 Thank you very much for joining us this afternoon. It's been a while since I've done a webinar actually, so it's really nice to be back doing a few bits. So today we're going to be looking at the Digital Economy Act and just a snapshot of some of the datasets that are available. As part of this, we're going to talk a little bit about the accreditation process. Just because I think that's something that people may not be so familiar with, and it's really good to have an understanding of the process before you look at applying for access. So my name's Deb Wilkshire. I'm the Data Access Services Manager at the UK Data Service. My colleague today is James Scott. We'll give us a little wave. He is one of the senior user support and training officers. I want to just give a shout out as well to Simon Whitworth from the UK Statistics Authority. Now, he can't join us today. Unfortunately, he's at another event. But we'd just like to express our thanks for his input into the content of today's webinar. I think most of us are probably aware of this, but it's worth just recapping. So we know there are various pieces of legislation that come into play when you're looking to access personal data or personal information. And which legislation applies will depend on where the data is coming from. Now, in practical terms, if you're applying for data, you may not need to know a great deal of information about that legislation. One of the main important points is that actually the data protection legislation is actually what supports and allows us to provide responsible access and use of data for research purposes. Today, we're going to focus specifically on the Digital Economy Act, and we're going to have a look at some of the data that are made available through that gateway. But just to be aware that there are other pieces of legislation that might be relevant. So the Digital Economy Act or the DEA came into being quite recently, so 2017. And this really aims to provide certainty and clarity for public authorities and researchers looking to access data for research. And the idea was to create a gateway for public authorities to access data in the public interest. Now, the term in the public interest is a very important one, and we'll come back to that. Now, the aim of the Digital Economy Act really is to provide a single gateway for secure data access. And it's really looking to replace the need for multiple gateways, which is what we've had over many years. Because this can be a very complex and lengthy process to navigate for researchers, and it can actually be quite confusing. So the aim is to replace this with a single gateway. Now, we're not quite there yet. For example, health data collected by public authorities isn't collected at the moment, included rather at the moment. But this is the aim. This is what the legislation is trying to achieve. So the gateway uses a trusted third party model where an accredited processor will link, de-identify, and make security available. Sorry, this is the typo at the end of that sentence. Secure data, and then make that available to an accredited researcher for an accredited project. So I'm going to dig into some of these terms a little bit more. But just to give you a little bit of background, so the Digital Economy Act gained royal assent in April 2017, and it made substantial amendments to the Statistics and Registration Service Act. Now, it has a number of governing principles on disclosure of data, and these are around the principles that you can see, so legal, ethical, etc., etc. The important bit for researchers is to be aware of this accreditation criteria. So these are our processes, projects, and researches that I mentioned. Now, the UK Statistics Authority, I mentioned them a little bit earlier, they are the accreditor, and there are a number of accredited data processes of which the UK Data Service is one of them. And we became an accredited processor back in March 2020, and we're just undergoing our first year review of that accreditation. Now, an accredited processor's role really is to carry out data linking, matching preparation, and or provide secure research facilities. The accredited projects and researchers. Now, if you've already accessed data through the ONS, you will be familiar with the approved researcher scheme. And the accredited projects and researchers is broadly consistent with that. One of the main differences is that ethical approval is now sought under accredited researchers. So here's where we get into the practicalities of applying for you as researchers. So the application process is a multi-stage process. At the moment, you must be based in the UK to apply. And there are two stages, really, so you need to become an accredited researcher. So that involves meeting the data owner's criteria and also attending a short training course. Then you will need to submit a research proposal. Now, we'll look at these in a little bit more detail, but essentially, you've got to show that your research is going to have a valid statistical purpose and that it's feasible. So to become an accredited researcher or an AR for short, you will need to submit an application form and you'll have to meet this accreditation criteria. Now, the UK Stats Authority manages the accreditation process. The good news is that AR status, once you've been given it, lasts for five years. So it's not something that you need to be doing for every project or every year. That status will cover you for five years. The other good news is that you can use that status across all accredited DEA processes. So if you go through the process with us at the UKDS and then later down the line, you decide that there's some data at