Impact analysis is concerned with identifying the additionality resulting from an initiative. Many initiatives designed to help people through welfare to work have been evaluated in this way. Inclusion promotes methods designed to test the impact of initiatives when all other factors are taken into account.
There are a number of different factors that reduce the direct, observed impact of an initiative. Some of these are more susceptible of analysis than others. The one which is most commonly analysed is deadweight. This is commonly referred to as 'would have got jobs anyway', although the alternative is more often people being helped by regular provision rather than the initiative. This is tested by analysing the outcomes for a group of people who are largely similar to those actually helped by the initiative. In order for this factor to be part of an evaluation, the 'control group' should be selected at the same time and researched in the same manner as those who participate.
A common method of selecting a control group is to have a control area (or several) in which the initiative is not offered, but this poses additional issues of matching the area's labour market characteristics as well as the individuals. An alternative is the use of random assignment, where people in the same area are randomly accepted onto the programme, and both those accepted and those refused are researched.
Other aspects that need investigation in impact analysis are leakage, where activities designed to help residents in an area in fact benefit other areas, substitution, where people who are helped substitute for other people who would have benefited in the absence of the programme, and displacement, where the initiative displaces economic activity from unsubsidised to subsidised without expanding the total. Some of these issues can be minimised by careful design of the initiative, but evaluation needs to test the success of these methods.
Previous projects using these methods include:
