Subjective Well-Being,Critical Race Theory,and the Assessment of Measurement Invariance Across Racial Groups

Abstract

The present study explores how Black and White youth respond to measures of subjective well-being within the context of critical race theory (CRT). Three levels of measurement invariance (i.e., configural, metric and scalar) were examined for indicators of subjective well-being. We hypothesized that there would be limited measurement invariance across groups based on the premise established by CRT that youth of color experience and perceive life differently than their White peers, which was supported. The findings of this study demonstrate that the measures work as expected within groups, but there is a considerable lack of invariance across groups. This study also provides some evidence that racial/ethnic differences cannot be taken for granted when assessing SWB in youth.

Keywords

critical race theory subjective well-being youth life satisfaction

Introduction

Diener (1984) theorized that subjective well-being (SWB) is comprised of positive affect, negative affect, and life satisfaction. Researchers later found that these three constructs interact in a way that affects a number of life outcomes (Long et al., 2012). Educational engagement (Antaramian et al., 2010), academic performance (S. M. Suldo & Shaffer, 2008), and severe psychological and behavioral disturbances such as suicidal ideation (Eryilmaz, 2011) are some of the outcomes that have been linked to SWB. Additionally, the constructs associated with SWB, such as life satisfaction, have been found to be predictive of future behavior (Diener et al., 2013). The consequences associated with such outcomes make the assessment of SWB in young people vital to helping them navigate the path to adulthood. The often-marginalized social status of youth of color, especially those living in economically challenged communities, indicates that SWB has the potential to be an important screening and outcome variable for those seeking to develop and implement interventions to address at-risk youth of color in schools and community settings. Additionally, the evaluation of the psychometric properties of potential measures of SWB is critical in determining the utility of the construct as a screening and outcome variable.

Life circumstances associated with race/ethnicity provide a social context that impacts how youths perceive SWB (Bradley & Corwyn, 2004). Yet, it is assumed that the constructs measured when assessing SWB (e.g., positive affect, negative affect, and life satisfaction) are being conceptualized similarly across racial/ethnic groups. Critical race theory (CRT) posits that youth of color are perpetually subjected to oppression that invariably impacts how they are socialized compared to their White peers. Therefore, conceptualizations of SWB and related constructs are likely to be varied across racial groups due to how they are engaged by social institutions that are inherently biased. Although comparisons across demographic characteristics are standard in research (e.g., race, gender, and class), methodologies that apply a critical race lens are not (Hood & Hopson, 2008). The present study explores how Black and White youth respond to measures of SWB using data from Wave I of the National Longitudinal Study of Adolescent to Adult Health (Add Health) data set. Our rationale for analyzing the data from the Add Health study is 2-fold. First, the variables observed in the study were operationalized in a way that allowed us to test the hypothesis and observe how well tenets of CRT are reflected in a nationally represented sample of adolescents. Second, even though the data were collected in 1994, few studies since have explored the potential implications of racial differences in the measurement of SWB (Vera et al., 2008), particularly using a nationally representative sample. We use a critical race lens to contribute to the idea that CRT can be useful in developing research that challenges academic and professional discourses that maintain institutional biases.

Literature Review

This study seeks to investigate how Black and White youth perceive SWB with consideration for the value and implications of assessing SWB in all youth. CRT is an appropriate framework due to the understanding it provides of systemic trends and practices that affect various social groups in different ways within hierarchies rooted in the social construction of race. The SWB of youth is affected at least in part by how they interact with social institutions (Durlak et al., 2011; Shin et al., 2010), which consequently impacts their development and how they respond to the choices and opportunities made available to them. Investigating this phenomenon is warranted given the implications of any assumptions that youth are being treated equally despite the evidence that they are not (Chambers, 2009; Gillborn, 2015; Leonardo, 2013). To explore these ideas further, we provide a brief overview of CRT and discuss how CRT can be used to explain trends that are prevalent within social institutions (e.g., schools). The discussion then expands to consider the challenges of measuring the constructs associated with SWB and how race further complicates those issues.

Theoretical Framework: Critical Race Theory

CRT was developed soon after the Civil Rights Movement in the mid-20th century yielded a number of legal victories for marginalized people in the United States. These successes were countered with bias in social institutions where lawmakers and other human rights duty bearers actively sought to disenfranchise people of color through policy and praxis (McCoy & Rodericks, 2015). A major tenet of CRT is that race is socially constructed. The widespread acceptance of race as a social construct has contributed to the maintenance of Whiteness as a standard that all other racial groups are expected to aspire to (Leonardo, 2013). This standard is presently enshrined in social institutions (e.g., legislative, educational, and criminal justice) and has been accepted by most people, regardless of racial or ethnic identity (Leonardo, 2012).

Rhetoric that endorses the neutral principles of colorblindness and meritocracy conceal racial inequities by promoting the assumption that individuals’ abilities are sufficient to succeed and acquire access to any needed resource (Solórzono & Yosso, 2001b). Furthermore, those who ascribe to colorblindness and meritocracy give no credence to ideas that individuals are systematically persecuted (e.g., law enforcement) or disenfranchised (e.g., redlining and voter suppression) based on their race or ethnicity (Riley & Ettlinger, 2011). The consequences of these inequities maintain a social system that institutionalizes a hierarchy based on privilege that venerates Whiteness at the expense of people of color, with additional implications regarding matters of gender and class (Gillborn, 2015; Ledesma & Calderón, 2015). CRT seeks to identify, explain, and inform methods to undo these disparities (e.g., Gillborn, 2006; Jay, 2003).

Critical Race Theory and Education

A number of scholars have examined education within the context of CRT (Ladson-Billings & Tate, 1995; Ledesma & Calderón, 2015; Leonardo, 2013; Riley & Ettlinger, 2011; Solórzono & Yosso, 2001a; Stovall, 2016). Gillborn (2013) explored and demonstrated the concept of interest-divergence, which occurs when White powerholders perceive an advantage in the deepening of racial inequalities at times of economic uncertainty. He hypothesized that interest-divergence would be present in educational policy development in the United Kingdom. His findings pointed to evidence of interest-divergence in English education system reforms. Jay (2003) applied a CRT lens when she discussed how education systems in the United States utilize a hidden curriculum to reinforce a hegemony that reinforces a social hierarchy rooted in White privilege. In other words, youth of color often receive implicit messages in school that promote the assumed cultural and intellectual superiority of the dominant group. These messages are built into the application of multicultural education modalities that set the values of the dominant group as the standard while otherizing those of groups deemed subordinate.

Consequently, efforts to deliver a more egalitarian educational experience are undermined by the implicit messages of White supremacy that maintain systemic oppression and disparities within education systems (Jay, 2003). Ledesma and Calderón (2015) found similar results in their review of literature that examined the development of CRT in elementary, secondary, and postsecondary education. They concluded that CRT “problematizes objectivity and exposes how colorblind and postracial ideologies, that envelope daily discourse, work to maintain privilege and protect White supremacy” (Ledesma & Calderón, 2015, p. 218).

