Schooling Experiences and Educational Outcomes of Latinx Secondary School Students Living at the Intersections of Multiple Social Constructs

Abstract

Based on the critical quantitative intersectionality framework, this study examined the relationships between Latinx students’ multiple intersecting social categorizations (i.e., gender, ethnicity, home language, socioeconomic status [SES], immigration status), their diverse schooling experiences, and educational outcomes. This study found that specific ethnic backgrounds for Latinxs were significantly associated with both their schooling experiences and educational outcomes. Latinxs’ SES was a critical factor in creating the intersectional effects associated with other social constructs for schooling experiences and educational outcomes. Living in poverty created greater challenges in experiencing school suspension for Latinos compared with their Latina counterparts.

Keywords

Hispanic students urban education high school programs intersectionality educational equity

Introduction

The Latinx population in the United States has significantly increased, and since 2000, it has accounted for more than half of the nation’s overall population growth. The term Latinx in this study refers to individuals who have origins in Mexico, South America, Central America, or the Caribbean (e.g., Cuba, Puerto Rico, the Dominican Republic), and who share ethnic, linguistic, and cultural backgrounds from Spanish colonization (Alemán, 2018). Furthermore, Latinx is an inclusive term that deconstructs gender binaries and includes any person who has a gender-fluid or gender-nonconforming identity (e.g., Castañeda, Anguiano, & Alemán, 2017; Castrellón, Reyna Rivarola, & López, 2017). Along with the growth of the Latinx population in the United States, the number of Latinx students enrolled in U.S. schools has also increased significantly. Although the percentages of White and Black students in U.S. public schools are projected to decrease from 2013 to 2025 (from 50% to 46% and from 16% to 15%, respectively), the percentage of Latinx students is projected to increase from 25% to 29% over this period (National Center for Education Statistics [NCES], 2017). Considering this, it becomes clear that providing adequate, equitable education for these students is a critical policy concern, both to improve overall educational outcomes and to ensure the long-term economic growth of the United States (Núñez & Kim, 2012).

Currently, however, numerous educational researchers have identified a series of critical issues involving inequities in the educational outcomes and schooling experiences of Latinx students in the United States (e.g., Núñez & Kim, 2012; Oakes, 1990; O’Connor, 2009; Ramírez, 2013; Reardon, Valentino, Kalogrides, Shores, & Greenberg, 2013). For example, Reardon et al. (2013), using data from the National Assessment of Educational Progress and the state accountability tests, reported significant achievement gaps between White and Hispanic students, and showed that these gaps were quite large across all 50 states. Their research showed that the disparity in reading achievement between White and Latinx students for the 2002 cohort in fourth grade was 0.64 standard deviations (SD). While criticizing discourses of achievement gaps in standardized tests, scholars (e.g., Carey, 2014; Milner, 2012, 2013) have also focused on the educational processes (i.e., opportunity gaps) that perpetuate gaps in outcomes. In particular, numerous studies on educational inequity have identified structural barriers encountered by Latinx students, not only in school (e.g., curriculum, access to school resources) but also outside school (e.g., institutionalized racism, colonialism).

Although many previous studies have identified educational inequalities for Latinx students in terms of achievement scores (e.g., Reardon & Galindo, 2009), college enrollment (e.g., Núñez & Kim, 2012), or degree attainment (e.g., Covarrubias & Lara, 2014), few studies to date have explored the diverse schooling experiences and educational outcomes of particular groups of Latinx students. Indeed, research that is based on one-dimensional social categories (e.g., race, ethnicity, gender, immigration status) typically homogenize the lived experiences of Latinx students by assuming that these people form a single undifferentiated group (Alemán, 2018; Covarrubias, 2011; Covarrubias & Lara, 2014). Due to the failure to consider the differing effects of multiple socially constructed categorizations (e.g., ethnicity, gender, immigration status), students who live in the margins among different Latinx groups may be rendered invisible, and therefore remain unseen in research and practice (Jang, 2018). Consequently, school leaders and policy makers may be limited in their capacity to challenge the particular forms of injustice, inequity, and oppression faced by students who belong to multiple marginalized student groups.

This study challenges the narrow perspectives on educational reform that are commonly found in dominant narratives, and which subvert or ignore the needs of particularly disadvantaged groups in urban education. It seeks to identify the requirements for equitable leadership and education, particularly for little-recognized groups of Latinx students. Diverse social constructs that include factors relating to ethnicity, gender, language, socioeconomic status (SES), and immigrant status are critical for discerning the structural inequalities experienced by Latinx students in their schooling experiences and outcomes (e.g., Alemán, 2018; Covarrubias, 2011; Covarrubias & Lara, 2014; Núñez & Kim, 2012; Ovink, 2014; Reardon & Galindo, 2009; Sólorzano & Bernal, 2001; Teranishi, 2007). More specifically, Teranishi (2007) has emphasized the importance of disaggregating racial categorizations into smaller groups by ethnicity to more fully capture the lived experiences of students considered in quantitative research. Similarly, a number of researchers have emphasized the heterogeneous lived experiences of differing ethnicities among the Latinx population (e.g., Reardon & Galindo, 2009). Immigration status is another critical factor that differentiates the schooling experiences of Latinx students, due to differing levels of language proficiency (Covarrubias, 2011) and of legal or economic security. The Trump administration’s recent anti-immigration policies toward undocumented students (Castrellón et al., 2017), at least 70% of whom are from Mexico or Central America (Hoefer, Rytina, & Baker, 2011; Passel & Cohn, 2011), have increased the difficulties these students face. This study also differentiates the schooling experiences and educational aspirations of female students from those of their male counterparts, given that various researchers have shown that significant gender gaps exist between Latino and Latina students (Covarrubias, 2011; Núñez & Kim, 2012).

Overall, this study aims to identify distinct patterns of educational inequities in Latinx students’ schooling experiences and math achievement, as they relate to these students’ multiple intersecting social categorizations (i.e., gender, ethnicity, home language, SES, immigration status). The term experience as used in this study broadly represents a series of events occurring in schools, and the students’ interpretations of those events. On the basis of that definition, this study focuses on the students’ exposure to high-quality teachers, their experiences of school suspension, and their schools’ structural characteristics (i.e., the percentage of students receiving free or reduced-price lunches), as a means of evaluating their schooling experiences. Furthermore, this study uses students’ math scores to assess their academic achievement.

This study corresponds with the criticism of urban education scholarship regarding the confined and distorted focus on the achievement gap to understand the diverse schooling experiences of Latinx students (e.g., Carey, 2014; Irvine, 2010; Milner, 2012, 2013). Solely focusing on gaps in test scores creates problems in fully understanding the nuances and complexities of learning, and thus limits educators’ capacity to support culturally diverse students (Carey, 2014). Describing and exploring the diverse schooling experiences of Latinx student populations can provide meaningful insights for urban educators, school leaders, and policy makers. Such insights are particularly important for urban educators seeking to challenge a system that remains a colonizing and oppressive space for this particular group of students. This is because urban educators confront the greatest challenges in working with the most recent immigrants, from lower socioeconomic families, and Latinx students in urban schools (Rong & Brown, 2002). In this study, urban education is conceptualized not only by geographical location (i.e., location in a large metropolitan area) but also by “a wide range of student diversity, including racial, ethnic, religious, language, [gender, immigration, and socioeconomic status]” (Milner & Lomotey, 2013, p. xv). In this context, the study further aims to enable urban educators to recognize systematic disparities in the educational experiences of Latinx students as well as potential “cultural conflicts” (Milner, 2012, p. 701), thereby diverting their implicit and often explicit color-blind practices and deficit mindsets toward Latinx students.

The critical contribution of this study lies in revealing the multiple structural inequities that are deeply associated with the ethnicity, immigration status, and gender of Latinx students. This study aims to generate greater understanding of these inequalities, and to focus the efforts of those working for social justice where needs are greatest. To discern these needs, we first need a new framework for looking at marginalized populations. The following section proposes such a critical, quantitative framework of intersectionality.

Theoretical Framework

To better understand the nuances and complexities in Latinx students’ schooling experiences in addition to their disparities in achievement, this study integrates different strands of scholarship. It draws on the literature illustrating the importance of diverse social categorizations and how they intersect to create nuances in Latinx students’ experiences. It further encapsulates scholarship on critical race theory, particularly Latina/Latino critical race theory (LatCrit), as well as the framework of critical quantitative intersectionality.

