Punctual or aperiodic

The structure survey will be carried out in the first half of 2024, while the "individual" survey will be carried out one year later, in winter 2025. This collection will take place over a period to be determined, but between January and March 2025 (no collection on Sundays).

The choice of survey period is dictated, on the one hand, by the time constraints weighing on the preliminary survey of reception site managers, and on the other hand, by the existence of weekly, monthly and seasonal variations in the opening of services and in the use made of them by homeless people.

In addition, the year 2025 was chosen to coordinate with the ES-DS survey, the next wave of which will take place at the same time and can therefore be used to calibrate the results.

Face to face by interviewer

A three-stage sample design. We have no sampling frame for the homeless population. The method used therefore consists in drawing individuals from a set of services they use, or services they "consume".

For the purposes of the survey, the services offered by the main services used by homeless people are selected: accommodation, meal distribution, and (depending on the results of the tests to be carried out) outreach services (maraudes) and day centers. Services are therefore the statistical units sampled. A service is defined by three parameters: its content (overnight stay, meal, reception), and the day and time at which it is consumed. As it would be difficult to list all the services consumed throughout the country during the reference period, services will be selected on the basis of a three-stage random draw: a certain number of agglomerations exceeding a certain population threshold (this threshold was 20,000 inhabitants in 2012, but has yet to be decided for 2025), then a certain number of facilities in these agglomerations, and finally a certain number of services in these facilities.

15 000