HMRC that you want to go and apply for, you take that AR status with you. You don't have to go through that process afresh for HMRC. And we think that's a real bonus. Now, as part of becoming an AR, you need to agree to your name being added to the UK Stats Authority website. Now, the accredited researcher criteria, I'm just going to include that because I think it's quite important to make sure that you are able to meet the criteria before you apply for secure data. So this is the criteria for the AR status. So you need to have an undergraduate degree or higher, which includes a significant proportion of maths or statistics. So there has to be some element of quantitative work. Or you need to be able to demonstrate at least three years quantitative research experience. The next stage is you have to successfully complete a safe researcher training course and I'll tell you about that in a moment. As I say, you have to agree to be included on a list of all accredited researchers. You have to agree to publish the results from all of your research projects completed through the accredited researcher scheme. You have to sign an adhered to a formal accredited research documentation, so that declaration. So that's the criteria. Now, the safe researcher training is, it's actually just part day now. We used to do this face to face obviously in the days before COVID, but we've now moved the training on to Zoom. So it lasts about three hours, three and a half hours depending on the group. The premise of the training really is that using sensitive controlled secure data, however you want to determine it, is largely about common sense. And researchers will have this common sense and we know that. However, there are some areas where specific knowledge about disclosure risk and how to mitigate it is required. And not all researchers will know that if they've not experienced working with secure data before. So this really is the, the aim of the training is to make sure all researchers have this specific knowledge. So essentially the course will look at the wider context. So we'll talk about how we understand data access. We talk about the five safe framework. We always have to talk about what happens if things go wrong so that people are aware of their responsibilities as a researcher access insecure data. We talk about what we mean by being a safe person and we talk about the research community and how, how that functions in this context. And then we move on to the more technical knowledge. So this is the section where we talk to you about statistical disclosure control, what that is and how it's applied. And it really varies with the people that we train some will come and they will have some knowledge of this already for others it will be completely new. I think most participants come on the training actually find it quite interesting which is which is good because obviously talking about this, some of the stuff can be quite dry. But people do tend to find it interesting and they go away. You know, quite happy with, with having gone through the training. At the end of it, all attendees have to take and pass an online test. And that's important. So it just turning up at the test isn't enough you have to pass the test in order to get your accreditation. So moving on to projects. Project applications will require approval by data owners and by the independent research accreditation panel or wrap. So we have a lot of acronyms so we tend to forget and throw those, throw those around a little bit. Now project applications have to be fully completed. So these aren't things that you can do in a quick five minutes over a cup of tea they need to be thought about. Because the, the rap will pay close attention to what you write. And what they really want to know is, are you meeting the public good, are you meeting this, this criteria. Can you demonstrate that. They also want good evidence of ethical approval. Now, most academics will be familiar with their, their institutional ethics processes. And most academic projects now have to get ethical approval from the universities or their institutions. The additional requirement here is that project proposals will have to complete a UK statistics authority self assessment ethics form as well. Now, some project proposals will require a bit more ethical scrutiny and those are reviewed by the NS deck. So they go over the ethics forms in great detail but it's worth just introducing you to the framework that the UK statistics authority use. So they've developed an easy to use framework for researchers to use to assess the ethics of their projects. I actually think this is quite a useful framework. It's a good tool I think to start to think about assessing your, your own projects regardless of whether you're applying through, through us for data available under the DA. And the idea of the framework is that it helps you to identify and think about how you would mitigate any ethical issues. I'm not going to go through them all because, you know, we don't really have a lot of time to do that today, but it's worth noting that there are six main principles and each principle is split into a number of items. So there's 22 items in total and the, the self assessment form will guide you through each of those 22 items and it will ask you to identify any potential issues. There is a lot of guidance on the UK statistics authority website and I can pop the link to that in the chat once I hand over to James. I want to just mention the research accreditation panels because I think they play a really key role. Now, obviously, wrap is the panel that assesses the applications for the Digital Economy Act. But if you go through other access pathways