While the Housing and Urban Development (HUD) study appears rigorous in employing four seemingly different approaches to arrive at its final estimates. The HUD Report justifies Ranally Metropolitan Areas's—as opposed to central cities or other smaller units—on the grounds that most of the homeless reside in the central city rather than the suburbs. But the Ranally Metropolitan Areas's (RMA'S) are not simply central city plus surrounding suburbs; the larger RMA's often include numerous different central cities, each one of which potentially has a homeless problem. The first approach is intended to provide a "worst case" estimate of homelessness, by accepting at face value the highest published figures that satisfy certain minimal standards. Two separate estimates of the non-sheltered homeless are obtained: one by extrapolation from the census "casual count," and one a projection of street counts conducted in Phoenix, Pittsburgh, and Boston at various times in 1983.