 Roedd yng nghymru ar y dyfodol y dyfodol yma? Yn y cwestiynau, ymwyntio'r OBSOS Mory oedd y cyfnodd y cyflwynffol i gyrsbwyntio gyrsbwyntio a'r gynhyrch o'r ddodol. Mae'r ddodol yma ymwneud yn ddodol, yw'r ddodol, ymwneud, ymwyntio, ymwneud, yw'r ddodol, housing, unemployment, poverty and inequality, crime, defence, education and low pay. I think that's such a fascinating list from a point of view of data because immigration. Immediately you're looking at the international passenger survey and the work of the ONS on migration stats. NHS constantly driven as we're about to hear by the need for data to manage our experience of the NHS and whether it's working or not, the endless debate about wasting times and so on, the economy for national accounts, housing, don't even begin to look at the mess of data around housing, private and public, unemployment, the labour force survey and so on and so forth. And the odd one out in that list which is being debated not far from here in the House of Commons is of course defence. Where actually the data is hard to come by and the certainties are few and far between and our understanding of what an intervention in Syria might do or whether there are in fact 70,000 moderate fighters waiting to join the club is very faint. And so that brings I think into sharp relief of value of having data that you can rely on and not just the official data so much of the ONS Government statistical service that I've mentioned but actually the prevalence of unofficial data in driving the decisions we make and our understanding of those issues that we say we care about, that we talk about in the pub. And yet with all this data floating around, Enora O'Neill put it brilliantly and her reflectors paradoxically in this new information order, it has become easier for people to make and pass on without checking in accurate and misleading assertions. And what's the balance to that? Well the balance to that is actually equipping people with the ability to check those claims to go back to primary sources and to make up your own minds as to whether they are being used appropriately and understood and what they tell us about the world and that's full facts job. As a fact checking organisation we monitor what politicians and journalists and pressure groups and business say in public debate and then we take it back to primary sources and we ask are these the appropriate sources, do they actually tell us what these people say they tell us? And we took that to its height during the election this year where King's College London very kindly hosted our 18 hour a day election centre starting at 6am every day going through to midnight starting to look at the next day's papers. We were servicing an extraordinary level of demand from journalists in particular. We worked with every broadcaster from CNN down to local TV. We worked all across the BBC. We were briefing not just on air but we were briefing programmes as they prepared to interview the party leaders and so on and so forth but also from the public. If you're on Twitter at the moment and you want to do a search for loveful facts you will see how profoundly important this is to people. We live in a time of gross distrust in our public life and actually the antidote to distrust is not as we sometimes think it is persuading people that they are wrong to be distrustful. It's giving people information to place well informed trust and that's what we try to do. Now how do you run an operation that can be run up by any major media outlet who need to know within five minutes whether the Prime Minister is being accurate in the way he's using figures or representing the law or whatever it might be or whether the leader of the opposition is or whether they can ask this question or not. Well it takes three things we've learned. The first is preparation. Thankfully or otherwise a lot of politics is repetitious and so we have the opportunity sometimes to get ahead of what data people are using and whether it's sound and do the heavy lifting before they come to us. The second is in-house expertise. We have a very talented but small team of researchers working in full fact and the third is connections and we were privileged particularly in the run-up to the election to build relationships with some of the leading research institutions in the country. So in education we work with the National Foundation for Educational Research, we work with the Institute for Criminal Policy Research, the Migration Observatory and so on. And through all this experience of understanding both the producer's point of view of the producers of official statistics of the Office for National Statistics sent us their statisticians to come and join in the election centre and see how their statistics were used and what a difference it made to present them clearly or otherwise. And working at the other end with the people whose job it is to tell the stories of that data in 30 seconds or less or in a tweet in real time while the party leaders are debating. We've really learned the value of what we've come to call data excellence. The entire chain of data from its basic management knowing where it is and how to find it and quality assurance of knowing that it actually represents the world that it is hoping to represent right through to professional presentation and being able to be relied upon to get the point across in 30 seconds or less in 140 characters or less. And that's brought us onto the eternal challenge of there are some things that can boil down in 30 seconds. And then there are some debates that you have to keep pushing out. And since we started five years ago one of the recurring things we've seen is households where people have never worked. It's an issue of great importance in an era in particular where we're talking a lot about benefits and whether we spend the appropriate amount and in the right places on benefits. This is an issue where there has been growth confusion for many years and we've been fact checking it for many years. And the data service came to mind recently when I saw a really nice small but important piece of research commissioned by JRF and done by NISA which was looking at these people and going back to the labour force survey and saying what can we know about these households and can we move this debate on a bit. And that actually exposed one of the very interesting findings from that was when we talk about households where nobody has worked many, many of them consist of one person. Many of them consist of groups of students where you wouldn't expect people to have worked. And so actually by going back to the data and finding out what it's really telling us we get the chance to move that debate on from a slogan to a deeper understanding. And that's something I was very pleased to see happening. Similarly, just this week I was at the National Foundation for Educational Research who just launched a new piece of research on another recurring and very important topic, teachers. And are they staying in the profession? And are they demoralised or are they going to leave? They too have gone back to the labour force survey and I asked them what was your experience? How did you do it? Obviously they went to the UK data archive and I said what did you think of it? I've got to come and talk here. And with real enthusiasm the researcher there said they're absolutely brilliant. You just go out and you check out the data and you can go and you can do this analysis. So now NFER have provided us with a real insight into not just what the teachers say they want to do but actually how many of them are leaving the teaching workforce and where are they going next. And we could not know that without the labour force survey and its accessibility to other researchers. So from our point of view both at the nub of having to respond at the absolute sharp end of the day-to-day business of political debate and actually bearing in mind that if you want political debate to be founded on sound information then you have to be able to provide the underpinning research that makes that possible. We've seen the work of the data archive and the data service just kind of permeating into the sides of our vision and so for that I'm extremely grateful for the chance to come and talk to you all. Thank you.