Definitional quandaries and other hazards in counting the homeless: an invited commentary.

Complexities of both definition and dynamics suggest that the unit of analysis in research on homelessness may need rethinking. Wright and Devine's corrective effort can be criticized on the grounds that it inflates the shortcomings of the admittedly flawed 1990 S-Night count, applies relative proportions from an unrepresentative local sample, and smuggles in an overly inclusive notion of homelessness.