Measurement of Subjective Well-Being

The conceptualization of SWB used in this study comes from Diener (1984). He described SWB as having three hallmarks. First, it is subjective. Second, it includes both positive and negative measures of affect. Third, it includes a global measure of the individual’s overall life satisfaction. Measurement of SWB has proven challenging given the subjective nature and varied conceptualizations of its components (e.g., life satisfaction; Diener et al., 2013; Eid & Diener, 2004), not to mention the existential difficulty of quantifying happiness (Diener, 1984; Haybron, 2011; Ozmete, 2011). In fact, there are many competing theories and conceptualizations of well-being across various disciplines (Jayawickreme et al., 2012; Theofilou, 2013). Nonetheless, the literature provides ample evidence of the critical role that SWB plays in the youth outcomes (Suldo & Shaffer, 2008; Suldo et al., 2011) and shows that “SWB deserves special attention for its role in positive youth development as an indicator, as a predictor, as a moderator/mediator, and ultimately a positive outcome” (Park, 2004, p. 36).

For instance, high SWB when compared to low SWB is associated with positive interpersonal functioning, lower presence of emotional or behavioral problems, and more adaptive cognitive and behavioral functioning (Busseri & Sadava, 2013). Eryilmaz (2012) found that satisfying the needs of high school students paralleled increased SWB. The constructs making up SWB have also been found to be associated with both academic performance and engagement (Antaramian et al., 2010; Long et al., 2012). Thus, research has shown that youth’s SWB affects their intentions to engage academically, which subsequently impacts their academic achievements. The value of enhancing and maintaining the SWB of young people can be seen in the guidelines from the Children’s Bureau of the U.S. Department of Health and Human Services, which recommend that the promotion of children’s well-being should be one of the primary goals for child welfare services (Lou et al., 2008). The concern raised by implementing these assessments in practice is whether or not the outcomes associated with them are generalizable across racial groups (Bradley & Corwyn, 2004), especially considering the various measures of SWB that exist (Pontin et al., 2013).

Race/Ethnicity and Subjective Well-Being

The socialization process for youth of color provides a context for their social and emotional development (Griffith et al., 2013; Neblett et al., 2012). Various adverse environmental conditions continually impact the SWB for young people of color (Oliver, 2006). The compound influence of disparities in educational attainment and economic mobility negatively affect outcomes in people of color (Hull et al., 2008; Vanderbilt et al., 2013). Poverty has been linked to truant behaviors in youth, increasing the likelihood of poor performance in school, school dropout, future unemployment, juvenile delinquency, and other high-risk behaviors (Newsome et al., 2008). The risk factors associated with poverty are exacerbated by racial identity in that poor youth of color on average have educational experiences that are qualitatively different than White youth when it comes to mental health diagnoses, assessment of learning needs, and punishment for disruptive behaviors in school (Tucker & Dixon, 2009).

Even when services are made available to address these issues, they usually are not provided in evidence-based ways that account for students’ cultural background (Koffman et al., 2009). This is likely due to the lack of psychometric measures that are culturally sensitive (Villodas et al., 2011). Few studies have investigated the multiple contextual influences that impact SWB in ethnically diverse youth (Vera et al., 2008). This is problematic because…

[i]t is known that there are cultural differences in life satisfaction that remain after controlling for objective conditions such as income…To parse these differences into artificial response tendencies that are unrelated to true life satisfaction versus real differences in life satisfaction is challenging, in part because the response tendencies in part reflect differences in how people feel and think about the world and themselves. (Diener et al., 2013, p. 517)

Furthermore, T. N. Brown (2008) discussed the issue of mental health assessment and the assumption of racial invariance using a CRT framework. He posited that “the meaning and measurement of mental health depends on race or varies because of racism…Thus, by ignoring the phenomenology of race and racism, the representation of psychological dysfunction [or well-being] may be skewed” (p. 57).

Subjective Well-Being and Measurement Invariance

Observing measurement invariance across groups begins with establishing that the pattern of relationships between the indicators and the constructs being measured across the groups are the same. This is the first in a series of multigroup analyses being used to evaluate the measurement invariance of the measure. When configural measurement invariance exists, there is evidence that on the face of it, the groups are conceptualizing the construct in a similar manner (Meredith, 1993). Metric measurement invariance tests the equality of the degree of relationship between items and the latent constructs (e.g., life satisfaction, positive affect, and negative affect). When metric measurement invariance is present, there is evidence that the items have a consistent interpretation across the groups (Byrne, 2008). Scalar measurement invariance builds on metric measurement invariance by adding equivalency of the item intercepts to the test of measurement invariance. Intercepts found not to be equivalent across the groups indicate that participants in one group tend to respond systematically higher or lower to items on the measure despite the fact that their level on the latent variable is the same (Meredith & Teresi, 2006). A significant lack of invariance within a measure calls for either different item scoring processes in order to validly represent participants’ item response sets or a complete reconstruction of the scale for completely novel item sets.

The utility of measuring SWB with adolescents is grounded in the assumption that the same underlying latent meaning of scales can be empirically established for all groups. This is necessary for the measurement relationship between the observed items and the true scores established for each group to have the same theoretical structure and psychological meaning across groups. The validity of the inferences that we can make when comparing Black youth to White youth on the three underlying constructs making up SWB (i.e., life satisfaction, positive affect, and negative affect) is dependent on the nature of the equivalence of the psychometric meaning of the measures across the two groups. Otherwise, observed differences in scale means and correlations cannot be attributed to CRT causes, but instead perhaps the differences in interpretation of the items between different subpopulations. Without empirical assessment, each argument is equivocally viable and requires further empirical validation. Ensuring the presence of invariance is critical because assuming that SWB is being measured similarly across racial groups could lead to the maintenance of practices that reinforce institutional biases that negatively affect youth of color. Measurement invariance and differential item functioning (DIF) are presented as differential item loadings and patterns expressed conditionally upon individual group memberships, such that each group has its own unique parameters. Therefore, participants who are otherwise identical in latent abilities measured by the items, except for possessing differential group memberships, are tested as to whether they respond differentially to equivalent items.

The lack of empirical investigation into the hypothesized between-group parallels in the interpretation of respondents’ perceived SWB results in a strong and previously unchallenged assertion about research with diverse samples. Namely, it is claimed that the scores between divergent gender and racial groups have identical and exchangeable meaning. Unless explicitly examined, however, the latent constructs which the items are presumed to reflect are assumed to be equivalently determined solely by the mean score differences themselves. Yet, if the items of specific scales are not parallel across the groups of interest, this would indicate that differences in local group interpretation misleadingly bias the perception of meaning for researchers, often with no readily apparent recourse for determining the presence of such biases. Finally, it is important to note that CRT not only calls into question how measures are constructed but how the results can be interpreted in ways that often support and maintain systemic inequity (Hood & Hopson, 2008).

