Advancing the Asian American and Pacific Islander Data Quality Campaign: Data Disaggregation Practice and Policy

Abstract

This study examines the impact of disaggregated data on shaping programs, services, and improving student outcomes for Asian American and Pacific Islander (AAPI) student populations at Coastline Community College (CCC). Using a mixed methods approach, including institutional data analysis and semi-structured staff interviews to examine the Asian American Native American Pacific Islander–Serving Institutions (AANAPISI) program and its use of disaggregated data to inform programmatic implementation and decision making, our findings indicate that disaggregated data largely shapes and improves the planning, implementation, and delivery of services to specific ethnic groups in order to enhance student experiences and outcomes.

Introduction

Decades of publications focusing on the Asian American and Pacific Islander (AAPI) student population have called for the collection and reporting of disaggregated data as a necessary tool for more accurately representing and understanding who students are, how they are performing, and how institutional student services can become more effectively targeted.[i] These publications cite the major differences that exist across various ethnic groups within the AAPI population as rationale for collecting more granular data.[ii] These concerns continue to be the motivation for AAPI data disaggregation efforts in the twenty-first century. The National Commission on Asian American and Pacific Islander Research in Education (CARE) reports that data disaggregated for individual subgroups—by race, ethnicity, gender, and other demographic distinctions—is critical for raising awareness about issues and challenges that impact subgroups disproportionately.[iii] The aggregation of data on AAPIs in higher education research, policy, and practice obscures the tremendous diversity that exists within the larger AAPI category.[iv] In reality, the AAPI population consists of at least forty-eight ethnic groups that differ greatly in cultural backgrounds, historical experiences, and socioeconomic and educational circumstances.[v]

As colleges and universities face increased concerns about accountability from internal and external constituents, data is an important tool for informing the work of practitioners and policy makers, which would allow them to target resources where they are most needed and to more effectively support an increasingly complex and heterogeneous student population. In order to improve data quality to more accurately represent and serve AAPI students, institutions must be supported by federal data policies; however, tangible change primarily occurs through the localized efforts of campuses and community groups. Thus, this paper aims to: (1) showcase the demonstrated success of data disaggregation efforts for improving AAPI student outcomes at a two-year institution; and (2) use this model to provide rationale for the advocacy of more localized data disaggregation efforts, as well as political support from federal data policy. 

Calls for AAPI Data Disaggregation

The need for disaggregated data that better represents the heterogeneity in the AAPI population is not a new concept. In fact, as early as 1980, before “disaggregated data” was widely used as a phrase, scholars were already pointing to the problem of lumping Asian Americans from different ethnic and cultural groups together into one category.[vi] Over three decades later, this issue is still at the forefront of the AAPI agenda in public health, social welfare, and civil rights.[vii] The aggregation of AAPI ethnic groups in data collection and reporting remains a prominent issue across all public sectors. The limited disaggregated data that is available reveals a startling reality. Median household income, for example, shows that although some ethnic subgroups, Asian Indians in this case, earn up to $22,000 over the median household income, Southeast Asians earn between $14,000 and $22,000 below the median.[viii] Disaggregated data also indicates lower levels of health care access and higher levels of welfare enrollment and incarceration for Southeast Asian and Pacific Islander subgroups.[ix] The differences between subgroup experiences and circumstance vary even more widely based on immigration history and language proficiency. Although research using disaggregated data that exists has been invaluable to the AAPI community, a sizable gap in scholarship remains to be filled and demand for data disaggregation in public services is greater than ever before. Most prominently, AAPI scholars in the education sector consistently and continuously call upon data disaggregation as a necessary tool for addressing AAPI student needs and improving educational outcomes.

In 1990, for example, Valerie Ooka Pang examined student scores in the California Achievement Test and found that more than 39 percent of Asian American students who took the test in the 1986-1987 academic year scored below the fiftieth percentile in reading. These findings, she concludes, point to the lack of research that examines the diverse needs of Asian American groups in the areas of language and readings skills.[x] In 2011, Pang, Peggy Han, and Jennifer Pang conducted a similar study examining the test scores of over one million seventh grade AAPIs in the California testing program. Their findings revealed that AAPIs, in aggregate form, performed significantly lower than White Americans in reading, but higher in math. Additionally, while Japanese and Chinese students scored over six points higher than the average AAPI reading scores, Cambodians and Laotians scored over eleven points and Samoans over fifteen points below the average.[xi]

