Measuring Latino Adolescents’ Coping With Ethnic-Racial Discrimination

Abstract

Few studies have examined coping strategies specific to ethnic-racial discrimination among Latino youth. The current study examined the psychometric properties of the Discrimination Coping Strategies Scale (DCSS) including its factor structure, validity, and longitudinal measurement invariance. Data came from four waves of a longitudinal study of 323 Latino adolescents (Wave 1 M_age = 15.31 years, SD = .76; 49.5% female). Exploratory factor analysis and confirmatory factor analysis suggested a factor structure that included a three-item latent factor and two manifest variables. Analyses suggested that the factor structure for the latent construct was invariant across four waves of repeated measures. Support for construct validity emerged with tests of convergent, divergent, and predictive validity. Taken together, findings provide support for the validity of the proactive coping subscale of the DCSS as a measure of Latino adolescents’ proactive coping with ethnic-racial discrimination.

Keywords

coping ethnic-racial discrimination Latino adolescent

Ethnic-racial discrimination, including prejudiced comments, negative stereotypes, and negative actions toward individuals based on their ethnic group membership (Brondolo, Ver Halen, Pencille, Beatty, & Contrada, 2009), is a key risk factor for ethnic-racial minority youth and increases during adolescence (Benner et al., 2018; Umaña-Taylor, 2016). Among Latino youth, experiences of and concerns about discrimination are salient stressors in their lives (Umaña-Taylor, Vargas-Chanes, Garcia and Gonzales-Backen, 2008). Prior research shows that up to 94% of Latino youth experienced or witnessed ethnic-racial discrimination by adolescence (Flores, Tschann, Dimas, Pasch, & de Groat, 2010), with many Latino youth reporting multiple experiences of discrimination (Edwards & Romero, 2008). Discrimination is related to increases in multiple forms of stress responses and these experiences during adolescence are negatively related to a variety of later outcomes including physical health, mental health, and academic achievement (DeGarmo & Martinez, 2006; Pascoe & Smart Richman, 2009).

However, the negative effects of discrimination can vary among youth, depending on their individual resources. For instance, coping (i.e., efforts to protect oneself from the adverse effects of stress) can buffer the adverse effects of discrimination on youth outcomes (see Brondolo et al., 2009 and Pascoe & Smart Richman, 2009 for reviews). Different ways of coping in adolescence may place individuals on distinct trajectories, and patterns of coping established in adolescence may be drawn upon throughout the life course (Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001). In the current study, we examined the use of various coping strategies among Latino youth, who currently comprise over 24% of the current youth population (U.S. Census Bureau, 2017) and will account for more than one third of the U.S. population by 2060 (Colby & Ortman, 2015). Given that estimates suggest that the prevalence of experiences with ethnic-racial discrimination is high among Latino adolescents (Flores et al., 2010), developing measures that examine how Latino youth cope with discrimination is important.

Theoretical Framework

We draw on the phenomenological variant of ecological systems theory (PVEST; Spencer, Dupree, & Hartmann, 1997) in framing this research. The PVEST model proposes that repeat exposure to discriminatory experiences during adolescence informs patterns of responses that shape individuals’ developmental trajectories and identities. By interacting with the world around them, adolescents form patterns for coping with stressful situations. These coping strategies, in turn, inform individuals’ identities and actions (Spencer et al., 1997). Here, we focus on primary and secondary coping strategies.

Primary control engagement strategies involve coping with the stressor or one’s emotions from the stressor directly (e.g., problem solving, emotional expression) and include strategies that are problem-focused (i.e., resolving problems related to the source of the stress), emotion-focused (i.e., emotions that were evoked by the negative event), and active (i.e., active behavior and cognitive attempts to deal with the threat, including talking to friends, disproving, self-affirmation; Pascoe & Smart Richman, 2009; Umaña-Taylor et al., 2008). Research suggests that these proactive coping strategies, in which individuals actively engage in efforts to resolve stressors (e.g., positive reappraisal, seeking social support), are generally related to lower levels of distress and can mitigate the effects of discrimination (Pascoe & Smart Richman, 2009). Moreover, as youth proactively cope with discrimination, they may have greater feelings of control over their lives leading to positive developmental outcomes (e.g., greater self-esteem, ethnic-racial identity, positive affect), including those related to academic success (Basáñez, Warren, Crano, & Unger, 2014; Dumont & Provost, 1999; Vera et al., 2012; You, Hong, & Ho, 2011).

In comparison, secondary control strategies include accepting or distracting oneself from the stressors (e.g., avoidance of the problem at hand, substance use; Compas et al., 2001; Edwards & Romero, 2008). Also known as passive or avoidance coping, these coping strategies are generally related to greater distress among adolescents (e.g., Pascoe & Smart Richman, 2009). Although these strategies may buffer the immediate negative effects of discrimination (e.g., negative emotions), they may be related to other health problems (e.g., substance abuse; Pascoe & Smart Richman, 2009). It is possible that individuals engage in these strategies when confronting the issue would take substantial time and energy (Feagin, 1991) or when they believe they have little control over their environments (Ojeda & Liang, 2014).