for, for data available through other legal gateways, you will meet their own individual research accreditation panels or approvals panels. Whatever panel reviews your application, they have a lot of similarity, so their aim is to review and approve your application. And they're tasked with doing that in a transparent and fair way. They meet periodically, most of them certainly all of the ones I know will meet monthly and I think that's, that's fairly standard, but just be aware that there might be some differences if you're going through other routes. And essentially their task is to look at your project, look at the information you've provided in your research proposal and make a decision. Is it legal? Is it ethical and is it feasible? So when I say about your research proposal needs to be filled in very comprehensively, this is why there has to be enough information for the panel to make a properly informed decision. Now, again, what materials you have to provide might vary a little bit, but these are the standard things. So your project application will need to be provided to them and your ethics assessments will need to be provided to them. And occasionally there might be some other supporting materials. Now, an outcome for a research approval panel or accreditation panel. You have a number of potential outcomes. So the best outcome for everybody involved is that there is a full approval at first go. So they have a look at your application, your research proposal, they're perfectly happy with it and they can approve it straight away. Or it might be that they will approve it, but on condition that a few amendments are made. There is, of course, the potential that a proposal might be rejected. It is actually rare. We don't see that very often. And I think the reason we don't see that is obviously because we do a lot of work with researchers before the panel see the research proposal to make sure that it's a good quality. So if you're applying for the UK data service, our access team will triage all of your applications before they are sent to the wrap. And the outcome really from doing this, this triage is that we don't get many applications that come back with anything other than a full approval. Now, full approval at first go is obviously the aim for all of us because if we have to do a lot of to and throw in and it has to go back to the panel for a second time, then obviously that will take a little bit of time. Okay, I'm going to hand over to James, give you a break from me talking. Okay, have you stopped sharing your screen? Okay, so we just want to have a look just throw an example really of some broad areas. Yeah, we kind of wanted to talk a little bit about some of the data sets that are available under the DA. And we thought we'd split them up into these five broad areas and give you an example for me just so that you get a reasonable spread. And clearly, this is a tiny snapshot really, absolutely tiny snapshot. But anyway, we thought this would be a reasonable way of doing it. So I'll start with the UK innovation survey. This in itself contributes to the Europe wide community innovation survey. And it's the main source of information on business innovation in the UK. And what it does really is it looks at general business information, good services and process innovation and the context for that innovation. It's based on firms with more than 10 employees and there are approximately 16,000 enterprises. And an enterprise is defined as being active in innovation. If it introduces a new or significantly improved good service or process. Excuse me. So it could be engaged in an innovation project or innovation projects that are not yet complete or in longer term innovation activities such as basic research and development and things like that. So, yeah, that's kind of gives you a little flavor of that one. And actually, what we're going to do the next few years also give you a few brief examples again is the tiniest of snapshots really about just to show the breadth of some of the publications that come off the back of using these data sets. So you can see here, this is literally just a tiniest of snapshots, but you can see here that we've got stuff about financial constraints on innovation. And we've got one on economic change in the labour market, the role of opening innovation and companies motivations and private and external benefits from investment and intangible assets. And, you know, there are obviously many, many more than that, but it gives you a sort of broad idea about, you know, what, what you can do with it. Okay, the second area that we're going to look at. Oh, sorry, I should say the second date the set we look at is the, it's what we call the ash, really the annual survey of ours and earnings. This is a 1% sample of individuals from any records about 140,000 to 185,000. You can construct a panel data set which obviously is quite useful. The employee completes so arguably having some people argue that it's more accurate than some other surveys. It includes data on wages, paid hours of work, pensions, arrangement, age, occupation, industrial classification codes, and more. And actually the low pay commission use ash to provide evidence on the impact of minimum wage, which obviously is, you know, fairly central for evidence based recommendations. So again, you know, it's a very useful data set to be able to access. In terms of ash based sort of publications, you know, there are a few here. Again, you know, tiny snapshot. So while on public sector results allocations, it's the financial crisis. That's a fairly recent one. Another really recent one, rising pay for performance among higher managerial and professional