Exploring pathways to independence: A data mining study to research predictors of long-term stay among homeless men in the New York City family shelter system
Item
-
Title
-
Exploring pathways to independence: A data mining study to research predictors of long-term stay among homeless men in the New York City family shelter system
-
Identifier
-
d_2009_2013:e443eef1be86:10505
-
identifier
-
10608
-
Creator
-
Rodriguez, Louis,
-
Contributor
-
Irwin Epstein
-
Date
-
2010
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Social work | Public policy | cox regression analysis | data mining | homeless men | long-term stay | survival analysis
-
Abstract
-
This research used a clinical data mining study to examine long-term stays of 811 homeless men in the New York City family shelter system. The overall goal of the project was to examine what predictors influenced long-term stay defined as more than 291 days in shelter. Survival analysis was used to measure how long it took for men to discharge from shelter. Cox regression analysis was used to assess whether predictor variables influenced length of stay of the men in the sample. Data was collected from administrative records. There were several key findings. Discharge patterns among the men slowed after 400 days in shelter. Exit disposition, age and family size were among the best predictors of long-term stay. Old men took longer time to discharge from shelter as compared to young men in the sample. Homeless men in large families also took longer to discharge from shelter as compared to men in small families. Efforts should be made to accommodate the service needs of large families. These families, identified at intake, need more support than others in finding housing, completing applications for housing, and minimizing barriers to relocation from shelter.
-
Type
-
dissertation
-
Source
-
2009_2013.csv
-
degree
-
Ph.D.
-
Program
-
Social Welfare