Coping With Ethnic-Racial Discrimination: Measures

Few studies have examined coping strategies specific to ethnic-racial discrimination (e.g., Yoo & Lee, 2005), with a limited subset of these studies reporting the psychometric properties of these scales. For example, Wei, Alvarez, Ku, Russell, and Bonett (2010) developed the Coping With Discrimination Scale (CDS) for use among an ethnic-racially diverse sample of college students. Psychometric evaluation of the 25-item CDS provided evidence of the validity of a five-factor structure, which included coping responses of education/advocacy, internalization, drug and alcohol use, resistance, and detachment (Wei et al., 2010). In other research, Noh, Beiser, Kaspar, Hou, and Rummens (1999) used principal components analysis of seven possible reactions to ethnic-racial discrimination among a sample of adult Southeast Asian refugees living in Canada. Analyses revealed two dimensions of coping strategies: active and passive. Although these measures represent important steps in understanding how individuals cope with discrimination, both measures were developed and validated on adult populations. Given the salience of ethnic-racial discrimination during adolescence (Edwards & Romero, 2008; Flores et al., 2010; Umaña-Taylor, 2016), measures that accurately assess these coping strategies among this age group are needed.

The Current Investigation

The current study examined the psychometric properties of the Discrimination Coping Strategies Scale (DCSS; Umaña-Taylor et al., 2008) among a sample of Latino adolescents. The DCSS was developed to assess both primary and secondary control coping strategies among adolescents (Umaña-Taylor et al., 2008). We examined the factor structure, validity, and longitudinal measurement invariance of the DCSS. First, we examined the factor structure of the DCSS. We expected three DCSS items (talking it out, thinking positively, and working to prove people wrong) to form a latent factor for proactive coping. In addition, we expected two secondary-coping strategies items from the DCSS (ignoring the situation; saying something rude) to form a separate latent factor. Then, we examined the longitudinal measurement invariance of the DCSS. Because discrimination increases during adolescence (Umaña-Taylor, 2016) and individuals’ strategies for coping continue to develop over the life course (Brondolo et al., 2009), it is necessary to ensure that measures are valid and reliable in longitudinal research.

We also examined the convergent, divergent, and predictive validity of the DCSS. To examine the convergent validity of the DCSS, we tested whether and how the DCSS related to more general coping strategies (i.e., those used for non-discrimination-related stress; measured by the Adolescent Coping Orientation for Problem Experiences Scale [A-COPES]; Patterson & McCubbin, 1987). We expected that the proactive coping latent factor would positively correlate with primary coping strategies measured by the family support subscale of the A-COPES, which focuses on talking through problems with family members (described in detail below). We expected that the secondary coping strategies latent factor would be positively correlated with measures from the A-COPES focused on avoiding problems and ventilating feelings, respectively (Edwards & Romero, 2008; Pascoe & Smart Richman, 2009).

We assessed divergent validity by examining correlations of each of the dimensions of the DCSS with the level of education in the adolescents’ neighborhoods. Prior research has shown that neighborhood structural qualities (e.g., levels of employment or education) may indirectly relate to Latino adolescents’ self-esteem, self-efficacy, and academic outcomes via perceptions of the neighborhood (e.g., Plunkett, Abarca-Mortensen, Behnke, & Sands, 2007). There is no conceptual reason, however, to anticipate that neighborhood education level would relate to adolescents’ coping strategies. Therefore, support for divergent validity would emerge if no statistically significant relation emerged between the DCSS measures and aggregate neighborhood education level.

In addition, we tested predictive validity by examining the relations between the DCSS and adolescents’ academic motivation. Prior theory and research supports a positive association between coping with discrimination and academic outcomes. First, according to PVEST (Spencer et al., 1997), youths’ strategies for coping are expected to inform their behaviors and identities through their self-system processes. Furthermore, empirical work with Latino youth has found that proactive approaches for coping (e.g., seeking social support) are positively linked with academic outcomes (DeGarmo & Martinez, 2006). Thus, we hypothesized that proactive coping strategies would be positively associated with academic motivation because these coping strategies would provide adolescents with adaptive solutions to stress that would enhance their positive behavioral outcomes, in line with the PVEST model (Spencer et al., 1997). For secondary coping strategies, we hypothesized that the use of these coping strategies would be negatively associated with adolescents’ academic outcomes as prior work has found these approaches to coping to be related to greater distress (Pascoe & Smart Richman, 2009), which can negatively affect adolescents’ academics.