occupations in Britain is eroding or enhancing the service relationship. And then there are the ones on living standards, poverty and equality and productivity investment and profits, you know, going back a little way. So you can see again, you know, quite a broad range, even just within that tiny snapshot of four there in terms of what people have been doing with it. Do excuse me. Sorry. So thirdly, though, what we're going to look at the five is what we call the words, the workplace in employment relations. So this is a cross is three cross sections, basically. So it's a cross section survey of managers. And that looks at things like recruitment and training, consultation and communication, employee representation, pay determination and payment systems, grievance of discipline, equal opportunities, work, life balance, health and safety, all these kinds of things. Secondly, includes a cross section survey. Sorry. Other employee representatives. And that includes information on the structure of representation at the workplace, time spent on representative duties, means of communication with employees, incident serving negotiation and consultation over pay and other kinds of things. And their involvement in things like redundancies, discipline and grievance matters, that kind of stuff. Okay. So, so that's the one for employee representatives. And the third one is a cross section survey of employees. Okay. And that third one is, you know, very much around sort of things like working hours, the amount of influence you have in your job and your job satisfaction. You're working arrangements, training and skills, information and consultation, pay that kind of stuff. Okay. So it's coming out from several different angles. Oh, God, excuse me. So yeah, again, you know, really useful, really useful though. And in terms of some of the research that's come out of it, you know, we can see here that there's different perspectives on multi-skilling and product market volatility, disability and earnings. You know, is it make a difference around, you know, the employee characteristics, work place performance and worker commitment and loyalty, gender pay gap. And even just within that fall, you can see that there's a fair breadth involved with that, you know, with what's being done with it. And I'm sure you as far broader than that, that's really just tip of the iceberg stuff really. The fourth one is a crime survey for excuse me, crime survey for Indian and Wiles. I started in 1981 and it's a victim survey. So it really asks whether the response that has been a victim of crime in the previous 12 months. Children aged 10 to 15 have started to be included from 2009. So the whole set of that data is held separately under its own catalog record. But after that children's data included in the main data set. There's some good stuff on the ONS webpage about the crime survey for Indian and Wiles, if that sounds like something that you would be interested in. So, you know, it's adult self completion. You know, there are mod modules on drinking behavior, drug use, stolen goods. The geography is low level as well. You get hate crime variables, interpersonal violence. You know, this can be domestic violence, sexual victimization and anyone that's done the SRT calls to another. We include an example around this. So, yeah, that's that's kind of gives you gives you a sort of overview of the main main aspects of that. And finally, there's, well, we all know what this is because we probably all did it in the last few days. We all know that was really, I think. So, you know, we know it happens every 10 years, you know, it goes to all the households in Wales and England and helps organizations to make decisions on planning and funding public services. I imagine that as with most of them people have been putting various sort of protest kind of answers in. I think it was the previous one that I had a lot of people, but they're really not a lot, a significant minority people, but they're religious Jedi, which is picked off. And I think a lot of people put in their nationalities, European at the moment. So yeah, but yeah, it includes stuff about transportation, education, health care, you know, obviously it's a really important thing to do for so many reasons really. So yeah, we have some other way. Do you think you might be able to do the last couple of slides? Thank you, let me. Can you just move the slides on for me though? Oh yeah, I can do it. Update anyway. And that was the last one. It's just the helpers. So we had a question about if you are in a position such as being a data protection officer or something similar. Do you still have to undertake the safe researcher training? Now, actually you could thinking about it, you can read that question a couple of different ways. So I'll answer it a couple of different ways. So if you wish to apply for data available under the DA, then yes, you will have to undertake the safe researcher training as part of the accreditation process, regardless of any previous training or job role that you would have had. If you look at it and think do you need the training to be something like a data protection officer, officially no. However, it actually probably would be quite a useful course to do for that. So it might be worth doing a safe researcher training course if that's something that you're looking at doing. So how does one negotiate with a DA processor and data provider to enable a... So how does one negotiate with a DEA processor and data provider to enable a data provider to make data available for a project? Good question. So in terms of the provider, in terms of the data owner, that's part of the various feasibility checks that the