A Scoping Review of Colorism in Schools: Academic, Social, and Emotional Experiences of Students of Color

: Students of color experience academic, social, and emotional challenges due to colorism in schools. The purpose of this scoping review is to compare the experiences with colorism of students from varying racial backgrounds (African Americans, Native Americans, Asians, and Latin) in U.S. public schools. It is predicted that the understudied group of Latinx and indigenous students of color will uniquely experience colorism in academic settings when compared to African American and Asian students. A 30 article literature review utilizing search dates from 1990 to 2020 was conducted employing a scoping review framework. Themes emerged that include: the privileging of lighter skin and more Eurocentric features in academic outcomes, the complicated social status created for students of color experiencing colorism in schools, and the increased potential for emotional challenges as a result of colorism. This review highlights possible school reform efforts to afﬁrm all skin tones, reduce colorist biases, and offer mediation to mitigate colorist experiences in the school environment.


Background
Globally, racial justice movements have been sustained by youth (Ortega-Williams 2017). In the 21st century, children and youth of color have continued to push for justice in a racially divided world that often judges people based on appearance rather than merit. While racial categories vary widely across the world, there is a global linkage of these movements that seek to confront anti-blackness, or the rejection of people who appear to be Black (i.e., shape of nose, lips, texture of hair, color of eyes, skin color, etc.) (Plummer 2020). One study found anti-blackness, in the form of colorism, or preferential treatment of same-raced people based on their physical proximity to Whiteness (Blow 2021). Researchers defined Whiteness as European characteristics, which may include: light skin, a narrow nose, thin lips, and/or smooth hair (Russell et al. 1993).
Additionally, scholarship affirms that students with darker skin and Afrocentric phenotypes (e.g., wide nose and full lips) encounter many environmental barriers solely because of those characteristics (Hunter 2016). This worldwide phenomenon is demonstrated through protests against skin lighteners in India, which also occurs in African nations. Moreover, darker-skinned people in both Brazil and the United States experience economic, social, and educational limitations. Ultimately, colorism strikes a global chord of discrimination with particularly long lasting impacts for youth of color. This scoping review seeks to advance the knowledge on colorism for educational stakeholders, including the experiences of students of color in schools, who face daily instances of discrimination based on their skin tone and features. This study provides a historical view on colorism within the communities of color, including African Americans, Native Americans, Asians, and Latinx. The rationale for the study is followed by the methodology. The findings demonstrate the myriad of negative experiences of students of color with colorism in schools, followed by conclusions about the usefulness of the evidence found in this scoping review.
Colorism results in bias due to the privileging of light over dark skin and the preference for certain physical attributes that appear to be White (e.g., shape of the nose, texture of the hair, eye color) (Keith 2009;Maddox 2004;Banks 2015). To date, an overwhelming majority of colorism research has discussed the phenomenon among adult populations of color (Banks 2015;Devaraj et al. 2018;Monk 2015;Perreira and Telles 2014). For example, Reece (2019b), a notable colorism scholar, explores the economic advantages of lighter skin for African Americans both by examining their marriage patterns and the impact of obesity on income (i.e., obesity income penalty), stratified by race and skin color (Reece 2019c). Reece (2021) also found lighter skin African American people perpetuated their privilege by marrying lighter partners so that their children would be more affluent, with more access to education and social clubs in a Jim Crow south. Similarly, Reece (2021) suggests color dictates much of the social outcome of Americans, regardless of race.
In addition, youth conceptualize skin tone differences at early ages (Adams et al. 2016). Due to their analogous encounters with colorism and discrimination, African Americans, Native Americans, Asians, and Latinx are frequently grouped together in research on students of color (Lee and Bean 2007). It is acknowledged that colorism transcends racial groups, and much of the literature explores colorism through a racial lens. Thus, this introduction is used to describe the historical development of colorism within each racial group.

African Americans
Regarding African American children's conceptualization of colorism, the landmark "Doll Test" was seminal, as it evoked conversations around skin tone bias. Experimenters conducted the Doll Test by using White and Black dolls to analyze racial identification and preference among African American children (Clark and Clark 1939). Evidence from the "Doll Test" informed the authors that the intersection of prejudice, discrimination, and segregation lowered African American youths' self-esteem (Billante and Hadad 2010). In 2010, this study was recreated and tested on White and Black children within two age groups: 4-5 and 9-10 years old (Billante and Hadad 2010). Findings indicated African American youths' preference for dolls with lighter skin tones (James 2010). Overall, scholars found African American children may exhibit preferences for light or medium skin tones, and also begin negatively stereotyping dark skin tones, starting in kindergarten (Adams et al. 2016).

Native Americans
Historically, the interactions and norms among Native Americans were shaped by Euro-American settlers, especially in regard to colorism (Brown et al. 2018). In school settings from 1872-1910, skin tone hierarchies were implemented based on: ancestry, appearance, and level of assimilation (Halliburton 1977). As a result, Native children exhibited skin tone preferences towards their classmates. Specifically, in the Cherokee Nation, dark skin Native American students, or "full-blood", were ridiculed by light students of Caucasian and Native descent, or "mixed-blood" (Brown et al. 2018). Colorism has also created major divisions among the Native community in regard to their behaviors towards dark skin individuals (Hall 2010).

Asian Populations
Scholars found an aversion towards dark skin individuals in Asian history, as well (Thompson and McDonald 2015). The preference for light skin was paramount across Asian communities and predates their contact with Europeans (Rondilla and Spickard 2007). An abundance of cultural differences exists in Asia, yet the preference for light skin is universal (Isa and Kramer 2003;Russell et al. 1993). Even Asian youth had desired lighter or whiter skin tones to obtain greater upward mobility (Li et al. 2008). Among Asian college students in the United States, they reported light skin as the most important physical attribute (Fegley et al. 2008).

