Factors Associated with Economic Integration of Refugee Households in the United States

Background and Purpose: Refugees are one of the most at-risk groups to experience poverty and economic exclusion in host countries due to the often abrupt and unplanned nature of their forced displacement. Social workers as front-line service providers need to have knowledge and understanding about the concept of poverty and economic integration among refugees. In this context, this study aimed to calculate poverty and factors associated with this problem among refugees in the U.S. to provide recommendations for refugees’ economic integration. Methods: Using the 2018 Annual Survey of Refugees (ASR) dataset and the monetary (Haughton & Khandker, 2009) approach to poverty we calculated households' income poverty by comparing their income with national poverty lines. Moreover, we used the capability (Laderchi et al., 2003) approach to poverty and an adjusted version of the global Multidimensional Poverty Index (aMPI; Alkire & Santos, 2010) to calculate households’ multidimensional poverty in three domains of health, education, and housing. To explore factors associated with poverty we used the multidimensional poverty framework proposed by Naseh et al (2020) and conducted bivariate and multivariate correlational analyses between poverty (income and multidimensional poverty) and households’ demographic characteristics (including biological sex, age, and ethnicity of the head of the household) and non-economic aspects of their adaptation (including household members’ English language proficiency at the time of arrival in the U.S. and at the time of data collection). Results: Around 21% (n= 324) of the surveyed households were income poor and around 79% (n=1,198) were multidimensionally poor. Results showed statistically significant, weak to medium, and positive correlations between households’ lack of English language proficiency (ELP) and poverty (income poverty & lack of ELP upon arrival: χ2[1]= 17.64, p= 0.000, Cramer’s V= .11; income poverty & lack of ELP at the time of interview: χ2[1]= 9.89, p= 0.002, Cramer’s V= .09; multidimensional poverty & lack of ELP upon arrival: χ2[1]= 100.98, p= 0.000, Cramer’s V= .23; and multidimensional poverty & lack of ELP at the time of interview: χ2[1]= 87.34, p= 0.000, Cramer’s V= .26). Moreover, correlations between poverty and demographic characteristics of the head of the households (age and ethnicity) were statistically significant and weak to medium (income poverty & age: χ2[6]= 22.79, p= 0.001, Cramer’s V= .12; multidimensional poverty & age: χ2[6]= 13.77, p= 0.032, Cramer’s V= .10; income poverty & ethnicity: χ2[11]= 32.60, p= 0.001, Cramer’s V= .15; and multidimensional poverty & ethnicity: χ2[11]= 88.35, p= 0.000, Cramer’s V= .24). Multivariate models of the study showed that that lack of ELP at the time of arrival was the best predictor for income poverty (Odds Ratio=1.98, p= 0.004), and lack of ELP at the time of the interview was the best predictor for multidimensional poverty (Odds Ratio= 2.7, p= 0.000). Conclusions and Implications: Findings of the study call for further attention to the importance of English language training for refugees. Access to English language classes for refugees could also be provided and promoted.