Method

Sample

Data for the current study came from a four-wave longitudinal study, with data collected once per year, focused on Latino adolescents’ ethnic identity development (Umaña-Taylor, Alfaro, Bámaca, & Guimond, 2009). The sample consisted of 323 Latino adolescents (49.5% female) who were in the ninth or tenth grade at Wave 1 (W1). Adolescents were recruited from one of the five high schools in non-metropolitan communities in Illinois. In all five schools, White students represented the numerical majority (75% to 91%) and Latino students comprised the numerical minority (8% to 18%) of the student body. At W1, participants were approximately 15 years of age (M_age = 15.31 years, SD = .76). The sample was diverse in terms of ethnic background (e.g., Chilean, Puerto Rican), but a majority of participants (n = 249; 77.1%) were of Mexican origin. Over the four waves of data collection, the proportion of Mexican-origin individuals in the study was stable (76.6% at W2; 76.4% at W3; 76.5% at W4). Although acknowledging the diversity among Latinos is important (Umaña-Taylor & Fine, 2001), our relatively small sample size for national groups other than Mexican-origin youth made it impossible to conduct analyses separately by group. Furthermore, our focus was on adolescents’ strategies for coping with ethnic-racial discrimination, and we did not have a theoretical reason to expect that the use of such coping strategies would vary as a function of national origin. Thus, the current study focused on Latino adolescents as a pan-ethnic population and did not examine specific national origin groups. With respect to participant retention across waves, of those who participated in W1, 85.4% participated at W2 (n = 276), 82.7% participated at W3 (n = 267), and 80.5% (n = 260) participated at W4.

Procedure

Ninth- and tenth-grade students were identified as Latino through school records. Adolescents also provided their ethnic self-identification label in the survey. With the assistance of school principals, students were invited to attend an informational session at each school where the purpose of the study was explained, and parental consent and youth assent forms were distributed. Adolescents who returned parental consent and assent forms completed a self-administered survey, which took approximately 45 min to complete. The same procedure was followed across all four waves of data collection (for a more detailed account of this procedure, see Umaña-Taylor et al., 2009).

Measures

DCSS

The five-item DCSS (Umaña-Taylor et al., 2008) was used to measure primary and secondary control coping strategies to deal with ethnic-racial discrimination. Adolescents were asked to record how often they used any of the five strategies listed (see Table 2 for items) on a 5-point Likert-type scale (1 = never to 5 = very often).

Generalized coping

The A-COPES (Patterson & McCubbin, 1987) was used to measure the frequency with which adolescents used behaviors to manage difficult situations or problems. Prior research shows the A-COPES is valid and reliable for use among adolescents in the Midwest (Patterson & McCubbin, 1987) and among Latino youth (Copeland & Hess, 1995). Three subscales of the A-COPES measure were used in the current investigation: Family problems, ventilating feelings, and avoiding problems. Adolescents were asked to record how often, on a 5-point Likert-type scale (1 = never to 5 = most of the time), they used specific behaviors to cope with difficulties or tense feelings across seven subscales. Six items measured solving family problems and included “Try to reason with my parents and talk things out; compromise” (α = .75). Ventilating feelings was measured with six items including “Say mean things to people; be sarcastic” (α = .76). Finally, avoiding problems was measured by five items including “smoke” (α = .73). These subscales of the A-COPES were selected because they aligned conceptually with the items and hypothesized subscales of the DCSS. For example, the solving family problems subscale aligns with the problem-focused nature of primary coping strategies included in the DCSS.

Neighborhood education

The level of education in adolescents’ neighborhoods was measured by one item that asked adolescents to assess the level of education in their neighborhood at W2. This item came from a measure focused on adolescent resilience in multicultural communities (Plunkett, Abarca, & Sands, 2004). Scores ranged from 0 = no education to 11 = graduate degree (M = 7.34, SD = 2.05).

Academic motivation

Academic motivation was assessed at W1 and W2 using five items measuring adolescents’ efforts toward school, the importance of grades and education, the extent to which they finish homework, and whether they like school (Plunkett & Bámaca-Gómez, 2003). Items were scored on a 4-point Likert-type scale (1 = strongly disagree to 4 = strongly agree). In the current sample, Cronbach’s alpha was .81 at W1 and .80 at W2. Analyses used a mean score across the five items (see Table 1 for descriptive statistics).

Table 1.

Descriptive Statistics on Variables of Interest Across Waves.

	n	M (SD)	Skewness (SE)	Kurtosis (SE)
Solving family problems W1	309	3.31 (0.81)	−.23 (0.14)	−.32 (0.28)
Solving family problems W2	268	3.11 (0.78)	.28 (0.15)	−.55 (0.30)
Avoiding problems W1	314	2.29 (0.85)	1.01 (0.14)	.45 (0.27)
Avoiding problems W2	264	2.16 (0.79)	1.24 (0.15)	1.20 (0.30)
Ventilating feelings W1	312	2.83 (0.82)	.19 (0.14)	−.25 (0.28)
Ventilating feelings W2	261	2.58 (0.72)	.38 (0.15)	.18 (0.30)
Neighborhood education W2	312	7.14 (2.26)	−.94 (0.14)	1.20 (0.28)
Academic motivation W1	273	2.60 (0.55)	−.40 (0.15)	−.11 (0.29)
Academic motivation W2	319	3.09 (0.59)	−.54 (0.14)	.21 (0.27)

Note. W1 = Wave 1; W2 = Wave 2.