Latinx Populations
In Latin America, skin color is a prominent factor (Telzer and Garcia 2009). Distinctions are made regarding skin tone, such as "Blanco" that refers to light skin Latinos and "Indio" or "Moreno" that refers to dark skin Latinos (Uhlmann et al. 2002). Since the United States has a history of stigmatizing dark skin tones, scholars reported Latinos living in the U.S. had a tendency to associate positive qualities with light skin tones (Montalvo 2005;Stephens et al. 2012). As for Latinx youth, Latinos possessing dark skin tones experienced more negative self-perceptions and lower self-esteem when compared to Latinos possessing lighter skin tones (Telzer and Garcia 2009;Shelton et al. 2005). These experiences have resulted in skin color discrimination between Latinos (Santana 2018). However, this study comprehensively explores the consequences for students of color regarding colorism circumstances in every area of their life.

Novelty and Scientific Contribution
The assessment of where the field stands on providing evidence of how colorism impacts youth in public schools is a meaningful scientific contribution. The cross-racial analysis of literature is also substantial for the field that has largely explored issues of colorism within specific racial groups but not across racial groups.

Rationale and Aims
While the literature provides information about discipline and academic disproportionality based colorism within schools, the totality of these experiences has yet to be explored in a systematic way that promotes continuity within colorism and educational scholarship. Additionally, there are several reports on the experiences of African American students with colorism, but fewer with Latinx and other indigenous people of color to which this scoping review will contribute. The main objective of this study is to answer the research question "What are the academic, social, and emotional experiences and outcomes linked to colorism for students of color in American public schools?".

Methods
The scoping review protocol outlines the planning and developmental stages that began in 2019 by the first and second authors (Keyes and Crutchfield 2021). This protocol is published after receiving rigorous feedback, which was incorporated in the study selection process. This protocol outlines in detail the step-by-step processes to be undertaken for this scoping review.
The Preferred Reporting Items for Systematic Reviews/Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) to examine the literature methodically was used (e.g., Tricco et al. 2018). Scoping reviews are well known in the medical health field (Arksey and O'Malley 2005;Levac et al. 2010) that is becoming increasingly popular in branches of social sciences. Scoping reviews are typically conducted to research a broad topic utilizing extensive questions to establish gaps in the literature. On the contrary, a priori assists when analyzing systematic reviews and fixed factors aimed at a more narrow precise question or questionnaire. Scoping reviews are helpful in aggregating research studies in various fields with emerging evidence, as they are appropriate to address questions beyond those associated with the experience or efficacy of an intervention or implementation (Peters et al. 2015). The five-phrase framework follows: (1) Identifying the research question; (2) Identifying relevant studies; (3) Study selection; (4) Charting the data; (5) Collating, summarizing, and reporting the results (Arksey and O'Malley 2005). This process was utilized to conduct this review. Soc. Sci. 2022, 11, 15 4 of 21

Five Phase Framework
The following is a detailed guide of how this study implemented the five phase framework from Arksey and O'Malley, as is the standard for conducting scientific scoping reviews.
Identifying the research question. As indicated in the objectives, a broad research question was applied concerning the totality of colorism experiences and outcomes that students of color in the U.S. may encounter.
Identifying relevant studies. A comprehensive search was conducted using ten EB-SCOhost databases: (1) Academic Search Complete, (2) Education Abstracts (H.W. Wilson), (3) Educational Administration Abstracts, (4) ERIC, (5) Primary Search, (6) Psychology and Behavioral Sciences Collection, (7) PsycARTICLES, (8) PsycINFO, (9) Social Work Abstracts, and (10) Race Relations. This electronic search consisted of an open date with no limiters, exact duplicates were removed automatically, and it generated six sources. Additional databases were set with publication dates 2000-2020; African American Communities and American Indian History and Culture both generated no sources. Final databases: JSTOR generated (1997) sources, Social Services Abstracts (364), Google Scholar (4); all were set with the same publication date. Keywords listed in all the electronic searches were <colorism or colorist or skin tone* discrimination or skin preference* or skin tone bias or skin tone stratification or skin color discrimination AND primary school* or k-12 school* or grade school*>. The last conducted search date was January 28, 2020.
Study selection. The article inclusion criteria for the study were as follows: the population sample was K-12 (ages 5-18) African Americans, Asians, Latinx group, and Native Americans. The literature included research conducted in the United States and peer-reviewed publications that included all research designs along with grey literature (magazines, newspapers, theses, and dissertations). However, magazines and newspapers did not populate in the searches. Studies were retained that had populations over the age of 18 that may have entered college and were explicitly stated (i.e., 18-23 year-old college students). Thus, we define students as those up to age 18 (which may include those in college). Articles were excluded if the population was four years or under (birth-age four) and above (age 19-100) unless explicitly stated as college students or adolescents including 18 year-olds. Articles were also excluded if they did not reference African Americans, Asians, Latinx groups, and Native Americans. Additionally, literature with countries/continents located in Canada, Mexico, Africa, Asia, Europe, Australia, South America, and Antarctica were excluded. Finally, if the research conducted was a book review or came from the following type of grey literature (blogs, bulletins, speeches, press releases, course materials, conference papers, book chapters, white papers, government documents, working papers, and/or research reports), it was excluded.
Charting the data. The data files, (2371) were imported into Covidence, a review management software tool utilized for the screening process. Covidence system automatically removed (37) duplicates during the imported phase indicating the final total of (2334) to screen. The screening process consisted of three phases, see Figure 1: Methods Flow Chart. The second and third authors screened titles, abstracts, and full-texts independently, while the first author resolved disputes by consensus subsequent to each phase. Two additional duplicates appeared that were not found through the initial process of importing data into Covidence, which left (2332) as an aggregate to screen. The title and abstract screening excluded (2228) sources, a remainder (104) potential studies were assessed through full-text screening. Subsequent to the screening, it excluded 64 sources and a remainder of 40 studies were included for the final phase of screening, where the back of the references list of those studies were assessed. The second and third authors completed that process, and an additional 18 studies were included, which added 58 studies to extract data from.
sources and a remainder of 40 studies were included for the final phase of screening, where the back of the references list of those studies were assessed. The second and third authors completed that process, and an additional 18 studies were included, which added 58 studies to extract data from. The typical research study characteristics, publication date, study population/location, methodology, aim of the study, and results were extracted. During the data extrac- The typical research study characteristics, publication date, study population/location, methodology, aim of the study, and results were extracted. During the data extraction, 28 studies were excluded not due to a fatal flaw, lack of evidence-based results, conclusion, or design issues, rather because they did not meet the overall inclusion criteria (Nippold 2011). This indicated that an aggregate of a final 30 studies were utilized as the data source.
Collating, summarizing, and reporting results. The data findings are arranged thematically according to the three prongs of the research question: academic, social, and emotional outcomes for students of color. Several of the individual articles fit into over-lapping categories and will only be summarized once. They will also be noted in the overlapping sections.

