial discrimination was associated with higher levels of depressive symptoms, more perceived stress, and lower levels of psychological well-being. Discrimination was unrelated to participating in problem behaviors. Similar to Wong et al. (2003), the present study utilized a longitudinal design in which the effects of prior levels of psychological adjustment were controlled in the analyses. Such an approach makes it less likely that the significant relationship between racial discrimination and psychological adjustment outcome is the result of the individuals' levels of previous psychological adjustment affecting their subsequent perceptions of racial discrimination. While the present analyses are unable to provide definitive evidence that experiencing racial discrimination causes lower levels of psychological adjustment, the use of instantaneous effect models with longitudinal designs does provide stronger evidence of the possible impact of experiencing racial discrimination.
Racial Socialization as a Compensatory Factor
The third objective of the present study was to investigate whether adolescents' patterns of racial socialization act in a compensatory manner. Our results indicate that, on average, adolescents who experienced High Positive patterns of racial socialization reported the most positive psychological adjustment outcomes, whereas adolescents in the Low Frequency and Moderate Negative racial socialization clusters reported the least favorable outcomes. For well-being, we even found group differences between the High Positive and Moderate Positive groups, again highlighting the importance not only of the content of the racial messages, but also the frequency with which parents convey the messages to their children. Our results are consistent with several other studies that have linked racial pride messages with a variety of positive outcomes such as greater feelings of closeness to other Blacks (Demo & Hughes, 1990), greater factual knowledge, racial and otherwise (Branch & Newcombe, 1986; Caughy et al., 2002), higher levels of domain-specific self-esteem (Constantine & Blackmon, 2002), and more positive academic beliefs (Smith, 1996). Messages that emphasize racial pride and self-worth may provide African American adolescents with a strong foundation upon which to build their personal self-esteem as well as a counter-weight to the many disparaging messages that they may receive about their racial group from the larger society. Messages regarding the existence of racial barriers may also help adolescents to appropriately attribute unfair treatment that they may receive because of their race to external sources (e.g., prejudice), thus reducing their risk for internalizing negative mood states such as depression (see Crocker, Voelkl, Testa, & Major, 1991; Peterson & Seligman, 1984).
While a pattern of racial socialization consistent with the High Positive cluster seems to help compensate for experiences of racial discrimination, our results also suggest that some patterns of engaging children around race may be detrimental to their psychological adjustment and even pose a risk to the development of an emotionally healthy child. In the study conducted by Bowman and Howard (1985), youth who were not taught anything about race had lower self-efficacy scores than those who received racial socialization messages. The absence of communication about race, implicit and explicit, may allow thoughts and stereotypes with which youth are grappling to go unchallenged (Spencer, 1983), act as a handicap (Stevenson et al., 1997), and effectively make adolescents more susceptible to the negative impact of discrimination experiences on their overall psychological health. As noted above, a dearth of communication around race may be symptomatic of a larger communication problem between parent and child which has negative consequences for adolescent outcomes. Adolescents whose parents are not involved in their lives are more likely to experience a range of negative developmental outcomes than those who offer support (Eccles, Early, Frasier, Belansky, & McCarthy, 1997).
Our findings indicate that adolescents who were in the Moderate Negative socialization cluster may also be placed at risk for less positive psychological adjustment. The combination of negative messages about African Americans and the dearth of racial pride and self-worth messages that these adolescents receive may be just as damaging as the negative impact of racial discrimination experiences on adolescents' psychological adjustment. It is likely that messages and behaviors that a child is exposed to from significant others such as parents are likely to have a greater impact on a child's developing sense of self than messages and behaviors that come from more distant others (Rosenberg, 1979). Because adolescence is such an important period for the development of a healthy racial identity and self-concept (Phinney, 1989), internalizing such a negative pattern of racial socialization may be especially detrimental to a psychologically well-functioning individual. Further research is needed that examines what factors serve to protect adolescents who are faced with this form of risk.
Racial Socialization as a Resilience Factor
Our fourth and final objective was to investigate whether particular patterns of racial socialization act as protective resilience factors. Our results suggest that High Positive racial socialization seems to buffer the negative influence of racial discrimination on African American adolescents' perceived stress. For those adolescents who experienced relatively little racial discrimination, there appeared to be no difference in the level of perceived stress for those in the High Positive, Moderate Positive, and Low Frequency clusters. For adolescents who experienced racial discrimination relatively more frequently, however, those in the High Positive cluster experienced much less stress than adolescents in the other three clusters. Adolescents who were in the Moderate Negative cluster reported higher levels of stress than the High Positive adolescents regardless of the amount of discrimination that they reported experiencing. Although the coefficient for the interaction term for problem behaviors only approached significance, a similar pattern of results was found in which High Positive racial socialization also seemed to act as a protective resilience factor.
