Toward a Multidimensional Conceptual Framework for Understanding “Servingness” in Hispanic-Serving Institutions: A Synthesis of the Research

Databases and Search Parameters

We searched the ERIC (EBSCO) database using the terms “Hispanic-Serving Institution” and “Hispanic-Serving Institutions.” The search identified a total of 293 publications that included academic journals (n = 169), ERIC documents and dissertations (n = 102), education reports (n = 6), and magazines (n = 16). Using the same keyword searches, we also manually searched the Journal of Hispanic Higher Education (JHHE) because the journal publishes work that is specific to Latinx college students and HSIs. We cross-checked our search of peer-reviewed journal publications by looking at several other databases including Academic Search Complete, EconLit, JSTOR, PsycINFO (ProQuest), ScienceDirect, Web of Science, and Wiley Online Library. Searching these additional databases and JHHE produced 14 academic journal articles that were not duplicative of the previous articles found when searching the EBSCO Host. Examples of the journal sources that were identified in our initial search are provided in Figure 2. Of note, JHHE was the source of a total of 51 publications that were identified in our initial search. Finally, we added book chapters from four books, three of which are specifically about HSIs and one which is about MSIs more broadly (n = 39). Collectively, our search yielded a total of 346 publications.

OPEN IN VIEWER

Figure 2.

Journals identified in the initial search.

Selection and Screening Process

The publications derived from this initial search were included in a systematic review and judged against an inclusion/exclusion criterion. Studies were included in the initial list if there was an implicit or explicit link to HSIs, including studies that explicitly explored concepts and phenomena with HSIs and those focused on other concepts and phenomena, but conducted at HSIs. This included studies that were not necessarily about higher education phenomena. For example, we included a study that had “HSI” in the title and sampled Latinx students enrolled at an HSI but was focused on health outcomes. Studies that were focused on understanding MSIs were also included in the initial selection if they explicitly addresses HSIs, even if periphery. Studies conducted at HSIs, but that were focused on students from non-Latinx backgrounds, were also considered.

Although we wanted to be as inclusive as possible in order to avoid the loss of potential sources of information, we ultimately decided to exclude ERIC documents and dissertations, education reports, and magazines (n = 124). We excluded dissertations since some had been transformed into multiple journal articles, and we did not want duplication. We also excluded education reports, policy briefs, and magazines, as the range and quality was too large to account for in a systematic way. Moreover, they had not been through an objective peer-review process.

Applying the initial inclusion/exclusion criteria reduced the sample of publications to be systematically reviewed to 222, including 183 journal articles and 39 book chapters. The selected studies represented different research designs including quantitative, qualitative, policy, historical, and conceptual approaches, among others. No limit was placed on the characteristics of the HSIs included; therefore, studies that focused on HSIs across institutional types (e.g., public, private, 2-year, 4-year, etc.) were included. In order to classify each of the 222 publications for analysis, we created a shared spreadsheet that captured and organized the following: (1) whether or not the HSI context was considered in the design and implementation of the study; (2) whether or not deficit framing was used in the analysis or findings; (3) the unit of analysis; (4) whether or not the study was empirical; and (5) the institutional characteristics of the HSI(s). We used this shared spreadsheet to review selected articles for quality criteria, to establish trustworthiness, to develop a list of codes that captured the distinct ways researchers conceptualized servingness, and to code the final set of publications included.

Quality Criteria

The 222 publications that were selected from the initial inclusion/exclusion procedure were systematically assessed for quality. The standard of each study’s quality was assessed using criterion measures (Caldwell, Henshaw, & Taylor, 2005; Petticrew & Roberts, 2006), represented by a series of questions. These questions were created by the authors to help them identify and assess the scientific quality of each publication (see Table 1). The data from these questions were added to the shared spreadsheet in order to help the authors organize the information obtained while assessing whether each study met the quality criteria. We responded to each criterion measure question using a 0- to 1-point scale, with 0 = no and 1 = yes.

