In the early hours of last Friday (Jan. 27), an army of 800 volunteers combed 13 cities in King County, block by block, in search of a very elusive demographic: people who are homeless and “roughing it” on the streets.

That search, called the One Night Count, found 2,594 people on the street, a three percent increase from last year, when volunteers counted 2,442 people. People were found “in cars, tents, all night buses, select hospitals and curled up in blankets under bridges or under doorways,” according to a press release sent out by the Seattle/King County Coalition on Homelessness (SKCCH), the organizers behind the One Night Count. 

But that number is not just meant to be some statistic — the organizers know that the number is fairly imprecise and largely inadequate  — but rather it also serves as a source of inspiration, an advocacy tool in Olympia, and a way to bring communities together to fix problems.

The number does not reflect homeless people staying in shelters that night. The count from the shelters will be released, along with a bigger report, toward the end of February. Last year the number of sheltered homeless people was 6,382, which put the total number at 8,824. With the little variation seen in the trends from year to year, it's reasonable to expect around the same number.

On paper, the One Night Count can sound like a cool espionage mission. A team of people tread quietly through the cold night wielding flash lights, keeping their eyes open, peering into dark spaces within trees and hard to reach places. Some people without prior experience expect to uncover some kind of hidden village of homeless people, or to become privy to some deep, dark secret of King County.

But the reality is far tamer, perhaps boring for some. You walk around in the cold, chatting with the rest of your team of four or five volunteers, and find very little sign of people living on the streets. In my team's search in Federal Way, we found a whopping total of 12 people, most of whom were in vehicles — by no means less important, but not quite as stark a situation as one might imagine encountering.

Much the same difficulty in finding homeless people was, if the statistics are any indication, prevalant in other areas. There are a number of reasons for this, not least of which is that homeless people are highly mobile, constantly looking for places to stay warm and, perhaps more importantly, stay out of sight for safety reasons.

"A lot of people go to great lengths not to be noticed, because that’s how they survive," SKCCH Executive Director Alison Eisinger said. "We respect that, and we know that’s a reality. We really made peace with the fact that we’re not going to count every single person."

To counter the shortcomings of a count done with volunteers, some communities around the nation have resorted to employing those who are or who have been homeless to help with their counts. But even those with street smarts have found it hard, because where homeless people were last year, or even last week, may not be where homeless people are now.

Additionally, to scour every corner of a community in search of the homeless, vast amounts of resources are required — too vast to be either practical or worthwhile. While SKCCH has managed to create a large regiment of volunteers dedicated to the cause through 30 years of partnering with local institutions, developing into what is now the largest community-organized effort in the nation, there are still holes in the map that they are unable to fill. Eisinger said that the count here has only been able to cover spots where there was adequate community interest, meaning that some places, even if a sizeable homeless population might exist, are for the time being uncounted.

So far Tukwila and SeaTac, which both undoubtedly hold a significant homeless population, are not part of the One Night Count. Eisinger said SKCCH is trying to work with communities to bring the count to those places in the future, but she was unsure of when that would be.

One extra difficulty is the timing of the event. It can be harder to find people in the inclement weather, and the count happens at the end of January, because the Department of Housing and Urban Development under the George W. Bush administration made that a requirement. Those who haven't retreated to some hidden, sheltered (and hopefully slightly warm) spot are found mostly in vehicles. Out of the 2,594 people counted on the street in King County, about 790 — or 30 percent — were people thought to be living in vehicles. Those living in impermanent structures, such as tents, numbered only 348 people.

The method for counting people in vehicles and structures causes further trouble. For every vehicle and structure where the exact number of inhabitants cannot be seen (which are most if not just about all, due to fog, ice, curtains, or cardboard blocking the windows), two people are counted. Considering the high percentage of the count that comes from these two categories combined, this method could lead to overcount or undercount. However, finding out the exact number within would be too troublesome and disruptive to bother with, so organizers are left with their hands tied, hoping to strike some semi-accurate middle ground.

There are other methods used to count homeless people, but King County, which has been able to develop its method for over 30 years, when "Operation Nightwatch" performed the count, has one of the most extensive counts in the nation.

Communities that have been unable to collect together such large resources have made do with other ways. In Travis County, Texas (where Austin resides), the point-in-time count is performed during the day, from 1 p.m.-7 p.m. Furthermore, they survey every tenth person to create a random sample of what the homeless population looks like. Last year, only 250 volunteers performed the count, a paltry number compared to King County, and they counted about 1,000 people.

Eisinger said that by counting at night, volunteers did not have to bother with issues of discerning who or who was not homeless as much, and they did not have to disrupt homeless people's lives by bothering them with survey questions. Furthermore, Travis County's method of counting seems unable to differentiate those who find shelter at night from those who rough it outside. Still, flaws aside, this writer (and several other volunteers, I'm sure) might have appreciated a count that did not extend from 1:30 a.m. to an hour at which people have already started morning commutes. 

However, Oregon's Multnomah County, has developed a few extra methods that could add more depth to King County's method. Outreach workers spend the week prior to the count, connecting to various homeless encampments and completing surveys on anyone who sleeps outside on the night of the count. Likewise, more than 150 agencies and programs conduct short surveys with anyone they encounter sleeping outside. Then, on the night of the count, shelters make a record of everyone they turn away.

It may be possible to learn from this example of  taking extra efforts to connect and survey homeless people roughing it the week before the street count takes place. This would appear likely to give a fuller and slightly less sparse picture of homelessness on the streets. But whether the amount of resources required for such actions justifies the end result is a different question. (Also, it should be noted that, despite the extra techniques employed, Multnomah County by no means has a more expansive count: they recorded only 1,532 people a year ago.)