Results

Preliminary Analyses

We present the analytic description prior to presenting the analyses. To examine the factor structure of the DCSS, the sample was randomly split using SPSS 24 built in procedures (approximately 50% of all cases). The split sample was used to conduct an exploratory factor analysis (EFA; n = 158) and a confirmatory factor analysis (CFA; n = 160). Five participants were missing data on all W1 DCSS variables and were excluded from all analyses. Although factor analysis is often conducted with large sample sizes, factor analysis has been shown to produce reliable results with sample sizes below 50 (de Winter, Dodou, & Wieringa, 2009). Subsequent analyses were conducted in Mplus Version 8. Table 1 presents descriptive statistics for the variables used in our tests of convergent, divergent, and predictive validity.

In all analyses, missing data were estimated using full information maximum likelihood (Muthén, Muthén, & Asparouhov, 2016). Preliminary analyses showed that those who were missing data at any wave had lower academic motivation, t(317) = 3.44, p ⩽ .01, mean difference = .24; were older, t(321) = −3.01, p < .01, mean difference = .27; and had higher scores for avoiding problems, t(318) = −2.74, p = .01, mean difference = .28 at W1 than those who were not missing data at any wave. No other differences emerged.

EFA

Using W1 data for half of the sample (n = 158), we conducted an EFA on the five DCSS items. To determine the best model fit, we compared iterative models containing one- and two-factor solutions using geomin rotation. Geomin rotation was selected because it allows the factors to correlate and has been shown to produce unbiased and easily interpretable solutions for less complicated factor structures (Sass & Schmitt, 2010). Acceptable model fit was indicated by a root mean square error of approximation (RMSEA) with an upper limit of .07; a standardized root mean residual (SRMR) of less than .05; and a comparative fit index (CFI) and Tucker–Lewis index (TLI) above .90 (Little, 2013). We used an item loading cut-off criterion of .45, which is considered an indication of a “fair” estimate, with lower loadings considered to be “poor” (Comrey & Lee, 1992).

The single factor EFA model was a poor fit to the data, χ² = 17.00, df = 5, p < .01; CFI = .76; TLI = .53; SRMR = .07; RMSEA = .12 (90% CI = [.06, .19]). Four items loaded onto this factor but factor loadings for talk, proud, and rude were mostly poor (i.e., ranged from .25 to .37) and the fourth item, work hard, had a factor loading of .94. The fifth item, ignore, did not load onto this factor.

The two-factor model indicated moderate model fit, χ² = 2.16, df = 1, p = .14; CFI = .98; TLI = .77; SRMR = .02; RMSEA = .09 (90% CI = [.00, .25]). Three items (i.e., talk, proud, work hard) loaded onto the first factor with loadings that met the cut-off criteria of .45. Conceptually, these items reflected proactive coping strategies. The remaining two items, “dealing with discrimination by saying something rude right back to the person” and “ignoring the situation” did not load onto either factor. Although both items were conceptually characteristic of secondary coping strategies, they did not collectively load on a single factor. We interpreted this to mean that these items (rude, ignore) should be treated as manifest variables. The eigenvalues steeply declined after the first factor (1 = 1.65, 2 = 1.18, 3 = .95, 4 = .67) suggesting that a single-factor model may fit the data. However, given that model fit improved from the single-factor to the two-factor model (e.g., CFI, SRMR), we proceeded to the next step (i.e., CFA) in which we tested a latent construct (i.e., proactive coping strategies defined by three items) and two manifest variables (i.e., rude and ignore). Table 2 presents the factor loadings for the two-factor EFA solution.

Table 2.

Factor Loadings for the Final EFA and CFA Models.

		EFA two-factor solution^a (n = 158)		CFA^b (n = 160)
		Factor 1	Factor 2	CFA^b (n = 160)
	“How often have you used the strategy of. . . ”
Proactive coping	“talking to the person to clarify the misconception or stereotype?”	.57 *	.00	.62**
	“just thinking about all the positive things that you have to offer and being proud of yourself?”	.54 *	−.05	.69**
	“working hard to prove discriminatory people wrong?”	.47 *	.25	.58**
Secondary coping	“dealing with discrimination by saying something rude right back to the person?”	.00	1.23	—
Secondary coping	“ignoring the situation?”	.12	−.09	—

Note. EFA = exploratory factor analysis; CFA = confirmatory factor analysis.

Geomin rotation.

Standardized estimates.

p < .05. **p < .01. The significance values in bold are indicated by *.

CFA

The second half of the sample (n = 160) was used to conduct a CFA to examine the validity of the factor structure that was identified in the EFA. We conducted a CFA that included all five items, mirroring the structure that emerged in the EFA—one three-item latent factor (i.e., identified by the three items assessing proactive coping strategies) and two remaining items (i.e., the items assessing secondary coping strategies) modeled as manifest variables. The latent factor and the two manifest variables were allowed to covary with one another, χ² = 7.90, df = 4, p = .10; CFI = .95; TLI = .88; SRMR = .04; RMSEA = .09 (90% CI = [.00, .16]).