Purpose and Hypothesis

The purpose of this study was to evaluate the measurement invariance of a set of indicators of the three domains making up SWB (i.e., life satisfaction, positive affect, and negative affect) for Black and White youth. We will also assess the measurement invariance of items that are indicative of differential or biased treatment at school in order to contextualize participant responses within a CRT framework. We hypothesize that there will be limited measurement invariance across groups on measures of SWB based on the premise established by CRT that youth of color experience and perceive life differently than their White peers. Therefore, we find it reasonable to expect that divergent perspectives would also exist on perceived acceptance at school within the sample. Three levels of measurement invariance: configural, metric (weak metric), and scalar (strong metric) will be examined for a set of indicators for life satisfaction, positive affect, and negative affect. The empirical presence of noninvariance within the measures will be empirically tested using the item response theory (IRT) based local goodness-of-fit statistic for between groups equivalence of models (Joe, 1997). Significant misfit across groups would thus represent differences in measurement model performance across subgroup strata, indicating the presence of biased interpretability under sum-score methodologies which assume a common scoring relation throughout the entire sample.

We acknowledge that intersectionality (i.e., the convergence of an individual’s multiple identities) impacts the way that people experience the world (Solórzono & Yosso, 2002). McCoy and Rodericks (2015) explain that CRT cultivates an understanding of how experiences of oppression related to racial identities intersect with other experiences of oppression that target subordinated identities (classism, xenophobia, transphobia, etc.). Our study aims to begin exploring potential differences across groups by limiting the differences observed to Black/White racial and gender (i.e., male and female) dichotomies in order to demonstrate that race and gender alone are enough to detect instances of measurement noninvariance. If that is the case, it stands to reason that observing additional demographic variables would increase the likelihood of measurement invariance occurring within a sample. We also acknowledge that utilizing secondary data presents a challenge in that the data were not collected for the purposes of this particular study. However, secondary data provide an opportunity to use a critical race lens to investigate the ways that assumptions about how race and the components of SWB are conceptualized could potentially contribute to the maintenance of inequity and hegemonic narratives.

Method

Sample

This study involved the analysis of data from the Feelings Scale included in the youth’s in-home questionnaire used in the first wave of the Add Health study (Harris & Udry, 2009). Add Health is a longitudinal study that began in 1994 with a nationally representative sample of 7th through 12th grade participants, examining social, economic, psychological, and physical well-being within the context of the family, school, and community. For the purpose of this study, the sample was restricted to youth who self-identified as Black or White in the middle adolescent years (between the ages of 15 and 17). Those restrictions resulted in a sample of 717 Black males and 766 Black females, as well as 1,972 White males and 2,066 White females. In the sample, each of the ages (i.e., 15, 16, and 17) was comparably represented in the sample to the national census. From the complete data sample of 6,504 individuals, 322 (4.95%) were removed for responding as being multiracial or of indeterminate race; of the remaining 6,182 subjects, a further 661 (10.7%) were removed as self-identifying as neither Black nor White. Thus, 15.11% of the total sample was removed from all further analyses. Approximately 82% of the parents reported having enough money to pay their bills and the average household income was approximately US$48,640 (standard deviation = 57,620). Table 1 provides information regarding the demographics of the study’s sample.

Table 1.

Descriptive Statistics.

Race … Gender	n	Approximate Percentage	Mean (SD)
Black males	717	.13	—
Black females	766	.14	—
White males	1,972	.36	—
White females	2,066	.38	—
Household income
Enough to pay bills	—	.82	—
Average household income	—	—	US$48,640 (57,620)

Measures

Subjective well-being measure (SWBM)

The SWBM consisted of three subscales assessing life satisfaction, positive affect, and negative affect.

Life satisfaction

Four items were selected from the Feelings Scale to operationalize life satisfaction. Participants were asked to use a 4-point rating scale to assess the extent to which he or she had experienced a particular feeling within the previous week on a scale ranging from 0 = never to 3 = most/all of the time. For example, participants were asked how frequently they felt hopeful about the future in the past week.

Positive affect

Nine items from the youth questionnaire were used to create the Positive Affect subscale. The first 2 items (“You felt that you were just as good as other people” and “You were happy”) were taken from the Feelings Scale. Participants responded to these items using a 4-point rating scale where they approximated how frequently they had experienced those feelings within the previous week with 0 = never and 3 = most/all of the time. The remaining 7 items were taken from the Personality and Family section of the questionnaire (e.g., Item 5, “You have a lot to be proud of”; and Item 7, “You feel you are doing everything just right”). Participants used a 5-point Likert-type scale, where 1 corresponded with strongly disagree and 5 corresponded with strongly agree. A high score on the Positive Affect subscale indicated that participants were reporting a high degree of positive affect.

Negative affect

Four items were selected from the Feelings Scale. Example items are “You felt that you could not shake off the blues, even with help” and “You felt like people disliked you.” Participants responded to these items using a 4-point rating scale where they approximated how frequently they had experienced those feelings within the previous week with 0 = never and 3 = most/all of the time. High ratings indicated more negative affective experiences.

Perceived bias

Five items were selected to assess whether respondents felt that they were accepted and supported by their teachers and student peers. Example items are “I feel like I am part of this school,” “The students at this school are prejudiced,” and “The teachers at this school treat students fairly.” Responses ranged from 1 = strongly agree to 5 = highly disagree.

Analysis Strategies/Data Analyses

Confirmatory factor analysis has been the primary method of measurement scale validation within the social sciences since its introduction in 1967 (see Jöreskog, 1967). IRT allows for latent variable measurement (Lazarsfeld, 1955), focusing upon the ability to interpret and score individuals equivalently between subpopulations, rather than upon the total score structure as a composite. By examining a scale as both a population approximation from the sample and as a collection of individuals, we are able to discount the model misspecification from operational utilization. To assess configural, metric, and structural invariance in the observed measures, the analyses were handled within two different procedures using R software (Version 3.5.1) mIRT (Version 1.29; Chalmers, 2012) package for all estimation, which compared the measurement structures of the empirical forms to the hypothetical model structure.

Our first analysis assessed the unidimensional item structure of the SWB Scale as described in the previous section, for which significance was set at α = .05 using the limited information goodness-of-fit statistic (Joe, 1997; see also Maydeu-Olivares & Joe, 2008), which performs equivalently to the likelihood ratio test within IRT frameworks. As a consequence of this empirical misfit, a second model was then tested within the established IRT framework, along with an additional two latent variables hypothesized for the additional items that assessed perceived bias. This structure was constructed as a purely exploratory analysis that enabled the unidimensional makeup of the items to be assessed and allowed for secondary loadings to be revealed. Once the measurement model from this exploratory analysis was established, measurement invariance was assessed to establish whether the latent makeup of the items under both the first and second model structures were equivalently interpretable and scorable across gender and racial identities. Evaluations of measurement invariance were conducted using the likelihood ratio statistic while using the standardized root mean square residual statistic to interpret local group differences in fit (see Holland & Wainer, 1993, for a more comprehensive overview of item level assessment of measurement invariance).

