Mapping Neighborhoods' Mental Health Vulnerability: Integration of Theory and Self-organizing Map.

Authors: Donghee Koh*, University of Tennessee
Topics: Medical and Health Geography, Human-Environment Geography, Urban Geography
Keywords: neighborhood, mental health, vulnerability, geo-visualization, self-organizing map, Los Angeles
Session Type: Paper
Day: 4/9/2020
Start / End Time: 1:45 PM / 3:00 PM
Room: Tower Court A, Sheraton, IM Pei Tower, Second Floor Level
Presentation File: No File Uploaded

Using Los Angeles County as a case study, this research aims to (1) develop neighborhood-level mental health vulnerability typologies and (2) examine their spatial distribution and its associations to racial/ethnic residential segregation patterns. The whole process of developing mental health vulnerability typologies was driven by consultations with the existing theoretical frameworks. A total of 16 variables was identified, and they can be divided into five neighborhood domains: socioeconomic structure, residential stability, disorder, physical environmental quality, and pollution burden. Using a Self-organizing map (SOM), an approach combining computational and geo-visualization techniques, the component planes were derived which were then used to systematically examine relationships among the variables. Next, a k-means clustering algorithm was employed to segment the multi-dimensional SOM output space into sub-typologies that neatly capture the neighborhood-level mental health vulnerability. Finally, the results were mapped in order to examine their spatial distribution patterns. By and large, the findings indicated that centrally-located and minority-concentrated neighborhoods tend to display a higher level of mental health vulnerability compared to suburban White-dominant neighborhoods. It was also found that there exist interesting partial and non-linear relationships among the variables. The findings of this study, we believe, offer useful insights that can help local governments devise more effective and efficient intervention plans.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login