OPEN IN VIEWER

Table 1

Quality criteria

Category	Criteria question
Empirical	1. Is the article empirical?
	2. Is the research purpose or objective clear?
	3. Is the literature review, conceptual, or theoretical framework appropriate and driving the research questions and/or methods?
	4. Is the method used appropriate for addressing the purpose or objective?
	5. Is there sufficient sample/data to address the purpose or objective?
	6. Is the research context adequately described?
	7. Is the analysis adequate or appropriate for addressing the purpose/objective?
	8. Are the results findings clearly presented and connected to the data?
	9. Are the methodological limitations and or trustworthiness stated?
	10. Are the conclusions drawn from or connected to the data and empirical evidence?
Nonempirical	1. Is the article nonempirical (descriptive or theoretical or program evaluation or trends analysis)?
	2. Is the research purpose or objective clear?
	3. Is the problem statement, introduction, literature review, conceptual or theoretical framework appropriate and connected to the purpose?
	4. Is the description or theoretical argument or evaluation methods sufficient for responding to the purpose?
	5. Is the context adequately described?
	6. Are the conclusions drawn from the description or theoretical analysis or evaluation?

Our selection of studies was further based on two different criterion scales—empirical and nonempirical publications. We purposefully created these two criterion scales because the articles that we labeled as nonempirical did not fit well into the first set of criterion measures we created. For example, one measure of quality was the use of a guiding theoretical or conceptual framework, yet we found that nonempirical publication did not usually include a theoretical or conceptual framework. We also assessed methods and analyses used, which were not always clear in nonempirical publications. As such, we developed a second set of questions and criterion scale that was better designed to assess the quality of nonempirical work. Once the scales were developed, each study was grouped according to whether they were identified as empirical or nonempirical. We then reviewed studies using the appropriate criterion scale. Empirical studies, or those that employed scientific methods and included data collection and analyses, were evaluated along 10 criterion measures and were included in the final database for analysis if the total score was equal to or greater than 6. Studies that were nonempirical, descriptive, or theoretical in nature were evaluated along 6 criterion measures and must have achieved a score a 4 or higher to be included in the final database for analysis.

Studies that did not meet quality criteria (i.e., below 6 for empirical studies or below 4 for nonempirical studies) were excluded from the final review. Empirical studies were often excluded for not meeting quality criteria because they had a misalignment between the title (“Effectively Serving AB540 Undocumented Students at an HSI”), the purpose (to explore the experiences of undocumented students), the research questions (no clear research questions listed), the conceptual framework (barriers and obstacles, resilience), the methods (mixed, surveys, interviews, and focus groups), and the unit of analysis (faculty and staff). Nonempirical studies that were excluded based on quality often were misaligned in many ways as well, such as title (“Hispanic College Students Library Experience”) and purpose (to examine academic libraries move toward electronic library materials). After completing this quality check procedure, 74 publications were determined to be below quality and therefore removed. This left a final total of 148 publications, which included 109 journal articles (empirical and nonempirical) and 39 book chapters (empirical and nonempirical), that met our quality criteria and were included in our systematic review. Figure 3 shows the entire search and screening process.

OPEN IN VIEWER

Figure 3.

Search and screening process results.

Coding and Intercoder Reliability

Using the 148 publications deemed acceptable quality, next we developed a coding structure to be used in our analysis. Each author read, reviewed, and developed initial codes for 20 randomly selected studies independently. Studies that were published by one of the authors were assigned to be reviewed by another author that did not work on the publication. Each author used the shared spreadsheet to develop initial codes and noted any questions that emerged when reading and reviewing each publication. After the first round of evaluations were completed, we met to discuss the codes that were emerging. During this meeting, we developed a list of 11 codes, discussed the meaning of each code, developed initial definitions of each code, and assessed the codes’ relevance to the research questions. Using the list of 11 codes and their definitions, we then independently coded the same 20 articles. We added an additional eight articles to this phase in an effort to ensure coding consistency and validity. Once coding was completed, we met a second time to discuss the usefulness of the codes and our understanding of the codes.

Through this systematic process of developing codes, we sought to ensure trustworthiness and minimize article selection bias between authors (DeCuir-Gunby, Marshall, & McCulloch, 2011). In addition to discussing codes during these meetings, we also discussed each of the other categories in the spreadsheet, including whether or not we felt the study framed HSIs as deficit, and if the researchers had considered the HSI context. We continued to interrogate each article while challenging our own assumptions about quality and usefulness of each article in conceptualizing servingness, and about what we believed was deficit framing. When discrepancies occurred, either about code definitions or about our assumptions as researchers, we discussed the issues until an agreement was reached by all the authors. This entire process was done to ensure trustworthiness, reliability, and consistency between the authors before coding the remaining articles and moving to analysis.

