![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
Inspired by the social vulnerability paradigm employed in hazard and disaster research and recent work connecting personal and housing vulnerabilities, this study uses the first wave of Los Angeles Family and Neighborhood Survey data to: (a) examine immigrants’ legal status as an independent social vulnerability that increases the risk of two or more of the following situations deemed to be precarious: renting, crowding, and unaffordable housing; (b) identify the individual-, household-, and neighborhood-level vulnerabilities associated with overlapping housing problems; and (c) identify the distribution of housing disadvantages across social groups. The sample comprises those born in the United States who are Black, White, and Latino, and three distinct Latino immigrant groups categorized by citizenship and legal status. The descriptive and multivariate regression results have implications for expanding hazard, disaster, and housing research and practice.
Acknowledgments
This research was supported by Grant R03 HD058915 from the National Institute of Child Health and Development, National Institutes of Health. An earlier version of the manuscript was presented at the 2015 Population Association of America annual meeting. The author thanks Rolf Pendall, Jeffrey Passel, Tate Desper, the Editor, and three anonymous reviewers for their suggestions on earlier versions of the manuscript. All errors of fact and interpretation are the sole responsibility of the author.
Notes
1. In contrast, the dominant paradigm in hazard/disaster research emphasizes that societies are at the mercy of nature and little can be done to shape risks to and impacts of natural hazards (Fordham, Lovekamp, Thomas, & Phillips, Citation2013).
2. See Fordham and colleagues (Citation2013) for differing perspectives about how vulnerability and resilience are connected.
3. Overlaps are particularly prevalent among groups such as African Americans, who, in part because of U.S. racial history and racialized social policies, are likely to be low income or poor and to live in poorer quality and highly residentially segregated neighborhoods than Whites are (e.g., Charles, Citation2006; Massey & Denton, Citation1993).
4. Although immigrants’ citizenship and legal status are typically treated as individual-level characteristics, immigration policy context and federal-level policies towards different groups delineate immigrants’ eligibility for citizenship and authorized means of entry and U.S. residence (e.g., Menjívar, Citation2011).
5. Under President Obama’s Deferred Action for Childhood Arrivals program, by September 2013, half a million unauthorized immigrant youth were approved for short-term relief from deportation and work permits, but given no path to U.S. citizenship (U.S. Citizenship & Immigration Services, Citation2014).
6. Unfortunately, the data used in the present study do not include information about other housing vulnerabilities that increase exposure to hazard risk, such as residing in mobile homes, or age of the housing unit (e.g., Morrow, Citation1999; Pendall et al., Citation2012).
7. The same could be said about crowding and housing affordability. For example, U.S. crowding standards are not objective metrics applicable to all groups in the country or in other parts of the world (e.g., Evans, Leopore, & Allen, Citation2000; Pader, Citation2002).
8. Previous analyses of L.A.FANS data use different samples, sets of covariates, and operationalizations of variables. For example, a prior study of housing affordability uses a low-income sample of respondents and fewer migration-related characteristics, and does not contrast Latino immigrants on the basis of both citizenship and legal status (i.e., naturalized citizens, authorized noncitizens, and unauthorized noncitizens; McConnell, Citation2013).
9. This study includes adults who filled out the adult module as the RSA or, in households with children under 18, as the RSA or the primary care giver of a randomly selected child. Respondents from a second nuclear family residing in the home were excluded, because of concerns about their correlated errors with the first family and the focus of this study on outcomes measured at the household level (e.g., rent, crowding, housing affordability). Respondents reporting that they spent 100% or more of their family income on housing costs were excluded, because of concerns about the quality of their housing cost or income data.
10. Results, shown later, indicate that no unauthorized noncitizen immigrants have zero precarious housing conditions. A binary dependent variable is preferable in this case to an outcome variable with a 0 to 3 distribution. Ancillary analyses using this latter operationalization, not shown, suggest the same substantive conclusions as reported here.
11. The mean variance inflation factor is below 2.2 for every model, suggesting that multicollinearity is not a serious problem biasing the results (Menard, Citation1995).
12. L.A.FANS did not ask renters or homeowners about utility or other housing-related expenditures. Data from an imputed file created by L.A.FANS staff (Bitler & Peterson, Citation2004) are used when there is missing information about income, housing costs, or the home value of owned homes. See McConnell (Citation2013) for the estimation of property taxes and homeowners’ insurance premiums for some homeowner respondents.
13. Ancillary analyses, not shown, indicate that nearly all respondents who had moved recently came from elsewhere in California.
14. Latinos also are hypersegregated from non-Hispanic Whites in Los Angeles (e. g., Wilkes & Iceland, Citation2004); however, census tract variables tapping into percentage Latino are too highly correlated with those about poverty and other variables to be included in the specifications. The indicator representing the concentration of African Americans is not highly correlated with other covariates, and is used to tap into the racial/ethnic composition of the neighborhood.
15. The author thanks an anonymous reviewer for suggesting the inclusion of this and other contextual variables.
16. The p value is .060 (not shown), just missing the standard .05 level of statistical significance.
17. p value of 0.159, not shown.
18. Ancillary analyses, not shown, including only the main effects of citizenship and legal status and the migration characteristics reveal that years in the U.S. is significant, suggesting that this relationship with the outcome operates via covariates in the fully specified model. Logistic regression analyses using alternative operationalizations of length of U.S. residence, such as a binary indicator of immigrant arrived since 1990, show the same lack of significance as reported results.