Counting Homeless People

What if the Census Counted People Like HUD?

I hate the Homeless Management Information System, because it is such bad data that it harms our community. Garbage in: garbage out is my biggest issue with this system. It is the law of the land that every shelter receiving public funds must submit data to the Department of Housing and Urban Development or they are at risk of losing federal and state dollars. The problem is that it misses so many people that using this data as any reflection of the number of homeless people diminishes the extent of the problem.

I was thinking, what if the Census were conducted in much the same way as the count of homeless people? What if there was no dedicated money going to the count every ten years, and it was all volunteer? What if there was very little training available to the volunteers who were expected to complete the Census count in every city? What if those volunteers made a pledge of confidentiality, but there was no oversight or law protecting the privacy rights of those who answered the Census questionnaire? What would the count look like if the government enlisted grocery store clerks, gas station attendants, and cable tv repair men to do their jobs but also oversee the Census count at the same time? That is the system we have with HUD. Case workers who all have full time jobs are expected to volunteer to submit data to the federal government to come up with a count of the number of homeless people. The shelters do not receive any additional funds or regular training about counting or confidentiality. While the bubbly and friendly young person who scans my milk and cereal at Dave's Supermarket may be good at her job, I don't want her to be in charge of the US Census. I also don't think the guys going out at night to save people from the cold should be the individuals in charge of the count of homeless people.

An all volunteer US Census would also allow for regional disparities in the count. San Diego city officials could decide that they are going to make sure that their grocery store clerks and cable tv staff were trained and spent a couple of weeks only on the US census. They may decide that the 10 year payoff in federal funds is worth spending a great deal of local money getting the count right. While Detroit would just not have the money to pay volunteers to count, and their numbers would suffer. So, San Diego would have a large growing population and Detroit would shrink in the US Census. This is what happens with the homeless count. Some communities put real money into the effort, and others rely entirely on volunteers. This creates huge disparities in the final numbers.

I have no problem with cable tv repair people, in fact, some of my best friends work for the local cable tv company, but I don't want them to ask me personal questions and then turn that information over to the government. I want someone who has to pledge that they will go to jail if they release any of my personal information while doing their job of counting the population. I want someone who is paid by the federal government to spend their full time counting people, and not volunteers working on other issues. The data generated from HMIS is unreliable and is not a good measure of the number of homeless people in a city.

NEOCH is working on our 2010 State of Homelessness report. We use the HMIS information in the report, but only to document the demographics of people within the shelters. It is only good information to say how many African Americans or veterans are using the shelters. It is good information for where people go after they leave a shelter, but we would never use this information as a measurement of the total number of homeless people in a community.

Brian
Posts reflect the opinion of those who sign the entry.

Census Update 2010

Ben Holbert of the US Census gives an award to Luke Drotar VISTA for NEOCH for his work in making sure that homeless people were counted in 2010. Then some guy in a Hawaiian shirt crashed the photo shoot.

Done for Another 10 Years
NEOCH and especially Luke Drotar did a ton of work on making sure that the US Census did not miss homeless people during the count. We are pretty sure that every shelter in Cleveland was counted, but most of those who were outside or those who only surface at a meal program were largely missed. We hope that the Census releases the shelter count as they had done in the 2000 Census. NEOCH staff have prepared a list of recommendations for the 2020 Census if we still have not got it together as a country and solved homelessness.

1. We need one staff person from the Census to be designated as liaison to NEOCH and the County Office of Homeless Services that will cover the whole county. This time there were three different offices that covered different service areas, and each had one supervisor.

2. We need more field staff counting, and fewer staff advertising the Census. Also, there were not enough administrators coordinating or directing the field staff that we could communicate with in the office. Those directing the field staff were also out in the field and we had a hard time getting in touch with Census staff.

3. We need to have Census clearly define to us the definition of shelter that they intend to count as part of the homeless count. Where do the mental health and treatment programs fit into the count?