The Importance of Diverse Social Categorizations in Perceiving Latinx Students’ Experiences

Extensive research in the United States has documented the inequality in Latinx students’ educational outcomes and schooling experiences, compared with those of their White or Black counterparts. These inequalities for Latinx students have been found to persist from kindergarten to postsecondary education. Starting at the kindergarten level, Latinx children show significantly less readiness to absorb the instruction provided than White or Black children (e.g., Duncan & Magnison, 2005). The rates of Latinx students’ school attendance (Cameron & Heckman, 2001; Covarrubias & Lara, 2014), graduation from high school (Perreira, Harris, & Lee, 2006), and graduation from college (Núñez & Kim, 2012; Stoops, 2004) are substantially lower than those of White students.

Various scholars have suggested a number of underlying reasons for the low educational outcomes of racial minority students, and particularly of Latinx students. Some researchers have argued that the lower performance of these students reflects their cognitive and motivational shortfalls (i.e., the deficit thinking model). Other scholars have criticized this interpretation, claiming that it fails to consider the structural barriers and forms of oppression that racial minority students experience (e.g., Milner, 2012; Valencia, 1997). These scholars point to a large body of evidence indicating that Latinx students experience a series of structural barriers that critically affect their educational outcomes. Scholars who study these structural barriers commonly argue that racial minority students face “accumulated differences in [gaining] access to key educational resources” (Darling-Hammond, 2010, p. 28). These scholars have identified race-related “opportunity gaps,” or gaps in the “opportunity to learn (OTL)” (Schmidt & McKnight, 2012, p. 13). Educators who are concerned with such opportunity gaps and the OTL (e.g., Carey, 2014; Irvine, 2010; Milner, 2012, 2013) often contend that inequalities in educational inputs are exacerbated over generations, so that historical, economic, and sociopolitical factors accumulate as an “educational debt” (Ladson-Billing, 2006, p. 5).

Many educational researchers have highlighted the institutionalized inequality that also prevails in school organizations (e.g., Delpit, 2012; Irvine, 2010). For example, Galindo and Reardon (2006) found that teachers of Latinx students typically have less teaching experience than teachers of White students, although these teachers are similar in terms of their credentials (i.e., certifications or advanced degrees). This issue of unequal teacher quality for disadvantaged students suggests that policy makers need to “define their ideal distribution of teacher quality” (Galindo & Reardon, 2006, p. 305).

Most studies on educational inequality have compared students on the basis of race, but understanding the effects of other social categorizations is also necessary for discerning the many inequalities in schooling experience and educational outcomes faced by diverse Latinx students. The more multidimensional social constructs proposed consider ethnicity, gender, immigration status, and possibly additional factors. Some studies on Latinx students have criticized the tendency to essentialize the experiences of Latinx people, or to overemphasize the importance of ethnic and national origins in differentiating between Latinx students’ experiences (Covarrubias, 2011; Reardon & Galindo, 2009). However, Reardon and Galindo (2009) compared the fourth-grade reading and math achievement scores of Latinx students, disaggregated by national origin, with those of White counterparts from a nationally representative sample (ECLS-K). They found that students who had Mexican and Central American origins showed significantly lower achievement scores (1 SD below) than White students, but Latinx students of other origins (e.g., Cuban or Puerto Rican) showed a smaller gap (0.5 SD below their White counterparts).

Another category for analyzing differences in educational experience is immigrant status (e.g., immigrant generation, or length of time in the United States). As most recent immigrants are located in urban communities (Rong & Brown, 2002), understanding the experiences of Latinx immigrant students and providing them with adequate support is particularly important in urban education. Some studies have differentiated the educational outcomes of Latinx students by both immigrant status and national origin (e.g., Covarrubias, 2011; Covarrubias & Lara, 2014; Reardon & Galindo, 2009). Extending the exploration of disparities in Latinx students’ educational outcomes, Conchas (2001) demonstrated that institutional mechanisms (e.g., school opportunity structure, guidance, and support from peers and staff members) are critical factors affecting Latinx immigrant students’ schooling experiences. In particular, Conchas found that these students did not receive sufficient guidance and support from staff members and peers, and they did not have positive role models. Another challenge experienced by Latinx immigrant students is rooted in the tension between language use at home and school (McLaughlin, Liljestrom, Lim, & Meyers, 2002). For example, McLaughlin et al. (2002) revealed that a lack of English proficiency among Latinx immigrant students and their parents challenged both the students’ schooling experiences (e.g., completing homework assignments) and their parents’ involvement in school activities (e.g., communication with teachers and schools).

The literature also pays particular attention to the role of gender in differentiating Latinx students’ schooling experiences. In particular, Ovink (2014) demonstrated that gender is a critical factor affecting Latinx students’ college-going behavior. Ovink found that the gendered familism of many Latinas influenced them to pursue higher education as a means of achieving independence under patriarchal power structures, and this motivation tended to increase their rates of high school completion and college enrollment. Furthermore, scholars have asserted that gender is an “institutionalized system of social practices” (Ridgeway & Correll, 2004, p. 510) established through cultural beliefs. The hegemonic gender beliefs in the United States, which describe women as more communal and requiring more subordinate behaviors (Ridgeway & Correll, 2004), often limit Latinas’ capacity to pursue postsecondary education (e.g., Ortiz & Santos, 2009; Ramírez, 2013; Sy & Romero, 2008). Expanding the binary gender categorizations (male, female), queer theorists focusing on Latinxs (e.g., Muñoz, 2010) highlight the marginalization of those with gender-fluid identities, such as queer, transgender, or gender nonconforming.

Scholars who focus on only one social categorization in analyzing Latinx educational experiences, or who consider dual categorizations, such as national origin plus immigration status, often apply the concept of culture to explain their findings. These scholars may contend that the performance gaps among Latinx students result from their cultural diversity. In particular, Delpit (2006) argued that cultural clashes between the students’ cultures and the school cultures can lead educators to misread their students’ abilities and aptitudes. More specifically, scholars such as Delpit have argued that discordance between cultural codes can negatively affect student performance. These negative effects result from the teachers’ use of instructional styles that differ from the social norms of the students’ communities. For example, Latinx students with Mexican or Central American origins (and students who have migrated to the United States recently) are more likely to have limited English language skills than Latinx students of other national origins (e.g., Puerto Rican). Therefore, these students typically experience greater cultural discontinuity in English-language classrooms, which affects their educational performance. However, the concept of culture has only a limited capacity to explain the diversity in schooling experiences (e.g., differences in teacher quality) that are faced by Latinx students with multiple intersecting social categorizations (e.g., gender). Indeed, the factor of power, or power relations, is critical to consider if we are to understand how privilege and structural oppression affect Latinx students in different ways, according to their multiple social categorizations.

Latinx Students Living With Multiple Social Constructs

Scholars who use the lens of critical theory, or the poststructural approach, typically consider the educational inequalities facing racial or ethnic minority students in terms of power. In particular, Shlapentokh and Beasley (2015) contended that “people can gain power over each other and then abuse this power because of an initial condition of inequality of resources among them—wealth or physical strength, intellectual capacity and information, sexual attractiveness” (p. vii). This kind of disparity in power relations means that different groups of people receive differing levels of access to resources, information, and opportunity. Such disparity can also mean that relatively disempowered people face abuse or violence from those who have more power in their schools, families, and communities—whether these disempowered people are racial minorities (Latinx), females, or immigrants.

Critical scholars have paid particular attention to the nuanced positionality of Latinxs by considering multiple social categorizations (e.g., immigrant status, national origin) which often involve unique power relations and struggles. In particular, Gloria Anzaldúa’s (1987) book Borderlands/La Frontera conceptualized social borderlands as mixed (mestiza or in-between) spaces. Anzaldúa spoke to the ambiguity, fluidity, and nuanced characteristics of Latinx people’s lives. The concept of a borderland also suggests that people living along a border (e.g., Latinx) are marginalized from the “center” of society. Borderland people experience limitations in their access to resources and power. In seeking to overcome those limitations, they are forced to confront inequalities in power (Callis, 2014).

Similarly, the concept of a middle ground (White, 1991) between people describes spaces where multiple social categorizations are created by complex relationships of power. White defined this middle ground as

the place in between: in between cultures, peoples, and in between empires and the nonstate world of villagers. . . . On the middle ground, diverse peoples adjust their differences through what amounts to a process of creative, and often expedient, misunderstandings. (p. x)

Various scholars have applied the concept of the middle ground to depict the experiences of marginalized people (e.g., indigenous, immigrants) in the United States. This concept helps to express the experience of people who live in the margins, in between different social worlds. The same concept can be used to situate the ideologies and practices of White, male, English speakers as central figures in the dominant culture (DeLeon, 2010). Like the concept of the borderland, the concept of the middle ground suggests that people living in the spaces between cultures and identities (e.g., Latinx) commonly experience dynamic inequality and social division in their relationships with others.

