Revisiting the Cost of Skin Color: Discrimination,Mastery,and Mental Health among Black Adolescents

Skin Tone and Social Inequality

Colorism theory provides a general framework for understanding the implications of skin tone, within racial groups, in U.S. society. Skin tone bias, or colorism, is a system of discrimination based on privileging individuals with light skin tone and devaluing individuals with dark skin tone (Hunter 2016). Importantly, colorism produces significant patterns of inequality among black Americans (Blake et al. 2016; Goldsmith, Hamilton, and Darity 2006; Hunter 2016; Monk 2014, 2015). While colorism has its roots in slavery, with black individuals with light skin tone being considered more beautiful and intelligent than black individuals with darker skin tone (Keith and Herring 1991), the significance and consequence of skin tone inequality persists in the current period (Arce, Murguia, and Frisbie 1987; Cobb et al. 2016; Monk 2014, 2015; Murguia and Telles 1996; Telles and Murguia 1990). For example, skin tone is significantly associated with black Americans’ household income, educational attainment, and occupational status, with darker skin toned black Americans faring worse than lighter skin toned black Americans in all cases (Keith and Herring 1991; Monk 2014). Furthermore, darker skin toned black Americans receive harsher prison sentences (Blair, Judd, and Chaplean 2004), are more negatively stereotyped (Maddox and Gray 2002), and experience discrimination at a rate of 11 times that of their lighter skin toned counterparts (Klonoff and Landrine 2000). More recently, research has looked at the relationship between skin tone and mental health outcomes. Findings suggest that, in general, darker skin toned black Americans fare worse than lighter skin toned black Americans in terms of self-rated mental health and depression (Monk 2015).

Everyday Discrimination

Much of the association between skin tone and health operates through experiences of everyday discrimination (Monk 2015). Everyday discrimination is defined as “chronic, routine, and less overt experiences of unfair treatment” (Williams et al. 1997:340). The 10-item measure of everyday discrimination captures day-to-day experiences such as being “treated with less respect,”“threatened or harassed,” and “called names” (Williams et al. 1997). These experiences are more common than experiences of major discrimination (e.g., housing discrimination, being fired because of race, or being denied a bank loan) (Kessler, Mickelson, and Williams 1999). Nonetheless, chronic stressors, such as everyday discrimination, are a significant threat to well-being because of their insidious, continuous, and persistent nature (Wheaton et al. 2013).

It is well established that experiences of everyday discrimination have detrimental effects on mental health. Perceived discrimination is directly related to psychological distress, depressive symptoms, anxiety disorders, and major depressive disorders (Banks, Kohn-Wood, and Spencer 2006; Kessler et al. 1999; Richard Banks, Eberhardt, and Ross 2006; Williams et al. 1997). Moreover, perceived discrimination during the transition to adulthood can have detrimental effects on mental health (i.e., depression and anxiety symptoms) decades later (Assari 2017). Discrimination, in part, contributes to poorer health by exposing individuals to trauma, stress, structural strain (perceptions of structural and institutional barriers), and maladaptive coping styles (Ahmed, Mohammed, and Williams 2007; Hill et al. 2017).

While experiences of everyday discrimination are commonplace for many black adolescents (Seaton et al. 2008), recent evidence suggests that the level and intensity of everyday discrimination varies by skin tone within racial groups (Monk 2015; Uzogara and Jackson 2016; Uzogara et al. 2014). For instance, Monk (2015) finds that darker skin toned black Americans perceive more everyday discrimination than lighter skin toned black Americans and that everyday discrimination predicts worse mental health outcomes. Thus, the impact of skin tone on mental health, in part, operates through experiences of everyday discrimination. From a social stress perspective, the causal mechanism linking darker skin tone to mental health disadvantages is thought to lie in the harmful effects of chronic experiences of race-based discrimination. Furthermore, discriminatory experiences set in motion physiological responses that eventually result in poor mental health outcomes (Mays, Cochran, and Barnes 2014; Williams and Mohammed 2009; Williams, Neighbors, and Jackson 2008). The perception of blocked opportunity based on skin tone may put darker skin toned black adolescents at particular risk for mental health problems because it signals to these youths that they are entering a world with structural barriers that may limit their ability to achieve their goals (Tatum 2010). Thus, experiences of discrimination at this stage in the life course may be particularly harmful due to the fact that adolescents are beginning to develop a sense of their position in the larger social hierarchy (Tatum 2010).

