Every year, King County officials issue the same warnings around the point-in-time count of homelessness: Take it with a grain of salt. It’s probably an undercount.
Nevertheless, the count is still the most consistent annual indicator of how well officials have responded to the homelessness crisis, with results potentially affecting critical funding and policy decisions.
The 2019 count found an overall 8% decrease in the county’s homeless population. But it also indicated such steep drops in homeless populations at the center of a raging local debate — chronically homeless people, and those with serious mental illness or substance use disorders — that some are questioning how the count squares with what they see on the ground.
The count showed a 38% decrease over last year in the number of chronically homeless people, defined as those with a disabling condition who’ve been homeless for a year or longer. For unsheltered, chronically homeless people — such as those living in tent camps, vehicles and on sidewalks — the estimate fell by nearly two-thirds.
“That is a wild fluctuation,” said Dan Malone, executive director of Seattle’s Downtown Emergency Service Center. “I mean, how on earth are there suddenly 1,600 fewer people outdoors who meet chronic homelessness criteria?”
The count also showed a 55% drop in unsheltered adults with a substance-use disorder and a 52% decline in unsheltered adults with a serious mental illness.
Where did they all go?, asked Brad Forbes with the mental-health advocacy group NAMI Washington.
“We’re working on [housing], but I have not heard from any providers that somehow there’s been this miracle,” Forbes said.
Seattle City Councilmember Lorena González said the estimates of unsheltered people in the report “don’t match what I see on the streets every day.”
“I think we have to be cautious, because the amount of money we have been able to invest in this space has not jumped in a way that would lead me to believe we have a 38% decrease in chronic homelessness,” González said.
Seattle Mayor Jenny Durkan in a statement credited city efforts to increase homeless shelters with more onsite services, as well as a strategy called diversion, which offers onetime financial aid. The city has also boosted its Navigation Team, which does outreach and cleanup of tent camps.
“This new One Night Count shows that our work and data-driven investments over the last year to prevent and address homelessness is having an impact,” she said. Durkan’s statement added that “we have a lot more work to do.”
Still, the numbers don’t necessarily reflect what some see in specific neighborhoods. Erin Goodman, executive director of the SODO Business Improvement Area, said she doesn’t see a reduction in homelessness on SODO streets.
“We want there to be a decrease,” Goodman said. “We want to hope that the city is making inroads on getting people inside and getting people contracted to services. But as long as we’re still dealing with the impacts of this crisis, it’s hard to see that decrease.”
Who was surveyed?
King County’s count relied on visual observation of hundreds of volunteers and paid “guides” — who were currently or recently homeless — who fanned out countywide on Jan. 25. It was followed up by a survey of 1,171 homeless people, which asked who they are and why they are homeless. The survey was conducted by for the third year in a row by the firm Applied Survey Research in California, which said the survey had a margin of error of 2.7%.
Those questioning the steep drops, however, pointed to the survey methodology, which estimates – does not literally count – the number of unsheltered chronically homeless people, and adults with substance abuse issues or serious mental illness.
“It suggests to me that it’s probably heavily influenced by whatever the surveying done was,” Malone said. “It could be that one year the surveyors end up connecting with a bunch of people who meet chronically homeless criteria, and then another year they talk to relatively few people, but I think having a wild fluctuation like that suggests there’s a difference in who’s being talked to in the surveys.”
Kira Zylstra, acting director of All Home King County, the agency that runs the count, said the count always strives for representative samples, but acknowledged fluctuations in the people encountered during the count. Survey answers are also self-reported, which has limitations, Zylstra said.
“It doesn’t mean all self-reported data is meaningless,” Zylstra said. “But that can be a factor, especially in more sensitive questions.”
Arthur Padilla, interim executive director at ROOTS Young Adult Shelter, said he and other service providers are trying to figure out the cause of the drop in numbers.
Some suggested that their interventions and diversion strategy were working. Padilla said it was hard to tell what was going on.
“There’s something around the point-in-time that is questionable,” Padilla said. “How could it not be? It’s a huge initiative, but it’s fiercely random.”
Zylstra said “fiercely random” was a “strong comment.”
“We 100% know the point-in-time is an estimation, is an undercount,” Zylstra said. “There is no perfect way to count communities experiencing homelessness.”
“Strive to improve”
The point-in-time count only reflects a snapshot of homelessness on one night. In 2018, the point-in-time count was 12,112, but over that same year, the county’s Homeless Management Information System counted 22,500 households accessing homelessness services.
Questions about the accuracy of the count arose last year as well, as Native organizations criticized the 2018 point-in-time as an undercount of American Indians and Alaska Natives.
In response, All Home King County worked with Native organizations to improve Native representation in the survey. In 2019, the representation of American Indians and Alaska Natives jumped to 10% of the overall total homeless count, up from 3% in 2018.
“[The count] is something we continue to strive to improve year after year,” Zylstra added, “but it is a reason to emphasize multiple data sources in a community.”