Through this process, we simultaneously developed a coding framework to be used for analysis. The framework included the following 11 codes: (1) academic outcomes, (2) nonacademic outcomes, (3) student experiences, (4) faculty and staff experiences, (5) external policy and governance, (6) internal leadership and decision making, (7) culturally relevant practices, (8) culturally relevant curriculum and/or pedagogy, (9) history and overview, (10) access, and (11) other. We also identified four initial themes, collapsing the 11 codes into these themes, which further expounded our investigation of how researchers had conceptualized servingness: (1) outcomes, (2) experiences, (3) internal organizational dimensions, and (4) external forces (see Table 2 for themes, codes, and definitions).

OPEN IN VIEWER

Table 2

Description of themes

Theme	Code	Definition	Density
Outcomes	Academic outcomes	Focused on academic outcomes as main variable of interest including GPA, graduation, 6-year grad rate, retention, persistence, course enrollment patterns, transferring, and course completion	30
	Nonacademic outcomes	Focused on nonacademic outcomes as main variable of interest including racial identity development, academic self-concept, engagement on campus, civic engagement, leadership identity development, and sense of belonging	24
Experiences	Student experiences	Focused on student experiences at institution(s) including those with campus racial climate, microaggressions, friendships on campus, same-race interactions, interactions with faculty, staff, and students, speaking Spanish	22
	Nonstudent experiences	Focused on faculty, staff, and administrator (nonstudents) experiences including those with campus racial climate, microaggressions, isolation, exclusion, inclusion, mentoring students, and serving as institutional agents	17
Internal organizational dimensions	Leadership and decision making	Highlights internal factors that affect the development and growth of Hispanic-Serving Institutions (HSIs), including institutional level decision making and policies, as well as leadership within HSIs; factors within the control of the institution	19
	Culturally relevant curriculum or pedagogy	Highlights curricular and pedagogical practices that faculty in the institution implement that take the cultural and racial ways of knowing of Latinx and other diverse students into consideration	22
	Culturally relevant practices	Highlights practices and programs that the institution supports and or implements that take the cultural and racial ways of knowing of Latinx and other diverse students into consideration	19
External forces	Policy and governance	Highlights external factors that have affected the development and growth of HSIs including decisions and policies of the local, state, and federal government; factors beyond the control of the institution	17

Data Analysis

In the next phase of analysis, the remaining studies of the 148 (n = 120) were randomly assigned to the authors and coded independently using the coding framework, bringing the total sample size used in the analysis to n = 148. Understanding how serving was conceptualized by other researchers involved a process of examining each study’s research design, framing of the research questions, data sources, analysis approach, and interpretation of findings. During the data analysis process, we considered the rigor of the research and whether there was strength in the evidence and interpretations made about HSIs. Specifically, we examined whether the research design was focused on understanding HSIs, as organizations or as a population, or the influence of HSIs. This included assessing whether the study drew from research on HSIs and effectively articulated a conceptualized framework to understand the organizational context and variation. We felt this was particularly important among studies that grouped HSIs as a whole despite there being contextual variation in their samples (e.g., 2-year vs. 4-year, public vs. private, rural vs. urban). For quantitative studies, we scrutinized whether appropriate variables were selected that could accurately be used to understand servingness. Among qualitative studies, we paid particular attention to the data collection and whether it detailed and accounted for the context of the HSI(s). Last, in inspecting the units of analysis, data sources, and sample sizes in each study, we determined whether a study overstepped a broader generalization about HSIs. In other words, we examined whether the results found in a study that sampled one HSI used these findings to make generalizations about all HSIs.

Outcomes

Outcomes were a common way that authors conceptualized HSI servingness, whether implicitly described as measures of servingness or not (n = 54). Outcomes included both academic and nonacademic variables, and were largely student-centered, meaning researchers made meaning of servingness based on student-level outcomes.