Skin Tone, Discrimination, and Mastery

The impact of a stressor, such as discrimination, on mental health in part depends on one’s ability to cope with stressful circumstances (Pearlin et al. 2007). Coping is the process of attempting to keep the strain of stressful events within “manageable bounds” (Pearlin 1989:250). While coping efforts can be both problem-focused and emotion-focused, their main function is to “master, tolerate, reduce, minimize the demands of a stressful environments” (Folkman and Lazarus 1980:223). The presence of coping resources may explain how similar stressful conditions can be more or less harmful to the mental health of individuals (Pearlin 1989, 1999). However, coping resources, like social stress, are socially patterned by status characteristics (Pearlin and Schooler 1978), suggesting that individuals with darker skin tone will have less access to coping resources relative to their lighter skin tone counterparts.

In this article, I evaluate whether mastery functions as a personal coping resource that mediates the impact of skin tone on depression and mental disorder. Mastery is characterized as a “generalized belief that most circumstances are under one’s personal control” (Thoits 2010:S46). Perceived control over one’s life is essential for inhibiting stress (Mirowsky and Ross 2003; Pearlin 1999). Indeed, research suggests that persons with higher levels of mastery view stressors as less threatening because mastery encourages active problem solving (Mirowsky and Ross 2003; Thoits 2010). Thus, individuals with high levels of mastery persevere in the face of challenges, anticipate and avoid problems, and seek the necessary social support in stressful times (Mirowsky and Ross 2003; Schieman and Plickert 2008; Thoits 2010). As such, it is unsurprising that those with higher levels of mastery are less depressed than those with lower levels of mastery, even when they are faced with persistent problems (Pearlin et al. 1981). One’s sense of mastery, however, may be undermined by persistent exposure to unfair treatment, which in turn may foster a sense of powerlessness (Keith et al. 2010). Individuals who feel powerless are unlikely try to attempt to actively solve their problems because they feel that they have little control over the circumstances in their lives (Ross, Mirowsky, and Pribesh 2001). Thus, discrimination can have detrimental effects on health by eroding psychosocial resources that would otherwise protect individuals from the impact of stressors on negative mental health outcomes (Keith et al. 2010; Ryff, Keyes, and Hughes 2003; Thompson and Keith 2001). Although there is a dearth of research on whether levels of mastery vary by skin tone in black adolescents, it is plausible, given the higher rates of discrimination experienced by darker skin toned black Americans (Monk 2015), that black Americans with darker skin tone will have lower levels of mastery relative to their lighter skin toned peers.

Adolescence, Skin Tone, Discrimination, and Mastery

The process by which adolescents learn the meaning and value of skin tone is undoubtedly complex. Monk (2015) suggests that individuals come to understand their place in the status hierarchy based on their treatment by others, which is “thoroughly relational and dynamic” (p. 412). As adolescents begin to gain greater autonomy from their family units, peer networks, in particular, become important to the development of their self-concepts (Call and Mortimer 2001). In addition to serving as a source of social support and entertainment (Call and Mortimer 2001), peer networks likely introduce important ideas about the meaning of skin tone. These ideas may be implicit or explicit. Through processes of social comparison, youths may formulate ideas about the meaning of their skin tone in relation to their peers. For instance, the disparate treatment of black Americans with darker skin tone in terms of teacher engagement as well as punishment (Blake et al. 2016; Tatum 2004) may signal to adolescents that having darker skin tone is a stigmatized status. Thus, youths may develop ideas about skin tone by observing how their peers are treated both by adults and by other peers. Adolescents also may learn about skin tone in face-to-face interactions with peers. Reflected appraisals, or how people think they appear to others, powerfully shape identity formation (Khanna 2004; Monk 2015). Thus, differential treatment between peers based on skin tone also may influence powerfully how youths understand the implications of their skin tone in the social hierarchy. For example, black Americans with lighter skin tone may be viewed as more attractive or smarter by peers (Stephens and Few 2007), thus reinforcing colorist notions of the superiority of individuals with light skin tone.

