.What Does the Homeless Census Really Tell Us?

It’s Jan. 30, and Joe Bishop, a volunteer with the Point-in-Time Count, or PIT Count, is driving at 10 miles per hour, while Amber Belcher rides shotgun and sifts through maps of three census tracts. The two met for the first time half an hour ago, when they were assigned this area.

Both Bishop and Belcher are rookies in the pre-dawn counting process aimed at better understanding the population of those without homes. Most of the counting happens from the car. Outside, the ground is wet from a nighttime rain. As we roll down a quiet Capitola street, the two volunteers confer with each other to make sure they follow the steps correctly. It’s 5:30 a.m. when Bishop turns around, peering toward me in the backseat, and says with a smile, “As you can see, Jake, the training was—”

“Extensive!” Belcher says, also smiling. Bishop and Belcher both watched a quick video a few days prior as part of their training, then were briefed for 10 minutes as a refresher at the Homeless Services Center before volunteers dispatched into small groups. They’re both happy to be volunteering and want to make sure they get everything right.

The information that volunteers are amassing this morning will serve as the first data point in the 2019 Santa Cruz County Homeless Census and Survey. Over the next couple of months, organizers will conduct about 400 interviews of homeless individuals, attempting to mirror the demographic breakdowns that volunteers like Belcher and Bishop found in the field last month. Researchers will then compile all of the findings in a report due out this summer. The study plays a pivotal funding roll for communities like Santa Cruz County, which receives federal money from the U.S. Department of Housing and Urban Development (HUD).

As Santa Cruz waits for the results of the bi-annual PIT Count, its struggle with the problem of homelessness shows no sign of abating. The homeless camp between the Ross department store and Highway 1, known by many as “the Ross camp,” will be closing in the next three weeks as city employees attempt to shift tent-dwelling residents to other shelters.

secure document shredding

Assumed Errors

Over the years, some skeptics have raised questions about counts like these, including the National Law Center on Homelessness and Poverty. The group authored a 2017 report titled “Don’t Count on It,” aimed at highlighting the ways that HUD’s PIT process underreports the actual number of homeless people nationwide.

Naomi Sugie, a social ecology professor at UC Irvine, participated in similar PIT counts, including one in New York City. She tells GT via email that she generally believes “these methods are assumed to have error.” That, she adds, makes it “not particularly useful to emphasize small differences from one year to the next.” Beyond that, if conditions change dramatically in a given year (i.e. a big winter storm), that could affect the count in more problematic ways, adds Sugie, who recently authored a report about how to use new technology and research methods to study “hard-to-reach groups.”

Here in Santa Cruz County, some leaders have grappled with what to make of the count for some time. “It is not real accurate, but you know what is more accurate? Nothing,” says Chip, executive director of the Downtown Association, who went on the 2013 count. He argues that survey results still provide valuable insight into the homeless community. “It’s the best data we have, and when you’re making policy decisions, data’s really important.”

The overall number of local homeless residents has jumped around in recent years without much explanation. Reports showed a 44 percent decrease from 3,536 people in 2013 to 1,964 in 2015. The total went back up again in 2017, to 2,249 people, still 36 percent less than the 2013 number.

Watsonville company Applied Survey Research (ASR) oversees this count and survey, as it does others across the San Francisco Bay Area. ASR Vice President Peter Connery finds that the data is actually “remarkably consistent” year-to-year, especially when it comes to survey results. Many social scientists, generally speaking, don’t like to rely heavily on self-reported data, but the reports’ statistics on reported drug abuse, mental illness struggles, reasons for becoming homeless, and duration of homelessness all hold rather steady. Also, since 2013, between 68-84 percent of respondents have said that they lived in the county before becoming homeless.

Each count, Connery says, is statistically accurate, and ASR tries to make sure as many volunteers as possible get paired with an expert guide from the survey company.

Foggy Math

Early on Jan. 30, after 15 minutes of circling mid-county, Bishop and Belcher spot the first person of the morning who appears to be homeless. From the car, Belcher bubbles in what she guesses is the person’s age bracket, 24-65. As Bishop drives around, Belcher keeps track of which streets we’ve already traveled down, and also wonders how the rain is affecting the count—if the downpour might have pushed people farther out of sight in search of dry hiding places. At the suggestion of organizers, Bishop and Belcher skip over a few streets in a more affluent neighborhood. Our census tracts include parts of Live Oak and Capitola. The 2017 count found that 1 percent of the county’s homeless population lives in Capitola.

Santa Cruz County’s PIT count happens every two years on one day during the last 10 days in January. Connery says HUD mandates that the counts happen at the end of the month because homeless people will often pool cash to get a hotel room when they can afford it, and that money is usually gone by the end of the month. This year, in order to count inhabitants of the Ross camp, Connery says that ASR had “embedded reporters” go undercover at the encampment.

No one knows exactly how variables like weather, date of the month or day of the week affect any given count, though Connery says he doesn’t see any of those factors having an impact. By virtue of its methods, the process involves making judgment calls and even some stereotyping. Belcher and Bishop are not allowed to disturb anyone or knock on car windows, so when they coast past parked vehicles, they’re instead looking for fogged-up windows—a telltale sign that someone’s been inside for hours. That can be tricky, because early in the morning after rain, many car windows look foggy. Later, when they see a tired-looking man with a backpack walking down a busy street at 9 a.m., they have to decide: homeless or not?

As we circle, Belcher wonders aloud if the counting process might be easier if each car was given a GPS tracker, so that every two years volunteers could see how the previous group covered the same ground. Chip suggests that surveyors could do a PIT count on two or more days for every homeless census to widen the sample size, but acknowledges that it would take more resources and may not be worth it.

Connery says that adding days probably would make the count more accurate. Other communities opt to do the counts every year instead of every other year. But he’s not sure what end such a change would serve. Instead of spending more money studying homelessness, he says that local governments should increase spending on services to put a real dent in both the size and the suffering of the homeless population.

He gets defensive when answering questions about the surveys. At the national level, many of the criticisms of HUD’s counts come from homeless advocates on the left, but locally, the critiques come more often from anti-homeless groups looking to undermine the survey’s results. “You don’t have a data problem,” says Connery. “You have a service delivery problem.”

Bishop and Belcher both participated because they know how important it is to fund solutions to homelessness. Bishop is the founder of the faith-based counseling nonprofit Respero, which has been increasing its homeless outreach. Belcher is a nurse who used to work at the Homeless Persons’ Health Project, and she sometimes misses her old job.

“You need funds in order to help people,” Belcher explains, recalling efforts to track down new funding sources on the phone. “It does break down to the dollar at some point. And if you don’t have the ability to provide resources, then you’re not gonna help anyone. I think it’s really important we count all the people we can.”

The two volunteers wrap up their count around 9:30 a.m. Belcher and Bishop have both been up since before 4 a.m., and they had second thoughts about coming out today. Once Belcher finally crawled out of bed after hitting the snooze button on her alarm a few times, she looked out her window at the rain, and thought that maybe she didn’t really want to leave the house, after all. “Then I was like, ‘You know what? You’re being lame because you’re gonna go out survey people who live out in this rain right now,” she says. “Suck it up!’”


  1. How expensive would it be to send hot-food carts out to the most populated areas regularly — and count the homeless who attend? There must be a reliable level of service that is possible, as well as an accurate counting process. We need more effective service delivery asap. I think wealthy donors need to be found, also.


Please enter your comment!
Please enter your name here

music in the park san jose
Good Times E-edition Good Times E-edition