background image
Accurate and economical targeting is one thing, assessing program effective-
ness quite another. Showing that most people who use a particular program do not
become affected by the condition the program is supposed to alleviate is insuffi-
cient: they might not have been affected in any case; the condition merely may
have been postponed, but not averted; or the ranks of those in need may simply
have been reshuffled (those allowed to "jump the queue" simply push back others
in line).
Finally, programs that focus on preventing new cases of something are said to
do primary prevention. Those that concentrate on the early identification and treat-
ment of current cases do secondary prevention. Secondary prevention efforts may
reduce the prevalence of a condition (total number affected at any time), but they
do not reduce the number of new cases.
Conceptual and Methodological Problems in Preventing Homelessness
Markers and Realities
For purposes of this review, people are homeless when they live without con-
ventional housing or take up residence in shelters. People are "at risk" of homeless-
ness when they have lost security of tenure in any residential setting, whether a
household or an institution. Typically, homelessness prevention programs are con-
cerned with preventing shelter entry, a criterion that is amenable to relatively easy
measurement and encompasses a major public cost even if it fails to capture private
burden. This definition is in accord with the conventions of the federal govern-
ment, most survey researchers (see Burt, 1996), and most of the programs we
review here.
Still, consider some important questions begged by this definition and how
they bear on assessment of prevention. The size of the shelter population is largely
driven by available beds and access rules (admitting criteria, limits on length of
stay, restrictions on freedom, etc.; see, e.g., Culhane, Lee, & Wachter, 1996). If a
shelter turns applicants away or evicts residents after some period, the official tally
of homeless people may be lowered. But it is not clear that those refused access or
put out are better off even if they do not end up on the street. Has homelessness
been prevented if those denied shelter find some arrangement--no matter how
makeshift--short of literal homelessness? Deterrence raises similar problems: if
officials intentionally make entry into a shelter so costly (in terms of eroded dignity
or cramped liberty) that people who would otherwise apply elect instead to stay in
overcrowded or deficient housing, has homelessness been prevented?
2
98
Shinn, Baumohl, and Hopper
2
In 1985, worried that hotel rooms drew people out of substandard housing and into the shelter sys-
tem, New York City made congregate shelters--where scores of families lived in a single, large room
with rows of cots--the entry point to the shelter system for families. Said New York's mayor at the time,