Underlying Mechanisms. As noted earlier, racial identity attitudes may be an important mediating mechanism by which certain patterns of racial socialization may act as a protective buffer against the impact of racial discrimination (Thompson, Anderson, & Bakeman, 2000). Racial identity attitudes that are consistent with the content of the racial socialization messages that characterize the High Positive cluster have been found to act as buffers against African American adolescents' experiences with racial discrimination. Wong et al. (2003) found that African American adolescents who held positive attitudes toward their race were buffered against the negative impact of racial discrimination on a variety of academic and well-being outcomes. In another study, a weaker relationship between racial discrimination and indicators of psychological well-being was found for African American adolescents who felt that other groups held more negative attitudes toward their racial group than those who felt other groups held more positive attitudes toward their racial group (Sellers et al., 2006). While several studies have linked racial socialization to racial identity (e.g., Demo & Hughes, 1990; Sanders Thompson, 1994; Stevenson, 1995), we are unaware of any study that has formally tested possible mediating relationships among racial socialization, racial identity, racial discrimination and psychological functioning outcomes. Such research is needed before any definitive conclusion can be made regarding the mechanism by which certain patterns of racial socialization act as a protective resilience factor.
Coping may be another potential mediating mechanism by which High Positive patterns of socialization may protect youth against the negative effects of discrimination over time. In addition to promoting racial identity, researchers have typically asserted that racial socialization prepares youth to cope effectively with racism-related experiences and accompanying stress and provides the necessary psychological strength to resist racial oppression (Ward, 1999). Surprisingly, however, few studies have assessed the relation between racial socialization and actual coping behavior. In one of the few studies to do so, Scott (2003) found that a measure of global racial socialization was related to approach coping strategies (i.e., strategies in which the individual actively deals with the stress). Specifically, racial socialization messages were positively related to greater use of social support strategies (e.g., telling family members about what happened) and self-reliance/problem solving (e.g., relying on personal resources and knowledge to deal with stress). Given that approach coping strategies have generally been related to greater feelings of self-efficacy and less distress (Moos, 2002), it may be that racial socialization messages consistent with those in the High Positive cluster may prepare youth to effectively cope with discrimination experiences in specific ways that lead to decreased stress and positive psychological adjustment. Qualitative work by Sanders (1997) and others (e.g., Lareau, 2003) has suggested that conversations between caretakers and adolescents about racial barriers—a component of the High and Moderate Positive Clusters—may contain not only discussion of the challenges youth face, but also ways to cope with these challenges. These conversations may, in turn, facilitate the development of advantageous coping strategies that buffer the negative effects of racial discrimination.
While the findings from the present study make an important contribution to the existing racial socialization literature, there are several important directions for future research. Future research needs to look at racial socialization within the context of other parenting behaviors to more effectively identify the mechanisms by which socialization influences child outcomes (Hughes, 2003). It is still unknown to what extent the nature of the parent–child relationship influences the effectiveness of the racial socialization messages. The use of two waves of data in the current study presents an advance over the preponderance of cross-sectional racial socialization investigations that characterize the extant research literature. Our analyses, however, do not fully address some of the interpretive limitations regarding the direction of causality that are characteristic of cross-sectional designs. Future research that employs prospective cross-lagged designs will help to further illuminate the causal direction of the associations among racial socialization, racial discrimination, and psychological adjustment. Another issue centers around the “representativeness” of the four clusters found in the present study. It is unclear whether the cluster patterns found in the present study are unique to the present sample or whether they represent fundamental patterns of racial socialization that most African American adolescents experience. The fact that we found only one marginal sociodemographic difference in the cluster membership for parental educational attainment provides very limited support for the possibility that these patterns of socialization are likely to generalize to other samples of African Americans. The family income of the sample is higher than the African American population in the United States, and only replication of the present method will definitively settle the issue. Finally, it should be noted that the present study has focused on adolescents' assessments of what socialization messages their parents have conveyed to them. It is important that as we digest the results from the present study, we recognize that we only have part of the picture. The little research that has looked at both parent and child accounts of racial socialization at the same time has found little congruence between the two (Scottham, 2003). This is not surprising given the nature of the interactions between parents and children (Coard & Sellers, 2005). Thus, future research is needed that focuses on the parents' perspective of racial socialization as well. Only then will we have a fuller understanding of the racial socialization process.