Experiences

A second way that authors conceptualized servingness was by exploring and describing experiences that people have at HSIs (n = 39). Experiences included those of both students and nonstudents (faculty, staff, administrators), suggesting that there is more to servingness than what happens to students. Instead, the experiences of all members within the organization shape servingness. Examples of experiences examined in the research included interactions with same-race peers, the presence of Spanish-speaking faculty, staff, and students, and perceptions of the campus climate at HSIs.

Internal Organizational Dimensions

The third way that researchers conceptualized servingness was by addressing internal dimensions in the organization (HSI) that had been altered in order to respond to the needs of Latinxs on campus (n = 60). Internal organizational dimensions included institutional policies and decision making as well as curricula, programs, and practices at the institution. We considered these elements within the control of the institution, and therefore essential for understanding, as these dimensions can be altered, as needed, to better serve Latinxs. We broke the publications that centered their analysis on internal organizational dimensions into three subthemes: leadership and decision making, culturally relevant pedagogy and curriculum, and culturally relevant programs.

External Influences

The fourth and final theme is external influences, such as state and federal legislation. Although authors did not necessarily write about policies and external forces as indicators of servingness, these influences may affect an institution’s ability to effectively serve minoritized students. Also, policy and governance are often beyond the control of the institution, yet institutions respond to the external environment, seek legitimacy, and act in ways that conform with institutionalized ways of knowing as determined by the field (DiMaggio & Powell, 1983; Gonzales, 2013), so it is important to regularly monitor these forces. In our search, the fewest number of articles and book chapters discussed external influences (n = 17).

Some literature focused on the federal legislation that led to the establishment of HSIs; however, most of it was historical, with a majority describing how HSIs became HSIs. In a thorough historical depiction of legislative actions leading to the 1992 decision that recognized HSIs, Valdez (2015) highlighted the advocacy work of the Hispanic Higher Education Coalition and the Hispanic Association of Colleges and Universities (HACU). He argued that Hispanic Higher Education Coalition, which included high-profile Latinx groups such as the Mexican American Legal Defense and Education Fund and the National Council of La Raza, and HACU were pivotal in advocating for the HSI designation during legislative hearings throughout the late 1970s and 1980s. Like previous historical analyses published by MacDonald, Botti, and Clark (2007) and Olivas (1982), Valdez stressed that the recognition of HSIs at the federal level was the result of self-determination and long-term advocacy efforts by Latinx community and educational leaders who sought out allies and resources that would help institutions enrolling the largest percentage of Latinxs better serve them.

Other publications have highlighted external forces that shape servingness. HACU, for example, has remained a constant advocate for Latinx college students in general, and the primary champion for HSIs, lobbying for greater political and economic support for these institutions (Calderón Galdeano, Flores, & Moder, 2012). Calderón Galdeano et al. (2012) published a descriptive article, stressing HACU’s essential role in increasing appropriations to HSIs under Title III and Title V, as well as lobbying for funding from agencies such as the U.S. Department of Agriculture, the Department of Housing and Urban Development, and the Department of Defense. In fact, the federal investment in HSIs increased annually from 1996 to 2004; however, it has since remained steady, at about $92 to $94 million each year, despite the fact that the number of HSIs has increased dramatically each year, and will continue to increase (Ortega et al., 2015).

Ortega et al. (2015) used data from the Delta Cost Project to analyze revenue trends for HSIs, raising concerns about these trends. For example, they showed that public 2-year and 4-year HSIs relied heavily on state, local, and federal grants and appropriations, are limited in their ability to increase tuition as a source of revenue, and are absorbing the declining investment in higher education from their states (Ortega et al., 2015). Even more concerning, research has shown that leaders at HSIs were ill-equipped to enhance their institutional advancement efforts (Mulnix et al., 2002, 2004; Ortega et al., 2015). Using survey data collected from 80 HSI leaders, Mulnix et al. (2004) found that 30% of the HSIs in their sample did not have offices that regularly monitored federal grant opportunities and 35% did not have staff that regularly interfaced with local, state, and private foundations.