Results
Results are presented by type of outcome (academic, social, and emotional), as well as across racial groups. Individual articles are summarized for their main findings in each type of outcome. This presentation is then followed by a synthesis of the main concepts within each outcome. Finally, the discussion of the results describes major themes that manifest from the summary of the evidence.

Results of Individual Sources of Evidence
Individual sources of evidence are summarized here with details about the specific population under study and outcomes as related to the subtitle (academic, social, or emotional). Several of the results are cross-listed because they explore outcomes in each of these three areas.

Academic Outcomes
The academic outcomes explored the relationship between skin tone and educational achievement across racial groups. Branigan et al. (2013) assessed the relationship between skin color and educational attainment for native born non-Hispanic Black and White adults. They found a significant relationship between skin color and educational attainment for Black men and women. The findings indicated a range between zero and two additional years of education for lighter skin Black adults when compared to their darker skin counterparts.
Moreover, Fergus (2017) used qualitative interviews with six Mexican students from sixth through twelfth grade to assess the importance of skin color in how they identify themselves while attending a predominantly White school in a midwestern urban city. Mexican participants who possessed White features perceived themselves to be able to rise above any structural inequalities based on skin tone, while those who possessed Mexican presenting features (e.g., darker skin, eyes, and hair) believed they would experience challenges when overcoming institutional bias against Mexicans. Hunter's (2016) work aligns with the previous study by exhibiting educators' preference for lighter-skinned students rather than those with darker skin tones. This bias was evident through the favorable treatment and encouragement that light skin students received from teachers. Kim and Calzada (2019) explored teacher impact on academic delay for Dominican American students. In the study, Dominican American boys whose phenotypic features appeared to be Black had lower academic achievement, which was mediated by lower teacher rated adaptive behavior in prekindergarten and kindergarten. Twenty-seven percent of the students in this group entered prekindergarten and kindergarten with academic delays. By the end of first grade, the percentage of the group that was academically delayed rose to 36%. This is compared to Hispanic boys with whiter phenotypic features, who had a higher percentage (30%) of academic delay before entering kindergarten. However, their delay decreased by 20% at the end of their first grade year, after they experienced high teacher rated adaptive behavior. This group of boys was uniquely impacted by teacher intervention, which was rooted in colorist biases.
In addition, Orozco and López (2015) used a sample of 90 students and collected survey responses to determine the association of skin color and perceived discrimination with grade point average. Through ordinary least squares regression, they found that with every 1 standard deviation increase in skin darkness, there was an associated 0.20 standard deviation decrease in grade point average. Ryabov (2016) used the ADD Health data to explore the educational attainment of Asian American and Hispanic American young adults in the U.S. as a function of skin tone, as well. He found a significant and strong positive association between skin tone and educational attainment. Meaning, participants' lighter skin tone was associated with a higher educational attainment and, therefore, grade Soc. Sci. 2022, 11, 15 7 of 21 point average (GPA). Ryabov (2013) used the same ADD Health data again to estimate the impact of skin tone on school-to-work and school-to-college transitions of African American youth. Through multinomial logistic regression, he found African American males with the lightest skin tones were more likely to find employment and to be enrolled in college than their same-race peers with darker skin tones. Additionally, the odds of finding a job were also significantly higher for African American females with the lightest skin tones. This is cross-listed as a social outcome.
Ryabov (2019) also used ADD Health data to explore Latinx adolescent educational outcomes. Those with whiter phenotypic features (e.g., light skin and blue eyes) had better academic outcomes than those with darker skin and brown eyes. Thompson and McDonald (2015) used ADD Health data to explore how perceived race and skin tone influenced grade point average, as well. They found significant variation across and within racial groups, with darker-skinned individuals receiving significantly lower grades than their lighter skin counterparts. This was also very noticeable in non-Black racial groups. Vasquez (2010) interviewed Latinx students about how their skin color influenced levels of support at school. Her findings suggest that protective support was given to lighter skin students, including mentorship, by school leaders when they were perceived to be White. However, when they discovered that the students were Mexican, this protective support was removed. Those who appeared to be more Mexican were consistently left without support. On the contrary, those who appeared more White could move in and out of the Mexican/White categorization, allowing them to receive additional support. This is cross-listed as a social outcome and an emotional outcome. In another study, Gullickson (2005) found that education and labor market outcomes were less influenced by skin color differences in 2005 than previously found. He used the National Survey of Black Americans Series (NSBA) wave one of adults aged 18 and older. This is cross-listed as a social outcome.
Monroe (2013) later wrote a conceptual piece for educators to consider the influence of colorism in their mentorship, professional experiences, school discipline, and when working with communities of color. Campbell (2009) found that multiracial youth had educational attainment more heavily linked to parental education and family income than skin color. Hannon et al. (2013) then conducted a logistic regression using the National Longitudinal Study of Youth (NLSY), which included a sample of 12-16 yearolds. The results indicated that darker-skinned Black girls were three times more likely to be suspended than lighter-skinned Black girls. Similar findings were discussed by Blake et al. (2017), using a multilevel logistic regression to control for background and school factors of an ADD Health sample of young African American female adolescents. The researchers found that colorism is a significant predictor of suspension and darkerskinned African American girls had the highest risk. On the other hand, Alarcón et al. (2000) discovered that skin color ratings were not significantly associated with grades, but that other characteristics, like the background of neighborhood, were associated with self-satisfaction among early childhood aged students.
Additionally, colorism can impact youth into adulthood through academic achievement and educational experiences. Biagas and Bianchi (2016) explored skin tone stratification among White and Hispanic undergraduates (with a sample that included 18 year-olds) at a large midwestern university. The authors viewed recordings of group interactions of three girls and three boys. The two groups included three individuals who were (1) White, (2) White-appearing Hispanic, and (3) Black-appearing Hispanic. In exploring who was the influencer of the group, Biagas and Bianchi (2016) found that both the White girl and boy were the major influencers of the group, followed by the White-appearing Hispanic, with the Black-appearing Hispanic having little influence on the group. This created in effect a three-tiered system of racial inequality. This is cross-listed as a social outcome. As for people aged 18 and over, Monk (2014) conducted over 6082 face-to-face interviews with persons of color aged 18 and over, which highlight the fact that skin tone stratification among African Americans still persists. The authors found that skin tone is significantly associated with Black Americans' educational attainment, household income, occupational status, and even the skin tone and educational attainment of their spouses. This is cross-listed as a social outcome.