Intersectionality and Latinx students

In considering the complex and nuanced positionalities of Latinx people, scholars of intersectionality have suggested a theory regarding the ways that multiple oppressions and power relations operate in creating Latinx students’ experiences (e.g., Collins, 1990; Covarrubias, 2011). The foundational concepts regarding intersectionality have their origins in Black feminist scholarship and activism, which challenges the converging systems of power that Black women experience. In line with Black feminists, Chicana feminist scholars have emphasized that the lives and experiences of Chicana women are distinct from those of White women, due to the converging intersections of gender, race, and national origin (e.g., Gonzales, 1980; Sandoval, 1991; Sólorzano & Bernal, 2001; Sólorzano, Villalpando, & Oseguera, 2005). Kimberlé Crenshaw (1991) proposed the umbrella term intersectionality as a basis for building coalitions among those who study issues of race, ethnicity, SES, and gender. Crenshaw suggested an interdisciplinary endeavor that includes diverse social identities and contexts beyond those of Black women (Dhamoon, 2010). As a critical legal scholar, Crenshaw showed a pattern in how court cases (which mainly operate on a single-axis framework) have failed to address the issues faced by women of color, who experience multiple interacting forms of discrimination and oppression based on their race, ethnicity, and gender.

Scholars of intersectionality have identified the dimensions of simultaneity, multiplicity, and power relations as essential aspects of a multisided challenge to marginalized groups. These three main premises regarding intersectionality are particularly important for understanding the multidimensionality of Latinx students’ schooling experiences and educational aspirations. First, the concept of simultaneity indicates the dynamic processes by which race, ethnicity, gender, immigration background, sexual orientation, and SES all operate at the same time to influence people’s lives. Studies in education that focus on only a singular social construct have a limited capacity to explain the experiences of Latinx who have multiple social locations and identities. Second, a concern for multiplicity reveals the interconnectedness among the factors of race, ethnicity, gender, and SES. Such a perspective embraces a “both/and” framework for analyzing the interconnections among multiple identities. Third, an intersectional approach enables us to understand the power relations involved in racism, sexism, capitalism, and heterosexism through “a lens of mutual construction,” instead of treating each dimension of life as “a static entity” (Collins & Bilge, 2016, p. 26).

Critical quantitative intersectionality

Studies focusing on intersectionality often use a qualitative approach, due to its potential for exploring the nuances of lived experiences among marginalized people, and for articulating their unique historical, cultural, and organizational contexts. Although most scholars admit the relative predominance of qualitative methodologies in studies of intersectionality, critical quantitative researchers insist that quantitative methods can have a powerful impact for informing educational policy, realizing social justice, and answering important questions regarding cultural or political issues (e.g., Covarrubias et al., 2018; Gillborn, Warmington, & Demack, 2018; López, Erwin, Binder, & Chavez, 2018; Stage, 2007; Teranishi, 2007). This study aims to explore the diversity of Latinx students’ experience with reference to a critical framework of positionality, and to do so with the use of quantitative methods (e.g., CRQI, QuantCrit).

This study applies the four principles of QuantCrit as proposed by Gillborn and his colleagues (2018, p. 169). These principles are as follows: (a) numbers are not neutral, (b) categories are neither “natural” nor given, (c) voice and insight are essential: data cannot “speak for itself,” and (d) numbers should be used for social justice. In addition to these main principles, this study integrates the critical quantitative approach and intersectionality theory to understand the complexity of Latinx students’ educational experiences, and to provide policy implications that are appropriate for this particular student group.

Several limitations need to be considered regarding the use of data from existing large-scale surveys for analyzing the intersectionality (especially the critical quantitative intersectionality) of diverse, multiple social categorizations. As surveys are typically not designed to examine the intersectionality of diverse social constructs, it should be acknowledged that using survey-derived data can raise critical issues. Applying a limited number of cases to quantitatively examine multiple social constructs can produce results that are based on insufficient statistical power (Hancock, 2007; McCall, 2005). In addition, the use of mechanistic categories can belie the fluid and complex nature of social constructs, which studies of intersectionality are meant to emphasize. Furthermore, as the researchers who conduct quantitative, survey-based studies do not actually speak with the participants about their personal experiences, the findings regarding identity, race, ethnicity, gender, or SES cannot be inter-related to determine how multiple intersecting social constructs are associated with particular life experiences, or with effects such as educational outcomes. The desire to overcome these limitations is a critical reason why researchers of intersectionality commonly prefer to use qualitative methodologies (e.g., counter-storytelling, narrative inquiry, ethnographic inquiry). These researchers find that qualitative methods “more naturally lend themselves to the study of complexity and [therefore they] reject methodologies that are considered too simplistic or reductionist” (McCall, 2005, p. 1772).

However, critical quantitative researchers often criticize scholars of intersectionality who use qualitative approaches in an overly subjective way, framing questions and results so they can find the data they expect to find (Stage & Manning, 2016). By providing objective evidence and using large-scale data sets, a critical approach to quantitative intersectionality studies can contribute to our understanding of intersectionality in the experiences of Latinx students, and can do so in a more rigorous way. This kind of hybrid approach can apply quantitative data regarding multiple social constructs in coordination with the findings of qualitative intersectionality studies.

Although the relative advantages of quantitative approaches are well accepted (in terms of generalizable knowledge), few quantitative studies of intersectionality using large data sets have been conducted. For example, studies exploring the intersectionality of gender and SES are often limited in their analyses, as they use only simple comparisons between single factors, rather than investigating various combinations of different social constructs. Even when scholars include multiple social constructs in their analyses, they often control for only one social construct, and neglect the nuanced, multiplicative nature of multiple constructs. Most previous studies have examined only certain aspects of the intersectionality in Latinx students’ educational outcomes (i.e., standardized test achievement scores), and they have generally failed to account for diversity in the students’ schooling experiences (e.g., exposure to high-quality teachers, the differing structural contexts of schools, or suspensions of students who do not fit in). Thus, in seeking to apply an inclusive, critical, quantitative intersectionality framework, this study proposes two main research questions concerning Latinx students:

Research Question 1: How is the intersectionality of multiple social constructs (i.e., ethnicity, gender, immigrant status, SES, and home language) associated with the educational outcomes of Latinx students?

Research Question 2: How is the intersectionality of multiple social constructs (i.e., ethnicity, gender, immigrant status, SES, and home language) associated with the educational schooling experiences of Latinx students?

The next section specifically describes the data and the variables of interest and provides explanations of how multiple social construct categories were quantified through the process of coding.

Data and Method

Data Source

This study applied restricted-use national longitudinal data provided by the NCES High School Longitudinal Studies 2009 (HSLS:09). The intent in using these data was to examine the multiple intersections of differing aspects of Latinx students’ lives, such as ethnicity, gender, home language, SES, and immigrant status. The HSLS:09 data set includes a nationally representative sample gathered from more than 23,000 ninth-grade students in 944 schools since 2009. An average of 25 ninth-grade students per school were randomly selected. In addition, a stratified, two-stage random sample design was used to acquire the sample schools, and to include both public and private schools. The type of stratified two-stage random sample design used in the HSLS:09 is often called a complex sampling design (Lumley, 2011).

To address the complex sampling issues (i.e., nonindependence among units or disproportionate sampling resulting in unequal selection probabilities), this study applied balanced repeated replication methods to ensure that the results were representative of the population, and to conduct the calculations of variance and parameter estimates (Hahs-Vaughn, McWayne, Bulotsky-Shearer, Wen, & Faria, 2011). Among the five sub-data sets of the HSLS:09 (base year, first follow-up, 2013 update, high school transcripts, and second follow-up), this study used the base year data. These data were collected in the fall term in 2009 and focused on ninth graders’ educational outcomes, as these outcomes were considered powerful predictors for students’ future educational attainments (Easton, Johnson, & Sartain, 2017; McCallumore & Sparapani, 2010).