Some researchers have considered the influence of state-level policies on the emergence of HSIs. M. Martinez (2015) analyzed state-level policy in Nevada, showing the political process undertaken as the state sought to embrace the emerging HSI status of Nevada’s public institutions. In response to declining financial support for higher education in the state, key stakeholders, such as state legislative representatives, the Latin Chamber of Commerce, and campus Offices of Diversity, explored the definition and implications of being HSIs (M. Martinez, 2015). Yet even with state-level support for HSIs, Martinez showed that there were still tensions present, such as those presented by a simultaneous push for a college completion agenda and pushback from academic deans who had concerns about their ability to hire faculty of color to match the changing student demographics (M. Martinez, 2015). Using an historical analysis, Doran (2015) documented similar tensions in Texas, as the University of Texas, San Antonio, made decisions that increased the enrollment of students of color, such as opening a downtown campus closer to the population, while simultaneously pushing for Tier 1 status, which in many ways negated their efforts to be broad access. These case studies highlighted clearly how external forces, such as state-level policies and decisions, may help or hinder an institution’s ability to serve racially minoritized students.

Flores and Morfin (2008) used descriptive enrollment data to analyze changes in the enrollment of students from various racial/ethnic backgrounds into less selective institutions (which are most likely to be HSIs) in California and Texas, since both states had policies banning affirmative action in admissions. They found limited evidence that the bans on affirmative action had led to an increase in the enrollment of Latinxs in less selective institutions (Flores & Morfin, 2008). More important, they found that Black, Asian, and white students have shown a greater overall increase in enrollment into these less selective institutions than Latinxs (Flores & Morfin, 2008). Núñez and Bowers (2011) came to similar conclusions, empirically showing an increase in the enrollment of racially minoritized students at HSIs (e.g., Black, Asian American, Native Americans), suggesting that external policies may have been the reason. Torres and Zerquera (2012) used U.S. Census data to identify “potential” HSIs in states with increasing numbers of Latinxs in the population, noting the importance of recognizing how changing demographics in the external environment must lead to changes internal to the institution. In general, the literature that focuses on external influences, such as legislation and policies, show that these forces indirectly affect institutions’ ability to serve Latinx students.

Framing HSIs in Research

Through our methodological analysis, we discovered two important considerations when exploring how researchers explicitly or implicitly frame the limitations and contributions of HSIs: (1) unit of analysis and (2) intentionality in considering the HSI context in research design and approach.

Deficit Framing

Of the 69 cases in which the HSI context was explicitly considered, we coded 10 as “deficit framing,” which we defined as researchers framing HSIs in a way that portrayed them as less effective than non-HSIs, sometimes making claims that HSIs are not actually serving Latinx students or arguing that they are simply “Hispanic-enrolling” (used by authors in a negative way, minimizing the fact that providing access to Latinxs is an important endeavor for postsecondary institutions). In one example, authors claimed that Latinx students might perform better at non-HSIs with higher transfer and graduation rates and more opportunities. This language reifies the normative standards of postsecondary institutions, and fails to recognize that the data used to make these types of claims are often flawed. Espinosa et al. (2017) found that using National Student Clearinghouse data provided a more positive outlook on HSIs, as it follows students through their educational journey, including transfer patterns, compared with federal data that only follows first-time, full-time enrollment, which is not the norm for students enrolled in HSIs. Although researchers should certainly be critical of HSIs, as HSIs have a significant amount of work to do in order to become fully functional spaces of inclusion and liberation (Garcia, 2018; Núñez, 2017b), scholars must be careful to recognize limitations in data and methods used, and be intentional in considering the complexities of sociohistorical contexts.

In some instances when HSIs were framed as deficit, the methods were rigorous and the findings substantiated, yet conclusions were made about HSI cultures and contexts without properly accounting for organizational variables in the research design. Hubbard and Stage (2009), for example, stated, “We found only a few differences between Hispanic Serving Institutions and those with fewer Latino students suggesting further evidence that HSIs do not have institutional missions that directly serve the needs of the Latino population” (p. 285). They went on to say, “These institutions demonstrate few differences with PWIs” (p. 285). Although it is true that HSIs do not have an historical mission for serving Latinxs, and the authors made valid comments about differences between HSIs, HBCUs, and PWIs, the authors’ overall claims were made to support their findings about differences in faculty attitudes, perceptions, and preferences, with little evidence of a connection between historical mission and faculty attitudes. Moreover, their study did not include measures of historical mission, so this connection was anecdotal, at best. The concern is that readers automatically make the assumption, whether the authors state it or not, that HSIs are not actually serving Latinx students because they really are white institutions. As evidenced by the four themes we found with regard to servingness, the concept is much more complex than a few variables measuring faculty attitudes, perceptions, and preferences can alone reveal.

