SFI and Rambøll conduct an evaluation of two different but interrelated projects in the period 2014-2017. The implementation and anchoring project is based on results from the National Homelessness Strategy, which was conducted in 2008-2013. In addition, The National Board of Social Services conducts a development and testing of interventions for homeless youth and youth at risk of homelessness (The Youth Project), which is also included in the evaluation. The project contributes with knowledge on interventions for homeless people, especially about which interventions work best for whom, i.e. different target groups among the homeless and how these interventions are best organized and implemented in the municipalities. For both projects the evaluation includes: 1) an implementation and fidelity evaluation 2) an impact evaluation, 3) an economic analysis, and 4) a method handbook. As the Youth Project is a model and method development project, a more formative evaluation approach is used in this part of the evaluation. The formative approach consists of a continuous focus on how the professional development of the model is strengthened and stimulated through the collected data. The implementation and fidelity evaluation involves the latest international research on the importance of fidelity in the use of floating support methods. Measurement of fidelity on program level is connected to individual data and economy data to illustrate the importance of fidelity for the effect of the methods and the economic gains. More extensive qualitative studies in the municipalities of the implementation processes and organizational factors that affect fidelity and efficiency are conducted. The effect evaluation helps to strengthen the evidence base for the use of the methods in the context of the Danish welfare state. The analysis is divided for both projects on target groups, to illustrate how the effectiveness varies with for example with age, addiction problems, and mental illness. Thus, it is analysed which interventions work best for whom in relation to the complexity of citizens' support needs. For both projects an economic analysis is conducted based on the approach Rambøll and SFI developed in the implementation of cost-benefit analysis of the Homelessness Strategy. The economic analysis includes the costs of a floating support intervention based on actual measurements, while the analysis is based on registerdata with the use of a control group to measure the cost effectiveness of the floating support methods in relation to use of public services, treatment and other services. A methodology handbook is prepared. The handbook will be targeted local practitioners and decision makers. The handbook unfolds the specific conditions for use of the methods and the associated organization and collaborative relationships that are important for the methods to optimize their effects.