Other studies pointing to the disproportional academic achievement within the AAPI community further demonstrate the need for data disaggregation to increase awareness about and target resources at bilingual education, poverty-related barriers, varying levels of English language proficiency, and AAPI-specific support programs and services, among other educational challenges and needs.[xii] The advocacy for disaggregated data is further supported by the release of the 2013 iCount report, which highlights the major postsecondary educational attainment disparities among AAPI subgroups. The report reveals, for example, that while 74.1 percent of Taiwanese, 71.1 percent of Asian Indians, and 52.7 percent of Koreans have a bachelor’s degree or higher, only 12.4 percent of Laotians, 14.1 percent of Cambodians, and 14.7 percent of Hmong have attained the same educational status.[xiii] As these studies show, the forty-eight AAPI ethnic subgroups range vastly above and below the average educational attainment rate and therefore cannot accurately be represented or served by aggregate data that masks their educational realities.

As evidenced by the three decades of research that have passed since the earliest recommendation for AAPI data disaggregation, this continues to be a persistent issue today and an increasingly critical one given the nation’s growing AAPI population. Increasing by 2.9 percent in 2012, the U.S. Census Bureau reported that Asians were the fastest growing demographic in the United States. Native Hawaiians and Pacific Islanders (NHPIs) grew 2.2 percent, falling only behind Latinos, the second fastest growing ethnic population.[xiv] Most of the growth in the Asian population was accounted for by international migration from various Asian countries to the United States, which highlights not only the rate of growth among AAPIs, but also the increasing heterogeneity experienced by this community.[xv] The impressive rise in the AAPI population indicates that it is more critical now than ever to find ways in which to accurately represent and understand the various histories, cultures, languages, and, most importantly, needs of these diverse communities. In order to achieve this, the AAPI data quality campaign must be nestled within a national context that influences sweeping change through political will. 

The National Context

The national dialogue regarding data disaggregation is complex, as it involves multiple levels of overlapping federal government branches, departments, agencies, and offices. The frequent reporting of oversimplified, aggregated data is exacerbated by the fact that federal databases do not consistently collect or report AAPI racial and ethnic categories. For example, while one database may aggregate AAPIs in one group, another may collect up to forty-eight different ethnicities. This patchwork of federal data collection methodologies can be difficult and frustrating to navigate, particularly as disaggregated data becomes increasingly important to the accurate delivery of culturally competent and meaningful services to ethnic specific populations.

In addition to the inconsistent reporting of ethnicities in various national data sets, many institutional administrators, researchers, and community leaders are faced with a number of challenges that hinder their ability to change their data collection methodologies. For example, at the iCount symposium, jointly hosted by CARE and the White House Initiative on Asian Americans and Pacific Islanders (WHIAAPI) in June of 2013, a group of administrators, institutional researchers, faculty, students, and other educational stakeholders from across the nation cited political will as one of the greatest barriers to data change. This finding echoes the sentiments of Professor Milbrey Wallin McLaughlin who poignantly notes, “Pressure from policy can be important even in settings that subscribe voluntarily to reform objectives simply because most institutions and individuals are allergic to change.”[xvi] Change, McLaughlin continues, “requires a combination of pressure and support from policy.”[xvii] In other words, despite the recognition that there is a deep need for better and more accurate data, institutions and administrators need to generate political will on campus to influence change. Data reform can be motivated by the application of pressure and support from education policy at the federal level. These policies may also be implemented by states, institutional systems, or local districts; however, federal policy is vital for change that universally impacts AAPI students across the nation.

To achieve this, ongoing efforts for data policy reform at the national level are taking place, and calls from members of Congress and officials within the Executive Branch have echoed similar sentiments. Congresswoman Judy Chu (U.S. House of Representatives and Chair of the Congressional Asian Pacific American Caucus), Congressman Mark Takano (U.S. House of Representatives), and Dr. Martha Kanter (Under Secretary of Education) attended the iCount symposium as keynote speakers and proponents of the data quality campaign, recognizing the importance of data disaggregation for the AAPI community. Additionally, one of the top educational priorities of the Congressional Asian Pacific American Caucus (CAPAC) is to “increase the reporting of disaggregated student achievement data based on ethnicity and increase the reporting of the school resources provided to communities that face educational challenges.”[xviii]

As evidenced by the inclusion of an increasing number of AAPI ethnic subgroups over the past twenty years, these efforts have been fruitful. For example, in 1990, the U.S. Census categories included ten Asian American and eight Pacific Islander ethnic groups. In 2000, the ethnic categories increased to eighteen Asian American and nine Pacific Islander ethnic subgroups. By 2010, the U.S. Census had expanded its subgroups to twenty Asian American and ten Pacific Islander categories—totaling thirty AAPI recognized subgroups as of today.[xix] These signs of incremental federal progress showcase the slow, but significant, success that the data quality campaign has experienced.