Social Outcomes
The impacts of colorism on social well being have been explored across racial and ethnic groups. Bailey (2000) used a case study approach to investigate the classroom experience of one Dominican American student, analyzing how his phenotype led other students to mistakenly classify him as African American until he used the Spanish language. Even within his own Hispanic culture group he was not automatically accepted or associated as a member, until he spoke Spanish, due to his skin tone. For Hispanic students, phenotype and language factor into their social school experiences. Park (2011) observed preschool students in their social interactions with peers at school, as well. In her analysis, she found that these peer interactions were heavily based on skin tone and their understanding of skin tone was associated with Native heritage. This is also cross-listed as a social outcome.
In Bellinger's (2007) study, the author used participant observation and interviewed 15 African American teens aged 16-18 about their concepts of 'good hair'. The findings of this study indicate that many of the young African American girls changed their hair as a social status because 'good' was a concept taught by their mothers. In addition, they learned the myth of straightening hair as a source of manageability. Bozo et al. (2018) found that Latina/o adolescents reported experiencing more microaggressions and having darker skin color than Asian American adolescents. The ANOVA analysis indicated, no main effect of gender was found for microaggressions, nor were there significant interactions between gender and ethnicity for microaggressions or skin color. Descriptive statistics showed participants reported having light brown facial skin color and experiencing microaggressions once per year or rarely. Facial skin color was positively associated with microaggressions (r = 0.12, p = 0.033). Nativity status significantly moderated the association between skin color and microaggressions. Among foreign-born adolescents, there is a positive association between skin color and microaggressions (B = 0.182, SE = 0.089, p = 0.046). Among U.S.-born adolescents, there is no association between skin color and microaggressions (B = 0.017, SE = 0.019, p = 0.377).
Reece (2019a) used ADD Health data, which included over 90,000 adolescent responses across four waves of data collection beginning in 1994. In this study, Reece (2019a) found that skin color plays a role in shaping identity changes and lighter skin adolescents chose non-Black single race as their identity more often, while darker-skinned adolescents chose Black as their identity. However, the interviewer identified all of this subsample as Black. This study implies that there are social consequences when there is a mismatch between chosen identity of adolescents and perceived identity by others, such as the interviewer in this study. Ryabov (2019) once again used ADD Health data to examine skin tone as a predictor of physical attractiveness and social status for Black men and women. The analysis shows that lighter-skinned young Black individuals attain a higher educational level, receive higher wages and enjoy better quality jobs than their darker-skinned coethnics. Moreover, the results show that more physically attractive young Black individuals, especially women, are advantaged in terms of educational attainment, wages, and job quality than their less physically attractive counterparts. These findings suggest that, among Black people, the skin color stratification coincides with that based on physical attractiveness to a large degree, with the implication being that the skin tone is a predictor of both physical attractiveness and social status for Black men and women. Eick (2010) used oral histories from Black and White high school students to determine how skin color limited access to social class post-Civil War. As poor Whites could dress up and be perceived as wealthy, poor Blacks could not do the same and were still seen as Black whether or not they had social class.

