摘要：Objectives. We sought to increase the accuracy of New York City’s estimates of its unsheltered homeless population. Methods. We employed 2 approaches to increasing count accuracy: a plant-capture strategy in which embedded decoys (or “plants”) were used to estimate the proportion of visible homeless people missed by enumerators and a postcount survey of service users designed to estimate the proportion of unsheltered homeless people who were not visible. Results. Plants at 17 sites (29%) reported being missed in the count, because counters either did not visit those sites or did not interview the plants. Of 293 homeless service users who were not in shelters, 31% to 41% were in locations deemed not visible to counters. Conclusions. Both plant-capture estimation and postcount surveys are feasible approaches that can increase the accuracy of estimates of unsheltered homeless populations. Communities are required to include estimates of local homeless populations in applications for US Department of Housing and Urban Development (HUD) Continuum of Care funding for programs responding to homelessness. Typically, few problems are involved in counting those individuals who are sheltered each night; methods to ensure that counts are not duplicated over time are also available. However, estimating numbers of homeless people living on the street, in parks, or in “unconventional” housing is another matter. The 2 strategies described here were used to assist New York City’s ongoing efforts to improve methods of estimating the size of its unsheltered homeless population. HUD endorses 2 methods of obtaining point-in-time counts of unsheltered homeless people: directly counting people in public places or screening those using selected services to determine whether they are homeless and without shelter. 1 Counts of visibly homeless individuals miss unsheltered people who remain out of sight during the counts. 2 , 3 They also depend on enumerators determining which individuals should be counted as homeless, and such judgments, whether made through observations or interviews, are subject to a host of inaccuracies. Conversely, many people using such services as soup kitchens are not homeless. As a result, surveys of their users must determine whether the arrangements of these individuals on the night in question meet operational definitions of homelessness. 4 For purposes of its annual street count, New York’s Department of Homeless Services divides the entire city, including transportation hubs and the subway system, into small zones (i.e., into a few contiguous blocks or a subway station). The department relies on service providers and police (as well as the experience gained during counts conducted in previous years) to classify each zone as high or low density, reflecting whether homeless people are likely to be found there in the middle of the night. Volunteer counting teams are sent to all high-density zones and to a random sample of low-density zones between midnight and 4 am on a single night in late February. Counting teams are instructed to interview all people encountered awake in their zones, to determine whether they are homeless and without shelter, and to count those found asleep. The final estimate of the number of unsheltered homeless people is the sum of those actually counted in high-density zones and a statistical extrapolation from those counted in the random sample of low-density zones. Street counts were conducted in Manhattan and some subways in 2003, expanded to include Brooklyn and Staten Island in 2004, and extended to all 5 boroughs in 2005. As part of the ongoing efforts of the Department of Homeless Services to improve the counts, the department invited independent researchers, with federal funding from HUD, to address 2 potential limitations of the street-count methodology: first, that volunteer teams are unlikely to find and count all visibly unsheltered people on their assigned circuits and, second, that some unsheltered homeless people are likely to be in places that would be missed during the Homeless Outreach Population Estimate (HOPE) survey. We conducted 2 studies to estimate the consequences of each of these limitations for the accuracy of New York City’s street count.