Social categorizations

As this study focused on identifying the intersectional inequities experienced by Latinx students, it included only students who identified themselves as either Latinx or White (with White participants used as a reference). To differentiate the national origins among Latinx students, five dummy variables (i.e., Mexican, Central American, South American, Caribbean [Cuban, Dominican, Puerto Rican], and Other) were used to compare the students of each ethnic origin with White students. These students were coded 1 as members of their respective categories and 0 for other categories. White students served as the reference, because educational research typically establishes a White group as the reference for exploring patterns of educational inequity among Latinx students. As this study excluded students who identified as having more than one race or ethnicity (6.2% of the sample) and it excluded other students of color (i.e., Black, American Indian/Alaska Native, or Asian students), its coding system for race or ethnicity included only 73.5% of the student population sampled. The information on Latinx students’ national origins was based on the HSLS:09 restricted-use data.

In terms of gender, the female variable was coded 1 for female students and 0 for male students (i.e., the reference group). The HSLS:09 data set includes three categories of immigration status: (1) born in the United States, (2) born in Puerto Rico or another U.S. territory, and (3) born in another country. This study created a dummy variable for immigrant status to further differentiate all students who were born in the United States (Categories 1 and 2) as a reference group for comparison with the non-U.S. born students (Category 3). The students’ home language was also a categorical variable for coding students who used other languages (e.g., Spanish) only at home. These students were coded 1, and those who used only English at home or who used English equally with other languages at home were coded as 0. Finally, this study used the SES index score as a covariate, which was calculated by the NCES in the HSLS:09 data set.

Dependent variables

This study used four dependent variables, namely the Latinx students’ math achievement scores, and three aspects of their schooling experience: suspensions, teachers’ years of teaching experience, and the percentage of students who were eligible for a free or reduced-price lunch (FRL) in their school.

Math achievement score

This study used ninth graders’ math standardized theta scores for the math achievement scores. The mathematics assessments measured ninth-grade students’ levels of achievement in algebra. These achievement scores are based on item response theory. Each score (X1TXMTSCOR) is a continuous variable and a ratio measurement, based on a 100-point test. These achievement scores were used as the only curriculum-related variable in this study because previous studies have shown a strong association between high school math achievement and students’ future academic success, such as their performance in college (e.g., Claesens & Engel, 2013; Lee, 2012).

Experience of school suspension

In exploring the patterns of Latinx students’ schooling experiences, school suspensions were considered an important indicator regarding the barriers and challenges that commonly push Latinx students out of school (e.g., Valencia & Black, 2002). The indicator variable for this factor equaled 1 if a ninth grader had experienced being suspended or expelled from school. Ninth graders who had not experienced a school suspension were coded 0 and served as the reference group.

Exposure to high-quality teachers

Including the factor of exposure to high-quality teachers was important in this analyses, because students’ academic performance and psychological characteristics can be affected by various factors related to teacher quality (Clotfelter, Ladd, & Vigdor, 2010; Goldhaber, 2002; Hanushek, 2007; Harris & Sass, 2011). For example, previous research has determined that teacher quality is often distributed unequally: Students of color typically have fewer high-quality teachers (with “high-quality” defined in the literature; for example, Goldhaber, Lavery, & Theobald, 2015). Recent literature that explores teacher quality has increasingly turned to value-added models (VAMs), which evaluate teacher quality by the increases in student achievement scores tied to that teacher. However, some scholars have argued that VAMs are not reliable for measuring individual teacher quality (Koretz, 2008; Darling-Hammond, 2015). Many scholars still use traditional measures (i.e., teaching experience) to investigate the effects of teacher quality (e.g., Goldhaber et al., 2015; Harris & Sass, 2011). I measured teacher quality based on information regarding teaching experience in math, which is also a traditional measure of teacher quality. In this study, a math teacher’s teaching experience is a continuous variable that indicates how many years the teacher has taught high school math.

Demographic composition of schools that Latinxs attend

Each schools’ demographic composition was measured by the percentage of students in the school who were eligible for FRL. Accordingly, this study used the data on this variable as reported in the school-level HSLS:09 data set. Although previous studies have included the percentage of FRL students as a school structural characteristic affecting Latinx students’ educational outcomes (e.g., Núñez & Kim, 2012; Alexander & Jang, 2018), this study explored how the intersections of multiple social categorizations of Latinx students were associated with the structural aspects of their schooling contexts.

Research models

This study used two statistical research techniques to answer the proposed research questions: multiple regression and logistic regression for the binary-dependent variable (i.e., experience of school suspension). I used multiple intersection effects in the statistical models to identify the unique contributions of converging multiple social categorizations, rather than using an additive approach that statistically considers only the main effects of various singular social categorizations (e.g., double jeopardy theory). The use of interaction terms showing multiplicative relationships among multiple social constructs was also meaningful for demonstrating the complexities and nuances of the lived experiences of Latinx students. In particular, this study applied interaction models focusing on either two-way interactions or three-way interactions, according to the numbers of independent variables included in the interaction terms. For example, if the coefficient for the interaction term of Mexican Origin × Female was statistically significant, then gender difference in the dependent variable was conditioned on whether a student was Latinx with a Mexican origin.

As Núñez and Kim (2012) rightly pointed out, a parsimonious statistical model is particularly critical for explaining the diverse schooling experiences and educational outcomes of Latinx students with consideration for multiple intersecting social constructs. Thus, this study eliminated nonsignificant intersectionality variables by a step-up approach (based on preliminary analyses), including only variables that were significant for Latinx students’ schooling experiences and educational outcomes.

This study used two statistical software packages for the different research models: SPSS and AM (developed by the American Institutes for Research). SPSS was used to analyze the descriptive statistics and to obtain the measures for reliability (Cronbach’s α). AM was used to conduct multiple regressions and logistic regression analyses, which allowed for the calculation of the appropriate standard errors based on the differences between estimates from the full sample and a series of created subsamples (see Hahs-Vaughn et al., 2011). This study also used base-year student-level weights to account for differential selection probabilities and differential patterns of response or nonresponse.

In addition, this study used three criteria for goodness-of-fit tests to compare the performance of the contrasting models (i.e., base models without interaction terms vs. full models with interaction terms). These tests included (a) the deviance test, (b) the Akaike information criterion (AIC), and (c) the Bayesian information criterion (BIC). In particular, a deviance statistic (−2Loglikelihood) for each base model and for the full model including interactions was computed for the deviance tests. The larger the deviance, the poorer the fit to the data: A model with lower values of AIC and BIC indicated a better fit.

Findings

This study attempted to identify distinct patterns of educational inequity in Latinx students’ math achievements and schooling experiences, as they relate to students’ multiple backgrounds, identities, and experiences. In particular, the study included interaction variables in the regression and logistic regression models to explore the convergence of students’ multiple social categories (ethnicity, gender, home language, immigrant status, and SES). The use of multiple statistical interaction terms enabled the examination of the core premises of intersectionality (i.e., simultaneity and multiplicity).

Association Between Intersectionality and Latinx Students’ Achievement

Table 1 presents the results of multiple regression analyses of the ninth-grade Latinx students’ mathematics achievement scores. Comparing Model 1 (without considering intersectionality) and Model 2 (with intersectionality) leads to the conclusion that Model 2 provides a better fit for explaining the relationship between math achievement scores and the Latinx students’ multiple social constructs. Model 2 indicates that the Latinx students’ SES, ethnicity, immigrant status, and intersecting social constructs (i.e., ethnicity, SES, and home language) were all significantly associated with math achievement. In particular, the higher the SES index scores, the higher the math scores that the students achieved, on average (β = 4.67, p < .001). In addition, “Other” Latinx students showed significantly lower math scores than their White counterparts, on average (β = −3.53, p < .001). Finally, students who were not born in U.S. territories showed higher math achievement scores, regardless of the students’ SES or ethnicity (β = 2.58, p < .001).

Table 1.

Results of Fitting Models Predicting Mathematics Achievement.

	Model 1	Model 2
SES	4.57*** (0.18)	4.66*** (0.16)
Ethnicity
Caribbean	−2.35*** (0.56)	−1.02 (0.62)
Others	−3.70*** (0.89)	−3.53*** (0.85)
Immigrant status (foreign-born)	2.34** (0.76)	2.58** (0.66)
Intersectionality
Caribbean × Language (non-English)		−2.82* (1.14)
South × SES		−2.98* (1.39)
(Intercept)	51.77*** (0.25)	51.81*** (0.17)

Note. For all models, the reference group for ethnicity is White student and the reference group for female is male. Standard errors are in the parentheses. The results of Model 1 and 2 report only significant variables. SES = socioeconomic status.