According to CARE’s 2008 report, however, there are at least forty-eight AAPI ethnicities in the United States, which highlights the work that has yet to be done to better represent the heterogeneity within the AAPI demographic.[xx] As the AAPI population continues to grow at an impressive rate, it is more important than ever that data disaggregation become a widespread practice among institutions. These efforts must be nested within a national context that (1) decreases the complexity that exists between various federal data sets, and (2) builds political will through education policy change.

In the current educational landscape, localized efforts have made positive gains to collect, utilize, and report disaggregated data; however, these successes are rarely spotlighted to make the case for data disaggregation at the national level. Therefore, while federal policy relies on institutions to make the case for data change, institutions simultaneously rely on federal policy to generate political will. This conundrum is the very site of data policy stagnation despite three decades of requests for data disaggregation. To overcome this stall in data change, case studies, which spotlight how the effective collection, use, and reporting of disaggregated data at the institutional level, must be used as a tool for advancing the data quality campaign. For example, the following case study, which demonstrates the successful data disaggregation efforts at one institution, provides a model and rationale for the advocacy of federal data policy that supports AAPI data change nationally.

Institutional Case Study: Coastline Community College

Coastline Community College (CCC) is a two-year public community college located in California. CCC was established in 1976 and comprises four separate centers located in different cities within the same county. Three of the centers, one of which includes a community art gallery, are instructional. The fourth facility houses administrative staff. CCC offers a variety of educational opportunities for students including associate degrees, career and technical education, and basic college readiness skills training and education programs, which have guaranteed transfer to a four-year university. One such program, in partnership with the University of Illinois, Springfield, offers a bachelor’s degree in computer science. CCC’s distance-learning program is extremely robust, as over half of its students take classes at a distance. CCC brands itself as a highly innovative institution that is capable of delivering a high-quality education, both in the classroom and through a variety of technologies. Located in four neighboring suburban cities, CCC caters to a diverse set of student constituencies, including traditional college students and nontraditional students (military, adult learners, students with intellectual disabilities, incarcerated students, and international students).

CCC has a total population of 12,577 students (fall 2013), 12 percent of whom are enrolled full-time, while the remaining 88 percent are part-time and noncredit students.[xxi] It should be noted that contracted military students (estimated at 3,800) are not counted among the total state-funded population. AAPIs account for 29 percent of the student population and are the largest ethnic group of color. Latino students make up 12 percent of the student body, followed by 8 percent African American students. White students are the largest ethnic group at 34 percent. CCC currently collects disaggregated ethnic data on ten Asian ethnic subgroups including Asian, Chinese, Indian, Japanese, Korean, Laotian, Cambodian, Vietnamese, Other Asian, and Filipino. In addition, CCC collects data on four Pacific Islander ethnic categories including Pacific Islander, Guamanian, Hawaiian, and Samoan. Vietnamese students account for the largest of the AAPI ethnic population making up 71.5 percent of the AAPI student body, followed by Chinese at 5.7 percent, Korean at 4.2 percent, and Japanese at 2.4 percent.[xxii]

The institution, located in Orange County, does not mirror the county demographics exactly, but generally reflects the population’s ethnic statistics. Orange County is located in Southern California with a population of just over 3 million residents; 17.9 percent of the population is AAPI, with Vietnamese being the largest AAPI ethnic group at 6.1 percent of the total population. The majority of Vietnamese Americans are foreign born, 70.3 percent, compared to 29.7 percent native born.[xxiii] The cities of Westminster and Garden Grove have the strongest concentration of Vietnamese Americans, in an area designated as Little Saigon. Within the Vietnamese American population (twenty-five years or older) of Orange County, 26.4 percent have less than a high school diploma, 19.7 percent graduated high school or equivalent, 25.3 percent have some college or an associate degree, 21.5 percent have a bachelor’s degree, and 7.2 percent possess a graduate or professional degree.[xxiv] Both CCC and Orange County have majority-minority demographics.

