 L if investigating completed it matters in any area because it allows us to find out uh... if we are having positive effects and positive impact as opposed to negative impact an also to make sure that we're not having any unintended consequences either positive or negative now that's important because andya's first of all we are often spending other people's money but private and public money in the pursuit of of initiatives which we believe will improve the climate, will improve forestry, air quality and what have you and we really need to know whether we are having those effects and we're not having negative side effects or unintended consequences. It's also important so that we know when to adapt that our policies often need tweaking or we need to change direction so that we don't keep on doing things that don't work and we start doing things that do work and that are effective. We should understand that when we do impact evaluations we're always working within the scope of probability and our aim is to make sure that we are as confident as possible that our impacts are positive and having societal benefits. We can never be absolutely definite because evidence is never definite. It's more than belief though because we establish the strongest designs we can so that for every intervention we have we also have a comparator or a control group and that allows us to see whether the initiatives in our test area are different significantly different from those in our control area so that's how we build up the knowledge that something is working it's not just working in and of itself it's in comparison with doing something else usually in a very closely defined and nearby area. We're learning a lot about how to get evidence into policy one of the ways to do that is to try and get policy involvement from the beginning is to be asking questions that are not just scientifically interesting and we need those but also that are scientifically interesting and have a connection to policy makers and to practitioners what are the problems they're dealing with either with policy development or with running an NGO or actually rolling out an initiative so I think some collaboration between the policy practice community and the research community at the beginning helps then we need to be independent of them it's very important we have to do this to independent standards that we're not interfere with by donors by policy makers or practitioners they may be on our advisory boards and what have you but the independence is absolutely crucial at the end of the day we've got to take very technical evidence and convert it or translate it into messages and languages and presentational formats that are useful to policy makers and practitioners they will not want table after table after table or graph after graph of data we come and examine that at events like this we need to have some take-home messages we need to have some one-page summaries for our ministers and senior policy makers good well-written three-page executive summaries for administrators of implementers and usually a version of our final reports that are readable in non-technical language of about 20 to 25 pages and that means we have different products for different audiences but you have to work evidence you have to take it and get into dialogue with people the word is to engage it's not dissemination it's engagement if you don't engage with the policy practice community you're not going to get evidence into policy or practice that engagement is very important