Flawed Number Released on Homeless Count

The biggest waste of money by the Department of Housing and Urban Development has to be the annual count of homeless people.  We always equate this with counting a gumball machine full of Skittles as you pour them out onto the floor.  The first question for both is "why?"  Why would you want to count the colorful Skittles while they are falling on to the floor and are still in motion?  And why would you count homeless people on one day as they traveled around the city?   The second question is how valid is a count in which untrained volunteers drive around trying to play this game of "is that guy a homeless guy sleeping or a pile of clothing?"  NEOCH never wastes its time with this silly undertaking in late January or February.  We do a count of the people sleeping in a small geographic area of the same streets in Cleveland on the same day every year to show the trend of number of people sleeping outside, but we certainly do not make the leap that this is a total count of the population.

The annual count figures were released by HUD this last week, and they show a decline in the United States on the number of homeless people.  The media reports never includes all the qualifications about the flawed data or how meticulous some communities are in the count (Cleveland) and how liberal they are in estimates in other communities (Los Angeles).  Cleveland/Cuyahoga County reported 2,100 people with zero families without shelter. We have to question that there were no families sleeping in cars over on the near West Side trying to avoid police detection. We also reported one of the lower rates of long term homeless in the country at 12.2%.  These are similar numbers as we had over the last three years, and is only a couple dozen above the total number of shelter beds in our community.  Assuming all beds are full and in fact some beds are used multiple times a night for the two largest shelters, the volunteer counters found only a few people outside in Cuyahoga County on that night. This does not match the spreadsheet we keep of homeless people who sleep outside in coordinating outreach workers.

It is hard to count people outside or in cars or in abandoned buildings.   In addition, these are not trained US Census workers doing the count or academics.  These are just volunteers who have an interest in maximizing the count to justify their jobs.   The media covers the story, but does not include any of the skeptics who can say that most of this information is junk. The Washington Post story on the decline in Veteran's homelessness down by 24%, but no mention about how bad the collection of the homeless data is by agencies that serve veterans in the United States.   I have read the report from the Department of Veterans Affairs contained in the HUD report, and can say that it is highly deceptive.  At least in the HUD report they are surveying the same facilities every year.  In the VA count, one third to one half of the facilities surveyed from 2010 are then just estimates in 2011.  This is useless information and there is no way to make a statement that the number of homeless veterans are down by 24% with so many variables.   This is truly picking a number out of the air and reporting it as fact.  

The only way that this data could ever be accurate is to:

  1. use data that is reliable like shelter bed usage or calls to the local first call for help number or kids who are declared homeless by the local school district.
  2. not use outdoor counts which are unreliable.
  3. find a new way to count homeless people in rural communities who never have enough beds to serve the population.
  4. have the data supervised by academics who can make accurate estimations using Census data.
  5. spend time and resources on representative cities, suburbs and rural communities and then make estimations on similar sized cities.
  6. develop some multiplier for total population designed by academics or limit the report to the shelter population because a physical count will never be accurate for a population that lives in the shadows.
  7. verify these numbers with on the ground secondary sources that do not have a vested interest in presenting a decline (government) or an increase in the numbers (advocates for expanded funding).

I never have understood why the media do not view these numbers with greater skepticism.  There is never a balance in these stories or a look behind the curtain.   No one ever reports that if HUD is reporting to Congress such flawed information, how do we expect them to oversee an end to homelessness?   How can they supervise affordable housing if they rely on such bad information from the field?  It is lazy for the media to report this information, and can only be seen as a reprint of a press release from the agency.

Brian Davis

Posts reflect the opinion of those who sign the entry