Achievement Gap Among Southeast Asian Refugees in the United States

• In Event: From Access to Belonging: Education Policy and Practice for Refugees and Migrants

Proposal

The displacement of millions of people due to political conflict, persecution, and economic crisis has created unprecedented waves of global migration. Among these are Southeast Asian refugees—primarily from Vietnam, Laos, and Cambodia—who arrived in the United States in large numbers following the end of the Vietnam War and the collapse of regional regimes in the late 20th century. This refugee population presents a unique opportunity to examine variation in integration outcomes, as these groups share broadly similar timelines and contexts of arrival but differ markedly in cultural background, education levels, and subsequent generational outcomes. This study investigates how cultural heritage, particularly the presence or absence of Confucian Heritage Culture (CHC), influences the educational attainment of Southeast Asian refugee populations in the United States.

Theoretical Framework and Motivation

Much of the literature on immigrant educational outcomes tends to treat Asian Americans as a monolithic group, often reinforcing the “model minority” stereotype. However, disaggregated data reveal considerable heterogeneity in educational outcomes across Asian ethnic subgroups (Sakamoto et al., 2022; Keo & Noguera, 2018). Vietnamese Americans, for example, have shown significantly higher levels of educational attainment compared to Cambodian and Laotian Americans, despite sharing the experience of forced migration during the same historical period.

To explore the underlying causes of this divergence, this study draws on Bronfenbrenner’s Ecological Systems Theory. The theory provides a multi-level framework that emphasizes the interaction between individual, familial, community, and societal factors in shaping developmental outcomes. Specifically, this research examines the role of the macrosystem (cultural norms and national traditions) and the chronosystem (timing and sequence of migration) in explaining variation in educational attainment. It also integrates micro-level variables such as family income, structure, and sibship size, as well as exosystem factors such as neighborhood effects and community social capital.

A central hypothesis of this study is that the Confucian Heritage Culture—common to countries like Vietnam, China, and Korea—imparts educational values that promote academic achievement and persistence. Previous work suggests that immigrants from Confucian-influenced societies often prioritize education, exercise parental authority over academic matters, and invest heavily in schooling as a form of upward mobility (Caplan et al., 1992; Jerrim, 2015; Holmlund et al., 2023). By contrast, Cambodian and Laotian cultures, more influenced by Indian traditions and shaped by distinct colonial and wartime experiences, may emphasize different pathways to social mobility and childrearing practices.

Empirical Strategy and Data Sources

The study treats the arrival of Southeast Asian refugees as a quasi-natural experiment, capitalizing on the exogeneity of forced migration waves to minimize confounding selection effects. Using individual-level data from the U.S. Census (1980, 1990, 2000) and the American Community Survey (2000–2023), accessed via IPUMS, the study identifies Southeast Asian refugees based on a combination of birthplace, language use at home, and self-reported ancestry. The analysis is restricted to individuals or household heads who migrated prior to 1995, the year U.S.-Vietnam relations normalized, to capture the period dominated by refugee migration rather than economic migration.

A binary treatment variable captures whether an individual belongs to a Confucian Heritage Culture, defined through ancestry (e.g., Vietnamese, Chinese) or language spoken at home (e.g., Vietnamese, Mandarin). A three-way fixed-effects regression model estimates the effects of CHC, arrival wave (Wave 1: pre-1979; Wave 2: post-1979), and generational status (first-generation, 1.5-generation, and second-generation) on educational attainment. Control variables include age, sex, birth order, family income, household head and spouse’s education, marital status of parents, and sibship size. Fixed effects by state, year, and age are incorporated to account for spatial and temporal heterogeneity.

To examine mechanisms, I complement the macro-level analysis with micro-level data from the Children of Immigrants Longitudinal Study (CILS), a dataset that includes measures of parental involvement, educational aspirations, homework time, and academic performance. These outcomes allow for a deeper exploration of behavioral and attitudinal pathways through which CHC may affect educational outcomes.

Findings

Preliminary analyses confirm that affiliation with CHC is positively associated with educational attainment. In Census data, the treatment group (CHC-associated) consistently outperformed the control group (non-CHC) across all model specifications. Even after controlling for socioeconomic status and family structure, the CHC variable remains significant, suggesting a robust cultural effect. In American Community Survey data, the treatment effect is smaller but still statistically significant.

The results also indicate significant interaction effects. First-generation immigrants in Wave 2 (1979–1995)—a group disproportionately affected by the trauma of war and extended stays in refugee camps—display markedly lower educational attainment, and this disadvantage is especially pronounced among non-CHC groups. Conversely, second-generation CHC-group individuals show substantial educational gains, highlighting the long-term cultural resilience and adaptation capacities of this group.

Mechanism analysis using the CILS dataset shows that CHC is associated with higher academic performance (e.g., math and reading scores), greater time spent on homework, and higher educational aspirations. Interestingly, CHC families report significantly lower levels of participation in school-based parental involvement activities (e.g., PTA meetings, volunteering). This suggests that cultural models of education in CHC households prioritize home-based forms of involvement and may differ from mainstream U.S. expectations of parental engagement.

Significance and Contributions

This research contributes to several strands of scholarship. First, it challenges the monolithic portrayal of Asian Americans in educational research by foregrounding subgroup heterogeneity. Second, it advances our understanding of how cultural heritage mediates integration outcomes in refugee populations. By treating culture not as a static trait but as an explanatory variable interacting with historical and socioeconomic contexts, this study offers a nuanced account of how educational inequality is produced and maintained.

Third, the use of quasi-experimental design and multi-level modeling allows for more credible inferences about causality than prior studies reliant on descriptive or correlational approaches. Fourth, the integration of Bronfenbrenner’s ecological systems theory provides a comprehensive framework for synthesizing micro-, meso-, and macro-level factors.

Finally, this study carries important policy implications. Refugee education programs often adopt one-size-fits-all models that ignore cultural diversity within racial categories. Findings from this study suggest that culturally responsive educational interventions must consider the differential legacies of historical trauma, migration timing, and heritage norms to be effective.

Author

• Thao Phuong Pham, The University of Wisconsin-Madison