Results

The original item construction of the SWBM was composed of three unidimensional subscales with summative scoring are reported in the descriptive statistics of Table 2 for gender and racial subgroups along with 99% confidence intervals, which was chosen due to the size of the sample. The items were fitted within a multidimensional IRT framework, which still poorly served to replicate the observed data patterns, $χ_{(d f = 1, 616)}^{2}$ = 4,067.74, p < .0001. These results demonstrate strong inconsistencies between the theoretical operationalization of the scale and the performance of the scale in our sample. Such findings substantiate the need for an alternative factor structure to be established and, alternatively, the introduction of bias should be the original operationalization persevere. Examination of the residual item correlation cross-loadings, for example, calls into question the strict unidimensionality of the items as posited by the SWBM.

Table 2.

Factor Loadings.

Item	Factor 1: Negative Affect	Factor 2: Social Acceptance	Factor 3: Life Satisfaction	Factor 4: Self-Esteem	Factor 5: Positive Affect	Factor 6: Life Satisfaction (2)	h ²
You felt hopeful about the future			−0.31		.26	−0.10	.24
You thought your life had been a failure	−0.53		0.17			0.17	.50
You enjoyed life			−0.76				.63
You felt life was not worth living	−0.54		0.15	0.11		0.15	.48
You felt you were just as good as other people					.56		.41
You have a lot of good qualities					−.10	0.79	.69
You are happy			−0.62				.51
You are physically fit			−0.10	−0.29		0.25	.26
You have a lot to be proud of						0.85	.85
You like yourself just the way you are				−0.89			.85
You feel you are doing everything just about right		−0.19		−0.40		0.10	.45
You feel socially accepted		−0.91					.85
You feel loved and wanted		−0.40	0.19			0.32	.64
You felt you could not shake off the blues…	−0.75						.57
You felt depressed	−0.88						.77
You are sad	−0.79						.61
You felt that people disliked you	−0.44	−0.15					.26

To explore this, we conducted an exploratory sequence of latent variables, ranging from two to six, with multidimensional IRT analyses utilizing the generalized partial credit model with oblique structure. The final latent model structure constructed consisted of six first-order latent variables which fitted the data with a limited information, $χ_{(d f = 8)}^{2}$ = 9.98, p = .27. The latent variables are proposed to represent perceived social acceptance and self-esteem as well as positive affect, negative affect, and life satisfaction. The third and sixth factors within the model were very similar in their composition and could arguably be interpreted as measures of life satisfaction, as substantiated by their positive correlation (r = .46; see Table 3).

Table 3.

Factor Correlations.

	Factor 1: Negative Affect	Factor 2: Social Acceptance	Factor 3: Life Satisfaction	Factor 4: Self-Esteem	Factor 5: Positive Affect	Factor 6: Life Satisfaction (2)
Factor 1: Negative affect	1.00
Factor 2: Social acceptance	0.34	1.00
Factor 3: Life satisfaction	−0.59	−0.39	1.00
Factor 4: Self-esteem	0.39	0.64	−0.38	1.00
Factor 5: Positive affect	0.31	0.17	−0.51	0.17	1.00
Factor 6: Life satisfaction (2)	−0.31	−0.68	0.46	−0.71	−0.29	1.00

The items that loaded on the first factor mostly captured respondents’ negative outlook on their own lives, indicating some aspect of negative affect as conceptualized by Watson et al. (1988). The items that loaded on the second factor provide some insight into respondents relationships to others and their perceived level of social acceptance, which has been found to be a mediator of SWB (Anderson et al., 2012). The items that loaded on the third and sixth factors illuminate how satisfied respondents were with their present circumstance. Taken together, the third and sixth factors demonstrate how life satisfaction is determined by the degree to which respondents “see themselves as making progress at personally valued goals, possessing capabilities necessary for performance, and likely to attain valued outcomes…” (Lent et al., 2005, p. 430). The items that loaded on the fourth factor called for respondents to reflect on their self-perceptions, which is indicative of their self-esteem (Nguyen et al., 2016). The items that loaded on the fifth factor captured how positively respondents viewed their lives and reflect Watson et al.’s (1988) definition of positive affect.

Next, we stratified the sample by gender and racial identity indicators (creating four mutually exclusive groups) as specified in the sample description to investigate the comparative parameter equivalence. Our results indicated that configural invariance failed to be rejected between the four groups, $M_{(32)}^{2}$ = 44.45, p = .07, indicating that the measurement structure and loading patterns replicated across the mixture of race and gender. Next, the equivalence of factor loadings across groups was tested and rejected, $M_{(293)}^{2}$ = 493.21, p < .0001, demonstrating a failure of both metric and scalar invariance. This pattern indicates that while the interpretative meaning of the items holds between groups (configural invariance), the relationship between individual items and their respective latent measure spaces are nonequivalent. Thus, the empirical findings suggest that the items themselves cannot be naively compared between groups in terms of mean comparisons with substantial adjustment.

These findings suggest that while the average response between groups for an item may have equivalent intercepts (i.e., the average manifest response is equivalent), the underlying meaning with which that average reflects the underlying true score is different between groups. Therefore, observed scores cannot be equated and interpreted using first-order comparisons (i.e., observed average score differences–based statistical methods, such as the generalized linear model). These conclusions specifically compromise the conventional meaning attributable to the differences reported in Table 4. Intercept invariance was tested, $M_{(206)}^{2}$ = 878.07, p < .0001, while freeing the slopes to vary across groups, and both were individually rejected. Scalar invariance (the equality of both the slopes between the latent traits and the observed items and the intercepts for each observed item) was also rejected, $M_{(467)}^{2}$ = 1,782.51, p < .0001.

Table 4.

Relative Invariance Between Racial and Gender Groups.

Race	M ²	df	p	SRMSR (95% CI)
Black	2,027.74	626	.00	[.050, .062]
White	1,221.40	496	.00	[.032, .034]
Gender
Male	1,935.76	626	.00	[.046, .054]
Female	1,045.62	496	.00	[.034, .035]

Note. CI = confidence interval; SRMSR = standardized root mean square residual.

A further sequence of hypotheses was evaluated to examine the relationship between the perceived bias measure, which assessed the level of acceptance participants felt at their respective school, and the SWBM. The null hypothesis of equivalent fit and linear relations between the perceived bias measure and the Feelings Scale was evaluated between all four groups. Under CRT, the argument would be that Black students would be significantly worse fitting under a common model, such that divergent measurement structures would reflect differences in operationalism between the latent variables of each group. Configural measurement equivalence was strongly rejected between all four groups, $M_{(992)}^{2}$ = 1,629, p < .0001. However, as per Table 4, it can be seen that these models are found to fit substantially worse for both male and Black participants relative to the larger demographic groups of comparison, indicating different operationalizations of the scale prompts.