Skin tone bias toward darker skin tone black Americans may have negative repercussions for the development of mastery. Black youths who perceive a disconnect between their efforts and their ability to achieve desired outcomes may feel ineffectual in shaping circumstances in their lives (Mirowsky and Ross 2003; Ross and Broh 2000). As such, discrimination based on uncontrollable characteristics, such as skin tone, may undermine personal agency or mastery (Keith et al. 2010). In the current study, it is hypothesized that both higher levels of discrimination and lower levels of mastery among darker skin toned adolescents will result in skin tone differences in depression and mental disorder.

Depressive Symptoms and Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) Mental Disorder

The current study employs two measures of mental health: depressive symptoms and any DSM-IV mental disorder, due to evidence that suggests that the use of diverse outcomes provides a more comprehensive understanding of the association between social risk factors and mental health in general (Aneshensel, Rutter, and Lachenbruch 1991; Erving and Thomas 2018; Turner, Wheaton, and Lloyd 1995). The skin tone patterning of mental health may differ depending on the outcome under study, which would be masked in a single-outcome study. In addition, depressive symptoms and any DSM-IV mental disorders are distinct measures of mental health. While depressive symptoms capture feelings of sadness and malaise (Mirowsky and Ross 2003), any DSM-IV disorder captures a large swath of the universe of mental illnesses including general anxiety disorder and attention deficit hyperactivity disorder, for example. Furthermore, the depressive symptoms measure shows the entire distribution of depression in the population (Mirowsky and Ross 2002), while the disorder measure is indicative of severe impairment and dysfunction beyond a given threshold (Horwitz 2007; Kessler 2002). Thus, the disorder measure is suggestive of those in the population who have severe mental health problems. It is important to note that the any disorder measure is not skewed by gender differences in the expression of mental health problems because it includes disorders common to both women (mood and anxiety disorders) and men (impulse control and substance use disorders) (Kessler et al. 2005). Thus, the any disorder measure serves as a generalized measure of disorder and balances out gender-specific pathways of disorder expression more than individual disorders.

The Present Study

This study determines whether there are significant associations between skin tone and depression/disorder in a population of black adolescents. Specifically, I examine the hypothesis that black Americans with very light skin tone have better mental health than their peers with darker skin tone. Second, I test the role of everyday discrimination and mastery as mediators in explaining the relationship between skin tone and depression and skin tone and mental disorder. Two research questions guide this inquiry: (1) Is skin tone associated with depression and mental disorder in a sample of black adolescents? and (2) What role does everyday discrimination and mastery play in explaining the relationship between skin tone and depression and skin tone and mental disorder in a population of black adolescents?

Sample

Data come from NSAL-A. The survey took place between 2001 and 2004 among a nationally representative sample of black Americans aged 13 through 17. Data collection was based on a multistage area probability sample using a stratified and clustered sample design (Jackson et al. 2016). In-person and telephone interviews were conducted with African American and Caribbean American adolescents who were attached to the National Survey of American Life adult households. The response rate was 80.6 percent. NSAL-A surveyed 1,170 adolescents. ¹ In the current analyses, 7 cases were excluded from the analytic sample because they were missing on the dependent variables. This resulted in an analytic sample of 1,163 respondents.

Mental Health Outcomes

Two measures of mental health were included: depressive symptoms and DSM-IV mental disorder. Depressive symptoms (α = .667) were measured using a 12-item measure of the Center for Epidemiological Studies Depression Scale (CES-D) (Radloff 1977)—a widely used and highly reliable index (Mirowsky and Ross 2003). The 12-item measure of CES-D depression is a continuous variable. Respondents were asked, “How often have you felt this way during the past week,” and responses were (1) I felt that I was as good as other people, (2) I had trouble keeping my mind on what I was doing, (3) I felt depressed, (4) I felt that everything was an effort, (5) I felt hopeful about the future, (6) my sleep was restless, (7) I was happy, (8) people were unfriendly, (9) I enjoyed life, (10) I had crying spells, (11) I felt that people disliked me, and (12) I could not get “going.” The items ranged from 0 to 3, where 0 = rarely or none of the time, 1 = some or a little of the time, 2 = occasionally, and 3 = most of the time. Items 1, 5, 7, and 9 were reverse coded so that higher scores would indicate higher levels of depression. Responses were averaged to produce an index of depression scored 0 to 36.