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

Although no particular three-way intersectionality was significantly associated with Latinx students’ math achievement, two interaction variables were significant in this regard: Caribbean × Home Language (β = −2.82, p < .05) and South American × SES (β = −2.98, p < .05). In particular, the intersectionalities associated with ethnicity, home language, and SES created heterogeneities in math achievement among Latinx students. The statistically significant interaction between Caribbean and home language indicated that the mean difference caused by a student’s home language was conditioned on the student’s ethnicity for Latinxs from Caribbean countries (Cubans, Puerto Ricans, or Dominicans). Thus, Latinxs who had Caribbean ethnic backgrounds (CE) and spoke non-English languages as their home languages (non-EH) showed significantly lower math achievement scores than the three other groups (non-CE and non-EH, non-CE and EH, and CE and EH). Similarly, the significant interaction between South American ethnicity and SES indicated that ethnicity moderated the impact of SES for this group: SES mattered less in terms of math achievement for Latinxs with South American ethnicity (SAE) than for other student groups (SES slope for SAE = 1.68 vs. SES slope for other student groups = 4.66).

Association Between Intersectionality and Latinx Students’ Schooling Experiences

This study examined the association between the intersectionality of Latinxs’ multiple social constructs and diverse schooling experiences, including their experiences of school suspension, their teacher’s levels of teaching experience, and the percentages of FRL students in each school.

Experience of school suspension

Table 2 reports the results of the logistic regression with a binary-dependent variable of the students’ experience of school suspensions, as predicted by the Latinx students’ multiple social categorizations. In comparing the overall model fits between Models 3 (without considering intersectionality) and 4 (with intersectionality), the smaller values of deviance in the AIC indicated that Model 4 provided a better fit to predict Latinx students’ experiences of school suspension than Model 3. In particular, the coefficient for SES in Model 4 (β = ‒0.79, p < .001) represents the change in the logit for every additional one-unit change of the SES index score. The exponent of the coefficient for SES is 0.45, which indicates that the odds of having a school suspension before the ninth grade with a one-unit higher SES index score was 0.45 times the odds for a student with a one-unit lower SES index score. Similarly, the logistic regression coefficient for Female (β = ‒1.24, p < .001) indicates how much the logit is expected to change when the value for Female changes by 1 unit (i.e., when a student is female compared with male). The exponentiated regression coefficient of Female, exp(–1.24) = 0.29, which indicates that the odds that a female Latinx student had experienced school suspension were 0.29 times the odds for a male. In other words, females were less likely to have school suspensions before the ninth grade than their male counterparts.

Table 2.

Logistic Regression Analysis of Students’ Experience of School Suspension Predicted by Their Multiple Social Categorizations.

Predictor	Model 3		Model 4
Predictor	β	e^β (odds ratio)	β	e^β (odds ratio)
SES	−0.87*** (0.08)	0.42	−0.79*** (0.09)	0.45
Female	−1.03*** (0.12)	0.36	−1.24*** (0.13)	0.29
Home language (non-English)	−0.86*** (0.23)	0.42	−1.00*** (0.27)	0.37
Immigrant status (foreign-born)	−0.60* (0.28)	0.55	−1.02* (0.51)	0.36
Ethnicity
Mexican	0.36* (0.16)	1.43	0.33* (0.17)	1.39
Caribbean	0.87** (0.28)	2.39	0.87** (0.29)	2.39
Others	0.82** (0.31)	2.27	0.79** (0.32)	2.20
Intersectionality
Female × SES			−0.41** (0.15)	0.66
(Intercept)	−1.80*** (0.07)	NA	−1.76*** (0.07)	NA

Note. For all models, the reference group for ethnicity is White student and the reference group for female is male. Standard errors are in the parentheses. The results of Model 3 and 4 report only significant variables. SES = socioeconomic status.

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

Finally, the coefficients for home language and immigrant status in Model 4 (β = ‒1.00, p < .001 and β = ‒1.02, p < .05, respectively) represent the change in the logit for using non-English as a home language (i.e., Spanish) and for being non-U.S. born, respectively. The exponents of the coefficients for these variables are 0.37 and 0.36, which indicates that students who spoke non-English as their home language and who were not born in the United States were overall less likely to have school suspensions than counterpart students in each other category. Three ethnicities were associated with higher experiences of school suspension: Mexican, Caribbean, and Others. These groups of Latinx students showed 1.39, 2.39, and 2.20 times higher likelihoods of school suspension experiences than their White counterparts.

The results in Table 2 show that one interaction term (intersectionality) between Female and SES was significantly associated with a Latinx student’s experience of school suspension (β = ‒0.41, p < .01). This significant interaction term, which was related to gender, created distinct patterns in the experience of school suspension across different SES index scores. In other words, the effect of SES on Latinx students’ experiences of school suspension was conditioned on their gender. As an illustration, Figure 1 gives calculations regarding the expected probabilities of experiencing school suspensions, which illustrate the differences between Chicanas and Chicanos who were U.S.-born and English speaking at home, on the basis of five SES index scores (+2 SD, +1 SD, average, ‒1 SD, −2 SD). As Figure 1 illustrates, Chicanos showed significantly higher disparities across different SES index scores in suspension rates compared with their Chicana counterparts. Chicanos who had an SES score that was 2 SD and 1 SD below average showed remarkably higher suspension rates compared with their Chicana counterparts, but this difference shrank significantly for their peers with higher SES scores (2 SD above average or more). As there was no intersectionality related to ethnicity, this pattern applied for all ethnicities. This finding may indicate that, in terms of experiencing school suspension, living in poverty raised particular barriers and challenges for Latinos compared with their Latina counterparts.

Figure 1.

Predicted probability of experiencing school suspension of Chicanas and Chicano students across SES index scores.

Math teachers’ teaching experience

As previously noted, this study used two statistical models to predict an association between students’ multiple social constructs and math teachers’ years of teaching experience: without intersectionalities (Model 5) and with intersectionalities (Model 6) (see Table 3). After comparing the model fits of both models, Model 6 was selected for this study to explain the association between students’ multiple constructs and math teachers’ years of teaching experience. Model 6 shows that a student who had a 1-unit higher SES index score was, on average, taught by a math teacher who had one more year of teaching experience, controlling for other social categorizations (β = 1.00, p < .001). Furthermore, Latinxs with Caribbean backgrounds were, on average, taught by math teachers who had about 29 months less teaching experience than the teachers for their White counterparts (β = −2.40, p < .01).

Table 3.

Results of Fitting Models Predicting Math Teachers’ Years of Teaching Experience.

	Model 5	Model 6
SES	1.00*** (0.23)	1.00*** (0.21)
Ethnicity
Caribbean	−2.48** (0.85)	−2.40** (0.69)
Intersectionality
Mexican × Female × SES		1.16* (0.56)
(Intercept)	9.99*** (0.32)	10.02*** (0.26)

Note. For all models, the reference group for ethnicity is White student and the reference group for female is male. Standard errors are in the parentheses. The results of Model 5 and 6 report only significant variables. SES = socioeconomic status.

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

Finally, the statistically significant interaction between Mexican, Female, and SES (β = 1.16, p < .05) showed that gender and ethnicity moderated the impact of SES on a math teacher’s teaching experience for Latinas with Mexican origin, compared with that for other groups. This result indicated that SES mattered more in enabling Latinas with Mexican origin to benefit from teachers with more teaching experience. In particular, a Latina with Mexican origin with higher SES experienced a greater benefit from having a more experienced teacher than did other student groups with higher SES.

School’s percentage of FRL students

In school finance studies, it has been widely accepted that a student’s eligibility for subsidies in the FRL program represents his or her level of access to material resources. For example, Alexander and Jang (2018) used the factor of FRL eligibility interchangeably with poverty and economic disadvantage. Thus, a higher percentage of FRL students in a school indicates a concentration of poverty in that school, and a challenging environment for schools with scarce resources. A school with a higher percentage of FRL students also presents a challenge for students attending that school, because that condition may limit “the access to material resources that could support quality educational programming” (Alexander and Jang, 2018, p. 26).

Table 4 shows the relationship between the percentage of FRL students in a school and Latinx students’ multiple social categorizations. Comparing the overall model fits between Model 7 without intersectionality and Model 8 with intersectionality leads to the conclusion that Model 8 (with its intersectionality among Caribbean, Female, and SES) provides a better fit. In particular, Latinx students with higher SES typically attended a school with a significantly lower percentage of FRL students than those Latinx with lower SES (β = ‒10.92, p < .001), regardless of their ethnicity, home language, or immigrant status. In addition, Latinas were found to attend schools with slightly lower percentages of FRL students than their male counterparts (β = ‒1.72, p < .01). All ethnic backgrounds, except South American, had higher percentages of FRL students in the schools they attended than their White counterparts by more than 10%. The categorizations by home language and immigrant status were not significantly associated with the percentages of FRL students in the schools they attended.