processor would do. So for instance, if for each, I think you mentioned earlier each application that comes in, the processor, so yourselves or O&S would be checking the project in terms of if there's a legal gateway. For instance, is the DEA the right gateway? Is it feasible to do that research? Is the data owner approval for that research? So we would normally go through the various feasibility checks, make sure that the application was being completed effectively prior to doing the data owner approval. I suppose it's very much up to the researcher themselves which processor they would wish to select and sometimes that's dependent on the data set. So for instance, we O&S holds certain data like the longitudinal studies that's only available in the SRS or Secure Research Service. Whereas there'll be other data sets that the UK data archive holds that the O&S doesn't. So it often depends on the, you know, say those data sets and often then it could well be a discussion if people want to link data sets that have multiple data owners and those data sets aren't already available as a linked data asset. Then there'll be, there may well be discussion between the processors as to where that research perhaps will take place in terms of those data owners. So that is sometimes, you know, a discussion that may well take place and often those discussions might happen before an applicant submits their application. So I think it's often sometimes done on the case-by-case to say depending on where the data set is held and likewise, you know, it depends sometimes on the type of user. So most government access may well take place in the Secure Research Service because that's very much sort of where our digital focus has been. But we do, we do have researchers from across all sectors, but to say it can depend on the case-by-case. I don't know if there's anything you want to add, Bill? I think Bill's on the call as well. Hello, can you hear me? There you are. Yes, thanks, Bill. Hi. So after the big build-up, I know I think Nick's covered that one. To be honest, I've probably not got much more to say on that one. Thank you. Sorry, Bill. I didn't announce you at the end. I hadn't realized you'd joined us. So that's fine. I was late. Don't worry. Apologies for not giving you the grand introduction. There is another one actually, though, that Bill and Nick, you might be able to answer, is to look at the link between, I'm presuming that's the ONS and ARD UK and UKDS. Would you be able to clarify what the link, if any, is between UKDS, ONS and ARD UK? The link between ADR UK and ONS, I think, is very clear. We are part of the partnership. ADR UK is a partnership. The trusted research environments and the devolved administrations are all part of that partnership. And ADR UK fund significant funding for the SRS, so they've helped us to grow the service. I guess the common link there is that the other parts of that network are all accredited under the DEA as data processors, and I guess that's then the link into the UK data service, or the data archive to be precise. Nick? Yeah, and I think just to add to what Bill said, you're right, Bill. So the ADR UK, which fundamentally is managed through ESRC, the Economic Social Research Council, and I know there's obviously a funding connection there in terms of also with UK data service. But they also fund, as Bill mentioned, to develop and enhance our service. So, for instance, the SRS, we've done additional enhancements in terms of the capacity and the performance of that. Likewise, we've developed a research accreditation service, which some of the people in the call may have used, which is our online system for people to submit applications to be an accredited researcher or to submit accredited projects. And then once your project's live, you can manage it through that system. And we are sort of putting further enhancements in place whereby that system will host any application to be a researcher or for a project for a DA processor. So we're engaging with the other program. We will be engaging with other processors to see if they want to use that system. And also, as you know, those enhancements, but particularly around bringing and acquiring and curating new data, linked data sources. So some of you on the call may be aware that the data first program, which is managed with the Ministry of Justice in terms of bringing in linked courts and magistrates data and subsequently prison and education data. So there's a program I've worked with you to bring in a number of additional linked data sets to support further analysis. And I think that's been instrumental in enabling us to bring in about 20 new data sets, both linked and additional COVID data sets over the last 12 months. Thanks, Bill. Thanks, Nick. Just one other comment on that is the ADR UK and UK DS are both funded by the Economic and Social Research Council. And thank you, Richard, for that one. At what point would the project need to achieve Research Ethics Committee approval? Is that after the project has been approved or before? Is the REC approval from a uni sufficient here or is there a specific REC approval from the data owner which will be needed? So you will need to provide your university approval and you will need to complete the UK Statistics Authority self assessment ethics form as well. These have to be done before the project goes to the RAP for approval. So that needs to be submitted at the same time ideally as your research proposal does. So if you're coming through the UK DS, the ideal would be that you send your research proposal and your ethics forms at the same time. And I would imagine that that would be the preference if you are going through