Journal of Official Statistics, Vol.9, No.4, 1993. pp. 747764

Current Issue
Personal Reference Library (PRL)
Personal Page

A Multiple Frame Approach to Sampling the Homeless and Transient Population

This paper describes the design of the homeless study component of the Washington, DC, Metropolitan Area Drug Study (DC*MADS). The study adopted a multiple frame approach to cover the homeless population. The frames included shelter and soup kitchen frames as well as street and encampment frames. The design also included temporal stratification with the selection of independent monthly samples of institutions, street blocks and sample days, and the random assignment of sample units to days. The paper discusses issues related to sampling homeless people, combining data across multiple overlapping frames, and the relative coverage of the different frames. It then examines the implications of these findings for future studies.

Rare mobile population; street sample; shelter sample; temporal stratification.

Copyright Statistics Sweden 1996-2018.  Open Access
ISSN 0282-423X
Created and Maintained by OKS Group