Table 4.

Results of Fitting Models Predicting a Percentage of Free or Reduced-Price Lunch Eligible Students in a School Where Latinxs Attend.

	Model 7	Model 8
SES	−10.47*** (0.59)	−10.92*** (0.60)
Female	−2.04* (0.78)	−1.72** (0.54)
Ethnicity
Mexican	8.73* (3.47)	12.32** (3.69)
Central American	9.25** (3.21)	10.01*** (2.62)
Caribbean	13.11** (4.54)	13.57** (3.98)
Others	16.29*** (4.55)	14.07** (5.02)
Intersectionality
Caribbean × Female × SES		−7.91** (2.63)
(Intercept)	32.58*** (0.83)	33.32*** (0.78)

Note. For all models, the reference group for ethnicity is White student and the reference group for female is male. Standard errors are in the parentheses. The results of Model 7 and 8 report only significant variables. SES = socioeconomic status.

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

Finally, a significant three-way interaction term indicated that the effect of SES on the percentage of FRL students in a school where Latinxs with Caribbean backgrounds (Cuban, Dominican, or Puerto Rican) attended was mediated by both ethnicity and gender. This significant intersectionality between gender and SES for Latinxs, particularly for those with Caribbean background (β = ‒7.91, p < .01), indicated gender-specific patterns related to the percentages of FRL students in the schools these students attended (see Figure 2). Latinas with Caribbean backgrounds and lower SES (1 SD and 2 SD below average) attended schools with higher percentages of FRL students than their Latino counterparts, which could indicate greater limitations in their access to quality educational programming than their male counterparts. In contrast, Latinas with CE and higher SES (1 SD and 2 SD above average) attended schools with significantly lower FRL rates than their male counterparts. This result indicated that Latinos with Caribbean backgrounds were not experiencing the same benefit from higher SES in attending schools with lower FRL rates as Latinas with CE and higher SES.

Figure 2.

Predicted percentages of free or reduced-price lunch eligible students in a school where Latinas and Latinos with Caribbean backgrounds across SES index scores.

Discussion

This study used a critical quantitative intersectionality framework (e.g., Covarrubias & Lara, 2014; Gillborn et al., 2018; Jang, 2018; López et al., 2018) to understand diverse schooling experiences and educational outcomes in spaces where Latinxs’ multiple social constructs intersected with each other and created complex power relations. The multiple social constructs used in this study were based on previous literature focusing on Latinx student populations. This section summarizes the major findings of the study and discusses implications of intersectionality, particularly for Latinx students.

In line with findings from the literature (e.g., Covarrubias & Lara, 2014; Reardon & Galindo, 2009), this study demonstrated that specific ethnic backgrounds for Latinxs were significantly associated with both their schooling experiences and educational outcomes. While quantitative studies typically assume that Latinx students’ educational outcomes would be the same using one quantified social construct, this study is particularly important to understanding how different ethnic backgrounds create heterogeneities among Latinxs. In particular, this study found that Latinxs with Mexican, Caribbean, and “Other” origins had a significantly higher probability of experiencing school suspension than their White counterparts. Furthermore, Latinx students with Caribbean origins (e.g., Cuban, Dominican, Puerto Rican) had significantly fewer years of teaching experience than their White counterparts, after controlling for other social constructs. In addition, all ethnic origins of Latinx, except South American, were associated with a higher percentage of FRL students in the schools they attended. Furthermore, only ninth-grade Latinxs with ethnic backgrounds of Others (not Mexican, South or Central American, or Caribbean origins) showed significantly lower math achievement scores than their White counterparts. Although Reardon and Galindo (2009) found disparities in the math achievement scores of kindergarteners with Mexican, Central American, Caribbean, and South American origins compared with their White counterparts, this study found no statistically significant difference in the math achievement scores of students with these ethnic origins compared with White students, perhaps because the disparities between White and Latinx students with those specific ethnic backgrounds significantly narrowed from kindergarten to ninth grade. Reardon and Galindo also found that the disparities narrowed from kindergarten through fifth grade.

Second, this study found that the intersectionalities between Latinxs’ multiple social constructs created heterogeneities of schooling experiences and educational outcomes within and across different student groups. Latinxs’ SES was a critical factor in creating the intersectional effects associated with other social constructs for schooling experiences and educational outcomes. Although students’ SES background was positively associated with all of the schooling experiences and educational outcomes examined in this study, the effects of SES were stronger or weaker for particular Latinx groups. In particular, three intersectionalities associated with Latinxs’ SES indicated that gender mediated the effects of SES on experiencing school suspension, having math teachers with more or less teaching experience, and the percentage of FRL students in a school. For example, Latinas with Mexican origin from higher SES did not experience the same benefits from SES in terms of access to teachers with more years of teaching experience compared with their male counterparts. Similarly, Latinas with Caribbean backgrounds did not experience the same benefit from SES in attending schools with lower FRL rates as Latinos with CE and higher SES. In addition, living in poverty resulted in more school suspensions for Latinos compared with their Latina counterparts.

These findings should not be interpreted as saying that Latinas do not experience the interplay of capitalism and sexism (Covarrubias & Lara, 2014). Rather, Latinos living in spaces at the intersections of ethnicity, gender, and SES have different experiences than Latinas, as “structures of domination [associated with Latinos’ and Latinas’ multiple social constructs] create distinct relations of power that can result in various forms of oppression and privilege for different individuals, depending on the context” (Covarrubias & Revilla, 2003, p. 461). That is, while O’Connor (2009) found that in higher education SES was significantly interrelated with Latinxs’ ethnic origins, this study further demonstrated that those intersectionalities between ethnic origins and SES were conditioned on gender categorizations. Identifying the reasons why the effects of SES depend on gender and ethnicity is beyond the scope of this study. Further studies, particularly those that adopt qualitative inquiries, must explore the particular mechanisms behind how gender categorizations mediate the effects of SES for Latinx student populations.

Third, whether they were foreign- or U.S.-born, Latinxs’ immigrant status was not significantly associated with the schooling experiences and educational outcomes examined in this study. However, Latinxs’ immigrant status may be more significant in different educational contexts. That is, while Covarrubias and Lara (2014) focused on the intersectional effect of immigrant status on high school dropout and enrollment rates in higher education, this status may not be salient before or during ninth grade. On the contrary, the intersectionality associated with home language use was significantly and negatively associated with Latinxs with Caribbean origins. Cuban, Dominican, and Puerto Rican students who speak mainly Spanish at home may be disadvantaged in accessing appropriate instructional resources. For example, Alexander and Jang (2017) found that districts with higher portions of English language learners (ELLs) had significantly lower expenditures per pupil for instructional practices. Latinxs with Caribbean origins may be located in districts with fiscal and structural inequities in support for ELLs. This potential explanation requires further exploration in future research.

Conclusion

Based on the critical quantitative intersectionality framework, this study examined the associations between the intersectionalities of Latinxs’ multiple social constructs and their schooling experiences and educational outcomes. Regardless of the preponderance of qualitative inquiries in intersectionality studies, the use of critical quantitative intersectionality analysis (e.g., Covarrubias & Lara, 2014) in this study provided a generalizable knowledge of the educational practices and schooling experiences of Latinx students. Similar to intersecting power relationships associated with multiple social categorizations of the concepts of borderlands and middle ground, Covarrubias and Revilla (2003) articulated that the power relations in spaces Latinxs occupy “at times position them as less powerful compared to others, and at other times privilege them in relation to others who are less empowered” (p. 463). Based on these conceptual and theoretical foundations of intersectionality, this study provides a better understanding of the Latinx student population than studies that have focused only on achievement gaps. In addition, by considering multiple social constructs associated with Latinx students’ educational outcomes from the literature (i.e., home language, immigrant status, SES, gender), this study comprehensively adds to the intersectionality discourse on how multiple social constructs affect Latinx students’ schooling experiences and outcomes. It contributes to our knowledge about intersectionality by focusing on the Latinx population, for which previous research has mainly adopted race-based or gender-based theoretical frameworks (Alemán, 2018). As this study includes multiple social constructs, it may go against the parsimony of statistical modeling and cause multicollinearity concerns. Thus, as noted, this study paid particular attention to these issues by including only significant social constructs and associated intersectionality variables based on extensive preliminary analyses, similar to Núñez and Kim’s (2012) approach. By focusing on access to quality teachers, the percentage of FRL students in the schools Latinxs attend, and experiences of school suspension, this study complements findings from previous critical quantitative intersectionality analyses that typically focused only on educational outcomes. Finally, as previous intersectionality literature exploring Latinx students has often concentrated on higher education (Alemán, 2018), this study provides intersectional knowledge for kindergarten to Grade 12 education, particularly at the high school level.

