 Information awareness raising campaigns have increased in popularity and more and more funding is being spent on these campaigns but there is no rigorous evidence that indicates that they are having their intended outcome. And so what we're trying to achieve with this conference, we want to take a step back and we want to reflect on what are the underlying assumptions of information awareness raising campaigns and what is the evidence behind it. And if we don't have any rigorous evidence behind it, how can we possibly create it? Several years we did a systematic literature review about the studies that are available to assess the impacts of different information campaigns and we found very little, we found very few evaluations that are rigorous in nature and now two years later we are here discussing with donors, discussing with researchers and reviewing the evidence that has emerged since. I think information can change behavior and we looked at some of the lessons today about settings in which that has happened. We still have a lot to learn about how we can sort of influence things more effectively. One of the keys to driving behavior change is understanding what information people already have and what some of the determining factors are and how they make decisions. I honestly believe that migrants are helped mostly if you provide objective information because migrants can decide for themselves whether they want to come to the EU or not. And I think it is our job to explain whether the expectations that migrants have or potential migrants have are actually realistic or not and then they decide whether they will come to the EU or not. There's lots of decisions that are taken in the course of migration but also perceptions of those who migrate and those who receive migrants. So the idea is to understand how people form beliefs, expectations, perceptions and how those are transformed into decisions. One way we can do so is to think about randomized evaluations and evaluating the impact of our information campaigns. So setting up large scale trials that examine how people who have received information vary from those who have not and sort of seeing over time whether that changes the way people behave. Another thing we might test is the different ways in which people receive that information and comparing whether receiving it in one way or in a different format might be more effective than others. So this is precisely what we do. We use experimental methodology, the same kind of methodology that we use to test new medicines. So what we are doing is we randomly assign these different policy interventions to different groups of people that are potential migrants and we follow them over time. We measure what is happening and this is the most rigorous possible way that we have to measure the impact of this intervention. What we are doing at the moment with the support of the European Union is we are testing different policies to try and provide alternatives to these potentially irregular migrants. The question now is that we need to tackle how do campaigns work in different contexts and which kind of campaigns work better than others. So there is a lot more work to do as we go forward.