Document Type


Publication Date



Background and Purpose: Approximately three million refugees have arrived in the United States in the past four decades. Literature suggests, after arrival in the country and in the short-term period thereafter, it is likely for refugees as New Americans to experience poverty and deprivation in different aspects of life. In this context, knowledge of factors associated with poverty among refugees is important and relevant for social workers as front-line service providers for refugees and economically disadvantaged individuals. Using social exclusion theory as a framework, this study hypothesized that the factors associated with integration in the refugee integration conceptual framework (Ager & Strang, 2008) would be associated with poverty among refugees. To test this hypothesis, bivariate and multivariate correlational analyses were conducted between poverty and English language proficiency, length of residency, and application for permanent residency among refugees. To provide a more comprehensive picture of poverty and in line with the “markers and means” indicators in Ager and Strang (2008) framework, refugee poverty was quantified using households’ income (monetary approach to poverty) and household level deprivation in three domains: education, health, and housing (capability approach to poverty).

Methods: The 2016 Annual Survey of Refugees (ASR) dataset was used in this study. The 2016 ASR was collected through structured phone interviews with 1,500 principal principle applicants for refugee status who arrived in the United States between the 2011 and 2015 fiscal years. Using Stata software program (version 15), Pearson’s chi-square test was conducted to measure bivariate correlations, Cramer’s V test was used to measure the strength of the bivariate correlations, and logistic regression was used for multivariate analysis. Results: Results showed statistically significant, negative, and weak correlations between income poverty and English language proficiency (χ2[1]= 7.40, p= 0.000, Cramer’s V= .10) as well as income poverty and length of residency in the United States (χ2[2]= 8.10, p= 0.02, Cramer’s V= .07). The correlation between multidimensional poverty and English language proficiency was statistically significant, negative, and strong (χ2[1]= 178.13, p= 0.000, Cramer’s V= .35). Moreover, the correlation between multidimensional poverty and application for permanent residency was statistically significant, negative, and weak (χ2[1]= 16.59, p= 0.000, Cramer’s V= .07). Multivariate models showed that lack of English language proficiency is the best predictor for income (Odds Ratio=1.5, p= 0.002) and multidimensional poverty (Odds Ratio= 4.4, p= 0.000).

Conclusions and Implications: These findings call for attention to English language training among refugees. Although English language classes are among the first services suggested for refugees upon arrival to the country, many cannot benefit from them due to lack of access to transportation, childcare, and long hours of work. Online or in-home English tutoring programs may be more practical for some refugees and investment is needed to enhance the scale of these programs and provide the required tools such as computers, smart phones, and Internet access for refugees to benefit from them. Additionally, access to English language classes could also be provided and promoted while refugees are in host countries waiting to be resettled in the United States.


Presented at the 2021 Society for Social Work and Research (SSWR) Virtual Conference.

Persistent Identifier

Included in

Social Work Commons