4. We need the privacy of the outdoor sites communicated to workers early in the process and this should be a United States policy and not just a regional policy.

5. One month before the count, outreach workers and homeless individuals should be hired and sworn in as Census employees to guide them to the outdoor locations on the day they count those resistant to shelter.

6. We need some guidelines for how they eliminate duplication with regard to the meal count since as many as 50-60% of the population are counted in other locations.

7. All the numbers should be released to the City and County after the count to compare to our February HUD count and to use for planning.

Brian

Posts reflect the opinion of those who sign the entry.

Census Count in Cleveland

Counting Homeless People as Part of the Census

It is an overwhelming task to try to enumerate the population, and it is especially daunting to count the transient populations. NEOCH did a ton of work over the last month to get as many homeless people as possible counted. We met with Census officials on a regular basis, and distributed Census gear to every homeless person who we met. The Census was at all three Stand Downs, and we gave out thousands of fliers.

Overall, I think that they got everyone at the shelters, but the other counts are unlikely to yield any good numbers. The Census officials counted shelters on Monday then soup kitchens on Tuesday and then the outdoor sites on Wednesday. The Census made the last minute decision to categorize transitional shelters as group quarters instead of shelters and so were planning not to count any of those facilities this week. We appealed this to the regional office to ask that they reconsider this decision, since transitional shelters are shelters and were counted in the 2000 Census. By skipping the transitional shelters, we would have a significant under count. They changed their minds about noon on Monday, but officials were not able to get to the transitional shelters until Tuesday or even some on Wednesday. NEOCH called every single shelter three times to make sure that a Census worker had visited their site. We found that four facilities were skipped, and we informed the census employees.

There seemed to be a lack of staff assigned to the homeless count and certainly a lack of clerical staff preparing for the count. It seems that were a lot of people advertising the importance of the Census, but not as many doing the actual count. We did not hear back from anyone else who said that they were missed, so we believe that every shelter was counted. We do not have much information on the second day of the count, when they were doing hot meals and mobile canteens, except that five or six of the homeless drop in/meal programs have Census staff on site for 15 to 20 hours a week to answer questions and encourage people to complete the application.

The last day was very bad and highly questionable as to accuracy. We gave the Census staff a list of all the outdoor sites that we were aware of locally. But it is nearly impossible to find these individuals in one 24 hour period. Basically, the outdoor count was a waste of time. Census staff really wanted to do the count at night, which every one of the professional staff said was a bad idea. We suggested early morning (6 a.m. to 9 a.m. and 5 to 8:30 p.m.), but they could not be convinced and went later into the night on Tuesday night. No matter if they have an army of counters, it is just impossible to count homeless people who live outside in 24 hours. So, they drove by a bunch of sites that we had given them, and tried as hard as they could in the dark to count, but this was not a good use of resources or time.

In 2000, only the shelter numbers were released and I am pretty sure that we did a good job in Cleveland with that count. It was a great deal of work trying to get the Census count correct, and most cities don't have an advocacy homeless Coalition to coordinate these activities.

Brian

Posts reflect the opinion of those who sign the entry.

Census Trouble?

Is Anyone Else Worried about the 2010 Census?

The New York Times had an editorial this weekend about the problems with the 2010 Census. I have been thinking about this for the last few months. What is happening with the next Census? The New York Times seems to say that there is some conspiracy here. I have no idea, but I do know that in 1998 we were meeting with the Census staff every month. We already had a detailed plan for counting homeless people by April 1998. The City was hosting meetings with all the social service providers in the City at that time. They were trying to build confidence, and map a strategy for counting everyone in shelter. It is almost the end of 2008, and we have not had one meeting yet with the Census. I know that they have hired some staff, but I have seen no plans yet. I am worried. This could be very bad for Cleveland. If we do not count all the poor people in Cleveland, we could face a 300,000 person city or a 150,000 drop in the population.

Brian
Posts by Northeast Ohio Coalition for the Homeless staff and Board.