Chill, Be Cool Man: African American Men, Identity, Coping, and Aggressive Ideation

Abstract

Aggression is an important correlate of violence, depression, coping, and suicide among emerging young African American males. Yet most researchers treat aggression deterministically, fail to address cultural factors, or consider the potential for individual characteristics to exert an intersectional influence on this psychosocial outcome. Addressing this gap, we consider the moderating effect of coping on the relationship between masculine and racial identity and aggressive ideation among African American males (N = 128) drawn from 2 large Midwestern universities. Using the phenomenological variant of ecological systems theory and person-centered methodology as a guide, hierarchical cluster analysis grouped participants into profile groups based on their responses to both a measure of racial identity and a measure of masculine identity. Results from the cluster analysis revealed 3 distinct identity clusters: Identity Ambivalent, Identity Appraising, and Identity Consolidated. Although these cluster groups did not differ with regard to coping, significant differences were observed between cluster groups in relation to aggressive ideation. Further, a full model with identity profile clusters, coping, and aggressive ideation indicates that cluster membership significantly moderates the relationship between coping and aggressive ideation. The implications of these data for intersecting identities of African American men, and the association of identity and outcomes related to risk for mental health and violence, are discussed.

African American males are disproportionately represented among victims of violence (National Center for Injury Prevention and Control, 2013) and have reportedly experienced a notable increase in depression incidence (U.S. Department of Health and Human Services, 2006) and suicide, especially for 15- to 24-year-old youth (Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2012) in recent years. Aggression is a well-noted correlate of violence, depression, and suicide (Brady, Gorman-Smith, Henry, & Tolan, 2008; Fowler, Tompsett, Braciszewski, Jacques-Tiura, & Baltes, 2009; Gvion & Apter, 2011; Riggs & Kaminski, 2010). Aggression is often described as hostile, injurious, or destructive behaviors born of frustration (Bandura, 1978; Goldstein, Young, & Boyd, 2008; Rippon, 2000). Researchers have frequently framed aggressive behaviors as coping responses to stressful situations that frustrate or disrupt goal achievement (Berkowitz, 1989), whereas aggressive ideation might be best understood as cognitions, beliefs, or attitudes related to carrying out hostile, injurious, or destructive behaviors, with or without the intention to follow through. A number of studies have focused on factors associated with aggressive behaviors among African American males. Far fewer investigations have focused on cognitive sequelae of African American male aggression and potential relationships with coping. Such insight is essential to broadening the evidence base needed to develop preventive mental health interventions among African American males. We address this gap by focusing on factors associated with aggressive ideation among African American emerging adult males. In addition, we examine the association between aggressive ideation and a coping strategy more frequently reported by males—distractive coping (Nolen-Hoeksema & Girgus, 1994; Li, DiGiuseppe, & Froh, 2006). In this study, the terms “Black” and “African American” are used interchangeably and reflect the research reports from which we derive the relevant literature. We define both terms as representative of individuals within the United States whose familial lineage includes members of the African diaspora.

It is important to study correlates of aggression among males because studies link poorly managed anger and hostility to physical health problems such as hypertension and serious heart conditions (Williams et al., 2000). Males are more likely than females to suffer from these health conditions, and are also more likely to openly express anger in direct, verbal, and physical aggression (Archer, 2004; Card, Stucky, Sawalani, & Little, 2008)—a factor attributable more to gender role socialization during childhood and adolescence than to biological sex differences (Brody & Hall, 1993; Ostrov & Keating, 2004).

Men’s social environments function to confine them to narrow masculine gender roles (Kilmartin, 2007) defined and circumscribed by gender-stereotyped emotional expression (Plant, Hyde, Keltner, & Devine, 2000). Male norms for expressing emotion make it an acceptable practice for anger to be expressed via physical aggression (Courtenay, 2000). In fact, recent literature demonstrated a higher likelihood for girls to engage in indirect, social, and relational aggression compared with males (Archer & Coyne, 2005; Ostrov & Keating, 2004), who employ direct aggression. Aggression in children is costly in terms of a trajectory of negative outcomes, including externalizing problems, poor peer relations, and low prosocial behavior (Archer, 2004; Card et al., 2008).

Identity construction is a key developmental task during emerging adulthood—the period between ages 18 and 29 (Arnett, 2000). Identity construction has been conceptualized as a coping strategy employed by African American males (Cassidy & Stevenson, 2005; Cunningham & Meunier, 2004; Spencer, Dupree, & Hartman, 1997). Spencer et al., 1997 specifically suggested that African American males adopt hypermasculine identities in response to stress confronted in high-risk environments. The period of emerging adulthood, marked by the active construction of identity, includes an additional challenge for African American males who seek to locate and/or concretize self-definitions of race. However, masculinity and racial identities are mutually constituted (Cren-shaw, 1991; Shields, 2008). In fact, intersectionality theory implies that hypermasculine identity adoption among African American males is shaped by broader sociopolitical, economic, and historical forces (e.g., structural racism). According to intersectionality theorists (Shields, 2008), research among marginalized males must also consider the impact of these forces on identity construction and enactment. For example, such forces might make masculinity feel more precarious (see Vandello, Bosson, Cohen, Burnaford, & Weaver, 2008), and thus create more pressure among African American males to recoup their manhood during this sensitive developmental period by enacting hypermasculine stress responses.