Testing explicit hypotheses of DIF for nonnested models is exceedingly difficult to establish because the models for each scale across racial strata are nonnested and would therefore require the implementation and construction of the Vuong (1989) test, which was not undertaken. Instead, comparisons of the item performance across racial groups were done to investigate whether individual strata interpreted items upon the SWB Scale differently than items with no expected differences as posited by CRT. We found that the fit across nonpooled item estimation (allowing for racial identity to interact with item response functions) demonstrated a considerable improvement upon 3 items: “you felt that you could not shake the blues, even with help from your friends,” “you were happy,” and “you felt that people disliked you.” For those items, both average and worst-case differences in misfit were markedly improved for scenarios in which DIF was allowed to be reflected in the item fitting procedure. This would indicate a relatively strong empirical estimate that racial identity does, in fact, appear to be a substantive factor in how SWB is perceived. The average difference in misfit between the item groups indicated an improvement of χ² = 146.86, which would be significant at the established α =. 05 for all degrees of freedom less than 120. Likewise, evidence is presented for the best fitting differences across racial groups, which indicated significant improvement in fit (χ² = 94.583) for all degrees of freedom less than 73; again indicating strong evidence of empirical improvement in establishing race as a necessary element in the measurement of SWB beyond alternative explanations for other DIF.

In summary, we have found the conventional three subscale structure of SWBM to be inconsistent across both gender and racial groups, indicating that the measurement structure is explicitly different and practically incapable of being compared. This has immediate implications for those who wish to demonstrate institutional differences in SWB, as responding individuals appear to conceptualize the questions in substantially different ways and relative contributions to decisions concerning SWB. These conclusions follow from both the superior fitting of the alternative measurement structure and are further under the relative nonequivalence of metric and scalar invariance, substantiating a common latent meaning of six conceptual clusters, but composed from divergent meanings across strata. Thus, we can conclude that while the alternative measurement structure of items consistent with the SWBM (i.e., configural invariance) that we introduced above does show evidence for equivalence between gender and racial strata, the relationship between the items and said configurations are not equitably interpretable across groups. This calls into question between group comparisons made based on sum-score comparisons, as the scores do not appear to fully represent the constructs commonly found in the literature.

Discussion

SWB has been connected to several outcomes for school-aged youth with the assumption that measures of the construct hold constant across racial groups. CRT posits that institutional bias is an omnipresent force that affects the experiences and perceptions of minority groups in lasting ways. Within the context of CRT, Black and White youth likely perceive SWB and its associated constructs differently due to the oppressive nature of social institutions. This study sought to explore perceived SWB across racial groups using data from the first wave of the Add Health data set. We hypothesized that measurement invariance across racial groups would be limited when observing SWB in Black and White youth. The findings of this study support our hypothesis.

While there is no consensus at which point the degree of measurement invariance should call into question the validity of a measure when used across groups (Byrne, 2008), the results of this study suggest that Black and White youth use measures of SWB differently. This finding could account for how their lived experiences impact what they expect from the people and institutions in their environments. Such an impact was demonstrated by Rose et al. (2017) when they applied a dual-factor model of mental health (which assesses SWB) in their observation of Black adolescent educational outcomes. Their findings indicated that school bonding and mental health were significantly linked. Black youth are often aware of institutional bias (Curtis, 2017; Hope & Jagers, 2014). This awareness would likely be reflected in their self-reported life satisfaction and affect despite any coping mechanisms they might utilize.

Socialization and environmental expectations play a similar role in how White youth report SWB. CRT states that social institutions promote Whiteness as the cultural standard of achievement and implicitly reinforce the notion of a colorblind meritocracy (Ledesma & Calderón, 2015). White youth who buy into the idea that social institutions are inherently fair and just will likely not see or acknowledge the benefits of their privilege (Curtis, 2017). For example, Grossman and Charmaraman (2009) explored racial–ethnic identity among a sample of White high school students and found that participants had an overall low rating of racial–ethnic centrality. This means that for a majority of the White youth observed, race was not viewed as a major factor in their experiences of the world around them. Such findings suggest that White youths’ assessment of their own SWB are least likely to be weighed down by the experiences of overt and covert oppression even when they belong to some oppressed group. For instance, gay and female White youth are susceptible to experiences of oppression and bias (Hurtado, 1989; Kleiman et al., 2015). However, their White privilege is a protective factor that gay and female individuals of color will never have access to (Hall, 2006; Han, 2007; Russo, 2001).

Solórzono and Yosso (2002) argued that hegemonic narratives and deficit thinking toward marginalized groups could be undone through critical race methodologies that created space for counterstories that amplified the voices of those who have been otherized. Years later, McGee and Stovall (2015) came to a similar conclusion after they observed the impact that experiencing institutional bias at school had on Black college students using CRT as a framework. Their findings led to a call for an interdisciplinary effort to develop appropriate means to help protect Black students from the negative mental health outcomes associated with discriminatory and traumatizing practices at their schools. They also challenged the notion of grit and perseverance as a primary way of explaining Black students’ capacity to endure oppression by suggesting a more holistic healing process that would help Black students to not only survive their negative experiences but also thrive beyond school (McGee & Stovall, 2015). It is surprising how little research has assessed SWB within the context of CRT considering its potential for improving how outcomes for youth (including SWB) are measured in diverse samples by accounting for the influence of institutional bias on the lived experiences of marginalized groups. Additionally, CRT could be helpful in understanding what barriers exist that inhibit the development of high SWB among youth of color by providing a framework that guides educators and youth workers in facilitating the development of antiracist pedagogies that support positive youth development and mental health.

Limitations

There are some limitations to this study that should be considered. As with any study using secondary data, we were limited to constructing the measures from the data collected in the Add Health study, which were likely not developed using a critical race lens. There were aspects of life satisfaction and positive and negative affect that were not captured by the items available in the data set. The Add Health study only assessed global life satisfaction. Various specific areas of life satisfaction (e.g., family, friends, school, and living environment) captured in measures like the Multidimensional Students’ Life Satisfaction Scale (Greenspoon & Saklofske, 2001) were not assessed. Additionally, this study did not use measures designed to capture the dynamic nature of individual’s positive and negative affective states. Durayappah (2010), for example, used both the Experience Sampling Method and the Positive and Negative Affect Scale to capture the complexity of youth’s positive and negative affect created by the dynamic nature of those self-perceptions. Lastly, the data used in this study are over 20 years old. However, institutional biases and their associated outcomes have held constant since the data were collected. Despite any observable cohort effects, disparities in emotional and behavioral health (J. Brown et al., 2007), criminal justice (Mears et al., 2016), educational attainment (Condron et al., 2013), and employment (Toldson & Snitman, 2010) that are generated by institutional bias have persisted since the early 1990s.

