Field Notes: Thinking about Homelessness Prevention

written by Norm Suchar
January 9, 2013

Improving homelessness prevention programs requires that providers and funders shift how they determine whom to award prevention funds. As the chart below will demonstrate, we can have the highest impact by serving the highest risk households, even though our success rate will be lower.

One of the biggest challenges with doing homelessness prevention well is targeting to the right group. For a prevention presentation I did recently, I created a chart that shows some of the paradox involved in targeting prevention assistance well. An expanded version of the chart is below. The chart shows success rates for low, high, and extremely high risk households and calculates the number of episodes of homelessness that would be prevented under each scenario. The figures I’ve used in the chart are for illustrative purposes. There are some resources at the bottom of this post if you’re interested in more data and discussion on this topic.

Here’s how to read the chart. The first row represents low-risk households, who have a 10 percent likelihood of becoming homeless if they do not receive prevention assistance. (Although it’s labeled low risk for the purpose of this chart, most prevention programs target even lower risk households.) If you assume that the intervention has a 90 percent likelihood of succeeding for the households who would have become homeless that would mean that for every 100 people served, 9 episodes of homelessness would be prevented and one person would become homeless. The way most people calculate the effectiveness of prevention programs, this program would have a 99 percent success rate.

The most important thing the table shows is that it’s possible to have a program that looks like it’s performing worse by traditional success measures, but is actually performing very well. The program targeting extreme risk households is actually preventing many more episodes of homelessness than the other programs.

Expecting that a program is going to prevent homelessness for 99 percent of the people served will discourage the program from serving higher risk households. But it is in serving those higher risk households that prevention assistance can have a bigger impact.

Additional Resources

Presentation on New York City targeting and prevention efforts

Presentation of New York City screening criteria

Prevention Targeting paper