Longitudinal Measurement Invariance

Given the findings above, we examined longitudinal measurement invariance of the latent construct that emerged for the three-item proactive coping strategies subscale of the DCSS. Longitudinal measurement invariance implies that the relation between the manifest indicators and the latent variable are invariant over time and is necessary to ensure that observed longitudinal differences in the latent variable represent actual changes in the measured construct (Little, 2013). We followed the procedures outlined by Little (2013) for longitudinal invariance testing. First, we fit an alternative null model to generate a baseline chi-squared analysis that included the assumption of no covariances among items and specified that the means and variances of the same items were equivalent over measurement occasions. The chi-squared analysis for the alternative null model is used in the computation of the CFI for subsequent invariance tests instead of the null model produced through the Mplus built-in procedures (Little, 2013). Second, we examined configural invariance (i.e., whether the pattern of fixed and free parameters was the same across time). Third, we examined whether the factor loadings were the same across waves (i.e., weak factorial invariance). Fourth, we constrained factor loadings and intercepts to be equal across time. A change in the CFI of .01 or greater and a statistically significant chi-squared analysis indicate that the measure does not meet the threshold for invariance (Little, 2013).

We examined the repeated measures invariance of the three-item latent factor measured by the DCSS across four waves of data. Table 3 compares the fit criteria for the null, configural, loading, and intercept models. The change of <.01 in CFI and the nonstatistically significant chi-squared analysis across these models indicated that the measure was invariant across four waves of data.

Table 3.

Analyses Examining Longitudinal Measurement Invariance (n = 323).

	χ²(df)	RMSEA [CI]	TLI	SRMR	AIC	BIC	CFI	Pass/fail (ΔCFI ⩽ .01)	Δχ²
Null	744.07 (84)**	.156 [.146, .166]	.209	.228	10,411.264	10,433.930
Configural	39.511 (30)	.031 [.000, .056]	.968	.035	9,814.702	10,041.361	.991
Loading	46.069 (36)	.029 [.000, .052]	.972	.041	9,809.260	10,013.253	.991	Pass (ΔCFI = .001)^a	6.558
Intercept	53.130 (42)	.029 [.000, .050]	.973	.044	9,804.321	9,985.648	.990	Pass (ΔCFI = .001)^b	7.061

Note. Model fit compares the longitudinal configural, loading, and intercept measurement invariance models for the three-item latent factor for proactive coping. The change in CFI column (ΔCFI) compares each subsequent test of invariance; RMSEA = root mean square error of approximation; TLI = Tucker–Lewis index; SRMR = standardized root mean residual; AIC = Akaike information criterion; BIC = Bayesian information criterion; CFI = comparative fit index.

compares the configural and loading models.

compares the loading and intercept models.

p < .01.

Convergent Validity

We examined the cross-sectional and longitudinal correlations of the DCSS and the A-COPES measures using W1 and W2, as the A-COPES was only administered at these waves. Consistent with conceptual notions, the proactive coping strategies subscale of the DCSS was positively and significantly correlated (within and across waves) with the six-item solving family problems A-COPES subscale (Table 4). The DCSS item of ignore had a small positive relation with the solving family problems subscale but only at W1. No other statistically significant relations emerged for this item. Rude had moderate positive relations with the five-item avoiding problems and the six-item venting feelings subscale within W1 and from W1 to W2. Furthermore, rude had a small negative relation to solving family problems within W2.

Table 4.

Longitudinal Correlations Between the DCSS (i.e., Proactive, Ignore, Rude) and the A-COPES Subscales and Neighborhood Education.

		Wave 1			Wave 2
		Family	Avoid	Vent	Family	Avoid	Vent	Neighborhood
Wave 1	Proactive	.41**	−.11*	.01	.24**	−.10	−.03	.07
	Ignore	.15**	−.11	−.06	.11	−.05	−.06	.02
	Rude	−.04	.31**	.41**	−.04	.25**	.33**	.02
Wave 2	Proactive	.18**	−.08	−.08	.26**	−.09	.02	.04
	Ignore	.08	.01	−.06	.00	.03	.04	.01
	Rude	−.12	.25**	.29**	−.15*	.30**	.32**	−.08

Note. n ranged from 249 to 318. DCSS = Discrimination Coping Strategies Scale; A-COPES = Adolescent Coping Orientation for Problem Experiences Scale; Proactive = Proactive Coping; Ignore = ignoring the situation; Rude = dealing with discrimination by saying something rude right back to the person; Family = solving family problems; Avoid = avoiding problems; Vent = ventilating feelings; Neighborhood = Neighborhood education.

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

Divergent Validity

Divergent validity was assessed by examining the cross-sectional and longitudinal relations of the DCSS and the single item measuring neighborhood level of education. Evidence of divergent validity emerged, as neighborhood level of education was not correlated with proactive coping at W1 or W2, ignore at W1 or W2, nor with rude at W1 or W2 (Table 4).