Implications and Future Research

It is important to address racial differences when creating interventions or developing preventive strategies to enhance SWB in the adolescent population because any measures used to assess SWB would need to establish a true baseline in order to determine their effectiveness. The results of this study demonstrate that, within groups, the measures work as expected; but across groups, there is a considerable lack of invariance. This indicates that the scales used in this study are not adequate for comparing Black and White youth on the three components of SWB (i.e., life satisfaction, positive affect, and negative affect), in which items from all three subscales were found not to be invariant across groups. Also, this study only investigated the construct validity and between-group measure invariance of components of SWB. Other important forms of validity, such as predictive and convergent validity and longitudinal measurement invariance were not investigated. Future research should investigate these forms of validity in order to provide additional evidence of the utility of the measures.

Future assessments of SWB would also benefit from more domain-specific measures of life satisfaction as well as more in depth measures of positive and negative affect. There are two potential approaches that could address the issues raised by our findings. First, researchers could substantively reevaluate and redesign measurement scales utilized in work with multiple racial groups. A second possibility is those wishing to assess SWB could accept that different scale scoring schema are necessary based upon relevant external information as part of the methods utilized in assessing relevant cultural and social structures in social institutions.

Conclusion

This study provides some evidence that racial/ethnic differences cannot be taken for granted when assessing SWB in youth. CRT provides a context for us to understand observed differences in SWB as an outcome. CRT can also be helpful in identifying ways to mitigate the circumstances that create and maintain these racial differences (McCoy & Rodericks, 2015), which influence SWB. However, such an undertaking is compounded by the complexity of SWB as a construct. It is important to note that the findings presented here are likely symptomatic of a much deeper issue. That is, race and racism influence how constructs are conceptualized and measured, not to mention how findings are interpreted. The influence of race on observed constructs and data is further complicated by the intersectionality characteristic of the human experience. For these reasons, Mertens (2015) challenged researchers to find novel approaches to investigate and address the complex issues faced by society that are rooted in difference and inequity. Wells and Stage (2015) echoed this sentiment while calling attention to the reality that even the term critical is not applied uniformly across the existing literature. With that in mind, a major question that comes out of this study is: Should investigative efforts be geared toward the pursuit of more accurate measures of SWB to adequately demonstrate the negative impact of biased social institutions as described in CRT, or would it be more prudent to address issues of inequity so that perceptions of SWB are more comparable across groups? Both approaches are worthy endeavors that could broaden our understanding of how race and SWB intersect.

Footnotes

Data Availability Statement

The data that support the findings of this study are openly available in the Inter-university Consortium for Political and Social Research repository at

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

Christopher A. Curtis

References

Anderson

C. A.

Krais

M. W.

Galinsky

A. D.

Keltner

(2012). The local-leader effect: Social status and subjective well-being. Psychological Science, 23(7), 764–771. https://doi.org/10.1177/0956797611434537

Antaramian

S. P.

Huebner

E. S.

Hills

K. J.

Valois

R. F.

(2010). A dual-factor model of mental health: Toward a more comprehensive understanding of youth functioning. American Journal of Orthopsychiatry, 80(4), 467–472. https://doi.org/10.1111/j.1939-0025.2010.01049

Bradley

R. H.

Corwyn

R. F.

(2004). Life satisfaction among European American, African American, Chinese American, Mexican American, and Dominican American adolescents. International Journal of Behavioral Development, 28(5), 385–400. https://doi.org/10.1080/01650250444000072

Brown

Meadows

S. O.

Elder

G. H.

Jr (2007). Race-ethnic inequality and psychological distress: Depressive symptoms from adolescence to young adulthood. Developmental Psychology, 43(6), 1295–1311. https://doi.org/10.1037/0012-1649.43.6.1295

Brown

T. N.

(2008). Race, racism, and mental health: Elaboration of critical race theory’s contribution to the sociology of mental health. Contemporary Justice Review, 11(1), 53–62. https://doi.org/10.1080/10282580701850405

Busseri

M. A.

Sadava

S. W.

(2013). Subjective well-being as a dynamic and agentic system: Evidence from a longitudinal study. Journal of Happiness Studies, 14(4), 1085–1112. https://doi.org/10.1007/s10902-012-9368-9

Byrne

B. M.

(2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20(4), 872–882.

Chalmers

R. P.

(2012). mirt: A multidimensional item response theory package for the R environment. Journal of Statistical Software, 48(6). 1–29. https://doi.org/10.18637/jss.v048.i06

Chambers

(2009). The “receivement gap”: School tracking policies and the fallacy of the “achievement gap.” The Journal of Negro Education, 78(4), 417–431.

10.

Condron

D. J.

Tope

Steidl

C. R.

Freeman

K. J.

(2013). Racial segregation and the Black/White achievement gap, 1992 to 2009. The Sociological Quarterly, 54(1), 130–157. https://doi.org/10.1111/tsq.12010

11.

Curtis

(2017). Observing Black youth sociopolitical group participation and high school democratic climate through the lens of critical race theory. Race Ethnicity & Education, 22(3), 410–427. doi:10.1080/13613324.2017.1395322

12.

Diener

(1984). Subjective well-being. Psychological Bulletin, 95(3), 542–575. https://doi.org/10.1037/0033-2909.95.3.542

13.

Diener

Inglehart

Tay

(2013). Theory and validity of life satisfaction scales. Social Indicators Research, 112(3), 497–527. https://doi.org/10.1007/s11205-012-0076-y

14.

Durayappah

(2010). The 3p model: A general theory of subjective well-being. Journal of Happiness Studies, 12(4), 681–716. https://doi.org/10.1007/s10902-010-9223-9

15.

Durlak

J. A.

Weissberg

R. P.

Dymnicki

A. B.

Taylor

R. D.

Schellinger

K. B.

(2011). The impact of enhancing students’ social and emotional learning: A meta-analysis of school-based universal intentions. Child Development, 82(1), 405–432. https://doi.org/10.1111/j.1467-8624.2010.01564.x

16.

Eid

Diener

(2004). Global judgments of subjective well-being: Situational variability and long-term stability. Social Indicators Research, 65(3), 245–277. https://doi.org/10.1023/B:SOCI.0000003801.89195.bc

17.

Eryilmaz

(2011). Satisfaction of needs and determining of life goals: A model of subjective well-being for adolescents in high school. Educational Sciences: Theory & Practice, 11(4), 1757–1764.

18.

Eryilmaz

(2012). A model of subjective well-being for adolescents in high school. Journal of Happiness Studies, 13(2), 275–289. https://doi.org/10.1007/s10902-011-9263-9

19.

Gillborn

(2006). Critical race theory and education: Racism and anti-racism in educational theory and praxis. Discourse: Studies in the Cultural Politics of Education, 27(1), 11–32. https://doi.org/10.80/01596300500510229

20.