Racial identity is a buffer against the deleterious effects of stress (Caldwell, Kohn-Wood, Schmeelk-Cone, Chavous, & Zimmerman, 2004; Sellers, Caldwell, Schmeelk-Cone, & Zimmerman, 2003) or an additional coping resource for ethnic minorities, including African American men (Cross & Vandiver, 2001; Mossa-kowski, 2003). However, more traditional or hypermasculine identity constructions may neutralize the positive benefits of racial identity, by encouraging African American men’s more rigid reliance on distraction as a stress-and-coping strategy. Our study considers this possibility within a developmental period when African American men are concretizing their sense of self and identity.

Masculinity Identity and Aggression

Men are stereotypically defined as more aggressive than women and more likely to respond aggressively in situations that create negative emotions such as frustration (Jakupcak, Tull, & Roemer, 2005). Definitions of traditional masculinity emphasize toughness, competitiveness, athletic prowess, decisiveness, violence, power, and aggressiveness (Cazenave, 1979; Franklin, 1985). Traditional definitions of masculinity identity have been associated with a variety of negative individual and social outcomes. Researchers have documented associations between more traditional masculinity identity and higher levels of anxiety (Courtenay, 2000; Zamarripa, Wampold, & Gregory, 2003), lower capacity for intimacy (Sharpe & Heppner, 1991), negative attitudes toward help-seeking (Good & Wood, 1995), low self-esteem (Cournoyer & Mahalik, 1995), and depression (Good & Mintz, 1990). Further, hypermasculine identity is related to an endorsement of a traditional masculine ideology (Mahalik, Good, & Englar-Carlson, 2003).

Stress associated with transgressions of the norms of traditional masculinity also contributes to men’s tendency to aggress (Franchina, Eisler, & Moore, 2001; Jakupcak, Lisak, & Roemer, 2002). Transgressions of these traditional norms lead to attempts at reaffirming masculinity in an effort to either relieve the resulting discomfort or regain the perceived control (Kinney, Smith, & Donzella, 2001). African American men are expected to conform to the dominant culture’s definition of masculinity while also navigating societal stigma related to their ethnic identity, even though systemic pressures make it difficult to meet those expectations (Cose, 2002; Lazur & Majors, 1995; Madhubuti, 1990). Therefore, there is reason to suspect that masculine identity could be related to variance in the amount of aggressive ideation expressed by men. What is less clear is the influence of gender and race on coping and aggressive ideation.

Racial Identity and Aggression

Racial identity has been associated with the mental health trajectory and other social outcomes of African Americans. As a core element of identity, it has been shown to have a modifying effect on the relationship between racial identity attitudes and several characteristics, including mental health among adolescents (Caldwell, Zimmerman, Bernat, Sellers, & Notaro, 2002). Widely accepted models of racial identity conceptualize racial identity as either stage or multidimensional processes. Stage models theorize that as persons develop, their attitudes and beliefs about what it is to be African American, and their endorsement of group identity, also change (Cross, 1971; Phinney, 1992). Other scholars hold a multidimensional view, theorizing that racial identity subsumes multiple facets of identity (Sellers, Smith, Shelton, Rowley, & Chavous, 1998).

The literature posits that endorsement of various racial identity attitudes has a protective effect against aggressive or violent behavior for African American adolescents exposed to violence, with some gender-specific influences (Caldwell et al., 2004). For example, involvement in multiple types of violent behaviors was associated with males for whom race was not a core element of their identity, but this was not true for females (Caldwell et al., 2004). Other researchers have identified associations between strong positive racial identity, active coping, and less aggression-endorsing beliefs and behaviors (McMahon & Watts, 2002). Caldwell et al. (2004) suggested that strong racial identity and belonging might moderate aggressive outcomes, as these “may offset the stigmatization and marginalization that being African American and male in this society often engenders” (p. 100).

Coping as a Moderator of Identity and Aggression

For individuals at risk for negative outcomes, individual strengths such as self-efficacy and coping skills have been shown to intervene in this trajectory (Bandura, 2001; Fergus & Zimmerman, 2005). Effective use of coping skills, whether cognitive or behavioral, may moderate the risk for deleterious mental health outcomes in African American males. Folkman and Lazarus (1985) have defined coping as cognitive and behavioral strategies employed in an attempt to manage or reduce stress. The repertoire of coping resources available to individuals varies by gender, race, education, socioeconomic status, and other demographic factors. Ruminative coping involves continuous focus on one’s negative mood, as well as the antecedents, consequences, and self-evaluations from this mood. Distraction, on the other hand, refers to deliberately channeling one’s attention away from negative mood, feelings, and experiences, and onto more neutral activities (Nolen-Hoeksema, 1991; Tice & Baumeister, 1993). Robust findings attribute gender disparities in the prevalence of mood disorders to men’s tendency to rely on distractive, as opposed to, ruminative coping strategies (Li et al., 2006; Nolen-Hoeksema, 1991). This is especially so for men who ascribe to a traditional male gender role, because rumination and perseverative thought run contrary to the restricted emotionality that defines masculinity (Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008).