Mental disorder was defined as any diagnosis of a DSM-IV mental disorder in the past 12 months, determined by psychiatric interviews. Respondents were administered a version of the World Health Organization Composite International Diagnostic Interview Version 3.0 for DSM-IV (Kessler et al. 2005). This variable is dichotomous, with 1 indicating that the respondent had met the criteria for mental disorder in the past 12 months and 0 indicating that the respondent had not met the criteria for disorder in the past 12 months. The four categories of diagnoses were mood disorders (major depressive episode, bipolar I, and bipolar II), anxiety disorders (panic disorder, agoraphobia, social phobia, general anxiety disorder, posttraumatic stress disorder, and separation anxiety disorder), impulse control disorders (oppositional defiant disorder, conduct disorder, attention deficit disorder, intermittent explosive disorder), and substance use disorders (alcohol abuse, alcohol dependence, drug abuse, drug dependence).

Self-assessed Skin Shade

A self-assessed measure of skin shade was used. Respondents were asked, “Compared to most black teenagers your age, what shade of skin do you believe you have—is that very dark brown, dark brown, medium brown, light brown, or very light brown?” The skin tone measure consists of five categories that were dummy coded with very light brown as the reference category. This coding preserves the possibility of threshold effects in the effect of skin tone that are not captured by the linear approach.

Everyday Discrimination

Everyday discrimination (α = .845) was measured using the 10-item everyday discrimination scale (Williams et al. 1997). Respondents were asked, “In your day-to-day life how often have the following things happened to you?” and responses were (1) you are treated with less courtesy than other people, (2) you are treated with less respect than other people, (3) you receive poorer services than other people at restaurants or stores, (4) people act as if you are not smart, (5) people act as if they are afraid of you, (6) people act as if they think you are dishonest, (7) people act as if they’re better than you are, (8) you are called names or insults, (9) you are threatened or harassed, and (10) you are followed in stores. The items range from 0 to 5 where 0 = almost every day, 1 = at least once a week, 2 = few times a month, 3 = few times a year, 4 = less than once a year, and 5 = never. All items were reverse coded so that a higher score indicates more experiences of everyday discrimination. Responses were averaged to produce an index of everyday discrimination score that ranged from 0 to 50.

Mastery

Mastery (α = .671) was measured using the seven-item Pearlin mastery scale (Pearlin and Schooler 1978). Respondents were asked, “How strongly do you agree or disagree with these statements about yourself,” and responses were (1) there is really no way I can solve some of the problems I have, (2) sometimes I feel that I’m being pushed around in life, (3) I have little control over things that happen to me, (4) I can do just about anything I really set my mind to, (5) I often feel helpless in dealing with the problems of life, (6) what happens to me in the future mostly depends on me, and (7) there is little I can do to change many of the important things in my life. The items range from 0 to 3 where 0 = strongly agree, 1 = somewhat agree, 2 = somewhat disagree, and 3 = strongly disagree. Items 4 and 5 were reverse coded so that higher scores indicate higher levels of mastery. Responses were averaged to provide an index of mastery that ranged from 0 to 21.

Control Variables

I control for gender, age, geographic region, and family income. Gender was coded as a dummy variable (1 = female, 0 = male). Age was included as a continuous variable ranging from 13 to 17. Family household income was defined in relation to the poverty line. Low income (reference group) was defined as less than 1.5 times the official federal poverty line; low-average income as 1.5 to 3.0 times the poverty line; high-average income as 3.0 to 6.0 times the poverty line; and high income as 6.0 or more times the poverty level. To measure family income in relation to the poverty line, participants’ parents/guardians were asked to self-report their family household income. Geographic region was defined as South and North (reference group). The weighted descriptive statistics for all variables in the analysis are provided in Table 1.

OPEN IN VIEWER

Table 1.

Descriptive Characteristics of the Study Population: National Survey of American Life–Adolescent Supplement, 2001–2003 (N = 1,163).

