The social domain has seen a growing debate on the effectiveness of publicmeasures. The realisation is taking shape that the legitimacy of socialpolicies such as the socialisation of vulnerable groups, anti-radicalisationprograms or regeneration projects in deprived neighbourhoods has to gobeyond the intuition that they `should work. But unlike methods in spheressuch as medicine and healthcare, social programs typically have no sharpdemarcation in time, intensity or target group and are implemented in arich context of unforeseen and unknown variables. This makes it difficult(if not impossible) to assess their impact with research methods thatcentralise a monocausal effect.