With its large AAPI student population, CCC applied for and received the Asian American Native American Pacific Islander–Serving Institutions (AANAPISI) federal designation and funding in 2010. The AANAPISI grant award totals $2 million over five years. For its proposal, CCC identified two main problems that needed to be addressed: (1) too few AAPI students were earning associate degrees; and (2) too few AAPI students were transferring to a four-year institution. In addition, CCC recognized that the low number of AAPI students enrolling in degree-applicable courses was a matter of concern. Thus, CCC’s project design focused on three goals:

1. Increase the number of AAPI students, originally underprepared for college work, in degree-applicable courses by 250 (baseline of 1,207 AAPI students)

2. Increase associate degree attainment by 150 more AAPI students (baseline of 49 AAPI students)

3. Increase transfer to a four-year institution by 100 more AAPI students (baseline of 27 AAPI students)

To achieve these goals, CCC developed thirteen unique activity components. Ten of these programs were developed for students, while three targeted college staff and faculty. Specific activity components include a supplemental instruction program where students receive extra tutoring and assistance with math and science courses, a volunteer faculty and staff mentoring program that fosters civic engagement as well as creates a campus community and provides an avenue for AAPI students to receive answers to all CCC-related questions. An unintended, yet positive, outcome of the peer-mentoring program is a revival of active participation in student government. This is particularly compelling as the majority of CCC’s students utilize the distance learning program or commute to campus for their classes. A new program, still in the developmental stages, is to create a culturally sensitive college success course for students.

Although the majority of these student programs are not explicitly targeted at specific AAPI ethnic groups, our findings indicate that with access to disaggregated data, CCC had the option to narrowly identify which student populations are facing the greatest educational barriers and, further, to target specific interventions that could respond to those students’ needs. CCC’s usage of disaggregated ethnic data, on multiple levels, informs the educational philosophy of the staff and faculty. In fact, one of CCC’s goals to fulfill their campus mission is to “improve its collection, analysis, and use of data to enhance the teaching, learning, and institutional effectiveness resulting in increased student success.”[xxv] It is clear that CCC, including the AANAPISI staff, rely on a culture of inquiry and data-use to understand the demographics of their student population as well as the surrounding community, and work to ensure their programs best meet the needs of their AAPI students. This sentiment was reaffirmed by CCC’s Title III activity coordinator, as he stated, “I believe it is very critical to the nature of what we are trying to do that we use disaggregated data to make immediate, informed decisions which allow us to redirect or refocus activities and services.”

CCC demonstrates the use of this data to target ethnic-specific student needs in its programs and services for its Vietnamese students. For example, the Title III project facilitator at CCC recognized that “the new immigrant issues are very important for the Vietnamese population. When we wrote the proposal, our number one objective was helping that population transition.” This analysis was based on his previously stated knowledge that Vietnamese students represent nearly 75 percent of the Asian population. Accordingly, CCC utilized the seven-level English as a second language (ESL) program that had been initiated before the grant began and implemented the Student Success Centers, which provide free embedded, online, and on-site tutoring through each of CCC’s centers. According to the project facilitator, these efforts “succeeded and went way beyond our objective for transitioning students from the highest-level ESL to regular college transfer courses.” In this example, disaggregated data informed CCC of the academic challenge and the population that faced that barrier; CCC administrators and staff utilized the data to implement programmatic change resulting in improved student outcomes (see Figure 1).

These efforts have shown promising signs of impact and increased student success. One student shared his experience with the Student Success Program in the quarterly CCC AANAPISI newsletter. He writes about his tutor, “Donna was instrumental in my success in this course, and I would like anyone who takes this course to have the same opportunity to succeed as I have.” More concretely, Figure 2 demonstrates the gains CCC has experienced in transitioning students to “degree-applicable courses” (see Figure 2).

CCC has also experienced success with its second objective focused on increasing the number of associate degrees awarded to AAPI students (see Figure 3).