Gillborn

(2013). Interest-divergence and the colour of cutbacks: Race, recession and the undeclared war on Black children. Discourse: Studies in the Cultural Politics of Education, 34(4), 477–491. https://doi.org/10:1080/01596306.2013.822616

21.

Gillborn

(2015). Intersectionality, critical race theory, and the primacy of racism: Race, class, gender, and disability in education. Qualitative Inquiry, 21(3), 277–287. https://doi.org/10.1177/1077800414557827

22.

Greenspoon

P. J.

Saklofske

D. H.

(2001). Toward an integration of subjective well-being and psychopathology. Social Indicators Research, 54(1), 81–108. https://doi.org/10.1023/A:10072192

23.

Griffith

D. M.

Ellis

K. R.

Allen

J. O.

(2013). An intersectional approach to social determinants of stress for African American men: Men’s and women’s perspectives. American Journal of Men’s Health, 7(4), 19S–30S. https://doi.org/10.1177/1557988313480227

24.

Grossman

J. M.

Charmaraman

(2009). Race, context, and privilege: White adolescents’ explanations of racial-ethnic centrality. Journal of Youth and Adolescence, 38(2), 139–152. https://doi.org/10.1007/s10964-008-9330-7

25.

Hall

R. E.

(2006). White women as postmodern vehicle of Black oppression: The pedagogy of discrimination in western academe. Journal of Black Studies, 37(1), 69–82. https://doi.org/10.1177/0021934705277286

26.

Han

C.-S.

(2007). They don’t want to cruise your type: Gay men of color and the racial politics of exclusion. Social Identities, 13(1), 51–67. https://doi.org/10.1080/13504630601163379

27.

Harris

K. M.

Udry

J. R.

(2009). National longitudinal study of adolescent to adult health (Add Health), 1994-2008 [Public Use]. Carolina Population Center, University of North Carolina-Chapel Hill [distributor], Inter-university Consortium for Political and Social Research [distributor], 2017-10-24. https://doi.org/10.3886/ICPSR21600.v18

28.

Haybron

D. M.

(2011). Taking the satisfaction (and the life) out of life satisfaction. Philosophical Explorations, 14(3), 249–262. https://doi.org/10.1080/13869795.2011.594959

29.

Holland

P. W.

Wainer

(Eds.) (1993). Differential item functioning. Routledge.

30.

Hood

Hopson

R. K.

(2008). Evaluation roots reconsidered: Asa Hilliard, a fallen hero in the “Nobody Knows My Name” Project, and African educational excellence. Review of Educational Research, 78(3), 410–426. https://doi.org/10.3102/0034654308321211

31.

Hope

E. C.

Jagers

R. J.

(2014). The role of sociopolitical attitudes and civic education in the civic engagement of Black youth. Journal of Research on Adolescence, 24(3), 460–470. https://doi.org/10.1111/jora.12117

32.

Hull

Kilbourne

Reece

Husaini

(2008). Community involvement and adolescent mental health: Moderating effects of race/ethnicity and neighborhood disadvantage. Journal of Community Psychology, 36(4), 534–551. https://doi.org/10.1002/jcop.20253

33.

Hurtado

(1989). Relating to privilege: Seduction and rejection in the subordination of White women and women of color. Signs, 14(4), 833–855.

34.

Jay

(2003). Critical race theory, multicultural education, and the hidden curriculum of hegemony. Multicultural Perspectives, 5(4), 3–9.

35.

Jayawickreme

Forgeard

M. C.

Seligman

M. P.

(2012). The engine of well-being. Review of General Psychology, 16(4), 327–342. https://doi.org/10.1037/a0027990

36.

Joe

(1997). Multivariate models and multivariate dependence concepts. Chapman & Hall/CRC Press.

37.

Jöreskog

K. G.

(1967). A general approach to confirmatory maximum likelihood factor analysis. ETS Research Bulletin Series, 1967(2), 183–202. https://doi.org/10.1002/j.2333-8504.1967.tb00991.x

38.

Kleiman

Spanierman

L. B.

Smith

N. G.

(2015). Translating oppression: Understanding how sexual minority status is associated with White men’s racial attitudes. Psychology of Men & Masculinity, 16(4), 404–415. https://doi.org/10.1037/a0038797

39.

Koffman

Ray

Berg

Covington

Albarran

N. M.

Vasquez

(2009). Impact of comprehensive whole child intervention and prevention program among youths at risk of gang involvement and other forms of delinquency. Children and Schools, 31(4), 239–245.

40.

Ladson-Billings

Tate

W. F.

IV . (1995). Toward a critical race theory of education. Teachers College Record, 97(1), 47–68.

41.

Lazarsfeld

P. F.

(1955). Recent developments in latent structure analysis. Sociometry, 18(4), 391–403. https://doi.org/10.2307/2785875

42.

Ledesma

Calderón

(2015). Critical race theory in education: A review of past literature and a look to the future. Qualitative Inquiry, 21(3), 206–222. https://doi.org/10.1177/1077800414557825

43.

Lent

R. W.

Singley

Sheu

H.-B.

Gainor

K. A.

Brenner

B. R.

Treistman

Ades

(2005). Social cognitive predictors of domain and life satisfaction: Exploring the theoretical precursors of subjective well-being. Journal of Counseling Psychology, 52(3), 429–442. https://doi.org/10.1037/0022-0167.52.3.429

44.

Leonardo

(2012). The race for class: Reflections on a critical race class theory of education. Educational Studies, 48(5), 427–449. https://doi.org/10.1080/00131946.2012.715831

45.

Leonardo

(2013). The story of schooling: Critical race theory and the educational racial contract. Discourse: Studies in the Cultural Politics of Education, 34(4), 599–610. https://doi.org/10.1080/01596306.2013.822624

46.

Long

R. F.

Huebner

Wedell

D. H.

Hills

K. J.

(2012). Measuring school-related subjective well-being in adolescents. American Journal of Orthopsychiatry, 82(1), 50–60. https://doi.org/10.1111/j.1939-0025.2011.01130.x

47.

Lou

Anthony

C. K.

Stone

Austin

(2008). Assessing child and youth well-being: Implications for child welfare practice. Journal of Evidence-Based Social Work, 5(1–2). https://doi.org/10.1300/J394v05n01_05

48.

Maydeu-Olivares

Joe

(2008). An overview of limited information goodness-of-fit testing in multidimensional contingency tables. In Shigemasu

Okada

Imaizumi

Hoshino

(Eds.), New trends in psychometrics (pp. 253–262). Universal Academy Press.

49.

McCoy

Rodricks

(2015). Critical race theory in higher education: 20 Years of theoretical and research innovations. ASHE Higher Education Report, 41(3), 1–117.

50.

McGee

E. O.

Stovall

(2015). Reimagining critical race theory in education: Mental health, healing, and the pathway to liberatory praxis. Educational Theory, 65(5), 491–511. https://doi.org/10.1111/edth.12129

51.

