When reading about the different ways homeless people are counted, I have found that there is a lot of room for error.  Starting with the street count, I believe there is no way to get an accurate count at any moment in time because there are so many people who would not be found.  It is a good idea, because without a number count of people who aren’t in shelters there is a huge lack of information, but I still feel there would be forgotten people.  One big area of forgotten people would be the homeless staying in their cars.  The people staying in their cars would not get counted in the shelter and institutional count or the street count.  As far as the surveys, random digit dials could be extremely ineffective.  There is no way of being accurate unless every household was contacted, which would be nearly impossible.  I do think the process of completing homeless surveys by homeless people would be very effective.  Because homeless people feel more comfortable with other homeless people, it makes sense to have them interviewing each other. The real question is how would there be a way of making it more accurate?  How is the ASR supposed to calculate all homeless people when there is no real way of talking to them all or surveying them all?

http://www.appliedsurveyresearch.org/homeless/