Distractive coping is more compatible with action-oriented masculine norms, in that the strategies help individuals shift attention away from unpleasant situations or thoughts (e.g., going for a run or a bike ride) to those that do not invoke stress (Nolen-Hoeksema, 1991). It is important to understand the role of distractive coping as a behavior that has been associated with protection from deleterious mental health across the life span, and with normative and clinical populations (Huffziger, Reinhard, & Kuehner, 2009; Jacobson et al., 1996; Li et al., 2006; Lyubomirsky, Caldwell, & Nolen-Hoeksema, 1998; Lyubomirsky & Nolen-Hoeksema, 1995; Nolen-Hoeksema & Morrow, 1993). Further, the relationship between masculine identity or gender intensification and aggression is moderated by life-events stress and gender-role stress (Cohn & Zeichner, 2006; Grour, Thomas, & Shoffner, 1992); therefore, it is important to understand how distractive coping is related to male identity as well as to aggressive outcomes.

Emerging Adults

In the current study, we explored the racial and gender identity of a sample of African American male college students. In 2012, African American male freshman enrollment in 4-year universities was 11 times lower compared with Whites and non-Hispanic Whites (U.S. Census Bureau, 2012). This may represent a reduced opportunity for positive identity exploration and construction during emerging adulthood.

Our understanding of African American men’s navigation of key elements of their identity (i.e., race and gender) during this critical developmental period is still limited and reactionary. Therefore, we examined their identity constructions during emerging adulthood (see Arnett, 2000, for details), while highlighting how specific coping skills function around identity. This has implications for African American male college students and for those without access to higher education. Understanding identity exploration and coping resources is key to developing a comprehensive model of developmental processes, particularly for African American males transitioning to adulthood.

Current Study

In the current study, we employed a person-oriented approach to explore identity and mental health issues in emerging adulthood. The person-oriented approach uses cluster analysis and is sensitive to individual differences often missed by more conventional methodologies that focus on the variables in question, rather than on the person or persons being studied (see Bergman & Magnusson, 1987; Rowley, Chavous, & Cooke, 2003). Spencer et al.’s (1997) phenomenological variant of ecological systems theory (PVEST) was utilized to better understand how African American males’ perceptions and experiences are shaped by cultural context. The PVEST theory combines Bronfenbrenner’s (1989) ecological systems theory with intersubjective experiences, including interpretive self-perceptions derived from interactions across systems and within cultural contexts (Spencer, 1982, 1985; Spencer et al., 1997). The PVEST framework therefore acknowledges that specific risk contributors, stress engagement, and reactive and stable coping responses may be required to facilitate positive life-stage outcomes for people of color, and specifically African American males, across the developmental life span (Spencer et al., 1997). In general, PVEST accounts for the likelihood that normative development, such as processes of self-perception, is more complex in contexts of race and high visibility.

The ongoing process of identity negotiation is key to the emerging adulthood stage of development. We hypothesize, therefore, that participants will cluster around the identity variables such that relatively higher private regard and centrality scores within any cluster would represent more positively developed identities. Comparatively higher scores on traditional masculinity or public regard would represent less positive identity development. We further hypothesize that the relationship between identity clusters and aggressive ideation will be moderated by the use of distractive coping, such that higher levels of distractive coping will be related to decreased levels of and aggressive ideation. In order to test these hypotheses, cluster profiles are included in both univariate and multivariate models that tested the association of cluster, coping, and aggressive ideation.

Method

Participants

This study is part of a larger investigation of race, ethnicity, gender, and coping with negative mood among diverse college students attending two large, predominantly White U.S. Midwestern universities. Participants were African American males (N = 128), selected from a sample of 654 students. The overall sample included 413 women (mean age = 19.5 years; SD = 2.11) and 241 men (mean age = 19.9 years; SD = 2.08). Racial composition was 52% African American and 48% White. Of the African American males, 98% ranged from age 18 to 25 years old, and 18- to 21-year-olds accounted for a majority of that number (78%). Self-identified African Americans accounted for 96% of the sample, whereas 3% identified as some other Black ethnicity (e.g., “Descendent of Africans,” “African parents,” “Nigerian though American born”). The sample was almost evenly split based on location, with 52% attending one of the universities (see Table 1). The most commonly reported income level was $21,000 to $60,000; 38% reported household incomes of $101,000 and above. Approximately 72% of the sample reported having mothers with at least some college education. Though there were demographic differences among participants based on the university they attended (i.e., location), there was no difference based on any key variable in this study. As a result, the two groups were combined into one sample, and the demographic distribution is presented in Table 2.