In addition to the focus on ESL to college-level courses and increasing the number of associate degrees awarded, CCC has taken measures to be more responsive to and inclusive of its large Vietnamese population by considering language barriers. In response to this need, the CCC AANAPISI grant partly funded the creation of a fast-track program brochure in Vietnamese, citing its efforts to better serve its largely Vietnamese immigrant students. CCC has also offered a faculty-led “Culturally Responsive Customer Service Workshop” for staff to gain communication skills to more effectively speak with English Language Learner students. This presentation included a guide to pronouncing Vietnamese first names, and several pieces of literature from the White House Initiative on Asian American and Pacific Islanders (WHIAAPI), Asian and Pacific Islander American Scholarship Fund (APIASF), and CARE. Staff also received a presentation about the cultural nuances of the Vietnamese community, which included pedagogical traditions within Vietnamese education and examples of Vietnamese student behavior and reactions, particularly among new immigrants and refugees. Informed by its access to disaggregated AAPI data, specifically the large Vietnamese student population, CCC has effectively put this data into practice.

The institutional data used at CCC, as evidenced by its data-driven program implementation and evaluation processes, is a valuable tool for examining the impact of its AANAPISI programs and services, as well as for capturing those student populations that need more attention and resources. CCC demonstrates an explicit culture of data use for strategizing AANAPISI program proposals, assessing the outcomes of those interventions, and developing future changes based on those results. CCC is but one example of several institutions across the nation that have collected and utilized granular ethnic data to effectively target resources and improve educational outcomes for their underserved AAPI students. The University of Hawaii System and the University of Guam are two other models.[xxvi]

In each of these cases, the institutions had the support of campus and/or system administrators that applied the needed pressure and support to generate political will. CCC, for example, is one of 112 community colleges in the California Community Colleges system (CCCS), the largest system of higher education in the United States. CCCS collects data on nine Asian American and four Pacific Islander ethnic categories and allows students to select multiple categories that reflect their identity. In their response to the WHIAPPI request for information on data methodologies, CCCS stated, “There aren’t any hurdles to disaggregating the groups since we have been collecting the separate Asian and Pacific Islander categories for twenty years.” Given this, CCC has a strong culture of inquiry in which to follow as it is nested within a system that also collects disaggregated data. A system of support, then, is key to addressing the unique needs of the most underserved students on campus through the use of data that best represents their educational experiences.

This case study sheds light on the benefits of having access to disaggregated data on college campuses and advances the AAPI data quality campaign as it spotlights the effectiveness of employing disaggregated data to impact student outcomes. While it is an excellent example of the positive work that can be achieved through using granular ethnic data, it must be noted that CCC is part of a system that has established a culture of collecting disaggregated data. This is not the case for many other institutions that are in need of data methodological changes. Therefore, policy makers play a critical role in the AAPI data quality movement.

Conclusion

Although AAPI scholars have recognized and cited the need for more accurate data for more than thirty years, it remains a bullet point that has been relegated to the recommendation or conclusion sections of AAPI scholarship. The stagnation between federal and institutional change must be overcome by using the point of intersection where policy and practice meet as a site for advancing this critical issue. Therefore, research that highlights the benefits of data disaggregation and its tangible impact on student outcomes are key to building rationale and promoting progress. Although seemingly sparse, the districts, systems, and institutions that currently collect and use disaggregated data are remarkable sites of innovation that should be used as models of success for future data policy.

In collaboration with the ongoing efforts of CARE, WHIAPPI, CAPAC, and other Congressional leaders, education and AAPI policy makers can support the advancement of this critical issue in three ways:

1. Advocate for the implementation of federal, state, and local policies that requires the collection and reporting of disaggregated ethnic data

2. Support efforts that make federal data sets more consistent across collection and reporting categories

3. Provide institutional supports for data disaggregation in the form of incentives and technical assistance

These three recommendations provide a necessary foundation to support the AAPI data quality campaign and to advance the call for data disaggregation.

AAPI students, who are a heterogeneous population that vary widely on the educational attainment pipeline, face educational barriers that are being masked by the common perception that they achieve universal academic success. Misrepresentation through inaccurate data is arguably the greatest, and certainly the most persistent, barrier for this group of students. As the fastest-growing ethnic group in the nation, education policy must find ways in which to respond to the unique needs of AAPIs and to serve those subgroups that are most in need of attention and resources. Aggregate data that masks the educational realities of various AAPI subgroups not only discounts their needs, but also fails to realize this growing population’s full potential. Data disaggregation is the first step for addressing this issue, but it is necessary to move beyond a research recommendation. In order to directly address the needs of underserved AAPI students across the nation, data disaggregation must be adopted as both a policy and practice.