Estimating census and survey undercounts through multiple service contacts

Abstract

During the past decade a tremendous amount of work has been done on studying the homeless. The 1980s saw many disconnected city‐specific counts of the numbers of homeless that sought to obtain some information on their characteristics. As these studies began to be published, there was a growing awareness of the inadequacy of the methods available and the variety of interests that prompted the studies. The 1980s also saw the growth of new methods and models for counting and the coordination of efforts for multidisciplinary studies. This paper describes the development of new models for counting and new methods of sampling in time and space.