Table 1.

Descriptive Statistics of Sample Demographic Characteristics

Variable	n	%
Age, years
18–21	100	78
22–25	24	19
25–35	4	3
Campus
EMU	66	51.6
UM	62	48.4
Instate
In	92	71.9
Out	36	28.1
Ethnicity
African American	124	96.9
Other Black	4	3.1
Employed
Unemployed	17	13.3
Full/part time	60	46.9
Class
Freshman	27	21.1
Sophomore	27	21.1
Junior	35	27.3
Senior 4th/5th	34	26.5
Mother’s education
Some H.S.	4	3.1
H.S. diploma	32	25
Some college	37	28.9
College diploma	55	43
Family income scale
$Under 20,000	27	21.1
$21,000-$60,000	46	36
$61,000-$100,000	25	19.5
$101,000 & over	23	18

Note. EMU = Eastern Michigan University; UM = University of Mch-igan; H.S. = high school.

Table 2.

Pearson Correlations, Means, and Standard Deviations of Study Variables (N = 128)

Variable	M (SD)	1	2	3	4	5	6	7	8	9	10	11
1. Age	20.39 (2.34)	—
2. Income	4.21 (2.1)	−.138	—
3. Mother’s education	2.12 (.89)	−.168	.424**	—
4. Location	1.48 (.502)	.176*	−.324**	−.356**	—
5. Distractive coping	5.89 (3.46)	.060	−.049	−.013	.012	—
6. Masculinity	4.13 (0.79)	.224*	.058	.204*	−.082	.139	—
7. Private regard	5.10 (1.04)	.200*	.119	.108	−.117	.125	.385**	—
8. Public regard	3.14 (0.97)	−.055	−.062	−.080	.246*	.076	.110	.188	—
9. Centrality	4.10 (0.82)	.199*	.204*	.108	−.093	.058	.369**	.755**	.148	—
10. ARIS	33.14 (8.73)	−.093	−.173	−.176*	.153	.143	−.276**	−.281**	.105	−.156	.083	—

Note. ARIS = Aggressive Ideation Subscale adapted from the Anger Inventory.

^*p < .05;

^**p < .01.

Procedure

Students at one university participated in the study through the “subject pool” of their respective psychology department. Others students were recruited via several methods: posters, and individual and group recruitment, including researchers operating from kiosks. Researchers also contacted student groups with predominantly African American membership. All students who agreed to participate signed consent forms. Most participants completed the full set of self-report measures in an average of 1 hr. Each student was given a debriefing form and offered either $10.00, food, food coupons of equal value, or course credit in return for their participation. The choice of incentive was contingent upon the method of recruitment used. Participants were also provided with information for the school counseling center in response to questions that suggested that some may have been experiencing negative moods.

Measures

Aggressive ideation

The Aggressive Ideation subscale is an adapted measure from the Anger Response Inventory (Tangney et al., 1996). The Anger Response Inventory used a scenario-based self-report system. Variation in aggressive ideation is assessed based on specific hypothetical scenarios. Representative items included “Someone you have just met treats you like you are not good enough” and “You are struggling to carry four cups of coffee to your table at a cafeteria, someone bumps into you, spilling the coffee.” Participants, on a 5-point Likert scale, indicated how they would feel, their likely behaviors, and their assessment of possible consequences of their actions. Responses chosen for factor analysis—for example, “I’d shove the friend against the wall” and “I’d kick a chair”—had to show a clear ideation toward an aggressive action. The responses represented their cognitions regarding the situation and therefore reflected aggressive ideation primed by the scenarios. Because this measure has not been widely tested or used in the literature, a factor analysis procedure was used to determine factor loadings of the scale items for the current sample.

Twenty-four items were selected because they conceptually demonstrated aggressive ideations. Of those, 17 items were extracted from the overall scale to form one factor labeled Aggressive Ideation. The Kaiser-Meyer-Olkin measure of sampling adequacy was .80, above the recommended value of .6, and Bartlett’s test of sphericity was significant, χ²(276) = 1134.81, p < .05. The resulting Aggressive Ideation subscale exhibited adequate internal consistency (α = .75) for the present sample. High scores indicate more aggressive ideation.

Manhood

Masculine identity was measured using an African American Manhood Scale developed specifically for this study. This measure assesses the salience of traditional masculine ideology to men’s overall identity. The measure consisted of 33 statements representing different ways of assigning meaning to one’s masculinity, using a Likert-scale from 1 = not at all important to 5 = very important. The statements were derived from findings from a study that investigated the meaning of manhood among African American men (Hammond & Mattis, 2005). Representative responses to the overall question “How important is each of the following to your beliefs about what it means to be a man?” included “Protecting myself and my family” and “Being resourceful and responsible.” In the current sample, the scale achieved good internal consistency (α = .94). A nine-item measure assessing the salience of traditional masculinity norms for African American men yielded an alpha of .79, and focused on statements that endorsed being physically strong, having power, and having courage as definitions of masculinity (Hammond, Matthews, Mohottige, Agyemang, & Corbie-Smith, 2010).