Variable	Mean	Standard Deviation	Minimum–Maximum
Dependent variable
CES-D depression	9.032	5.465	0–30
DSM-IV mental disorder	0.216	0.412	0–1
Focal independent variable
Very light brown	0.067	0.251	0–1
Light brown	0.231	0.422	0–1
Medium brown	0.459	0.499	0–1
Dark brown	0.184	0.388	0–1
Very dark brown	0.058	0.235	0–1
Mediator
Everyday discrimination	11.144	9.376	0–45
Mastery	15.131	3.880	0–21
Control
Age	14.985	1.405	13–17
Female	0.500	0.500	0–1
Household income
< 1.5 times poverty	0.477	0.500	0–1
> 1.5 and ≤ 3.0 times poverty	0.273	0.450	0–1
> 3.0 and ≤ 6.0 times poverty	0.177	0.383	0–1
> 6.0 times poverty	0.031	0.175	0–1
South	0.526	0.500	0–1

Note. CES-D = Center for Epidemiological Studies Depression Scale; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, fourth edition.

Analytic Approach

Data modeling proceeded in several steps. First, I present weighted descriptive statistics to describe the analytic sample (see Table 1). In the next set of analyses, I present the average levels of discrimination and mastery in the sample by skin tone (see Figure 1). Next, I used ordinary least squares regression to specify the association between skin tone and depression. Linear probability modeling was used to estimate the skin tone–mental disorder relationship because linear probability models are not limited by a fixed variance of the residual and thus allow for comparisons across models (Mood 2010). I ran three models for each outcome: (1) a model that entered skin tone, (2) a model that entered discrimination, and (3) a final model that added skin tone, discrimination, and mastery (see Table 2 and Table 3). All multivariate analyses account for age, gender, household income, geographic region, and incorporated appropriate survey weights.

OPEN IN VIEWER

Figure 1.

Discrimination and mastery by skin tone.

OPEN IN VIEWER

Table 2.

Center for Epidemiological Studies Depression Scale Depression Regressed on Skin Tone, Discrimination, and Mastery (N = 1,163).

	(1) + Skin Tone	(2) + Skin Tone, Discrimination	(3) + Skin Tone, Discrimination, Mastery
Panel A	b (SE)	b (SE)	b (SE)
Light brown a	1.020 (0.698)	1.002 (0.679)	0.473 (0.600)
Medium brown a	0.759 (0.659)	0.766 (0.641)	0.266 (0.567)
Dark brown a	1.256 (0.723)	1.168 (0.704)	0.610 (0.622)
Very dark brown a	2.826** (0.907)	2.504* (0.883)	1.372 (0.782)
Discrimination		0.137***(0.017)	0.078***(0.015)
Mastery			−0.661***(0.037)
Household income
> 1.5 and ≤ 3.0 times poverty	0.070(0.380)	0.054(0.369)	0.503(0.326)
> 3 and ≤ 6 times poverty	−1.685***(0.443)	−1.754***(0.431)	−0.904*(0.383)
> 6 times poverty	−1.451(0.931)	−1.395(0.906)	−0.586(0.800)
Female	0.241(0.324)	0.632*(0.319)	0.234(0.283)
Age	0.241*(0.114)	0.127(0.112)	0.178(0.099)
South	−0.302(0.322)	−0.016(0.316)	−0.226(0.279)
Constant	4.810**(1.841)	4.700**(1.792)	15.125***(1.683)
R 2	.027	.079	.284
Panel B
Skin tone differences
Very dark versus light brown	1.805*(0.743)	1.502*(0.724)	0.898(0.640)
Very dark versus medium brown	2.067**(0.704)	1.738**(0.686)	1.106(0.606)
Very dark versus dark brown	1.570*(0.758)	1.336(0.738)	0.762(0.652)
Dark brown versus light brown	0.236(0.504)	—	—
Dark brown versus medium brown	0.497(0.446)	—	—
Medium brown versus light brown	−0.261(0.407)	—	—

Note. Unstandardized regression coefficients with standard errors are shown in parentheses. Dashes indicate that mediators are not considered.

a

Compared to very light brown skin tone.

*

p < .05. **p < .01. ***p < .001.

OPEN IN VIEWER

Table 3.

Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Mental Disorder Regressed on Skin Tone, Discrimination, and Mastery (N = 1,163).