Emotional Outcomes
Various emotional challenges have been identified for students experiencing colorism. Araujo and Borrell (2006) reviewed 14 articles about outcomes for Latinx in education and employment. Their review largely demonstrates the links between skin color discrimination and mental health outcomes. For example, Araujo and Borrell (2006) reported darkerskinned Mexicans who had a heightened experience with colorism also grappled with stress, leading to higher levels of depressive symptoms. Ayers et al. (2013) examined a five-year randomized control trial of a substance abuse prevention intervention of 1150 Mexican students in eighth grade from 28 public schools in Phoenix, Arizona. Through their logistic regression, they demonstrated that ethnoracial appearance (i.e., phenotypic features like hair texture, eye/hair color, etc.) increases vulnerability to substance use and that this relationship is moderated by the strength of ethnic identity and generation status. Calzada et al. (2019) investigated the association between skin color as a representation of racial phenotype and the mental health of young Latinx children. By the end of the first grade, children perceived as Black had higher ratings on internalizing and externalizing problems including anxiety, depression, and anger compared to those Latinx children perceived as White. This study found that darker-skinned Latinx kids are at increased risk for emotional health challenges. Monk (2015) reported on how skin tone was associated with perceived discrimination among African American young adults. The author revealed that skin tone was linked with issues of depression and even other health-related outcomes where Black/White health disparities exist. Additionally, Mathews and Johnson (2015) found that, for young African American college students, there were no significant associations between self-esteem and skin tone, nor social capital. Their results suggest a buffer of colorism for medium brown skin as opposed to very light or very dark-skinned African Americans. Ramos et al. (2003) found that when Afro-Latina females were compared with European American, African American, and Latino adolescents, they reported statistically higher rates of depression due to their dual racial and ethnic identities. This is often impacted by their physical features like skin tone that lead others to identify them as Black. Vasquez (2010) and Monk (2014) are cross-listed as academic and/or social outcomes and are summarized in those sections. Moreover, view the Appendix A: Table A1 Scoping Review Results.

Synthesis of Results
The majority of the studies identified in the scooping review illustrate experiences with colorism that influence academic outcomes for students of color. Whether through teacher bias towards those with darker skin and Afrocentric features or through the lack of perceived academic support in schools, the academic experiences of students of color are impacted by bias. This implicit and explicit bias against darker skin results in the privileging of lighter skin and more Eurocentric features. There were more articles on Latinx student populations than there were other student groups of color (e.g., African American, Native American, and Asian American). One of the most significant findings across all racial groups is that lighter skin and features are significantly associated with better academic achievement, including additional years of schooling.
Results of this scoping review also indicate the complicated social status and relationships for students of color facing colorism in schools. Namely, a focus on appearance dominated the articles ultimately included in the review, revealing that students might even change their hair and/or speech to avoid the consequences of colorism. Further, the experiences of students of color appear to be less warm and supportive among peers and teachers when they have more Afrocentric features. Even when students identified themselves as non-Black, others often identified them as Black due to their skin color and other phenotypic features, setting up conflict in their peer relationships. In essence, the studies found a three-tiered hierarchy in peer relationships in schools, where White students maintain a higher social status than lighter-skinned students of color, but lighter-skinned students of color maintain a higher social status than darker-skinned students of color on interracial campuses. Additionally, the fluidity of movement between social classes and privilege within schools for White or lighter-skinned students of color was apparent in the study findings when compared to darker-skinned students of color.
The results of the emotional outcomes analysis largely identified darker skin as a heavy risk factor for substance abuse and emotional problems, including externalizing and internalizing symptoms such as anger, anxiety, and depression when compared to lighterskinned Latinx students or those Latinx students perceived as White. These outcomes were seen in children as young as six years-old and also included an increased risk of substance abuse, in addition to other emotional problems.

Discussion
Colorism is a form of oppression that is expressed through a hierarchical treatment of individuals where, typically, favoritism or superiority are associated with those of lighter complexions, and those of darker complexions experience rejection and mistreatment in areas such as the educational system (Jackson-Lowman 2013). The objective of this scoping review was to assess the academic, social, and emotional experiences and outcomes linked to colorism for students of color in American public schools. The value of a scoping review that synthesizes the experiences of students with various racial/ethnic/skin tone is useful not only to researchers exploring the total impact of colorism but also to educators across schools with varying demographics. Rather than assuming that colorism only impacts certain groups, it is clear from this review that colorism is a shared experience across many cultural groups. Therefore, professional development for teacher and school staff (including leaders, counselors, social workers, and school resource officers) should address this important topic that impacts student outcomes. These staff have a large influence on the student experience and must be aware of the pervasiveness of colorism in schools.
While there are limitations to the study, several themes arose from this review. The first theme was confirmed in previous literature across ethnic and racial groups and was also included in the review: lighter skin tones and more Eurocentric features were linked to better academic outcomes, including higher GPAs, additional years of schooling, improved academic performance, more teacher/peer support, decreased rates of suspensions, and better school to work/college transitions. Thompson and McDonald (2015) suggest this occurs because of mediating factors such as teacher bias and exclusionary rules. In general, social inequalities stemming from colorism are produced through relationships such as teacher-student relationships within schools' punitive sanctions. As such, darker-skinned students face the greatest barriers to optimal educational outcomes due to differential treatment. This finding underscores the need for implicit bias and multicultural competence training that exposes teachers to the causes and consequences of stereotypes and cognitive biases (Quiocho and Rios 2000;Thompson and McDonald 2015). This supports the need for alternative disciplinary strategies that preclude students from missing key instructional time (Thompson and McDonald 2015).
The second theme was that students of color may engage in any number of behaviors to mitigate the experiences of colorism they face in schools. The review indicated how some students of color change their physical features (e.g., hair, speech) to simulate Eurocentric traits and characteristics. For students of color, and in particular darker-skinned female students, this was done to avoid colorist experiences and discrimination. Hunter (2016) revealed that empirical evidence suggests that skin color modifies outcomes and produces advantages in education for lighter-skinned students of color. In this way, lighter skin color and more Eurocentric features grant privilege and social capital in education, which often cannot be accessed or created by darker-skinned more Afrocentric-appearing students across racial groups. School staff and administrators who better understand colorism can advocate for the development of critical consciousness among students of color, for them to resist harmful phenotype-associated ideologies and messages. This allows for a psychological safe space in which students of color, in particular darker-skinned students, can form and sustain positive racial and academic identities (McGee et al. 2016). Furthermore, school staff can be intentional in their inclusion and promotion of belongingness for all the range of skin tones, with a focus on equitable treatment for darker-skinned more Afrocentric-appearing students across all racial groups.
The third theme was mental health and substance abuse outcomes which varied by skin tone. The review found that darker skin students of color experienced a higher likelihood of mental health and substance abuse problems when compared to lighterskinned students of color within the same racial group and then when compared to their White counterparts. This is an area that warrants more attention in future research because this finding may suggest that skin color impacts self-concept and self-esteem, which in turn influences overall mental health ). An example of this claim was observed by Perry et al. (2013), where the researchers found that in African American females, increased suicidal ideations and behaviors were associated with the identification of medium or darker skin tones. In addition, ethnic affirmation (i.e., positive feelings about one's race) was found to be a key protective factor. The literature reviewed for this study continues to show the shortcomings of the available research that provides depth of knowledge about the student experiences of multiple racial groups with colorism in schools. The result of the review is that multiple studies show how colorism in schools inhibits academic, social, and emotional outcomes for students of color.
A higher concentration of the literature focused on academic outcomes and colorist experiences among the African American and Latinx populations. Although colorism has been observed in Asian and Native American populations, the literature search yielded limited studies focusing directly on these populations during the last two decades. Beyond this, the majority of the studies were quantitative research using secondary datasets such as ADD Health, NLSY, and NSBA. The current study's findings confirm the outcomes in the current literature and help to fill some knowledge gaps to better understand colorism in schools for students of color.