The Multidimensional Inventory of Black Identity (MIBI)

Racial identity was measured using the MIBI (Sellers, Rowley, Chavous, Shelton, & Smith, 1997). This is a self-report measure measuring the three stable dimensions of identity in African American college students and adults. This pencil-and-paper measure consisted of 56 items measured on a 7-point Likert scale. For the present study, the Centrality, Private, and Public Regard scales were used. These three scales targeted the individual’s perception of their racial identity.

Centrality scale

This scale is comprised of eight items and measured the degree to which an individual views race as critical to their self-concept (current sample, α = .61). Representative items included “Being Black is an important reflection of who I am” and “I have a strong sense of belonging to Black people.”

Private Regard scale

This six-item scale examined the extent to which the individual feels positively or negatively about their race membership and to other African Americans (current sample, α = .64). Representative items included “I feel good about Black people” and “I am proud to be black.”

Public Regard scale

This is a six-item scale measured the individual’s perception, negative or positive, of how African Americans are viewed by others. The scale consisted of representative items including “In general, others respect Black people” and “Society views Black people as an asset” (current sample, α = .68).

Coping Response scale

Distractive coping responses were measured using an adapted Daily Emotion Report (Nolen-Hoeksema, Morrow, & Fredrickson, 1993). Adaptations included asking participants to focus on a recent distressing event and to describe the event (type, duration, and severity and type of resulting distress). On the Distractive Coping scale, participants checked off items from a list of 16 distractive responses from the Daily Emotion Report (Nolen-Hoeksema et al., 1993). High scores on the Distractive Coping scale indicated engagement in thoughts and behaviors that took their minds off their negative feelings. Examples of responses included behaviors such as “Go to sleep to escape how I feel,” and thoughts such as “These feelings won’t last.” Items on the Distractive Coping scale exhibited adequate internal consistency (α = .75) for the current study sample.

Demographics

Age, maternal education (as a proxy for family of origin socioeconomic status), income level, and location or current enrollment were control variables in this study. Age was measured in years and on a continuous scale. Because participants were recruited from two different universities with varying demographic profiles and slightly different study recruitment procedures, location was included in these analyses.

Analytic Strategy

Cluster analyses were used to determine commonalities among respondents and develop identity profiles based on race (Centrality, Public Regard, and Private Regard) and masculine identity. Based on previous research with similar data (Banks & Kohn-Wood, 2007; Seaton, 2009), variables were first standardized as z scores. These scores were used to delineate meaningful clusters using Ward’s method (see Lorr, 1983). Hierarchical cluster analysis was utilized based on recommendations in the literature (Chavous et al., 2003).

Bivariate analyses were used to explore (a) how the clusters differentiated on the four identity variables, and (b) whether the clusters differed on coping and aggressive ideation. Cross-tabs and one-way ANOVAs were conducted to examine group differences and group characteristics. Finally, hierarchical multiple regression analyses were conducted to ascertain whether the clusters or identity profiles varied from each other based on aggressive ideation. Interaction terms were created using the centered versions of the relevant variables. Coping was entered as a moderator in the analysis, and demographic characteristics of the sample (age, location, mother’s education and income level) were controlled for in the model.

Results

Clusters were conceptualized based on the patterning of their z-score loadings for the four identity variables. Descriptive information of final clusters is provided in Table 3, and a bar graph representation of the profiles is illustrated in Figure 1. The first cluster was labeled Identity Ambivalent (n = 10; 10%), with negative scores on all variables, with the highest negative scores being for Private Regard (z = −2.09), Centrality (z = −1.61), and Masculinity (z = −1.44). The second cluster, Identity Appraising (n = 58; 57%), shared relatively strong scores on Public Regard (z = .86). Private regard and masculinity scores in this cluster were lower relative to other clusters (.14 and .12, respectively). The third cluster was identified as Identity Consolidated (n = 34; 33%). This group reported that being African American was a crucial component of their identity (Centrality, z = .49). They also held members of their race in high regard (Private Regard, z = .46).

Table 3.

Means, Z Scores, and Standard Deviations of Cluster Variables

Cluster	Identity traits	Z score (SD)	M (SD)
Identity Ambivalent (n = 10)	Masculinity	−1.44 (1.38)	3.00 (1.09)
	Private regard	−2.09 (.96)	2.92 (.99)
	Public regard	−.65 (.80)	2.47 (.75)
	Centrality	−1.61 (.76)	2.79 (.62)
Identity Appraising (n = 58)	Masculinity	.12 (.88)	4.23 (.69)
	Private regard	.14 (.62)	5.25 (.64)
	Public regard	.86 (.48)	3.87 (.45)
	Centrality	.10 (.67)	4.18 (.55)
Identity Consolidated (n = 34)	Masculinity	.25 (.70)	4.33 (.55)
	Private regard	.46 (.30)	5.58 (.31)
	Public regard	−.93 (.61)	2.19 (.57)
	Centrality	.49 (.50)	4.51 (.41)

Figure 1.