	(1) + Skin Tone	(2) + Skin Tone, Discrimination	(3) + Skin Tone, Discrimination, Mastery
Panel A	b (SE)	b (SE)	b (SE)
Light brown a	−.018(.053)	−.019(.052)	−.036(.051)
Medium brown a	−.070(.050)	−.070(.049)	−.086(.048)
Dark brown a	−.064(.055)	−.070(.054)	−.088(.052)
Very dark brown a	.043(.068)	.023(.067)	−.015(.066)
Discrimination		.009***(.001)	.007***(.001)
Mastery			−.022***(.003)
Household income
> 1.5 and ≤ 3 times poverty	−.025(.029)	−.026(.028)	−.010(.028)
> 3 and ≤ 6 times poverty	−.020(.033)	−.024(.033)	.004(.032)
> 6 times poverty	.019(.070)	.023(.068)	.050(.068)
Female	−.023(.024)	.001(.024)	−.012(.024)
Age	.035***(.009)	.027**(.009)	.029***(.008)
South	−.046(.024)	−.028(.024)	−.035(.024)
Constant	−.214(.139)	−.221(.136)	.126(.142)
Panel B
Skin tone differences
Very dark versus light brown	.061(.056)	—	—
Very dark versus medium brown	.113*(.053)	.093(.052)	.072(.051)
Very dark versus dark brown	.107(.057)	—	—
Dark brown versus light brown	−.046(.038)	—	—
Dark brown versus medium brown	.006(.034)	—	—
Medium brown versus light brown	−.052(.031)	—	—

Note. Estimates are based on the linear probability model. Unstandardized regression coefficients with standard errors are shown in parentheses. Dashes indicate that mediators are not considered.

a

Compared to very light brown skin tone.

*

p < .05. **p < .01. ***p < .001.

Research Implications

The research findings from this study have practical implications. The finding that black adolescents with very dark skin tone experience higher levels of discrimination and have higher levels of depression than their lighter skin toned peers suggests that policies should aim to reduce skin tone discrimination. One potential way to reduce discrimination against black adolescents with very dark skin tone would be to include more black Americans with darker skin tones in media images. Increased exposure to diverse skin tone imagery has important implications for how we think about particular groups (Louie and Wilkes 2018; Moore 2008). For example, increased exposure to positive messaging about stigmatized groups helps to reduce the stigmatization of these groups (Hu et al. 2015). The classroom context also has been identified as an important environment where skin tone discrimination takes place (Hunter 2016). As such, policy should seek to inform teachers about the mental health consequences of skin tone bias, especially among black adolescents with very dark skin tone (Blake et al. 2016; Hunter 2016). Bringing a critical lens in teacher training to the consequences of colorism on the mental health of black adolescents may have practical implications for the way teachers interact with students of color (Hunter 2016).

Limitations and Future Directions

There are several limitations that deserve mention. First, the measure of skin tone is self-reported; thus, we do not know the respondent’s objective skin tone. While some scholars have argued (Monk 2015) that self-assessed skin tone is preferred because it reflects an “embodied social status” based on how others have treated an individual over his or her life course, it is also possible self-assessed skin shade may be confounded with exposure to discrimination and/or depression/disorder. Thus, self-assessed measures of skin tone make it difficult to determine whether the skin tone respondents report is caused by experiences of discrimination and/or depression/disorder or whether skin tone results in greater exposure to discrimination and/or depression/disorder (Klonoff and Landrine 2000). Indeed, some research suggests that individuals who experience higher levels of depression identify as darker than their interviewer-rated skin color, suggesting that how one assesses his or her skin tone may be influenced by one’s social context (Monk 2015). It also is plausible that greater exposure to discrimination could influence one’s self-assessed skin shade, indicating that black adolescents who experience higher levels discrimination perceive their skin tone to be darker than black adolescents who experience lower levels of discrimination (e.g., Penner and Saperstein 2008). Because darker skin tone is associated with devalued status (Hunter 2007), individuals who experience higher levels of discriminatory treatment may identify as darker, regardless of their actual skin tone. Low levels of mastery may both directly and indirectly influence skin tone evaluations. For instance, individuals who feel like they do not have control over the course of their life may rate their skin tone as darker than those with higher levels of mastery because they have more negative attributions about their life chances (Penner and Saperstein 2008). As well, individuals with lower levels of mastery often feel powerless in their social environment (Mirowsky and Ross 2003), which may in turn increase attributions of discriminatory treatment, resulting in these individuals’ rating their skin tone as darker than those with higher levels of mastery. This would suggest that mental health status, discriminatory experiences, and mastery play a role in shaping skin tone identification (Penner and Saperstein 2008).

