One of the challenges to measuring the performance of homelessness assistance programs is comparing performance between programs that serve different populations. Risk adjusted performance measures can help. Risk adjustment means that different performance standards are set for programs depending on the population they serve. For example, a program that serves people with no income would have fewer people exiting to permanent housing than a one that serves people who are currently employed.
One of the easiest ways to do risk adjustment is to base it on the barrier screening tools your community is using. Barrier screening tools are used to identify how many barriers a household has to moving into, and stabilizing in, permanent housing. Many communities are already using barrier screening tools them as part of a centralized intake or coordinated assessment process, and the new CoC regulations brought about by the HEARTH Act require that communities use some sort of a coordinated assessment process. (For those who are interested, here is an example of a barrier screening tool.)
If your barrier assessment tool has three categories, you can then use those categories to do the risk adjustment. For example, you could compare performance serving people in each category to other programs or benchmarks for that category. You might get an extra point for being above average, 2 extra points for being more than 10 percent above average, etc. You can also combine the categories using a weighted average.
Without something like a uniform barrier assessment process, things get a little trickier. An alternative would be to look at characteristics in your HMIS data, such as income or chronic homelessness. This will likely require an analysis of your data to determine what characteristics need to be adjusted for.
If you haven’t read it, we have a good publication on performance measurement called What Gets Measured Gets Done: A Toolkit on Performance Measurement for Ending Homelessness, and there’s a section on risk adjustment that starts on page 41.
Image courtesy of S@Z.