Bar chart of z scores of cluster variables.

Several demographic differences between clusters were apparent. Of note, a majority (56%) of the participants in the Identity Consolidated and the Identity Appraising (33%) groups reported having mothers who had college diplomas, compared with 10% of mothers for Identity Ambivalent group members. Almost 50% of the Identity Consolidated group reported annual household incomes between $60,000 and $150,000, compared with 20% and 34% in the Identity Ambivalent and Identity Appraising groups, respectively. The Identity Ambivalent group included approximately 10% of its membership within the 22- to 34-year age group compared with 25 and 30% in the Identity Appraising and Identity Consolidated groups, respectively.

A one-way ANOVA examined between group differences on the four identity variables, F(2, 99) = 44.01, p < .01; centrality, F(2, 99) = 120.22,p < .01; public regard, F(2, 99) = 78.49, p < .01; private regard, F(2, 99) = 15.11, p < .01; and masculinity (see Table 3). Tukey’s post hoc tests show that the Identity Consolidated group had the highest level of centrality (M = 36.06, SD = 3.26). The Identity Appraising group had the highest level of public regard (M = 23.21, SD = 2.69), and the Identity Consolidated group had the highest level of private regard (M = 33.47, SD = 1.86). For the masculine identity variable, the Identity Consolidated group (M = 142.8, SD = 18.22) and the Identity Appraising group (M = 139.57, SD = 22.81) had significantly higher levels than the Identity Ambivalent group (M = 99.0, SD = 35.95, p < .0001).

An ANOVA was conducted to examine whether the clusters differed significantly on coping and aggressive ideation. The results were nonsignificant for coping but significant for aggressive ideation, F(2, 99) = 5.25, p = .007. The Identity Ambivalent group (M = 41.30, SD = 8.91) differed significantly, with higher levels of aggressive ideation than the Identity Consolidated group (M = 31.29, SD = 7.78, p = .005).

Demographic characteristics were controlled in a model with independent variables regressed on the outcome variable of aggressive ideation. Main effects for all variables and interactions between distractive coping and the identity profiles were assessed (see Table 4). Results indicated that distractive coping was related to aggressive ideation, such that more use of distractive coping was related to less aggressive ideation. Significant interactions were found for the Identity Ambivalent and Identity Consolidated profiles as related to the effects of coping on aggressive ideation. The interactions were graphed to facilitate interpretation. The procedure recommended by Aiken, West, and Reno (1991) used intervals of one standard deviation below and above the mean and the mean itself. The results showed an effect size of .100 for the inclusion of the interaction in the model in the second step of the regression procedures. In the Identity Consolidated group, higher levels of distractive coping were related to lower levels of aggressive ideation. For the Identity Ambivalent group, higher levels of distractive coping were related to higher levels of aggressive ideation (see Figure 2).

Table 4.

Parameter Estimates for Regression With Interaction of Cluster Membership Associated With Aggressive Ideation Scores

	Model 1		Model 2
	B	SE	B	SE
Intercept	46.95***	8.93	41.76***	9.04
Age	−..42	.37	−..44	.36
Location	−.1.74	2.04	−..85	1.99
Mother’s education	−.1.41	1.17	−.1.78	1.15
Household income	−..29	.51	−..24	.49
Distractive coping	34	.28	1.01*	.39
Ambivalent identity	6.76*	3.44	7.94*	3.67
Consolidated identity	−.1.84	2.06	−.1.12	2.01
Distractive Coping × Ambivalent	—	—	.75	3.77
Distractive Coping × Consolidated	—	—	−5.40**	1.97
Adjusted R2	.052		.12
ΔR2	.125		.08
F statistic	1.701		4.223*
Effect size based on ΔR2	.143		.100

^*p < .05.

^**p < .01.

^***p < .001.

Figure 2.

The interaction between distractive coping and aggressive ideation across the Ambivalent and Consolidated identity profiles. The Identity Appraising profile was the referent group for these analyses

Discussion

The present study uniquely contributes a developmental and intersectional approach to racial and gendered identity and the relation of intersecting identity profiles to coping and aggressive ideation. The combination of both race and gender as identities that African American male emerging adults must make meaning from and organize into self-perceptions are an important, yet understudied, aspect of developmental processes. As articulated by Spencer et al. (1997), normative development such as processes of self perception, are more complex in contexts of race, ethnicity, and visibility. College-age African American males in our sample are negotiating multiple milieus, including academic, social, socioeconomic, and a myriad of cultural dynamics, while solidifying a dual status of race and gender in a context in which they are highly visible because of their reduced numbers. The participants in our sample are not just male and not just African American, and our approach allowed us to explore how a group of college-age males self-organize in the context of the ecological systems within and among which they must negotiate (Spencer et al., 1997). Our sample participants exist in a larger context in which African American and male identities are often ascribed negative stereotypes and expectations, and thus are particularly vulnerable to maladaptive reactive coping methods in response to various social, academic, and cultural challenges to their personhood (Cunningham & Spencer, 1996). Our analyses suggest there is variation in how this organization of identity occurs, and that this variation has an impact on coping and aggression.