Predictive Validity

A structural equation model (SEM) was used to examine whether the proactive coping latent factor and the manifest items of rude and ignore predicted adolescents’ later academic motivation. Adolescents’ age, gender, GPA, and academic motivation at W1 were used as covariates. Model fit was acceptable, χ² = 30.44, df = 14, p = .01; CFI = .95; TLI = .89; SRMR = .03; RMSEA = .06 (90% CI = [.03, .09]). In support for predictive validity, results of the SEM showed a significant positive relation between proactive coping at W1 and the five-item academic motivation scale at W2, above and beyond the effects of the covariates. There was no statistically significant relation between rude or ignore and adolescents’ later academic motivation. This model is presented in Figure 1.

Figure 1.

Longitudinal relations between the DCSS measures and academic motivation (n = 323).

Discussion

Coping with ethnic-racial discrimination is a salient developmental task for Latino adolescents (Umaña-Taylor, 2016). Moreover, these coping strategies may buffer the effects of discrimination on youth’s academic, health, and economic outcomes (e.g., Brietzke & Perreira, 2017; Brondolo, Blair, & Kaur, 2018; Edwards & Romero, 2008). However, few measures have been developed to examine coping specific to ethnic-racial discrimination (e.g., Noh et al., 1999; Wei et al., 2010) and none, to the authors’ knowledge, have been validated for use with adolescents. In the current investigation, we sought to fill this gap by examining the factor structure, longitudinal measurement invariance, and validity of the DCSS measure among a sample of Latino adolescents. Results provide support for the use of the DCSS to assess proactive coping strategies with a three-item subscale. Support for the items capturing secondary coping strategies was less clear.

Primary Control Coping Strategies

The latent factor representing primary control engagement strategies (i.e., the proactive coping subscale) was supported by EFA, CFA, and longitudinal measurement invariance analyses, aligning with prior theory and research (Pascoe & Smart Richman, 2009). As hypothesized, the proactive coping subscale was positively correlated with the family support subscale of the A-COPES, which focuses on talking through problems with family members. The proactive coping subscale also was negatively related to avoiding problems, aligning conceptually with the problem-focused nature of primary control strategies (Pascoe & Smart Richman, 2009). However, this relation was only present at W1. Further research is needed to understand why this relation did not persist over time. Finally, the SEM (i.e., our test of predictive validity) showed that there was a positive relation between proactive coping and academic motivation. These findings offer support to the idea that positive, stable strategies for coping with ethnic-racial discrimination may inform individuals’ actions and are consistent with the PVEST model (Spencer et al., 1997). Moreover, these findings demonstrate one potential pathway by which proactive coping with ethnic-racial discrimination may affect adolescents’ academic outcomes.

Secondary Control Coping Strategies

Our hypothesis that the items capturing ignoring the situation and saying something rude would load onto a latent factor was not supported. It is unclear why these items did not reflect a latent factor. However, results from our EFA and CFA analyses did support the idea that these strategies for coping with ethnic-racial discrimination were distinct from the proactive coping latent factor. Given that these two items did not form a latent factor, we included them as manifest variables to examine their validity. Results of our test for divergent validity supported our hypothesis (i.e., that neighborhood education was not related to these items). Although we hypothesized that rude and ignore would be positively correlated with avoiding problems and ventilating feelings, our hypotheses were only partially supported. Whereas rude had moderate positive relations with avoiding problems both cross-sectionally and longitudinally, ignore was unrelated to the measures of interest except for a small positive correlation with solving family problems at W1. It is possible that whereas rude taps into coping strategies representative of dysregulated emotions (Marceau et al., 2015), ignore taps into a behavioral dimension of avoidant coping distinct from substance use (which is a main component of the A-COPES avoiding problems subscale). This idea is further supported by the relation of rude to the A-COPES ventilating feelings subscale.

In addition, our hypotheses regarding predictive and convergent validity were not supported for rude or ignore. Results indicated that that neither rude nor ignore were related to academic motivation. Notably, there was a negative correlation between rude and ignore. It may be that the use of these strategies, and the outcomes associated with their use, is context dependent. For example, although ignoring the situation could lead to greater internalizing problems (Compas et al., 2001), it may also be an adaptive strategy in situations where youth have less power or control over their environment (e.g., Ojeda & Liang, 2014). Alternately, being rude may lead to social ostracization or punishment depending on the context (e.g., school), which may create cascade effects in other domains of the adolescents’ life. Further research is needed to better understand the processes by which these coping strategies may relate to developmental outcomes.

Despite limited support for the secondary coping strategies items, the current findings underscore the need for measures specific to ethnic-racial discrimination among adolescents, and replicate prior research showing only moderate correlations between coping strategies specific to ethnic-racial discrimination and more generalized measures of coping (Wei et al., 2010). Indeed, our findings indicate that these more generalizable coping strategies may not be interchangeable with situation-specific strategies (Connor-Smith, Compas, Wadsworth, Thomsen, & Saltzman, 2000; Wei et al., 2010). Furthermore, the longitudinal measurement invariance of the proactive coping subscale provides evidence supporting the use of this latent factor over time. Future research may examine whether and how adolescents’ strategies for coping with ethnic-racial discrimination develop over time and in response to contextual factors (i.e., discriminatory events).