Nelson Laird, Bridges, Morelon-Quainoo, Williams, and Holmes (2007) made similar comparisons between HSIs, PWIs, and HBCUs using regression analyses and data from the National Survey of Student Engagement, stating that because Latinxs were engaged at HSIs in similar rates as they are at PWIs, HSIs, “We suggest that HSIs are in the midst of a shift from having white-oriented institutional cultures to cultures inclusive of Hispanic student and their educational needs” (p. 51). Again this suggestion is valid, and although this may be the case, their analysis was at the student level, meaning they accounted for student perceptions of the environment, yet their claims about the “white-oriented culture” were at the organizational level. The authors did not include indicators of culture in their statistical models, such as those institutional organizational variables we discussed as indicators of servingness (e.g., curriculum, pedagogical approaches, support programs). Again, readers may take this conclusion as valid, without being critical of the fact that this claim was beyond the scope of the research design.

In other instances where we coded articles as deficit, the authors made generalizing claims about HSIs based on limited evidence. For example, Contreras et al. (2008), following a review of the mission statements of 10 HSIs, found that none of the 10 made reference to the institution’s mission to serve Latinx students. They went on the state that the HSI identity is a “manufactured identity that is highly variable” (p. 74), which is certainly true; in fact, it is an identity manufactured by the federal government (Santiago, 2012). After a more thorough review of these 10 institutions’ website, the authors stated, “Based on the website search we were not able to discern a Latina/o agenda across the Hispanic-serving institutions assessed” (p. 78). Yet this claim is based on a review of 10 mission statements/institutional websites, despite the fact that there are nearly 500 HSIs in the United States. Moreover, it assumes that mission statements and websites are key indicator of servingness. Again, our thorough review of the literature shows that servingness is more complicated than can be revealed in a mission statement or website.

Multidimensional Conceptual Framework of Servingness in HSIs

Evidence from our research review shows that servingness in HSIs cannot be reduced to one (or even two) factor(s). Rather, it manifests in multiple ways and should be conceptualized along multiple dimensions. Based on the findings from the literature review, we added the consideration of systemic, political, and historic forces to the consideration of individual and organizational forces, in order to advance a multidimensional conceptual framework of servingness in HSIs that is grounded in extant research to date (see Figure 4). It is informed by a comprehensive analysis of how researchers to date have framed servingness based on the main phenomena they analyzed. In other words, we came to understand how servingness has been understood in research based on the variables and elements observed, outcomes and experiences explored, and overall findings of current research centering HSIs. Outcomes and experiences tend to address the individuals (both students and other personnel, including faculty, staff, and administrators) as the units of analysis. External influences on serving are more distal, shaping the capacity for HSIs to develop structures for serving Latinx students. Thus, structures for serving and external influences on serving, in contrast to outcomes and experiences, are likely to focus on organizations as units of analysis.

OPEN IN VIEWER

Figure 4.

Multidimensional conceptual framework of servingness in HSIs.

Implications for Research, Practice, and Policy

The importance of this multidimensional conceptual framework for understanding servingness in HSIs is that it illustrates the complexity of what it means for an institution to be or become an HSI, to embrace or enact an HSI organizational identity, and to come to serve Latinxs in HSIs. Importantly, the framework operationalizes a broader array of contexts and variables that can and must be used to define servingness in HSIs. Moreover, we found that there is an extant body of literature that centers HSIs that can and should be used by scholars, practitioners, and policy makers taking on the task of conceptualizing servingness. The foundational HSI research must be used in research and practice, and should influence future policy directions and decisions.

For researchers, the multidimensional conceptual framework for understanding servingness provides a way to conceptualize, develop, and implement studies with and for HSIs and the people within them. It provides a systematic way to structure and isolate the variables researchers are seeking to explore, while providing language about the complexities of servingness and the limitations of defining servingness by one element. As scholars develop studies about servingness, we encourage them to consider as many variables in the framework as possible, rather than reifying servingness to a single dimension.