In conclusion, the present study utilized an innovative research approach to build upon the extant racial socialization literature in investigating the role of racial socialization messages in adolescent psychological outcomes. This analytic approach is consistent with conventional wisdom that socialization practices are implicit as well as explicit and do not occur in isolation. Our results indicate that although experiencing racial discrimination is a significant obstacle in the psychological development of African American adolescents, certain racial socialization experiences provide adolescents with some protection. Thus, racial socialization seems to be an important factor to consider in understanding the resilience of African American adolescents. We hope that future research will build upon our current findings to further illuminate the complex processes by which African American parents transmit messages to their children about the meaning of race.
This research was supported through grants from the National Institute of Mental Health (NIMH 5 R01 MH61967-02) and the National Science Foundation (BCS-9986101). The sixth author is the principal investigator of the grant. We thank the members of the African American Family Project for their help with data collection.
- 1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. , & (
- 1980). The black community as a source of positive self concept for black children: A theoretical perspective. In R.Jones (Ed.), Black psychology (pp. 106–138). New York: Harper and Row. (
- 1985). Race-related socialization, motivation, and academic achievement: A study of Black youths in three-generation families. Journal of the American Academy of Child Psychiatry, 24, 134–141. , & (
- 1985). Black child socialization: A conceptual framework. In H. P.McAdoo (Ed.), Family ethnicity: Strength in diversity (pp. 245–270). Thousand Oaks, CA: Sage Publications. , & (
- 1986). Racial attitude development among young Black children as a function of parental attitudes: A longitudinal and cross-sectional study. Child Development, 57, 712–721. , & (
- 1999). Effects of religiosity and racial socialization on subjective stigmatization in African-American adolescents. Journal of Adolescence, 22, 223–242. , & (
- 1997). Race-related factors and life satisfaction among African Americans. Journal of Black Psychology, 23, 36–49. (
- 2002). The influence of racial socialization practices on the cognitive and behavioral competence of African American preschoolers. Child Development, 73, 1611–1625. , , , & (
- 2004). Brief report: Initial psychometric properties of the everyday discrimination scale in Black adolescents. Journal of Adolescence, 27, 363–368. , , & (
- 2005). African American families as a context for racial socialization. In V. C.McLoyd, K. A.Dodge, & N. E.Hill (Eds.), African American family life: Ecological and cultural diversity (pp. 264–284). New York: Guilford. , & (
- 2004). Towards culturally relevant preventive interventions: The consideration of racial socialization in parent training with African American families. Journal of Child and Family Studies, 13, 277–293. , , , & (
- 1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. , , & (
- 2002). Black adolescents' racial socialization experiences: Their relations to home, school, and peer self-esteem. Journal of Black Studies, 32, 322–335. , & (
- 1991). Social stigma: The affective consequences of attributional ambiguity. Journal of Personality and Social Psychology, 60, 218–228. , , , & (
- 2003). The association of physical maturation with family hassles among African American adolescent males. Cultural Diversity and Ethnic Minority Psychology, 9, 276–288. , , , & (
- 1990). Socialization and racial identity among Black Americans. Social Psychology Quarterly, 53, 364–374. , & (
- 1997). The relation of connection, regulation, and support for autonomy to adolescents' functioning. Journal Of Adolescent Research, 12, 263–286. , , , , & (
- 1989). Multiple problem youth: Delinquency, substance use, and mental health problems. New York: Springer-Verlag. , , & (
- 1999). African Americans' mental health and perceptions of racist discrimination: The moderating effects of racial socialization experiences and self-esteem. Journal of Counseling Psychology, 46, 395–407. , & (
- 2000). Discrimination distress during adolescence. Journal of Youth and Adolescence, 29, 679–695. , , & (
- 2002). Racial socialization messages and the quality of mother/child interactions in African American families. Journal of Early Adolescence, 22, 200–217. , , & (
- 1991). Resiliency and vulnerability to adverse developmental outcomes associated with poverty. American Behavioral Scientist, 34, 416–430. (
- 1995). The crime of punishment: Racial and gender disparities in the use of corporal punishment in U.S. public schools. Journal of Negro Education, 64, 454–463. (