Limitations and Conclusions

Given the relations between ethnic-racial discrimination and youth outcomes, accurately measuring and understanding the strategies youth use to cope with these experiences is important. The DCSS has the potential to advance the literature by providing a short, psychometrically sound measure for examining proactive strategies for coping with ethnic-racial discrimination throughout middle to late adolescence. Nevertheless, limitations include the use of a convenience sample of Latino adolescents in the Midwest who were attending schools in which they were a numerical minority, which limits generalizability. In addition, due to the small sample size for different Latino groups, our findings are based on a Latino pan-ethnic categorization. It will be important for future studies to consider whether the findings are consistent for different Latino ethnic groups in the United States. Future research examining the DCSS with larger, more diverse samples (i.e., youth from other ethnic-racial backgrounds and different geographical and schooling contexts) is needed.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Elana R. McDermott

References

Basáñez

Warren

M. T.

Crano

W. D.

Unger

J. B.

(2014). Perceptions of intragroup rejection and coping strategies: Malleable factors affecting Hispanic adolescents’ emotional and academic outcomes. Journal of Youth and Adolescence, 43 1266-1280. doi:10.1007/s10964-013-0062-y

Benner

A. D.

Wang

Shen

Boyle

A. E.

Polk

Cheng

(2018). Racial/ethnic discrimination and well-being during adolescence: A meta-analytic review. American Psychologist, 73, 855-883.

Brietzke

Perreira

(2017). Stress and coping: Latino youth coming of age in a new Latino destination. Journal of Adolescent Research, 32, 407-432. doi:10.1177/0743558416637915

Brondolo

Blair

Kaur

(2018). Biopsychosocial mechanisms linking discrimination to health: A focus on social cognition. In Major

Dovidio

J. F.

Link

B. G.

(Eds.), Handbook of stigma and discrimination (pp. 219-240). New York, NY: Oxford University Press.

Brondolo

Ver Halen

N. B.

Pencille

Beatty

Contrada

R. J.

(2009). Coping with racism: A selective review of the literature and a theoretical and methodological critique. Journal of Behavioral Medicine, 32, 64-88. doi:10.1007/s10865-008-9193-0

Colby

S. L.

Ortman

J. M.

(2015). Projections of the size and composition of the U.S. population: 2014 to 2060. Washington, DC: U.S. Department of Commerce Economics and Statistics Administration, U.S. Census Bureau.

Compas

B. E.

Connor-Smith

J. K.

Saltzman

Thomsen

A. H.

Wadsworth

M. E.

(2001). Coping with stress during childhood and adolescence: Problems, progress, and potential in theory and research. Psychological Bulletin, 127, 87-127. doi:10.1037/0033-2909.127.1.87

Comrey

A. L.

Lee

H. B.

(1992). A first course in factor analysis (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Connor-Smith

J. K.

Compas

B. E.

Wadsworth

M. E.

Thomsen

A. H.

Saltzman

(2000). Responses to stress in adolescence: Measurement of coping and involuntary stress responses. Journal of Consulting and Clinical Psychology, 68, 976-992. doi:10.1037/0022-006X.68.6.976

10.

Copeland

E. P.

Hess

R. S.

(1995). Differences in young adolescents’ coping strategies based on gender and ethnicity. Journal of Early Adolescence, 15, 203-219.

11.

DeGarmo

D. S.

Martinez

C. R.

(2006). A culturally informed model of academic well-being for Latino youth: The importance of discriminatory experiences and social support. Family Relations, 55, 267-278. doi:10.1111/j.1741-3729.2006.00401.x

12.

de Winter

J. C. F.

Dodou

Wieringa

P. A

. (2009). Exploratory factor analysis with small sample sizes. Multivariate Behavioral Research, 44, 147-181. doi:10.1080/00273170902794206

13.

Dumont

Provost

M. A.

(1999). Resilience in adolescents: Protective role of social support, coping strategies, self-esteem, and social activities on experience of stress and depression. Journal of Youth and Adolescence, 28, 343-363.

14.

Edwards

L. M.

Romero

A. J.

(2008). Coping with discrimination among Mexican descent adolescents. Hispanic Journal of Behavioral Sciences, 30, 24-39.

15.

Feagin

J. R.

(1991). The continuing significance of race: Antiblack discrimination in public places. American Sociological Review, 56, 101-116.

16.

Flores

Tschann

J. M.

Dimas

J. M.

Pasch

L. A.

de Groat

C. L.

(2010). Perceived racial/ethnic discrimination, posttraumatic stress symptoms, and health risk behaviors among Mexican American adolescents. Journal of Counseling Psychology, 57, 264-273. doi:10.1037/a0020026

17.