In terms of intersecting identities, the findings suggest that three distinct profiles are present in the sample: Identity Ambivalent, Identity Appraising, and Identity Consolidated. These three identity profiles may represent differing processes of identity development (i.e., race and gender) in African American males’ transi-tioning to adulthood. The Identity Ambivalent profile describes relatively nascent development in identity construction during the emerging adulthood period, in which young men’s identity is in the greatest state of flux. Preliminary decisions about individual identity regarding both race and gender are being made away from parents and other influential figures. The Identity Appraising profile suggests a stage when young men may be attempting to actively construct their identities in a way that best suits them. They may exhibit much incertitude as they try to reconcile personal expectations with social, familial, and community influences. An ongoing appraisal process of societal and cultural demands, and the primacy that they assign to the views of others, may be evidence of a still relatively underdeveloped identity. The Identity Consolidated profile, on the other hand, characterizes young men who have achieved relative permanence in their definitions of their identity. Although identity construction remains an ongoing process for individuals throughout their lifetime, the Identity Consolidated profile may represent a more stable plateau of identity. These young men report greater awareness of societal views of African Americans than those in the other cluster groups, but they also report racial identity salience. Their endorsement of masculine identity traits is tempered or complemented by their racial identity.

Identity during the transition to adulthood may be influenced by distorted hypermasculine and aggressive images of African American males (Caldwell et al., 2004; Cassidy & Stevenson, 2005), particularly for those with ambivalent or similarly unresolved identity profiles. This could explain the higher levels of aggressive ideation among the Identity Ambivalent group relative to the Identity Consolidated group. There is also variation in the link between coping and aggressive ideation across these identity groups.

The relationship between coping and aggressive ideation is buffered by a more consolidated identity. Participants who have organized their identities related to race and gender in ways that are stable and moderate, yet are aware of potential negative perceptions of their group, are less likely to report aggressive ideation. Although this finding supports the literature that suggests that positive racial identity may buffer against poor mental health outcomes (Utsey, Giesbrecht, Hook, & Stanard, 2008), our analyses suggest that holding moderate and consistent views of both one’s race and gender is related to less potential aggression. Given the body of work that associates direct aggression among males to negative outcomes (Courtenay, 2000), it is possible that consolidated identity is protective for African American males during an important developmental stage of life, when patterns of behavior are becoming concretized (Arnett, 2000).

In contrast, the relationship between coping and aggressive ideation was exacerbated by membership in the Identity Ambivalent profile. It is possible that in this state of ambivalence about one’s identity, short-term adaptive strategies such as distractive coping reinforce one’s tenuous and unequal social position, leading to greater frustration and possible aggression (Bandura, 1978; Berkowitz, 1989; Goldstein et al., 2008; Rippon, 2000). It may also be useful to interpret our findings in the context of broader anger-rumination-coping models. For example, it may be the case that identity-ambivalent participants were more likely to engage in anger rumination, which would correspond to an increased likelihood of engaging in aggressive ideation (or the aggressive ideation could be the content of the anger rumination mental coping process).

The literature suggests that distractive coping strategies may reduce the effect of daily and life stressors. In this sample, Identity Consolidated youth experience less aggressive ideation when they use distractive coping. Youth in the Identity Ambivalent profile, on the other hand, do not benefit from distractive coping strategies. Identity Consolidated youth have more adaptive identities, but their demographic characteristics (e.g., household income, parents’ education) may also reduce exposure to some stressors and make it easier to navigate others. This conglomeration of resources may help to protect such youth from negative developments, whereas for Identity Ambivalent youth, distractive coping may be one of the few protective factors available. Overreliance on this strategy, therefore, may well reduce its efficacy, especially in high-risk, stressful situations. As conceptualized by PVEST, African American adolescent males’ appraisal of self and other may result in either reactive (maladaptive) or stable coping responses, which have implications for life stage outcomes (Spencer et al., 1997).

Our findings have implications for mental health professionals who work with this population. As part of both the assessment and treatment, service providers should be aware of the influence of these multiple identities and their intersection on outcomes for African American males. Understanding how identity intersects in emerging adult African American males, and the effect on coping decisions, would be an important inclusion to the diversity training offered to mental health providers.