- 1997, May). Development and initial validation of scales to measure racism-related stress. Poster presented at the 6th Biennial Conference on Community Research and Action, Society for Community Research and Action, South Carolina. (
- 2003). Correlates of African American and Latino parents' messages to children about ethnicity and race: A comparative study of racial socialization. American Journal of Community Psychology, 31, 15–33. (
- 1997). When and what parents tell children about race: An examination of race-related socialization among African American families. Applied Developmental Science, 1, 200–214. , & (
- 1999). The nature of parents' race-related communications to children: A developmental perspective. In L.Balter (Ed.), Child psychology: A handbook of contemporary issues (pp. 467–490). Philadelphia: Psychology Press/Taylor & Francis. , & (
- 1993). Using focus groups to facilitate culturally anchored research. American Journal of Community Psychology, 21, 775–806. , & (
- 2006). Parents' ethnic-racial socialization practices: A review of research and directions for future study. Developmental Psychology, 42, 747–770. , , , , , & (
- 1996). Changes in African American resources and mental health. In J. S.Jackson (Ed.), Mental health in Black America (pp. 189–212). Thousand Oaks, CA: Sage Publications Inc. , & (
- 1999). The prevalence, distribution, and mental health correlates of perceived discrimination in the United States. Journal of Health and Social Behavior, 40, 208–230. , , & (
- 1993). The socialization of cooperative, competitive, and individualistic preferences among Mexican American children: The mediating role of ethnic identity. Hispanic Journal of Behavioral Sciences, 15, 291–300. , , & (
- 1996). The schedule of racist incidents: A measure of racial discrimination and a study of its negative physical and mental health consequences. Journal of Black Psychology, 22, 144–168. , & (
- 1996). Bootstrapping goodness-of-fit measures in categorical data analysis. Sociological Methods & Research, 24, 492–516. , , & (
- 2003). Unequal childhoods: Class, race, and family life. Berkeley, CA: University of California Press. (
- 2006). The racial socialization questionnaire—teen (RSQ-t): A new measure for use with African American adolescents. Unpublished manuscript. , , , & (
- 1983). Covariance structure models: An introduction to LISREL. Beverly Hills, CA: Sage Publications. (
- 2004). Latent class models. In D. W.Kaplan (Ed.), The Sage handbook of quantitative methodology for the social sciences (pp. 175–198). Thousand Oaks, CA: Sage Publications. , & (
- 1995). Ethnic socialization of African American children: Implications for parenting, identity development, and academic achievement. Journal of Youth and Adolescence, 24, 377–396. (
- 2002). Sensitivity to status-based rejection: Implications for African American students' college experience. Journal of Personality and Social Psychology, 83, 896–918. , , , , & (
- 1999). Racial socialization and racial identity: Can they promote resiliency for African American adolescents. Adolescence, 34, 493–501. (
- 1999). Promoting resilience in urban African American adolescents. Social Work Research, 23, 159–169. , & (
- 2002). Life stressors, social resources, and coping skills in youth: Applications to adolescents with chronic disorders. Journal of Adolescent Health, 30, 22–29. (
- 2001). Racial discrimination as a moderator of the links among stress, maternal psychological functioning, and family relationships. Journal of Marriage and Family, 63, 915–926. , , , , & (
- 2006). African American adolescents' discrimination experiences and academic achievement: Racial socialization as a cultural compensatory and protective factor. Journal of Black Psychology, 32, 199–218. , , , & (
- 1993). Socialization of Chinese American children. In H. P.McAdoo (Ed.), Family ethnicity: Strength in diversity (pp. 245–270). Thousand Oaks, CA: Sage Publications. , & (
- 1993). The relationship of demographic and background factors to racial identity attitudes. Journal of Black Psychology, 19, 7–24. , & (
- 1985). Racial socialization of young Black children. In H. P.McAdoo & J. L.McAdoo (Eds.), Black children: Social education, and parental environments (pp. 159–173). Thousand Oaks, CA: Sage Publications, Inc. (
- 1984). Causal explanations as a risk factor for depression: Theory and evidence. Psychological Review, 91, 347–374. , & (
- 1989). Stages of ethnic identity development in minority group adolescents. Journal of Early Adolescence, 9, 34–49. (
- 1995). Parental ethnic socialization and adolescent coping with problems related to ethnicity. Journal of Research on Adolescence, 5, 31–53. , & (
- 2004). The impact of ecological risk and perceived discrimination on the psychological adjustment of African American and European American youth. Journal of Community Psychology, 32, 375–389. , , , & (
