Findings from a local independent report released by C. Peraro Consulting, LLC assessing the quality of the annual homeless count reported that unsheltered homelessness may be underestimated by as much as 40 to 59 percent.
The one-day street count in Oahu was reported earlier this year by the State of Hawaii as 1,475 persons.
Results from the study show a potential undercount of between 587 and 875 persons, putting the total well over 2,000 and as high as 2,350 persons.
These findings are part of a paper presenting measures of quality that can be implemented by communities to produce more accurate homeless counts. Currently, the U.S. Department of Housing and Urban Development requires communities to complete an unsheltered “street” count at least once every two years. The State of Hawaii has completed an annual count every year since 2009.
“Everyone knows the homeless count is a minimum number, since it is not possible to find every unsheltered person when the count is performed. This paper presents ways to measure quality and make estimations to account for the persons missed in the physical enumeration,” said Carlos Peraro, co-author of the study. Mr. Peraro currently provides consultation and administration for the State of Hawaii’s Homeless Management Information System (HMIS).
The report compares data on persons listed as homeless and unsheltered in the HMIS with persons found through the physical canvassing of persons. The count date was January 22, 2013. The study found that only 59% of the persons found in the street canvass had prior records in the HMIS. In addition, up to 875 persons not found in the street count had open and active records in the HMIS and were recorded as homeless and unsheltered.
The report makes recommendations to measure the quality of the Point-In-Time count in several ways including:
1) Computing the percentage of persons enumerated where full names were collected that can enable linkage of count information to HMIS data
2) Calculating the percentage of persons found in the PIT count with prior and/or active records in the HMIS
3) Using encounter information on unsheltered persons recorded in the HMIS to identify persons who are likely to have been missed by the street canvass performed by homeless service staff and volunteers.
The report is being prepared for submission to a peer-reviewed journal covering homeless and/or methodological research.