Limitations
Despite strong findings that support the understanding of the experiences of students of color with colorism in schools, this study also had notable limitations. First, this study did not employ the framework that some scoping reviews use, a sixth phase, where consultation which is optional (Arksey and O'Malley 2005). Stakeholders were also not consulted, and as displayed in the protocol, this was consistent with others' scoping review process (e.g., Perreira et al. 2018;Lehning et al. 2017). The lived experiences informed by such people as public school students and educators could have possibly added more insight to the review. It is suspected that one of the keywords, discrimination, listed in the search generated much of the literature regarding racial discrimination, which did not meet the study criteria prior to the full-text screening method. For this review study, Covidence specifically generated reasons for excluding data sources during the entire screening procedure based on incorrect setting, publication date, and incorrect population age. The back of reference list limitation was consistent with the latter reasons. This study had 28 sources that did not meet all-inclusive criteria. These exclusion criteria were indicative of the incorrect type of study topics on racial discrimination. In contrast, this review illuminates an exhaustive scope of data collection and methodological rigor aligned with the guidelines of conducting a scoping review (Arksey and O'Malley 2005). Finally, although Asians and Native Americans experience colorism, the data revealed there is a dearth of literature in the child and adolescent category. More of the data collection in this study generated research on African Americans and the Latinx group.

Conclusion and Future Directions
The objective of this scoping review was to assess the academic, social, and emotional experiences and outcomes linked to colorism for students of color in American public schools. The current study's findings revealed three themes: (1) Across ethnic racial groups, lighter skin tones and more Eurocentric features were linked with better academic outcomes; (2) Students of color, in particular darker skin students, may alter physical features and characteristics to mitigate the consequences of colorism and discrimination in schools; (3) Darker skin students across racial groups had more experiences related to mental health and substance abuse problems. The evidence to date leaves advocates in this area with some optimism. Future directions for school reform leaders can focus on a commitment to culturally relevant pedagogy, curricula, and programming. Since "whiteness" is affirmed by default throughout U.S. society, educators and administrators must not be afraid to proactively affirm other skin tones that more resemble "blackness." In addition, other future directions should center on the constant monitoring of school and classroom discipline by checking for implicit or explicit colorist bias in patterns of both punishment, rewards, and achievement. Beyond this, providing targeted counseling and mediation to address and prevent colorist experiences for students of color is warranted. There is still much work to be done on reducing experiences with colorism in schools. On the contrary, all stakeholders in the success of students of color can take what has been found in this study as an impetus to engage in work that reduces colorist behaviors in public schools, which greatly affect darker skin students of color across racial groups.

Data Availability Statement:
The articles listed within this scoping review are available in online databases through most university library systems.

Conflicts of Interest:
The authors declare no conflict of interest. Table A1. Scoping review results.

3-year longitudinal, quantitative survey study
To examine the psychosocial development, physical health, and behavioral adjustment of Puerto Rican children Children's ratings of their skin color were not associated with their sex, school grade, ethnic identity, the child's or the parent's nativity, or the racial or ethnic compositions of 3 social contexts: their neighborhood, their classmates, and their closest friends. Puerto Rican children did not show a preference for light-colored skin. Moreover, there were no significant differences in self-esteem based on the child's self-reported skin color. The lack of association between self-esteem and skin color was interpreted in light of a developmental tendency prevalent in early to middle childhood to place a positive value on different aspects of one's self. Ayers et al. (2013) Data come from the sixth wave of a 5-year randomized controlled trial of a substance-use prevention intervention that surveyed 1150 Mexican heritage students, then in eighth grade, from 28 public schools in Phoenix, Arizona

Regression model
To examine how Mexican students' ethnoracial appearance predicted adolescent substance use, and whether the relationship differed by generation status and strength of ethnic identity Results of the logistic regression reveal that generation status and ethnic identity moderate the relationship between ethnoracial appearance and substance use after adjustment for demographic covariates. For use of alcohol, ethnoracial appearance does not have a significant main effect; only the interaction between second-generation status and ethnoracial appearance is significant. For second-generation adolescents, increases in a more indigenous ethnoracial appearance decrease the predicted odds of using alcohol. Bailey (2000) 11th grade student at Central High School, a Providence City school of 1350 students, which is over 20% Dominican Case study; ethnography observation, interviewing, and video-recording To describe on one 17 year-old Dominican American's negotiation of ethnic/racial identities during a single class period at school Analysis of Wilson's talk and interaction reveals much about the local roles of language and phenotype in the negotiation of identity, showing that language can situationally precede phenotype as a criterion for racial classification and that racial identities can shift across linguistically, interactionally constructed contexts.