- 1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. (
- 1979). Group rejection and self-rejection. Research in Community Mental Health, 1, 3–20. (
- 1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57, 1069–1081. (
- 1997). Overcoming obstacles: Academic achievement as a response to racism and discrimination. The Journal of Negro Education, 66, 83–93. (
- 1994). Socialization to race and its relationship to racial identification among African Americans. Journal of Black Psychology, 20, 175–188. (
- 2003). The relation of racial identity and racial socialization to coping with discrimination among African American adolescents. Journal of Black Studies, 33, 520–538. (
- 2003). What we tell our sons and daughters: Parent-child race socialization among African American adolescents. Unpublished doctoral dissertation, University of Michigan. (
- 2003). Racial identity, racial discrimination, perceived stress, and psychological distress among African American young adults. Journal of Health and Social Behavior, 43, 302–317. , , , & (
- 2006). Racial identity matters: The relationship between racial discrimination and psychological functioning in African American adolescents. Journal of Research on Adolescence, 16, 187–216. , , , & (
- 2003). The role of racial identity in perceived racial discrimination. Journal of Personality and Social Psychology, 84, 1079–1092. , & (
- 2002). Discrimination, crime, ethnic identity, and parenting as correlates of depressive symptoms among African American children: A multilevel analysis. Development and Psychopathology, 14, 371–393. , , , , , & (
- 1996, March). Racial-ethnic socialization: Family, school, and community influences. Paper presented at the biennial meeting of the Society for Research in Adolescence, Boston, MA. (
- 1983). Children's cultural values and parental child rearing strategies. Developmental Review, 3, 351–370. (
- 1997). A phenomeonological variant of ecological systems theory (PVEST): A self-organization perspective in context. Development and Psychopathology, 9, 817–833. , , & (
- 1994). Validation of the Scale of Racial Socialization for African American adolescents: Steps toward multidimensionality. Journal of Black Psychology, 20, 445–468. (
- 1995). Relationship of adolescent perceptions of racial socialization to racial identity. Journal of Black Psychology, 21, 49–70. (
- 1997). Managing anger: Protective, proactive, or adaptive racial socialization identity profiles and African-American manhood development. Journal of Prevention and Intervention in the Community, 16, 35–61. (
- 2002). Development of the teenager experience of racial socialization scale: Correlates of race-related socialization frequency from the perspective of Black youth. Journal of Black Psychology, 28, 84–106. , , , & (
- 2002). “Mitigating instigation”: Cultural phenomenological influences of anger and fighting among “Big-Boned” and “Baby-faced” African American youth. Journal of Youth and Adolescence, 31, 473–485. , , , & (
- 1997). Racism stress management: Racial socialization beliefs and the experience of depression and anger in African American youth. Youth and Society, 29, 172–222. , , , & (
- 1997). Why are all the black kids sitting together in the cafeteria? New York: Basic Books Inc. (
- 1990). Relationship between internalized racism and marital satisfaction. Journal of Black Psychology, 16, 45–53. (
- 2000). Effects of racial socialization and racial identity on acculturative stress in African American college students. Cultural Diversity and Ethnic Minority Psychology, 6, 196–210. , , & (
- 1996). Perceived experiences of racism as stressful life events. Community Mental Health Journal, 32, 223–233. (
- 1990). Sociodemographic and environmental correlates of racial socialization by Black parents. Child Development, 61, 401–409. , , , & (
- 2005). Latent gold 4.0 user's guide. Belmont, MA: Statistical Innovations Inc. , & (
- 1999). Resilience and resistance. In A.Garrod, J. V.Ward, T. L.Robinson, & R.Kilkenny (Eds.), Souls looking back: Life stories of growing up Black (pp. 173–185). Boston: Routledge Kegan Paul. (
- 1999). Race, socioeconomic status, and health: The added effects of racism and discrimination. In N.Adler & M.Marmot (Eds.), Socioeconomic status and health in industrial nations: Social, psychological, and biological pathways. Annals of the New York Academy of Sciences (Vol. 896, pp. 173–188). New York: New York Academy of Sciences. (
- 2003). The influence of ethnic discrimination and ethnic identification on African American adolescents' school and socioemotional adjustment. Journal of Personality, 71, 1197–1232. , , & (
- 1994). Resiliency research: Implications for schools and policy. Social Policy Report, 8, 1–18. , & (
- 2002). Natural mentors and adolescent resiliency: A study with urban youth. American Journal of Community Psychology, 30, 221–243. , , & (