Second, the data are cross-sectional, and thus it is impossible to make definitive statements about the causal ordering of discrimination, mastery, and depression/disorder. Although it is possible that individuals who are more depressed perceive more discrimination and have lower levels of mastery, the majority of the evidence suggests that discrimination is a strong predictor of mastery and depression/disorder (Brown et al. 2000; Kessler et al. 1999; Schulz et al. 2006). As such, I invoke the often-used assumption that chronic stressors affect resources more than vice-versa (Pearlin et al. 1981). In addition, because of the cross-sectional nature of the data I do not know if the association between skin tone, everyday discrimination, mastery, and depression/disorder will change as the adolescents in this sample age. There is some indication that mastery decreases as individuals age (Mirowsky and Ross 1992, 2007), yet if mastery decreases at the same rate in all groups, then the findings reported in the current study will not change. Finally, it should be noted that these results are limited to adolescents aged 13 through 17, and thus the generalizability of these results may not reflect the skin tone patterning of depression/disorder in adult populations.

Future research should explore other coping resources that may explain the association between skin tone and health in black adolescents. It is possible that coping resources, such as religious involvement, self-esteem, and social support, also may be socially patterned by skin tone ² (Keyes 2009; Mouzon 2013, 2017; Sellers 2006). If adolescents with darker skin tone have fewer of these coping resources than do their peers with lighter skin tone and these coping resources are associated with lower levels of depression/disorder, then these coping resources may mediate the impact of discrimination on depression/disorder between black adolescents with very dark brown skin tone and black adolescents with lighter skin tones.

Second, scholars should aim to more extensively measure the association between skin tone and different types of stressors. It is likely that this study underestimates the role of stressors in explaining skin tone differences in depression/disorder by measuring only discrimination (Taylor and Turner 2002). Thus, future research should be attentive to chronic stressors such as financial strain and ambient threats (Taylor and Turner 2002) as well as traumatic event stressors (Turner et al. 1995) and ambiguous discrimination (Thomas 2015). A more comprehensive measure of the “stress universe” (Wheaton 1994) will better capture whether the costs of skin tone extend beyond experiences of everyday discrimination.

Future research should be attentive to contextual factors that may influence the skin tone patterning of depression/disorder. For instance, experiences of colorism likely differ depending on both the racial and the socioeconomic composition of one’s neighborhood. It is plausible that adolescents with darker skin tone in primarily middle-class white contexts may experience higher levels of discrimination than their lighter skin toned peers (Lacy 2004), yet these youths also may have well-developed personal coping strategies in part due to the parental work that middle-class black parents do to foster self-esteem and racial pride in their children (Dow 2016; Lacy 2004). These concerted strategies may reduce the impact of discriminatory experiences on the mental health of black adolescents with very dark skin tone residing in these contexts. It should be noted that I tested a skin tone by racial composition of school ³ interaction in predicting depression/disorder, but no significant interactions were observed.

The study of skin tone and mental health in adolescents would benefit from the use of qualitative methods. In-depth interviews may provide further insights into the contexts in which skin tone is salient in the lives of black youths and the contexts in which skin tone is less salient (Brubaker 2002; Loveman 1999). For example, skin tone may be particularly relevant in peer-to-peer contexts, where social comparisons and romantic courtship take place (Adams et al. 2016; Hunter 2016), but not as salient within the family context where physical appearance may be less relevant. In addition, attention to the contexts in which black adolescents perceive the most skin tone discrimination will provide useful information to policy makers about where interventions to reduce skin tone discrimination should be implemented.

Finally, there is some indication that the implications of skin tone may vary by gender (Keith et al. 2010; Thompson and Keith 2001). Due to data sparseness, I was unable to test a skin tone by gender interaction using the five skin tone categories employed in the current study. However, in supplementary analyses (not shown) I recoded skin tone into three categories: (1) light brown skin tone, consisting of very light brown skin tone and light brown skin tone; (2) medium skin tone, consisting of medium brown skin tone; and (3) dark skin tone, consisting of dark brown skin tone and very dark brown skin tone. I tested a skin tone by gender interaction in predicting depression and mental disorder using these three categories, but no significant skin tone by gender interaction interactions were observed. However, future research should attempt to elucidate whether the relationships examined here are gender-specific using all five skin tone categories.