Participant observation and open-ended interviews
To explore the definition of "good hair" and to understand the reasons young African American women choose to change physical attributes Respondents agree that "good hair" for Black men and women is an unattainable myth perpetuated by generations of (mostly) Black women. Younger African American women in the study change their hair because their mothers did it with the hopes of giving one better opportunities for new jobs or promotions. However, the downside to changing their hair is that, to others, it appears they have lost their racial pride. Biagas and Bianchi (2016) Whites and Latino/as with varying skin tones; Midwestern University Experimental methods analyzed through ANOVA To examine whether Whites exhibit social status, in a manner consistent with this three-tier system (Whites, Honorary Whites, and Collective Blacks) White was influenced the most by White confederates; Whites were influenced less by their light-skinned Latino/a confederates, and even less by dark-skinned Latino/a confederates. This finding supports the notion that skin tone and ethnic background have the equivalent capacity to invoke processes to create the tripartite system of racial/ethnic inequality. To investigate the association between skin color, as a salient marker of racial phenotype, and the mental health of young Latinx children By the end of first grade, "collective black" children had higher ratings on several indicators of internalizing and externalizing problems compared to their "honorary white" peers; this pattern was particularly pronounced for girls. Moreover, the association between externalizing behaviors at baseline and first grade was stronger among children with dark, relative-to-light, skin color.
Campbell (2009) 7th-12th graders (white, black, Asian, or Native American) in the United States in 1995 Ordinary least squares regression for (GPA), logit model for (enrollment in an advanced math course) To investigate how multiracial groups fit into the system of racial oppression The educational outcomes of Latinos and single-race groups are significantly associated with skin color and the racial perceptions of observers. Multiracial young adults' high school and college educational outcomes are not consistently related to either measure of appearance. Parental education and family income are the most important predictors of educational outcomes for multiracial respondents across different types of outcomes. Eick (2010) Black and White high school students; Baltimore, MD Oral histories of graduates from Miller Town in Baltimore, MD To explore the inter and intraracial relationships of students in schools during desegregation Skin color and a white middle-class look shaped white students' perception of class status. White girls of poor background who could assume a middle-class appearance could gain access to myriad extracurricular clubs where they mingled with the richer youth. On the other hand, young black women of middle-class background in the early years of desegregation were marginalized. Similarly, for the black male youth of some means attending the general track, there was no escaping the racialized class constructions.

Author(s)/Year Study Population/Location Methods Aims of the Study Results
Fergus (2017) Six 9th-12th grade students; Central American and Mexican student/Midwestern city

Phenomenological framework
This explores how six Mexican students signal the meaningfulness of skin color in their identification as Latino/a/Mexican and how such experiences interact with their notions of the opportunity structure Six Mexican students who discussed teachers identifying them as "White-looking" or "Hispanic/Mexican-looking". The "White-looking" students perceived themselves as being able to permeate structural barriers; meanwhile, the "Hispanic/Mexican-looking" students believed such barriers affected their ability to "make it", regardless of their aspirations Gullickson (2005) Black Americans/18 and up from the National Survey of Black Americans

Multivariate models
To examine the changes across cohorts in differentials within educational, occupation, and spousal attainment The authors found significant declines in skin tone differentials for younger cohorts, in terms of educational and labor market outcomes, but not in terms of spousal attainment. These declines begin with cohorts born in the mid-1940s. In addition, there is evidence of period declines of skin tone differentials in occupational attainment in the 1980s. I discuss possible explanations for the declines. Hannon et al. (2013) African American youth aged 12-16 years old/National Longitudinal Survey of Youth

Logistic regression
To examine whether skin darkness affects the likelihood that African Americans will experience school suspension This was disproportionately driven by the experiences of African American females. The odds of suspension were about 3 times greater for young African American women with the darkest skin tone compared to those with the lightest skin tone.
Hunter (  To investigate the association between skin color and academic achievement in young Latino students The findings showed that for Dominican-origin (Dominican American = DA) boys, being "collectively black" was indirectly associated with lower academic achievement in first grade, mediated by lower teacher-rated adaptive behavior in prekindergarten or kindergarten. Specifically, when the "collective black" boys entered prekindergarten or kindergarten, 27% were academically delayed, and by the end of first grade, the percentage at academic risk had increased to 36%. In contrast, when the "honorary white" boys entered school, 30% were delayed on academic school readiness, but only 21% were at risk by the end of first grade. Thus, given that only DA children were rated in the darkest skin color category, DA boys who are "collective black" emerged as the group at possible highest risk for academic underachievement. For DA girls, skin color was not significantly associated with academic achievement in first grade.

