Library Digital Collections

Data from: Overcrowded Housing Reduces COVID-19 Mitigation Measures and Lowers Emotional Health Among San Diego Refugees from September to November of 2020

View Collection Items

Collections »

Data from: Overcrowded Housing Reduces COVID-19 Mitigation Measures and Lowers Emotional Health Among San Diego Refugees from September to November of 2020

About this collection


1 digital object.

Cite This Work

Hassani, Ashkan; Omaleki, Vinton; Erikat, Jeanine; Frost, Elizabeth; Streuli, Samantha; Sahid, Ramla; Yusufi, Homayra; Fielding-Miller, Rebecca (2023). Data from: Overcrowded Housing Reduces COVID-19 Mitigation Measures and Lowers Emotional Health Among San Diego Refugees from September to November of 2020. UC San Diego Library Digital Collections.


Study design and setting:

The Partnership for the Advancement of New Americans (PANA) is a research, public policy, and community organizing hub that serves refugees and asylum seekers in San Diego County. Data for the present cross-sectional study were collected between September 2020 and November 2020 as part of PANA’s biannual community survey. The survey was designed in English and translated into Arabic and Spanish by bilingual study staff. Interviewers were provided with a list of PANA members and the survey was subsequently administered over the phone and in person by trained research assistants in Arabic, Burmese, Dari, English, Karenni, Oromo, Pashto, Somali, Spanish, and Swahili. Up to two household members could be interviewed. However, interviewers tracked whether an individual had completed a survey or not. Data were collected as part of a programmatic report intended to provide a general overview of the community. As such, we did not conduct a priori power calculations focused on one specific outcome.

Interviewers entered participant answers using Qualtrics software (Qualtrics ver. 2021). The survey contained 83 questions regarding demographics, COVID-19, housing, employment, health, children, belonging, and resilience. The survey took approximately 30 minutes to complete and up to 1 hour if it was being translated. The study team met weekly to discuss any issues with translation and address questions raised by survey administrators. Consistent with our participatory action approach, all data collection and analyses were conducted in conversation with PANA staff and guided by their lived expertise and organizational policy priorities. Variables for the present analyses were chosen in consultation with PANA leadership to address the primary research question: What is the association between housing and COVID-19 vulnerability? Vulnerability was broadly understood as both healthcare access and the broader mental health impact of the epidemic. The full report with all variables can be found elsewhere.

Date Collected
  • 2020-09 to 2020-11
Date Issued
  • 2023
Principal Investigator
Co Principal Investigators


Participants were recruited using a convenience sampling approach. Individuals were eligible to participate in the study if they were over the age of 18, could speak one of the languages included in the study, willing to provide informed consent, and part of the refugee community. “Refugee community member” was construed broadly and included individuals who had arrived as refugees, asylum seekers, or their American-born children. PANA staff contacted all individuals for whom the organization had provided services or with whom they had engaged in community organizing efforts in the previous 5 years. Each staff interviewer was assigned a list of people based on language. Research assistants also recruited participants in person in shopping areas that members of the refugee community typically frequent. Staff interviewed and conducted the survey by phone with each individual from the convenience sample - ensuring no duplication. Interviews were limited to no more than two individuals per household (typically a parent and an adult child). All participants who completed the survey were given a $20 gift card to thank them for their time and expertise.

Measures and analyses:

Within the survey, participants were asked about the total number of individuals living in their homes and the total number of rooms in the home. In accordance with the California Department of Public Health (CDPH), overcrowding was defined as more than one individual per room. Severe overcrowding was defined as 1.5 individuals per room. For example, a traditional 2-bedroom apartment would have 4 distinct rooms, including the kitchen and living room, and it would require 6 inhabitants to be classified as “severe overcrowding.” Other explanatory variables included access to affordable housing (Are you in an affordable housing unit, or receiving a section 8 voucher to help with the rent?), access to a diagnostic test for COVID-19 (Have you ever gotten a test for COVID-19?), and self-rated emotional health with 5 potential responses ranging from “never” to “always” (How often have you been bothered by emotional problems such as feeling anxious, depressed or irritable?). In consultation with PANA leadership, the study team decided to dichotomize the emotional health variable for analysis and compare participants who reported ‘never’ experiencing emotional problems vs all others. We also included demographic variables such as income from the previous week (Have you worked for money in the previous week), age, year of arrival to the United States, gender (male or female options only), cohabiting with a partner, number of children (How many children under the age of 18 are you responsible for?), and family size.

We first examined basic univariate frequencies as a study team to develop a basic understanding of sample demographics and the prevalence of indicators of interest within the sample. We then used chi-square and student’s t-tests to test the hypothesis that our primary predictor and covariates of interest were significantly associated with living in severely overcrowded housing. We built simple logistic regression models to measure the unadjusted odds of association with overcrowded housing. Descriptive and bivariate statistical analyses were conducted using Stata 16.

After assessing bivariate associates for both statistical and theoretical significance (the latter based on both the literature and discussions with the PANA leadership team) we constructed a structural equation model based on our hypothesized pathways between crowded housing, access to affordable housing support, reported emotional health, and the likelihood of accessing a COVID-19 test in the summer and fall of 2020. The model was fit to the data with the Mplus software package, using a weighted least squares (WSMLV) estimator with probit link function to account for the mixture of binary, ordinal, and continuous variables. Model fit was assessed based on individual covariate statistical significance along with global fit indices as recommended by Kline.

Ethical considerations and IRB:

Informed Consent was verbally obtained from all participants by trained PANA staff before administering the Qualtrics survey. Data gathered from the questionnaires was confidential. This study was reviewed and approved by the University of California, San Diego Institutional Review Board (UCSD IRB) under project #201601SX. In consideration of the vulnerable status of our study population, we adopted a participatory action approach to this community-led project and our team from UCSD only provided technical assistance.


Partnership for the Advancement of New Americans



View formats within this collection

  • English

Identifier: Ashkan Hassani:

Identifier: Rebecca Fielding-Miller:

Identifier: Samantha Streuli:

Identifier: Vinton Omaleki:

Related Resource

    Primary associated publication