Mears

D. P.

Cochran

J. C.

Lindsey

A. M.

(2016). Offending and racial and ethnic disparities in criminal justice: A conceptual framework for guiding theory and research and informing policy. Journal of Contemporary Criminal Justice, 32(1), 78–103. https://doi.org/10.1177/1043986215607252

52.

Meredith

(1993). Measurement invariance, factor analysis, and factorial invariance. Psychometrika, 58(4), 525–543. https://doi.org/10.1007/BF02294825

53.

Meredith

Teresi

J. A.

(2006). An essay on measurement and factorial invariance. Medical Care, 44(11), S69–77. https://doi.org/10.1097/01.mlr.0000245438.73837.89

54.

Mertens

D. M.

(2015). Mixed methods and wicked problems. Journal of Mixed Methods Research, 9(1), 3–6. https://doi.org/10.1177/1558689814562944

55.

Neblett

E. W.

Rivas-Drake

Umaña-Taylor

A. J.

(2012). The promise of racial and ethnic protective factors in promoting ethnic minority youth development. Child Development Perspectives, 6(3), 295–303. https://doi.org/10.1111/j.1750-8606.2012.00239.x

56.

Newsome

W. S.

Anderson-Butcher

Fink

Hall

Huffer

(2008). The impact of school social work services on student absenteeism and risk factors related to school truancy. School Social Work Journal, 32(2), 21–38.

57.

Nguyen

A. W.

Chatters

L. M.

Taylor

R. J.

Mouzon

D. M.

(2016). Social support from family and friends and subjective well-being of older African Americans. Journal of Happiness Studies, 17(3), 959–979. https://doi.org/10.1007/s10902-015-9629-8

58.

Oliver

(2006). “The streets”: An alternative Black male socialization institution. Journal of Black Studies, 36(6), 918–937. https://doi.org/10.1177/0021934704273445

59.

Ozmete

(2011). Subjective well-being: The science of happiness and a proposal for a national index. American Psychologist, 55(1), 34–43. https://doi.org/10.1037/0003-066X.55.1.34

60.

Park

(2004). The role of subjective well-being in positive youth development. Annals of the American Academy of Political and Social Science, 59(1), 25–39. https://doi.org/10.1177/0002716203260078

61.

Pontin

Schwannauer

Tai

Kinderman

(2013). A UK validation of a general measure of subjective well-being: The modified BBC Subjective Well-Being Scale (BBC-SWB). Health & Quality of Life Outcomes, 11(1), 1–9. https://doi.org/10.1186/1477-7525-11-150

62.

Riley

Ettlinger

(2011). Interpreting racial formation and multiculturalism in a high school: Towards a constructive deployment of two approaches to critical race theory. Antipode, 43(4), 1250–1280. https://doi.org/10.1111/j.1467-8330.2010.00825.x

63.

Rose

Lindsey

M. A.

Xiao

Finigan-Carr

N. M.

Joe

(2017). Mental health and educational experiences among Black youth: A latent class analysis. Journal of Youth and Adolescence, 46(11), 2321–2340. https://doi.org/10.1007/s10964-017-0723-3

64.

Russo

(2001). Taking back our lives: A call to action for the feminist movement. Routledge.

65.

Shin

R. Q.

Morgan

M. L.

Buhin

Truitt

T. J.

Vera

E. M.

(2010). Expanding the discourse on urban youth of color. Cultural Diversity and Ethnic Minority Psychology, 16(3), 421–426. https://doi.org/10.1037/a0018693

66.

Solórzono

D. G.

Yosso

T. J.

(2001a). From racial stereotyping and deficit discourse toward a critical race theory in teacher education. Multicultural Education, 9(1), 2–8.

67.

Solórzono

D. G.

Yosso

T. J.

(2001b). Maintaining social justice hopes within academic realities: A Freirian approach to critical race/LatCrit pedagogy. Denver University Law Review, 78(4), 595–621.

68.

Solórzono

D. G.

Yosso

T. J.

(2002). Critical race methodology: Counter-storytelling as an analytical framework for education research. Qualitative Inquiry, 8(1), 23–44. https://doi.org/10.1177/107780040200800103

69.

Stovall

(2016). Out of adolescence and into adulthood: Critical race theory, retrenchment and the imperative of praxis. Urban Education, 51(3), 274–286. https://doi.org/10:1177/0042085915618718

70.

Suldo

S. M.

Shaffer

E. J.

(2008). Looking beyond psychopathology: The dual-factor model of mental health in youth. School Psychology Review, 37(1), 52–68.

71.

Suldo

Thalji

Ferron

(2011). Longitudinal academic outcomes predicted by early adolescents’ subjective well-being, psychopathology, and mental health status yielded from a dual factor model. The Journal of Positive Psychology, 6(1), 17–30. https://doi.org/10.1080/17439760.2010.536774

72.

Theofilou

(2013). Quality of life: Definition and measurement. Europe’s Journal of Psychology, 9(1), 150–162. https://doi.org/10.5964/ejop.v9i1.337

73.

Toldson

I. A.

Snitman

(2010). Editor’s comment: Education parity and economic disparities: Correcting education-attainment discrepancies among Black people in the United States. The Journal of Negro Education, 79(1), 1–5.

74.

Tucker

Dixon

(2009). Low-income African American male youth with ADHD symptoms in the United States: Recommendations for clinical mental health counselors. Journal of Mental Health Counseling, 31(4), 309–322. https://doi.org/10.17744/mehc.31.4.j451mx7135887238

75.

Vanderbilt

A. A.

Isringhausen

K. T.

VanderWielen

L. M.

Wright

M. S.

Slashcheva

L. D.

Madden

M. A.

(2013). Health disparities among highly vulnerable populations in the United States: A call to action for medical and oral health. Medical Education Online, 18, 20644. https://doi.org/10.3402/meo.v18i0.20644

76.

Vera

Thakral

Gonzales

Morgan

Conner

Caskey

Bauer

Mattera

L.-A.

Clark

Bena

Dick

(2008). Subjective well-being in urban adolescents of color. Cultural Diversity and Ethnic Minority Psychology, 14(3), 224–233. https://doi.org/10.1037/1099-9809.14.3.224

77.

Villodas

M. T.

Roesch

(2011). Examining the factor structure of the positive and negative affect schedule (PANAS) in a multiethnic sample of adolescents. Measurement and Evaluation in Counseling and Development, 44(4), 193–203. https://doi.org/10.1177/0748175611414721

78.

Vuong

Q. H.

(1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57(2), 307–333. https://doi.org/10.2307/1912557

79.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070.

80.

Wells

R. S.

Stage

F. K.

(2015). Past, present, and future of critical quantitative research in higher education. New Directions for Institutional Research, 2014(163), 103–112. https://doi.org/10.1002/ir.20089