Mathews and
Johnson (2015) African American women (between the ages of 18 and 23)/Historically Black College/University at a Southern University Mixed methods survey research and qualitative responses, ANOVA To explore participants' perceptions and preferences for skin complexion ANOVA results of skin tone and the self-esteem index indicate that the participant's skin tone did not have a significant effect on their level of self-esteem, F(2, 84) = 1.196, p = 0.307. Women who measured high in self-esteem would also measure high in social capital, irrespective of skin complexion.
Monk (2014) Black Americans, 18 and over 6082 face-to-face interviews To explore how skin tone stratification among black Americans persists into the 21st century The authors found that skin tone is significantly associated with black Americans' educational attainment, household income, occupational status, and even the skin tone and educational attainment of their spouses.
Monk (2015) African American sample is nationally representative of black households in the 48 coterminous states with one adult age 18 and over

Multistage probability design
To examine the relationship between skin tone, discrimination, and health among African Americans Skin tone is a significant predictor of multiple forms of perceived discrimination (including perceived skin color discrimination from whites and blacks) and, in turn, these forms of perceived discrimination are significant predictors of key health outcomes, such as depression and self-rated mental and physical health. Intraracial health differences related to skin tone (and discrimination) often rival or even exceed disparities between blacks and whites as a whole. The author also finds that self-reported skin tone, conceptualized as a form of embodied social status, is a stronger predictor of perceived discrimination than interviewer-rated skin tone.

Monroe (2013) Black and White populations Literature review
To explore how and why colorism motivates imbalances within communities of color The author recommends that educators consider colorism during mentorship, professional experiences, school discipline, and in when working with additional communities of color. Perceived discrimination and skin color are both negatively related to grades, whereas maintaining Spanish is positively related to grades, collectively accounting for almost 20% of the variability in grades. For skin color, each 1 SD in skin darkness increase is associated with a decrease of 0.20 SD in grades. There is a moderating effect of discrimination on attachment, accounting for approximately 13% of the variability in attachment scores.
Park (2011) Six young children ages 3.5 to 5.5; 3 European, 3 multicultural groups/metropolitan area on the West Coast

Qualitative observations of regular classroom activities and discussions
To explore how children learn about diversity First, students made claims to identities that demonstrated the significance of physical markers commonly associated with race including skin color. Students were stretched in new directions in their social interactions with peers and had valuable opportunities to use experiences beyond their own to gain new insights about racial and ethnic diversity. Ramos et al. (2003) Add Health, a nationally representative survey sample of over 90,000 adolescents across four waves of data collection over 20+ years Ordinary least squares regression model To compare depressive symptoms among European Americans vs. African Americans, Latinos, and Afro-Latinos Afro-Latina females, who exhibit characteristics like skin tone that are often associated as "Black", reported statistically higher rates of depression than males of all of ethnic groups. This suggests that Afro-Latina females face higher levels of stress due to their dual ethnic and racial identities of African American and Latina.

Reece (2019a)
Add Health, a nationally representative survey sample of over 90,000 adolescents across four waves of data collection over 20+ years

Multinomial logistic regression
To examine the effect of skin tone on racial identity change, exploration of adolescent racial identity formation Results reveal that lighter-skinned adolescents are more likely to change their identification to a non-black single race, while darker-skinned adolescents are more likely to change their identification to black only. Color seems to play a large role in shaping identity changes. Ryabov and Goza (2014) National Longitudinal Study of Adolescent Health (sample of adolescents and young adults)

Multivariate analyses
To examine the impact of phenotyping on academic and employment outcomes among Latino adolescents/young adults Results indicate that, among Latinos, light skin and blue eyes are associated with better academic outcomes than having dark skin and brown eyes, while those with darker skin enter the labor market earlier than their light-skinned coethnics.
Ryabov (2013) Add Health resulted in a nationally representative sample of adolescents between the ages of 12 and 21 years

Multinomial logistic modeling
To estimate the impact of skin tone on school-to-work and school-to-college transitions of African American youths The findings suggest that African American males with the lightest skin tone were more likely to find a job and to be in college than their coracial peers with darker skin tones. The odds of finding a full-time job were also significantly higher for African American females with the lightest skin tone. Family background factors, marital status, prior achievement, and average school socioeconomic status matter the most. To test the hypothesized relationship between skin tone, physical attractiveness, gender, and status attainment, while controlling for a range of individual socio-economic and family effects The analysis shows that lighter-skinned young Blacks attain a higher educational level, receive higher wages and enjoy better-quality jobs than their darker-skinned coethnics. Among blacks, the skin color stratification coincides with physical attractiveness to a large degree, with the implication being that the skin tone is a predictor of both physical attractiveness and social status for black men and women.

Ryabov (2016) Add Health Regression
To evaluate the educational attainment of Asian American and Hispanic young adults in the United States as a function of skin tone and other covariates Results consistently point to a strong association between educational attainment and the lightness of skin tone. The findings also suggest that the aforementioned relationship is the strongest among U.S. young adults of Filipino and Puerto Rican descent. Thompson and McDonald (2015) Add Health Regression models To examine the extent to which self-identified race category and perceived skin tone influence educational performance independently and simultaneously Analyses show significant skin tone differences in grade point average (GPA) both across and within racial groups, with darker skin-toned individuals receiving significantly lower grades than their lighter skin-toned counterparts. Net of controls and skin tone differences in GPA are essentially flat among African Americans but are notably stronger among other race/ethnic groups. These findings highlight the interplay between racial categorization and colorism by revealing the categorical disadvantage of racial stigma versus the more fluid colorism boundaries of nonblack groups.
Vasquez (2010) Twenty-nine third-generation Mexican Americans ages 17 and up/Northern and Southern California

Qualitative
To explore how the lives of middle-class third-generation Mexican Americans both racialized and gendered Findings show that Mexican women are afforded more "flexible ethnicity" than Mexican men. Accordingly, Mexican men are more rigorously racialized than Mexican women. They are racialized through exoticization, whereas Mexican men are racialized as threats to safety. Lighter-skinned Mexican individuals escaped consistent racialization.