As this study and other intersectionality studies (e.g., Jang, 2018) have consistently demonstrated critical differences in schooling experiences and access to educational resources within Latinx populations, collecting detailed quantitative data for ethnic subgroups is important to challenging hidden structural inequalities and power structure. Still, the State of Connecticut legislature’s recent act of banning data collection for ethnic subgroups reflects the totalizing assumption that the lives of all ethnic groups within a certain race group are the same. Such an assumption of policy makers is problematic for understanding the intersecting inequities experienced by minoritized student groups, revealing the chronic privilege and oppression attached to particular social constructs and challenging the injustice experienced by marginalized students, including Latinx students. A continuous effort to collect ethnic backgrounds in large-scale quantitative data sets (e.g., Census, NCES) is critical to understanding the complexities and nuances of Latinx students’ diverse schooling experiences and educational attainments. Furthermore, to explore structural inequalities related to gender categories in quantitative intersectionality analyses, future quantitative data collection procedures for large-scale data sets should expand the current binary gender categorizations (male, female) to include gender nonconforming and transgender students.

The important limitation of using quantitative data in intersectionality analyses should also be addressed. Using existing surveys, this study was unable to illustrate how diverse power relations attached to Latinxs’ multiple social categorizations intersected with each other, thereby oppressing and confining their educational opportunities and attainments. For example, the statistically significant interaction terms found in this study were mainly related to SES, but not intersections between qualitative categorizations (e.g., Gender × Immigration Status), except for the interaction term between Caribbean background and home language. The reason for the statistically insignificant intersections between qualitative social constructs might have been the lack of additional effects above and beyond the sum of the main effects of singular social categorization. It might have also been the limitation of the critical quantitative analyses using complex statistical modeling (e.g., small sample size of particular student groups with multiple social constructs, limited available variables). For this reason, critical quantitative intersectionality researchers have often confronted a challenge in applying more complex statistical models (e.g., Núñez’s, 2014, multilevel model of intersectionality) to obtain meaningful rather than superficial findings. In this regard, future studies using qualitative inquiries may supplement the quantitative findings of this study by exploring how the qualitative social categorizations intersect with each other in creating educational inequities and power structures among Latinx students.

This large-scale critical quantitative intersectionality study should significantly help policy makers to design educational policies. Policy makers often exhibit a strong preference and desire to rely on numbers from research. In particular, this study should challenge policy makers’ common assumption of “synonymizations” (Alexander & Jang, 2018). The synonymization of social constructs involves the melding of ethnicity and other social constructs, such as assuming Latinxs are immigrants, poor, and non-English speakers. Alexander and Jang (2018) emphasized the danger of policy makers comprising a particular racial group of students with different identities in addressing the structural inequalities of communities with multiple marginalized social constructs. For policy makers, the critical implication of this study is its call to challenge dangerous assumptions about Latinx students and to recognize the intersectional influences of multiple social constructs on diverse educational access and schooling experiences. This increased awareness can serve to influence policy in directions that result in more educational equity and social justice.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Sung Tae Jang

Author Biography

Sung Tae Jang is an assistant professor in the Department of Education Policy and Leadership at the Education University of Hong Kong. He was nationally selected as one of the graduate fellows and policy associates sponsored by the University Council for Educational Administration. As a critical quantitative researcher and social justice scholar, his research and teaching interests include educational equity, intersectionality, school culture, and social justice leadership.

References

Alemán

S. M.

(2018). Mapping intersectionality and Latina/o and Chicana/o students along educational frameworks of power. Review of Research in Education, 42, 177-202.

Alexander

N. A.

Jang

S. T.

(2017). Balancing act for English learners: Efficiency and equity in Minnesota over the past decade. Education Policy Analysis Archives, 25, 1-31.

Alexander

N. A.

Jang

S. T.

(2018). ‘Synonymization’ threat and the implications for the funding of school districts with relatively high populations of black students. Race Ethnicity and Education, 22, 151-173.

Anzaldúa

(1987). Borderland/la frontera. San Francisco, CA: Aunt Lute.

Callis

A. S.

(2014). Bisexual, pansexual, queer: Non-binary identities and the sexual borderlands. Sexualities, 17(1-2), 63-80.

Cameron

S. V.

Heckman

J. J.

(2001). The dynamics of educational attainment for Black, Hispanic, and White males. Journal of Political Economy, 109, 455-499.

Carey

R. L.

(2014). A cultural analysis of the achievement gap Discourse: Challenging the language and labels used in the work of school reform. Urban Education, 49, 440-468.

Castañeda

Anguiano

Alemán

S. M.

(2017). Voicing for space in academia: Testimonios of Chicana communication professors. Chicana/Latina Studies: The Journal of Mujeres Activas en Letras y Cambio Social, 16, 158-188.

Castrellón

L. E.

Reyna Rivarola

A. R.

López

G. R.

(2017). We are not alternative facts: Feeling, existing, and resisting in the era of Trump. International Journal of Qualitative Studies in Education, 30, 936-945.

10.

Claesens

Engel

(2013). How important is where you start? Early mathematics knowledge and later school success. Teachers College Record, 115(6), 1-29.

11.

Clotfelter

C. T.

Ladd

H. F.

Vigdor

J. L.

(2010). Teacher credentials and student achievement in high school a cross-subject analysis with student fixed effects. Journal of Human Resources, 45, 655-681.

12.

Collins

P. H.

(1990). Black feminist thought: Knowledge, consciousness, and the politics of empowerment. Cambridge, MA: Unwin Hyman.

13.

Collins

P. H.

Bilge

(2016). Intersectionality. Malden, MA: John Wiley.

14.

Conchas

G. Q.

(2001). Structuring failure and success: Understanding the variability in Latino school engagement. Harvard Educational Review, 71, 475-504.

15.

Covarrubias

(2011). Quantitative intersectionality: A critical race analysis of the Chicana/o educational pipeline. Journal of Latinos and Education, 10, 86-105.

16.

Covarrubias

Lara

(2014). The undocumented (IM) migrant educational pipeline: The influence of citizenship status on educational attainment for people of Mexican origin. Urban Education, 49, 75-110.

17.

Covarrubias

Nava

P. E.

Lara

Burciaga

Vélez

V. N.

Solorzano

D. G.

(2018). Critical race quantitative intersections: A testimonio analysis. Race Ethnicity and Education, 21, 253-273.

18.

Covarrubias

Revilla

A. T.

(2003). Agencies of transformational resistance. Florida Law Review, 55, 459-477.

19.

Crenshaw

(1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241-1299.

20.

Darling-Hammond

(2010). The flat world and education: How America’s commitment to equity will determine our future. New York, NY: Teachers College Press.

21.

Darling-Hammond

(2015). Can value added add value to teacher evaluation? Educational Researcher, 44, 132-137.

22.

DeLeon

A. P.

(2010). How do I begin to tell a story that has not been told? Anarchism, autoethnography, and the middle ground. Equity & Excellence in Education, 43, 398-413.

23.

Delpit

L. D.

(2006). Other people’s children: Cultural conflict in the classroom. New York, NY: The New Press.

24.

Delpit

L. D.

(2012). Multiplication is for White people: Raising expectations for other people’s children. New York, NY: The New Press.

25.

Dhamoon

R. K.

(2010). Considerations on mainstreaming intersectionality. Political Research Quarterly, 64, 230-243.

26.

Duncan

G. J.

Magnison

K. A.

(2005). Can family socioeconomic resources account for racial and ethnic test gaps? The Future of Children, 15, 35-54.

27.

Easton

J. Q.

Johnson

Sartain

(2017). The predictive power of ninth-grade GPA. Chicago, IL: University of Chicago Consortium on School Research.

28.

Galindo

Reardon

S. F.

(2006). Hispanic students’ educational experiences and opportunities during kindergarten: A report to the national task force on early childhood education for Hispanics. Stanford, CA: Stanford University Graduate School of Education.

29.

