In HUD’s PIT Count Methodology Guide, HUD establishes several standards CoCs must adhere to when conducting their PIT counts. Standard 11 states:
“CoCs must be able to verify that unsheltered homeless people identified in the count are unsheltered on the night designated for the count as defined at 24 CFR 578.3 of the Homeless Definition Rule: ‘An individual or family with a primary nighttime residence that is a public or private place not designed for or ordinarily used as a regular sleeping accommodation for human beings, including a car, park, abandoned building, bus or train station, airport, or camping ground.’”
Because RVs are “ordinarily used as a regular sleeping accommodation,” HUD does not believe that all persons sleeping in RVs should be included in the PIT count. However, there are several factors that a CoC can consider when determining whether to include occupants of RVs in their unsheltered PIT count data, including:
- the RV is in disrepair (e.g., holes, broken windows, flat tires, removed or broken siding);
- the RV occupants do not have access to sewer, water, and electricity connections; and
- the RV is parked in an unusual place, including non-RV designated lots, non-residential areas, or areas where RVs are not allowed to park, and the RV is clearly occupied.
CoCs must exercise discretion when counting persons in RVs because the inclusion of RVs often involves understanding the specific circumstance for RVs to determine if their occupants should be counted as unsheltered.
If a CoC chooses to count RV occupants, they must either talk to all RV occupants on the night of the PIT count to verify their housing situation or talk to enough RV occupants in the CoC on the night of the PIT count to derive a reliable sample. If a CoC chooses to use a sample approach they must consider several factors:
- How do you create a reliable sample size? This requires a consistent approach to counting RV occupants. Ideally, the CoC would create a list of criteria (like the list above) for determining which RVs they might include. Then the CoC would require the PIT count teams to conduct a random sample. HUD recommends requiring PIT count teams to engage the RV occupants in every third RV.
- Does the CoC need to create different samples for RV occupants in different kinds of areas in the CoC? If RV occupants appear to have different numbers and characteristics in cities as opposed to more rural areas, the CoC should consider creating a sample for cities different than rural areas.
- How will the CoC need to adjust to remove RV occupants who are not homeless? If the CoC learns that people in RVs that meet the CoC’s standard for including in the PIT count include people who are not actually homeless, then it must adjust its overall sample to remove these people. It is not unusual to find that not all people in the types of RVs we might think are housing people who are homeless are actually homeless. If the CoC finds that a certain number of people in the sample are not homeless, they should adjust their total sample. For example, if 1 in 10 RV occupants were found not to be homeless, then the CoC should assume that 10 percent of all RV occupants counted are not homeless – and should remove them from the PIT count.