Who Counts? Doubts Raised Over Homeless Data

On Tuesday, the Department of Homeless Services Commissioner Gilbert Taylor announced he would resign on Jan. 1, amid plans by the mayor to review and restructure the department. The administrative upheaval, coming just before the annual homelessness street count at the end of January, will likely draw closer scrutiny to the way the city collects and manages its homelessness data.

Homeless populations are notoriously difficult to count. Shelter head counts are seen as the most reliable metric, while street counts are the most contentious. But regardless of which part of the homeless population is being counted, government funding – and ultimately, funding for direct service providers – often hinges on that data.

According to federal statistics released at the end of October, a rising number of New Yorkers are homeless. The U.S. Department of Housing and Urban Development, using data from the city’s street count, shelter head counts, and other sources, estimated the city’s homeless population rose 11 percent in one year. The report said more than 75,000 individuals were homeless in February 2015. 

But whether homeless numbers are rising or falling is often a source of heated debate.

In November, New York City Police Commissioner William Bratton rankled administration officials by saying street homelessness had “exploded” over the last two years, and the city responded too slowly to fix the problem. 

“That narrative is incorrect,” Taylor responded at a City Council hearing Dec. 9. According to the DHS’s annual Homeless Outreach Population Estimate (HOPE) Count, the number of people living on the street has actually gone down by 5 percent.

Yet Taylor acknowledged – although he said he couldn’t quantify it – he believed “there has been an increase,” citing department caseloads, anecdotal evidence from the public, and observations by homeless services providers.

Many nonprofit services providers have long been skeptical of the city’s street count estimates.

“As we’ve said year after year, there’s just no accurate way to count the number of homeless individuals living on the streets,” said Giselle Routhier, policy director for Coalition for the Homeless. “Particularly with the HOPE count – it’s nearly impossible to accurately compare that over time,” she said, noting that weather conditions in a given year could raise or reduce the count.

Dr. Kim Hopper, a medical anthropologist and leading researcher on homelessness, helped design the HOPE count. Hopper says HOPE was only ever meant to measure a limited subset of the homeless population — the most “recalcitrant” individuals.

“You have this minimalist count which is designed to get at those people who are most resistant to coming indoors,” said Hopper. The count is taken over one night in the dead of winter in which volunteer counters ask homeless individuals questions off a checklist to help classify if they are a veteran or are HIV-positive, mentally ill, or have other health problems. Hopper explained that in the worst weather, many homeless people are able to find a friend’s couch, a car, a vacant property, or some other shelter – places that are out of view of the small teams of volunteer HOPE counters. 

For many reasons, Hopper said, to “transform (the HOPE count numbers) into an accurate assessment of how many people spend some time on the street during the course of a year would be a mistake.”

However, critics say that is exactly what the city has done, by portraying the HOPE count as a comprehensive survey of all homeless people living on the city’s streets.

DHS insists it always clarifies that their HUD-certified count estimates only those who are “chronically homeless.” Chronically homeless individuals are a subset of the homeless population and are defined by HUD as including those who have lived on the street for at least a year or many times in one year, and can be diagnosed with drug abuse, serious mental illness, brain damage or other serious chronic physical illnesses or disabilities.

But a review of the department’s public presentation of HOPE data tells a different story.

For the last decade, press releases and presentations by DHS publicizing the results of the HOPE survey have not explained the limitations of the count. A presentation of the 2015 results simply stated the count intends to determine “the number of people living unsheltered across the city.” 

“I – and a lot of others – have concerns about how accurately (the city data) represents the homeless population,” said Christy Parque, executive director of Homeless Services United, which represents nonprofit homeless services providers. While the numbers are a starting point, Parque said, “I think the count could be improved. I think there’s a lot of valid criticism on the fact that we do it in winter when you’re least likely to find people.”

DHS officials said they would review the issue after questions from New York Nonprofit Media about how the department presents its data.

While people living on the street are among the most visibly homeless, other vulnerable portions of the homeless population are often overlooked and left uncounted, advocates say. The statistics are important because they help determine funding levels for programs. If people aren’t counted, they may not get the help they need, advocates explain.

“We’ve been concerned that by not counting them they essentially become invisible,” said Carol Corden, president of New Destiny Housing, a nonprofit that provides housing to victims of domestic violence and their children. “And if they're invisible, then they’re less likely to have access to homeless resources.”

That’s exactly what’s happening to her domestic violence clients, she says.

Despite a network of specialized programs serving thousands of domestic violence victims in New York, HUD’s official report on New York City homeless populations this year showed zero homeless victims of domestic violence in New York City.

The New York City Coalition on the Continuum of Care, a coalition of homeless housing and shelter providers, consumers, advocates, and government representatives, has not submitted any data on homeless domestic violence survivors to HUD since 2012, when they were the third-largest homeless subpopulation in New York City, totaling 7,676. Unlike most other subpopulations, like veterans or those with HIV/AIDS, reporting data on homeless domestic violence survivors is optional – and that lack of data was reported as zero in the HUD report.

A spokesperson for HUD acknowledged the error, saying it was a misprint that would be corrected. It in no way reflected a belief at the agency that New York City has no homeless domestic violence survivors, HUD press officer Brian Sullivan said.

“I can assure you, nobody in this department, not a soul believes that,” he said.

Corden said she understands that the zeroes were a mistake. “But the question still remains why the city of New York, their Continuum of Care, has chosen not to count victims of domestic violence who are homeless,” she said. “The idea is to get as accurate a count of the homeless population as possible, right?”

Dr. Benjamin Charvat, who co-chairs the steering committee at Continuum of Care and oversees all research and data analytics for DHS, would not offer New York Nonprofit Media an explanation on the record as to why homeless domestic violence victims are not included in the report to HUD. 

“If you’re not even counting people, it’s easy to say, ‘This is not an important population,’” Corden said. “The fact that it’s largely families … I think it’s egregious, because you’ve got all these kids who are in that mix.”

“It has an impact,” said Cecilia Gastón, executive director of the Violence Intervention Program Inc., which works with Continuum of Care on behalf of domestic violence housing providers. Without the numbers being provided to HUD, she said, “we just don’t get the housing.”

Advocates agree that it’s critical to collect all the information possible about homeless groups. 

“If we’re really trying to get our arms around homelessness, then we need to know how many homeless people there are. That’s the whole point of the HUD count,” Corden said. “It’s hard to push for accurate solutions – to really advocate for appropriate approaches – if you don’t have data.”