 ..to start by raising the issue about impact. Impact is something which is part of the way of life of higher education at the moment and for very good reasons. It is also a thing of government. Felly, we are particularly interested as the UK data service in understanding how data which we hold on behalf of others is used to create impact, because we understand how we sort of understand how we as an organisation have impact both with other data services in the curation and the dissemination and the provision of access to data. But what we don't tend to know about, or know about quite so well, is how the data that we hold on behalf of others is used for impact in some measurable way. So I'm going to give the, since the UK data service is funded by the Economic and Social Research Council, I'm going to use the standard research council's definition of impact, try keeping this in your mind as we go through this afternoon. They define it as the demonstrable contribution that excellent research makes to society and the economy. So this isn't about academics talking to each other, this isn't about government, local or national governments speaking to each other, it is about the contribution that research generally makes upon society and the economy. So these are multiple layered, impact is multiple layered. The ESRC defines impact in three ways. It talks about instrumental impact which is about influencing policy or practice or service provision. It's actually getting stuff done. Conceptual impact which is contributing to the understanding of policy and issues around policy. But also impact is about capacity building and skills development. And this is something which tends to slip off the edge of the impact discussion. How do we help other people, how do we collectively help other people learn how to do things better? And large investments which are funded by the Economic and Social Research Council have had the opportunity in the past to bid for extra funding to invest in impact officers. And we as the UK Data Service has one and that's Victoria Moody who's sitting over here and she's the linchpin behind this event. And one of the things I hope to get out of this and she hopes to get out of this is more good examples of how data is used to provide this impact. It's not just the impact from the UK Data Service holding data on behalf of government departments. So I suspect I need to tell some of you about what the UK Data Service does. And I'll do that very briefly. The nearest obvious analogy to the UK Data Service that everyone will clearly understand is a library. But there are loads of difference, a data library. But there are loads of differences. We provide an end-to-end service which benefits all of the key players in the data life cycle. And sometimes the benefits that we provide to those key players is a little difficult to articulate. But essentially most of the government departments in the UK and in Scotland provide to the UK Data Service survey or census or some data which we curate and provide access to for research. And I really mean for research in its broadest sense. I'm not just talking about research within higher education. Most of the data that we hold is data which can be used by anybody and for any purpose. But it's not open with a capital O. It's not always open with a capital O. So we acquire data from a wide range of data creators, national government, local government, researchers that are funded by the ESRC and international government organisations. But we also deal with the third sector and commercial organisations as well. And we benefit the data owners by providing access to those data on their behalf. So there's a legal relationship between the UK Data Service and the data controller. And there is also a legal relationship, sometimes very minor legal relationship between us and you, the potential user or the potential researcher. All data owners need to do is to make sure that data comes to us and to finalise the legal side. When we receive data we will carry out all manner of checks in terms of both quality control and in terms of risk assessment in terms of disclosure control. So no post codes in survey data, no pregnant men, all the usual problems with data. We will check those before we make them available and have a dialogue with the data owner in order to make sure that what we're giving out is the right thing. And there's no cost to the data owner in this. We provide this service for nothing. The ESRC pay for this because they believe that the sharing of data is important and that the investment that is made by government and the ESRC and others in the production of data is something which is greater than the cost of making it available. So it's an additional part of the cost, the original investment is for our service to make these data available. In the workflow we then ingest data into our preservation systems which also provide, which is also modelling to provide access to these data. And we ensure that anything which is passed to us has the potential to be maintained forever. So we're a digital preservation activity as well as a social science data access and distributor. And for some data we provide online interfaces, for some data researchers have to come and use our systems and our hardware because they are so disclosive or there is so much risk of disclosure. But in general most of our data is available through a straightforward registration system. And we tailor access mechanisms really to suit the needs of researchers rather than based on the type of data that we're dealing with. There's more than 7000 data collections that we hold and what we're doing is to try part of our scope is to make sure that these data are properly described so that they can be found and used with very little additional effort on behalf of the researcher. And we have a question bank for example and it provides access to the full text of over 700,000 questions asked in various different surveys over about the last 45 years. All of our data collections are version DOI'd and so there's easy to use citations. We believe very strongly in the reproducibility of research and in total we have something more than 7000 data collections in one place so all accessible through the same access point. The UK data service is also one of the UK pioneers of providing secure or controlled access to data, access to data which has a known risk of disclosure and data which certainly can't be made open within secure environments. We run industry standard software in order to provide remote access to researchers not just in higher education to personal and sensitive data and researchers have to go through a standard process of being approved. They have to be trained, the projects that they undertake have to be vetted and we've just forgotten the word, we've been pioneering a framework called the five safes as five areas where we can try and guarantee at least to the best of our ability that the data controller is not in any way breaking. Laws in making sensitive personal information available. There's a lot more in terms of what we do as an organisation and we are a service. We're providing a service to both data controllers and to data users. We're strongly influenced by what users want us to do but we are funded on a five yearly basis in order and we have to plan in five year cycles. But part of our remit is to talk about the benefits of data sharing and the impact of data use. There are many other activities that we are involved in within the data space. They vary from dealing in high level into international groups looking at the ethics and the legalities around the reuse of big data. We also have members of staff who are looking at building an open data platform, a platform which holds and provides access to a much wider range of materials but using it through pathways that we're legally able to provide. Part of the purpose of this event is not just for us to say that there are huge and demonstrable benefits associated with having a robust and end to end data service infrastructure. We're going to hear about how others use data to create impact and to create positive and successful interventions in society and the economy. Doing it in a way in which individuals' rights are protected and individuals not only are the rights protected but personal information is respected and personal information is not just about people, it is about organisations and companies as well. But I think that the primary value of the data service infrastructure is the influence that we can have across the whole of the life cycle of data, the whole of the life cycle of data from its creation to its original primary use to its secondary use. I think we need all to step up to make sure that the ideas around reuse in data should be put as firmly as possible into the planning part of the capture of these data in the first place. Running a survey without thinking about how it may be made available at the end is now something that nobody should be tolerating. The management of data creation within higher education is now pretty much based upon data management planning so that it's clear that these data can be reused after the primary purpose for their creation has been spent. So impact to me is truly an economic impact and how can we do better research using the data that we've got but using the data that's been created for another purpose.