Limitations

The psychometric properties of the measure of masculinity used in this study, although based on the findings of a qualitative study by Hammond and Mattis (2005), have not been formally evaluated; thus, we may not have accurately measured the construct. The structure of the measure, however, does assess how closely participants’ beliefs about masculinity link to traditional masculine beliefs that have been described in the literature. This study demonstrates the utility of this masculinity measure in accounting for another dimension of identity. Further, important variables that comprise significant aspects of identity, such as sexual orientation and religious affiliation, were not measured and therefore could not be included in the conceptualization of intersectional identity profiles used in this study. It is likely that inclusion of additional domains of identity would influence the association of relationships among the variables in this study.

The measure used for aggressive ideation in this study did not make distinctions between verbal, physical, and other forms of aggression, and relied upon self-reported ideations of physical aggression. It is unclear how likely reported ideations in this study would represent actual behavior. Because the measure required participants to prognosticate what they may do in certain situations, participants may have over- or underreported aggressive responses, knowing there were no actual or aversive consequences. Indicators of internal consistency (Cronbach’s alpha coefficients) for the scales of the MIBI used in this study were low to medium in magnitude, although, generally, the alphas for these scales fall between .60 and .77 in this sample. The alphas in this sample for centrality, private regard, and public regard are, however, comparable with alphas reported for these scales in other studies (see Sellers et al., 1997, 1998; Stevenson & Arrington, 2009). Cronbach’s alphas are one way of determining a measure’s internal consistency (Sijtsma, 2009). However, statisticians argue that, taken alone, Cronbach’s alphas do not fully capture scale reliability. In fact, the stability of the MIBI subscales across studies, as well as their theoretical grounding, provide additional confidence in their functionality. It is also important to note that our data were collected among males attending a predominantly White institution (PWI) that could have an impact on the scale psychometrics.

This sample, by nature, is anomalous. Demographic indicators for this sample (e.g., household income) show that these young men have access to resources that most other African American males may not. This may represent a different contextual environment for identity development due to relatively reduced or different life stressors. Therefore, caution should be exercised when interpreting these findings, especially for the population of African American males. Because our sample had a more limited socioeconomic range, we were unable to fully assess how variations in socioeconomic position or status might shape aggressive ideation and coping.

Building on intersectionality theory, it will be important to test our hypothesis in larger, nationally representative samples in which data could be stratified by socioeconomic status. Another sociodemographic limitation of our data lies in the exclusion of questions designed to assess sexual identity or orientation, which are important, mutually reinforcing aspects of personhood. This oversight should be addressed in future studies, as doing so will have important implications for how we evaluate stress and coping, and how we develop comprehensive strategies to intervene upon problematic aggressive ideation among African American males.

We relied primarily on the PVEST as a study framework. This model has been widely used in studies investigating connections between identity, stress, and coping among African American males. However, the model has some limitations. PVEST relies solely on individual perception of the ecological system and provides a more explicit model for assessing proximal stress-and-coping processes. There is less explicit guidance provided in PVEST for the operationalization and assessment of broader so-cioeconomic and structural forces that shape identity construction and coconstitute aggressive ideation. Future research should consider examining the role of policies and practices (e.g., racial profiling) in eliciting hypermasculinity. Few existing theories take these or other cultural factors into account, and thus treat aggressive ideations and behaviors as wholly “irrational choices.” A critical next step in this area of research would be to document the ecological precipitators and onset of aggressive ideation in real time, as well as to develop upstream policies designed to create more therapeutic landscapes for African American males.

Finally, our use of person-centered methodologies offers strengths. However, it also has some drawbacks that should be taken into account. Most importantly, person-centered approaches overlook intraindividual variation across time and different contexts. In other words, although they generate a more holistic profile, they do little to predict whether response profiles will emerge each time African American males confront sources of frustration. We did not differentiate between more stable personality traits (e.g., neuroticism) and situational responses. Assessing traits like neuroticism, which increase an individual’s sensitivity to negative emotion, will help to make these kinds of trait–state distinctions.

Implications for Future Research

This study continues a trend of research that links the masculine gender role to initiating aggression (Jakupcak, 2003); however, the present findings provide evidence for a relationship between identity and aggressive ideation for young African American men. Research should further examine the development of race and gender identity in African American boys throughout the life course to better appreciate the unique effects of identity processes for a variety of positive and negative outcomes. Also, additional aspects of identity should be assessed to understand how statuses such as sexual orientation vary the relationships under investigation. Further, although our findings are interesting, it would be informative to better understand what other resources could be developed to shield African American males against the effect of negative experiences. This may prove especially critical to research on racial discrimination, efficacy, and resilience.

The possible buffering effects of racial identity in relation to traditional gender expectations should be examined as well. The direct and indirect relation of identity to men’s well-being and mental health should be explored by examining the role of multiple forms of coping (e.g., community connectedness, religious beliefs) as moderators. Possible differences in identity development for African American males attending PWIs versus those at Historically Black Colleges and Universities should be investigated.