Mark Rendell
Mark Rendell

Counting the Homeless: What’s the Point of Point-in-Time?

Was the City's count of the street community last week better or worse than nothing?

One well-known advocate for Yellowknife’s homeless community is calling the City’s homelessness point-in-time count last week “random and haphazard,” and questioning the ethics of releasing numbers that she claims greatly distort the scope of the problem.

“I don’t know how somebody can ethically go out there and put these numbers out, because of the potentially devastating effect they could have,” Lydia Bardak, executive director of the John Howard Society, told EDGE. “That’s what’s really frightening. If you use an unreliable data method, how is that going to influence policy decisions?”

The preliminary tally from last week’s count showed that around 150 surveys were filled out. They were gathered during a four-hour session on Wednesday afternoon in which tents were set up in the 50/50 parking lot and on the south side of Frame Lake. People were served a free lunch and invited to fill out a survey asking demographic questions.

“What they found was 150 people who were willing to come for lunch and fill out a survey. It doesn’t tell us anything about the population,” said Bardak. She added that she’d talked to people in the street community who had openly scoffed at the idea of completing the City’s survey.

Bardak places the number of homeless people far higher. According to data collected by the Dene Ko Day Shelter, the predecessor of the current Safe Harbour Day Centre which Bardak managed, 365 individuals had accessed the shelter up to March 31, 2014. A 2009 report by the Yellowknife Homelessness Coalition indicated an even greater number: 936 people, or five percent of Yellowknife’s population, had accessed homeless shelters in 2008 alone.

Dayle Hernblad, the City’s homelessness coordinator, said that the City’s count was not intended to provide the definitive picture of homelessness in Yellowknife. “No single method can give a finite number,” she told EDGE, adding that the data is still being analyzed by professionals. Stephen Gaetz, a York University Professor hired by the City to help analyze the count’s findings, also said the figure was likely on the conservative side, according to NNSL reports.

The main point of the count, said Hernblad, “was to get a baseline of data for going forward, because Yellowknife has committed to embrace the housing-first model. In the community plan for the federal government, we committed to establishing a baseline number to help inform us in different ways when we’re looking at pilot projects.”

Even with this as a goal, Bardak isn’t sure why the City is getting baseline data in a manner she claims is “so unreliable the flaws next year could be different than the flaws of this year.”

“We know the number of people who seek addiction counseling, who access income support, who use the shelters, the number of kids in child services, how many people are accessing health and social programs, police and corrections statistics… If we had a researcher who could compile this, we’d have a much better picture of who’s out there and what critical needs they have.”

Jeff Barichello, an economic statistician with the GNWT, told EDGE that the City’s count is better than nothing, though no point-in-time count is likely to paint a statistically accurate picture whichever demographic you’re looking at.

“It’s a completely different survey than what you’d normally do… In a normal population count, we would go around a look at the houses, make a map of houses, then draw from a sample of the houses to estimate the population.“

Without phone numbers and fixed addresses for many people in the street community, this is clearly impossible. Barichello suggested that to approach something more statistically valid with point-in-time counts you really need to do multiple counts.

“If you had two or three points-in-time, you’d start to surround the truth,” he said.