The Center for Social Data Analytics Colloquium

Google map

"Big and Small Data for Improving the Measurement and Lives of People Experiencing Homelessness"

Biography: Zack W. Almquist is currently an Associate Professor in the Department of Sociology, Adjunct Associate Professor of Statistics, and Senior Data Science Fellow at the eScience Institute at the University of Washington. Before coming to UW in 2020, Prof. Almquist held positions as a Research Scientist at Facebook, Inc and as an Assistant Professor of Sociology and Statistics at the University of Minnesota. Dr. Almquist is a recipient of the American Sociological Association’s Section on Methodology’s Leo Goodman Award. He is also a recipient of the NSF’s CAREER Award and the ARO’s Young Investigator Program Award. He is currently the Editor-in-Chief of the Journal of Mathematical Sociology. His research centers on the development and application of mathematical, computational and statistical methodology to problems and theory of social networks, demography, homelessness, and environmental action and governance.

Project 1 - Traditionally, unsheltered Point in Time (PIT) Counts are the result of volunteers conducting an in-person head-count of individuals experiencing homelessness on a single night. This resource-intensive method is widely understood to be an undercount. It also fails to capture essential qualitative information about what people living unsheltered experience and need. This past spring, the King County Regional Homelessness Authority (RHA), in coordination with Professor Zack W. Almquist (University of Washington) and Lived Experience Coalition (LEC) , took a novel approach to the PIT. The RHA conducted the 2022 unsheltered PIT count as a combined qualitative interview process and quantitative survey over the course of a month. The respondent selection for both the qualitative and quantitative surveys followed a Respondent Driven Sampling (RDS) protocol. RDS provides a sampling strategy for estimating size and percentages of hard-to-reach populations which lack an administrative sampling frame. We are running a second study, including a cell-phone only sample over the month of April. Results of the ongoing study will be discussed.

Project 2 - Homelessness has short- and long-term negative impacts on both physical and mental health, and represents a significant burden to communities throughout the US. According to the Alliance to End Homelessness, Washington State has 22,000 people experiencing homelessness, of which nearly 12,000 live in King County. Within the County, the number of unsheltered people has increased by about 150% in the last 13 years. Temporary strategies used to address the homelessness crisis include permanent or nighttime congregate shelters, sanctioned encampments and parking lots, hotel rooms and small shelters. However, we currently lack objective measures on the effectiveness of each of these strategies in improving the well-being and health outcomes of people experiencing homelessness. The non-invasive assessment of daily sleep timing represents a reliable marker of a person’s well-being. We present preliminary data as a proof-of-principle of our ability to measure sleep and assess its timing and quality in unhoused people sleeping in a nighttime shelter. For the current proposal we seek to provide further evidence of our ability to monitor sleep in people experiencing homelessness in different sleeping environments, and connect sleep timing to additional objective and subjective measures of health, sleep quality, and well-being. This evidence will be critical to seek funding from federal agencies with the overarching goal of using sleep as a critical metric for the effectiveness of any intervention to address the current homelessness crisis, and as a predictor of health outcomes in affected people.

Project 3 - A spatiotemporal analysis of environmental hazard exposures and health risk to people experiencing homelessness in Seattle, Washington. We describe the environmental stressor and exposure risks to people experiencing homelessness in Seattle, Washington. We report the short- and long-term effects of environmental risk and exposures to the growing and vulnerable population experiencing homelessness. To analyze the spatiotemporal variability of encampment locations throughout the Seattle area, we use identifiable individual- and encampment-level data gathered by social workers at REACH Evergreen Treatment Services, the largest homelessness resource provider contracted by the government of Seattle-King County, Washington. Additionally, we build geostatistical models with space-time smoothing to estimate areas of greater environmental exposure potential by risk factor of interest. Preliminary results indicate tent encampment locations have concentrated in geographic areas at high risk for harmful environmental exposure over the past +6 years. Compared to the median King County deciles of concentration, tent encampment locations are increasingly located in areas with greater potential for harmful environmental exposure, including close proximity to traffic, Superfund sites, and in areas with high lead exposure, toxic industrial facilities, and toxic air release.