Gillborn

Warmington

Demack

(2018). QuantCrit: Education, policy, “Big Data” and principles for a critical race theory of statistics. Race Ethnicity and Education, 21, 158-179.

30.

Goldhaber

(2002). The mystery of good teaching. Education Next, 2(1), 1-7. Retrieved from http://www.nuatc.org/articles/pdf/mystery_goodteaching.pdf

31.

Goldhaber

Lavery

Theobald

(2015). Uneven playing field? Assessing the teacher quality gap between advantaged and disadvantaged students. Educational Researcher, 44, 293-307. doi:10.3102/0013189X15592622

32.

Gonzales

(1980). Toward a feminist pedagogy. Frontiers: A Journal of Women Studies, 5(2), 48-51.

33.

Hahs-Vaughn

D. L.

McWayne

C. M.

Bulotsky-Shearer

R. J.

Wen

Faria

A. M.

(2011). Methodological considerations in using complex survey data: An applied example with the Head Start Family and Child Experiences Survey. Evaluation Review, 35, 269-303.

34.

Hancock

A. M.

(2007). When multiplication doesn’t equal quick addition: Examining intersectionality as a research paradigm. Perspectives on Politics, 5, 63-79.

35.

Hanushek

E. A.

(2007). The single salary schedule and other issues of teacher pay. Peabody Journal of Education, 82, 574-586.

36.

Harris

D. N.

Sass

T. R.

(2011). Teacher training, teacher quality and student achievement. Journal of Public Economics, 95, 798-812.

37.

Hoefer

Rytina

Baker

(2011). Estimates of the unauthorized immigrant population residing in the United States. Washington, DC: Department of Homeland Security.

38.

Irvine

J. J.

(2010). Foreword. In Milner

H. R.

(Ed.), Culture, curriculum, and identity in education (pp. xi-xvi). New York, NY: Palgrave Macmillan.

39.

Jang

S. T.

(2018). The implications of intersectionality of race, gender, and socioeconomic status on Southeast Asian female students’ educational outcomes: Critical quantitative intersectionality analysis. American Educational Research Journal, 55, 1268-1306.

40.

Koretz

(2008). Measuring up: What educational testing really tells us. Cambridge, MA: Harvard University Press.

41.

Ladson-Billing

(2006). From the achievement gap to the education debt: Understanding achievement in U.S. Schools. Educational Researchers, 35(7), 3-12.

42.

Lee

(2012). College for all: Gaps between desirable and actual P–12 math achievement trajectories for college readiness. Educational Researcher, 41(2), 43-55.

43.

López

Erwin

Binder

Chavez

M. J.

(2018). Making the invisible visible: Advancing quantitative methods in higher education using critical race theory and intersectionality. Race Ethnicity and Education, 21, 180-207.

44.

Lumley

(2011). Complex surveys: A guide to analysis using R (Vol. 565). Hoboken, NJ: John Wiley.

45.

McCall

(2005). The complexity of intersectionality. Signs: Journal of Women in Culture and Society, 30, 1771-1800.

46.

McCallumore

K. M.

Sparapani

E. F.

(2010). The importance of the ninth grade on high school graduation rates and student success. The Education Digest, 76(2), 60-64.

47.

McLaughlin

H. J.

Liljestrom

Lim

J. H.

Meyers

(2002). LEARN: A community study about Latino immigrants and education. Education and Urban Society, 34, 212-232.

48.

Milner

H. R.

(2012). Beyond a test score: Explaining opportunity gaps in educational practice. Journal of Black Studies, 43, 693-718.

49.

Milner

H. R.

(2013). Rethinking achievement gap talk in urban education. Urban Education, 48, 3-8.

50.

Milner

H. R.

Lomotey

(Eds.). (2013). Handbook of urban education. New York, NY: Routledge.

51.

Muñoz

(2010). Brown, queer and gendered: Queering the Latino/o “street-scapes” in Los Angeles. In Browne

Nash

C. J.

(Eds.), Queer methods and methodologies: Intersecting queer theories and social science research (pp. 55-68). New York, NY: Routledge.

52.

National Center for Education Statistics. (2017). Status and trends in the education of racial and ethnic groups. Retrieved from https://nces.ed.gov/programs/raceindicators/highlights.asp

53.

Núñez

(2014). Advancing an intersectionality framework in higher education: Power and Latino postsecondary opportunity. In Paulsen

M. B.

(Ed.), Higher education: Handbook of theory and research (pp. 33-92). Dordrecht, The Netherlands: Springer.

54.

Núñez

A. M.

Kim

(2012). Building a multicontextual model of Latino college enrollment: Student, school, and state-level effects. The Review of Higher Education, 35, 237-263.

55.

Oakes

(1990). Multiplying inequalities: The effects of race, social class, and tracking on opportunities to learn mathematics and science. Retrieved from http://eric.ed.gov/?id=ED329615

56.

O’Connor

(2009). Hispanic origin, socio-economic status, and community college enrollment. The Journal of Higher Education, 80, 121-145.

57.

Ortiz

A. M.

Santos

S. J.

(2009). Ethnicity in college: Advancing theory and improving diversity practices. Sterling, VA: Stylus.

58.

Ovink

S. M.

(2014). “They always call me an investment” Gendered familism and Latino/a college pathways. Gender & Society, 28, 265-288.

59.

Passel

Cohn

(2011). Unauthorized immigrant population: National and state trends. Washington, DC: Pew Hispanic Center.

60.

Perreira

K. M.

Harris

K. M.

Lee

(2006). Making it in America: High school completion by immigrant and native youth. Demography, 43, 511-536.

61.

Ramírez

(2013). Examining Latinos/as’ graduate school choice process: An intersectionality perspective. Journal of Hispanic Higher Education, 12, 23-36.

62.

Reardon

S. F.

Galindo

(2009). The Hispanic-White achievement gap in math and reading in the elementary grades. American Educational Research Journal, 46, 853-891.

63.

Reardon

S. F.

Valentino

R. A.

Kalogrides

Shores

K. A.

Greenberg

E. H.

(2013). Patterns and trends in racial academic achievement gaps among states, 1999−2011. Stanford, CA: Stanford University Graduate School of Education.

64.

Ridgeway

C. L.

Correll

S. J.

(2004). Unpacking the gender system a theoretical perspective on gender beliefs and social relations. Gender & Society, 18, 510-531.

65.

Rong

X. L.

Brown

(2002). Immigration and urban education in the new millennium: The diversity and the challenges. Education and Urban Society, 34, 123-133.

66.

Sandoval

(1991). US third world feminism: The theory and method of oppositional consciousness in the postmodern world. Genders, 10, 1-24.

67.

Schmidt

W. H.

McKnight

C. C.

(2012). Inequality for all: The challenge of unequal opportunity in American schools. New York, NY: Teachers College Press.

68.

Shlapentokh

Beasley

(2015). Power and inequality in interpersonal relations. New Brunswick, NJ: Transaction Publishers.

69.

Sólorzano

D. G.

Bernal

D. D.

(2001). Examining transformational resistance through a critical race and LatCrit theory framework: Chicana and Chicano students in an urban context. Urban Education, 36, 308-342.

70.

Sólorzano

D. G.

Villalpando

Oseguera

(2005). Educational inequities and Latina/o undergraduate students in the United States: A critical race analysis of their educational progress. Journal of Hispanic Higher Education, 4, 272-294.

71.

Stage

F. K.

(2007). Answering critical questions using quantitative data. New Directions for Institutional Research, 2007, 5-16.

72.

Stage

F. K.

Manning

(2016). Research in the college context: Approaches and methods. New York, NY: Routledge.

73.

Stoops

(2004). Educational attainment in the United States: 2003. Washington, DC: U.S. Department of Commerce.

74.

S. R.

Romero

(2008). Family responsibilities among Latina college students from immigrant families. Journal of Hispanic Higher Education, 7, 212-227.

75.

Teranishi

R. T.

(2007). Race, ethnicity, and higher education policy: The use of critical quantitative research. New Directions for Institutional Research, 2007, 37-49.

76.

Valencia

R. R.

(1997). The evolution of deficit thinking: Educational thought and practice. Abingdon, UK: Routledge Falmer.

77.

Valencia

R. R.

Black

M. S.

(2002). “Mexican Americans don’t value education!” On the basis of the myth, mythmaking, and debunking. Journal of Latinos and Education, 1, 81-103.

78.

White

(1991). The middle ground: Indians, empires, and republics in the Great Lakes region, 1650-1815. Cambridge, UK: Cambridge University Press.