A Crowdsourced Homelessness Data Collection System and Its Econometric Analysis: Strengthening Inclusive Public Administration Policies

Authors: Praniil Nagaraj

Abstract:

This paper proposes a method to collect homelessness data using crowdsourcing and presents an approach to analyze the data, demonstrating its potential to strengthen existing and future policies aimed at promoting socio-economic equilibrium. This paper's contributions can be categorized into three main areas. Firstly, a method for collecting homelessness data is presented, utilizing a user-friendly smartphone app (currently available for Android). The app enables the general public to quickly record information about homeless individuals, including the number of people and details about their living conditions. The collected data, including date, time, and location, are anonymized and securely transmitted to the cloud. It is anticipated that an increasing number of users motivated to contribute to society will adopt the app, thus expanding the data collection efforts. Duplicate data are addressed through simple classification methods, and historical data are utilized to fill in missing information. The second contribution of this paper is the description of data analysis techniques applied to the collected data. By combining these new data with existing information, statistical regression analysis is employed to gain insights into various aspects, such as distinguishing between unsheltered and sheltered homeless populations, as well as examining their correlation with factors like unemployment rates, housing affordability, and labor demand. Initial data are collected in San Francisco, while pre-existing information is drawn from three cities: San Francisco, New York City, and Washington D.C., facilitating the conduction of simulations. The third contribution focuses on demonstrating the practical implications of the data processing results. The challenges faced by key stakeholders, including charitable organizations and local city governments, are taken into consideration. Two case studies are presented as examples. The first case study explores improving the efficiency of food and necessities distribution, as well as medical assistance, driven by charitable organizations. The second case study examines the correlation between micro-geographic budget expenditure by local city governments and homelessness information to justify budget allocation and expenditures. The ultimate objective of this endeavor is to enable the continuous enhancement of the quality of life for the underprivileged. It is hoped that through increased crowdsourcing of data from the public, the Generosity Curve and the Need Curve will intersect, leading to a better world for all.

Keywords: Crowdsourcing, homelessness, socio-economic policies, statistical analysis.

References:

[1] The 2023 Annual Homeless Assessment Report (AHAR) to Congress. The U.S. Department of Housing and Urban Development 2023. https://www.huduser.gov/portal/datasets/ahar/2023-ahar-part-1-pit-estimates-of-homelessness-in-the-us.html.
[2] San Francisco Homeless Count and Survey, 2022 https://hsh.sfgov.org/wp-content/uploads/2022/08/2022-PIT-Count-Report-San-Francisco-Updated-8.19.22.pdf
[3] Willison, C., Unwala, N., Singer, P. M., Creedon, T. B., Mullin, B., & Cook, B. L. (2024). Persistent Disparities: Trends in Rates of Sheltered Homelessness Across Demographic Subgroups in the USA. Journal of racial and ethnic health disparities, 11(1), 326–338. https://doi.org/10.1007/s40615-023-01521-9
[4] Johnson, G., Culhane, D., Fitzpatrick, S., Metraux, S., & O’Sullivan, E. (Eds.). (2024). Research Handbook on Homelessness. Cheltenham, UK: Edward Elgar Publishing. https://doi.org/10.4337/9781800883413
[5] HUD Office of Policy Development and Research. (2024). Quantitative and Qualitative Analyses of Unsheltered Homelessness at the Community Level. Accessed Sep 2024. https://www.hud.gov/programdescription/pdr
[6] Yoder Clark A, Blumenfeld N, Lal E, Darbari S, Northwood S, Wadpey A. Using K-Means Cluster Analysis and Decision Trees to Highlight Significant Factors Leading to Homelessness. Mathematics. 2021; 9(17):2045. https://doi.org/10.3390/math9172045
[7] Gov.UK, (2019) Review of Models of Homelessness (2019) https://assets.publishing.service.gov.uk/media/5c989f23e5274a3ca4216021/Homelessness_Models.pdf
[8] Segal MR, Bloch DA. A comparison of estimated proportional hazards models and regression trees. Stat Med. 1989 May;8(5):539-50. doi: 10.1002/sim.4780080503. PMID: 2727474.
[9] Byrne, T., Montgomery, A. E., & Fargo, J. D. (2019). Predictive modeling of housing instability and homelessness in the Veterans Health Administration. Health services research, 54(1), 75–85. https://doi.org/10.1111/1475-6773.13050
[10] Kowarsch, D., Yang, Z. (2022). A System Dynamics Approach on Modeling Homeless Prevention Strategy: A Case Study of LA County. In: Tareq Ahram, Waldemar Karwowski, Pepetto Di Bucchianico, Redha Taiar, Luca Casarotto and Pietro Costa (eds) Intelligent Human Systems Integration (IHSI 2022): Integrating People and Intelligent Systems. AHFE (2022) International Conference. AHFE Open Access, vol 22. AHFE International, USA. http://doi.org/10.54941/ahfe100986