Little

T. D.

(2013). Longitudinal structural equation modeling. New York, NY: Guilford Press.

18.

Marceau

Zahn-Waxler

Shirtcliff

A. E.

Schreiber

J. E.

Hastings

Klimes-Dougan

(2015). Adolescents’, mothers’, and fathers’ gendered coping strategies during conflict: Youth and parent influences on conflict resolution and psychopathology. Developmental Psychopathology, 27, 1025-1044. doi:10.1017/S0954579415000668

19.

Muthén

B. O.

Muthén

L. K.

Asparouhov

(2016). Regression and mediation analysis using Mplus. Los Angeles, CA: Muthén & Muthén.

20.

Noh

Beiser

Kaspar

Hou

Rummens

(1999). Perceived racial discrimination, depression, and coping: A study of Southeast Asian refugees in Canada. Journal of Health and Social Behavior, 40, 193-207.

21.

Ojeda

Liang

C. T.

(2014). Ethnocultural and gendered determinants of coping among Mexican American adolescent men. Psychology of Men & Masculinity, 15, 296-304. doi:10.1037/a0033293

22.

Pascoe

E. A.

Smart Richman

(2009). Perceived discrimination and health: A meta-analytic review. Psychological Bulletin, 135, 531-554. doi:10.1037/a0016059

23.

Patterson

J. M.

McCubbin

H. I.

(1987). Adolescent coping style and behaviors: Conceptualization and measurement. Journal of Adolescence, 10, 163-186.

24.

Plunkett

S. W.

Abarca

Sands

(2004). Neighborhood qualities related to youth outcomes in Latino youth. Unpublished Manuscript, California State University Northridge.

25.

Plunkett

S. W.

Abarca-Mortensen

Behnke

A. O.

Sands

(2007). Neighborhood structural qualities, adolescents’ perceptions of neighborhoods, and Latino youth development. Hispanic Journal of Behavioral Sciences, 29, 19-34.

26.

Plunkett

S. W.

Bámaca-Gómez

M. Y.

(2003). The relationship between parenting, acculturation, and adolescent academics in Mexican-origin immigrant families in Los Angeles. Hispanic Journal of Behavioral Sciences, 25, 222-239. doi:10.1177/0739986303025002005

27.

Sass

D. A.

Schmitt

T. A.

(2010). A comparative investigation of rotation criteria within exploratory factor analysis. Multivariate Behavioral Research, 45, 73-103.

28.

Spencer

Dupree

Hartmann

(1997). A phenomenological variant of ecological systems theory (PVEST): A self-organization perspective in context. Development and Psychopathology, 9, 817-833.

29.

Umaña-Taylor

A. J.

(2016). A post-racial society in which ethnic-racial discrimination still exists and has significant consequences for youths’ adjustment. Current Directions in Psychological Science, 25, 111-118. doi:10.1177/0963721415627858

30.

Umaña-Taylor

A. J.

Alfaro

E. C.

Bámaca

M. Y.

Guimond

A. B.

(2009). The central role of familial ethnic socialization in Latino adolescents’ cultural orientation. Journal of Marriage and Family, 71, 46-60. doi:10.1111/j.1741-3737.2008.00579.x

31.

Umaña-Taylor

A. J.

Fine

M. A.

(2001). Methodological implications of grouping Latino adolescents into one collective ethnic group. Hispanic Journal of Behavioral Sciences, 23, 347-362. doi:10.1177/0739986301234001

32.

Umaña-Taylor

A. J.

Vargas-Chanes

Garcia

C. D.

Gonzales-Backen

(2008). A longitudinal examination of Latino adolescents’ ethnic identity, coping with discrimination, and self-esteem. Journal of Early Adolescence, 28, 16-50. doi:10.1177/0272431607308666

33.

U.S. Census Bureau. (2017). The Hispanic population in the United States: 2014. Retrieved from https://www.census.gov/data/tables/2014/demo/hispanic-origin/2014-cps.html

34.

Vera

E. M.

Vacek

Blackmon

Coyle

Gomez

Jorgenson

. . . Steele

J. C.

(2012). Subjective well-being in urban, ethnically diverse adolescents the role of stress and coping. Youth & Society, 44, 331-347. doi:10.1177/0044118X11401432

35.

Wei

Alvarez

A. N.

Russell

D. W.

Bonett

D. G.

(2010). Development and validation of a Coping With Discrimination Scale: Factor structure, reliability, and validity. Journal of Counseling Psychology, 57, 328-344. doi:10.1037/a0019969

36.

Yoo

H. C.

Lee

R. M.

(2005). Ethnic identity and approach-type coping as moderators of the racial discrimination/well-being relation in Asian Americans. Journal of Counseling Psychology, 52, 497-506. doi:10.1037/0022-0167.52.4.497

37.

You

Hong

(2011). Longitudinal effects of perceived control on academic achievement. The Journal of Educational Research, 104, 253-266. doi:10.1080/00220671003733807