the ONS SRS as well. It makes life a lot easier if we have all the paperwork up front and then we don't have to keep going back and forth to chase various bits up. Somebody's commented that they can see a list of projects on the UK SA site. Is there a list of accredited researchers? And if so, where are they? I can answer that if you want to. Please Nick. Yeah, sure. So all projects that have been accredited under the Digital Economy Act are listed on the UK Statistics Authority website. I'm happy to send a link to them following this meeting. And recently on that same landing page further down. And you can also find a list of all accredited researchers. So that's been added quite recently. I think we're probably going to be another iteration of that in the next few days because we've sent the list to the Statistics Authority in terms of the list we maintain. And that includes certainly everybody, I think, has been accredited by ourselves. So an SA or the UK Data Service and those applications. Similarly, I think it includes one or two of the other processes where as Deb said earlier, the desire is to recognise researchers from different processes, but providing that there's consent from those initial researchers that we get those lists from the relevant processes. So where we have that then that sort of transfer transfer really between the recognition. If you are accredited under UK Data Archive is set to via this you apply to them and then that you are accredited. Then you'll be recognised for the DA and you'll be able to use the other services from across the DA. So that list is maintained on the Statistics Authority website and I can send a link for those to Deb after this meeting. Brilliant. Thanks Nick and then I can send them around after the event. Just a quick question from someone about training. So in the form of time does the UK Data Service intend to offer a series of courses on access, which could become a required element of research training? Yes, I have to say we have been discussing this and I think it would be really, really useful for people, especially people that are coming to this process you and don't have a lot of experience or background with secure data. I think it's essential that we do that and we will be doing that over the coming months. So really at the moment we've just started to dip our toe into that with this webinar and the previous webinar I did last month where we talked a little bit about the application process. But yes we certainly do intend to do some further training on that. Whether it becomes a required element of research training, that's probably a question that's sort of above my scope really, but I think it would be an excellent element to all research training if I'm honest. I think especially as more and more researchers are finding that they want to be accessing these data. So I think more training is definitely the way to go. So do keep your eye on our website for future events coming up. Okay, so I think the last question we've got, again, Nickham Uprade, Bill, Nickham Bill, it might be back to you. A DA process may provide linkages, etc. between data sets. Data linkage slash entity resolution is a complicated area which has an implication on results. Are there any plans to expand on the documentation of the methods of linkage? Okay, shall I take this on Nick? So I mean I think it's one important distinction to make is that when we talk about accreditation under the DEA, you can be accredited for essentially hosting the data and then that's what the SRS is accredited for. And then there's another type of accreditation which is the linkage using identifiers essentially. So the SRS won't hold identifiers. We can't do the kind of linkage that can be done elsewhere within ONS because a separate ONS has been accredited for that purpose. I just mentioned that. In terms of guidance, I don't know. What I do know is that internally in ONS we've recently set up a sort of, I can't remember its name, but it's like a centre for linkage. And I think this is really a recognition of the fact that there is increased demand for this kind of work. And what would imagine that as part of that, then guidance is clearly going to be part of that. And the only other point to make really is linkage. We don't really offer a service where we do linkage for individual projects because, I mean, as the questions kind of alluded to it, it's a kind of very specialist area. It's kind of very resource, you know, intensive takes a long time. The model really under the ADI partnership is that where there is demand for two datasets to be linked for sort of multiple research use, i.e. not just one project, but if you can demonstrate that this dataset will be used over again, then that sort of feeds into the process and we'll do an assessment and hopefully be able to meet that demand. The other thing I would say is there is an awful lot of demand for linkage within ONS at the moment. So I hope that's answered the question. Okay, no, I think we've answered all the questions. So I will close for the afternoon then, but I'd like to say thank you very much to all of you who came along this afternoon. I hope you found that useful. Any comments after the event then obviously please do drop us an email and we'll try and help in any way we can. So thank you to James for joining me. He's put his picture on so I'm hoping he's still okay behind the picture. I'm alive. Perfect. And a big thank you to Nick O'Donnell and Bill South from the ONS for all your input this afternoon and for answering those questions. Enjoy the rest of your afternoon everybody and I'm sure we will speak to some of you as you go through your research in the future.