With this study, we also caution researchers from making claims about HSIs and servingness, without fully accounting for the assumptions underlying the framing, limitations, and contributions of the study. For example, interpretations made about individuals (often students) as units of analysis should address what variables or qualities have or have not been examined in the data, such as organizational structures for serving or external influences on the capacity to serve. Researchers should explicitly recognize and state in methodological limitations sections the full range of variables or qualities that are missing from their analyses, and be careful in making broad claims about servingness in their discussion and implications sections. Subsequently, they should incorporate and speak to findings from extant literature about these missing or exogenous variables and qualities when interpreting their results. Utilizing the multidimensional conceptual framework of servingness in research is critical, as it will enhance methodological rigor of the research centering HSIs. Incorporating various elements of the framework into research will more fully contextualize the limitations and contributions of HSIs.

In practice, the multidimensional conceptual framework can help administrators, faculty, and staff in HSIs better understand how to transform their institutions in order to better serve Latinx students. This makes the framework essential, as the question about how to serve Latinx students is one of the most pressing questions in HSIs, with organizations such as the Alliance for HSI Educators hosting annual conferences specifically for practitioners seeking guidance and advice about best practices and approaches for serving students. The framework helps practitioners within HSIs understand the complexities of servingness; yet it provides a practical guide for understanding the various elements to consider. First, they may want to isolate key outcomes and experiences that they can measure, and determine how they will use those as indicators of serving on their campus, recognizing that not all HSIs should want their students to have the same experiences and outcomes. These variables will necessarily vary by mission, purpose, and institutional type (Garcia, 2018; Núñez et al., 2016). Then administrators, faculty, and staff may want to observe the structures in place that are influencing the desired outcomes and experiences, either positively or negatively. They may also conduct a SWOT analysis or ask key stakeholders to complete an inventory to help them identify these structures. Based on what they find, they may then hone in on some of the most commonly cited variables for better serving Latinxs, including curricula, pedagogy, services, and support structures.

The framework provides key policy advocacy groups such as HACU, Excelencia in Education, American Council on Education, The Education Trust, and the Campaign for College Opportunity with a more holistic understanding of servingness that can inform the evaluation of the “performance” of these institutions. Specifically, the framework more clearly operationalizes servingness, making explicit the meaning of servingness, which will help advocacy groups and policy makers better understand how to evaluate HSIs’ contributions to important societal outcomes such as economic mobility (Espinosa, Kelchen, & Taylor, 2018; NASEM, 2018) and STEM workforce development (NASEM, 2018). Moreover, the framework can help multiple audiences understand the extent of generalizability of research to all HSIs. We urge legislators and advocacy groups to consider the complexities of servingness as they evaluate HSIs. These groups must also see their role in defining and conceptualizing servingness within the framework, as external influences on serving, asking them to consider the best approaches for positively influencing and motivating those within the structures to better serve Latinx in HSIs.

HSIs play a critical role in the nation’s economic and scientific development (NASEM, 2018; NSF, 2017). Thus, the framework can also inform legislators’ and state, local, and federal agencies’ efforts to construct policies to address the needs of HSIs and, in turn, support these stakeholders to engage more thoughtfully and deeply with institutions that are critical to the development of regional and national workforces. For example, federal agencies that are interested in the advancement of science, such as NSF and NIH, can identify indicators from the framework that are salient to national goals of cultivating the science workforce. They can then employ associated indicators or dimensions to guide the development of new or existing program areas to support HSIs or to craft requests for grant proposals. State legislators, for example, might use the framework to guide their understanding of local HSIs’ contributions to local workforce development, and allocate resources to invest in the economic growth that could result from strengthening local HSIs’ capacities to train future employees in their respective regions.

The recent NASEM’s (2018) report on MSIs’ and HSIs’ contributions to the science workforce in the United States highlighted “intentionality” as a key feature that enables MSIs to cultivate talent in STEM among marginalized groups that in the past have had limited, if any, chances to pursue higher education. In addition to the directions described above, the framework for HSIs outlined here might also provide a departure point for scholars of national and global MSIs to better operationalize dimensions and indicators of intentionality in these kinds of institutions, as they seek to identify how diverse MSIs can best serve groups that historically have been underserved and outright discriminated against by higher education institutions and systems. Moving forward, it is critical to situate such analyses appropriately in salient local, state, federal, national, or global contexts (Hallmark & Gasman, 2018).