 Hello, welcome to the ANS webinar. I'm sorry about that technical delay. I'm Frances Watson and this webinar will provide an overview of the main elements of an institutional research data policy and suggest some approaches to creating a policy that supports the implementation of research data management. I will use examples from policies that are currently in place in Australian universities. Now, in order to do this and to make it, I think it's more effective, I'm going to go in and out of the presentation to websites. So if I miss a button and we have to go back over a slide, I hope you'll bear with me. I understand that ANS has provided funding to a number of institutions to implement seeding the commons projects and one of the key deliverables in those projects is a data management, a statement of a data management policy. Now, I'm assuming that the people who are involved in these projects may have had policy experience but they may not have. So I've prepared this webinar to take into account that this is a basic overview and really not one for an experienced practitioner. So now we'll start. I'll take the questions at the end of the presentation because I have to admit to you, I'm not particularly experienced with webinars and so it's probably easier for me to deal with the questions at the end. Thank you. Hi. Now, here's the standard definition for data management. It lists all the different activities that you have to do to protect, deliver and enhance the value of data and information sets. Why the university, why managed data? The university of Bath says now the management is organising and preserving research data so it can be available for reuse. Now what I've done during this presentation is I've put links at the bottom, references and links, so that when you get your copy of the presentation you'll be able to just click on these things and see what other information there is available. When I started with ANS in 2009 a review of the data management policy showed that essentially research was done, as you all know, the researcher did this research, analysed the data and materials and produced a publication. After that the data and material were attained for the period required under federal and state legislation and were then either left on the shelf or disposed of. Since then ANS has introduced the concept of research data management in the Bath quote, the aim is to encourage data to be retained, described and made available. Essentially research data model in any university varies, it can vary in bits and pieces, but it will be that the research data management during any project will lie with the principal investigators and the faculty will be responsible for the researcher. The university will be responsible for the provision of training, support and advice on the research management data as well as the provision of backed up storage and service. Now this statement comes from the University of Exeter. I understand that not all Australian universities have repositories, but as you move towards research data management these repositories and other storage advice and assistance will be available. Data management implementation involves all of the main operational areas of the university. Research grants offers because they have the data on the grants, the library for metadata, the IT for the IT support services you need to manage and move the data around, HR for the information about who is the researcher so that you can build metadata and descriptions, the heads of departments and of course the researcher. Implementation requires the development of data management planning tools, data management templates, training packages. You have to be able to have access to advice and assistance because the researcher needs to know how to do these things. Storage and curation, but most of all an effective communication strategy to ensure that all parties are aware of the policy and understand their role. I put this slide in from the Digital Curation Centre because it shows you just how many different activities or activities are involved in research data management to provide it, tools, support, metadata, storage and policies. Now policies are very important because you can't do much without it. Now this is an interesting presentation and I'll put the link in in case you'd like to look at it later. Implementation, well a decision has to be taken within the university to implement data management. The institution's approach is developed by consultation between the key internal stakeholders listed in the earlier slide. They do the planning and at different levels of the university. Of course they would do the planning and the coordination of the overarching processes and resource allocation and this commitment is expressed in a policy document. Why have a research data policy? Well if you are in a university, if you don't have a policy that states the university has committed to this activity it won't happen. So if you want to organize and preserve research that can be used for the future, you have to have a policy. The role of policy. Policy provides an operational framework with in which the institution functions and what it does is it helps align the efforts and achieve the institutional vision. Now a university is a legal entity and as such it is subject to federal and state legislation. It is required to conform to national codes and practices and it has to take into account community expectations and then it has its own internal mission and value statement. All of these provide the context for the institution's governance. So you can see this. This is the external world. Then within the university it has to acknowledge and act on those. The policy is the statement about what the university is going to do and the procedures and guidelines for the how. And you unfortunately have to hear me say that a couple of times in this presentation. We are now going to go onto the University of Newcastle Australia site. Now all universities draw a breath. All universities have governance and policy or policy units. They may be called different things but they're all there to manage that. It's a very important part of the organization structure. They will all have a policy library or a policy index or a policy register that will be accessible by the website. This definition is really quite good. As policy plays an important role it promotes the principles which dictate how the members of the university will act. In this case the researchers will manage their data. Those principles are derived from and shaped by the law, regulations that cover the university, national standards as I said previously. And the role of the policy is to translate values in legislation into operations and to manage risk. One of the key things to think about policy is it's not just like a memo. It's actually a legal document. And so the process for drafting it and approving it is quite detailed and structured and every university will have it. This page is on the policy framework. If I'm lucky enough for it to go. Now it takes us through policies. It gives you a definition. There are two levels of policy, governance and strategic and operational. And above them are the acts of parliament. And below them are and attached to them are procedures which stipulate how the policy is implemented and guidelines which provide additional information on policy or any aspect of the policy. And then it goes on to give you all the definitions including that profile. So as you can see, your university will have one of these pages and it will have these things. And you don't have to be in command of all of them but you would probably need to go through them to familiarize yourself if you're going to coordinate policy. Policy development. As a way is the development and approval procedure. You can't do anything without a procedure and a development guidelines. And basically it's identify the issues and this is quite logical. Conduct analysis, develop the draft, consult, gain approval, communicate and implement and then review. Policies are generally reviewed every one to two years. Well three years, that's about three years as long as it goes. The policies approved and alerted. There are policy reforms. There's a glossary. And this will give you the definition of what these things are so that you can understand what it is, where your policy fits and what sort of statement it has to be. University will provide you with templates and there's a template for all these aspects of policy. So if you want to get the policy and procedure approval sheet, there's a policy procedure blank template and also has a template which is an outline of what issues you might need to consider in developing your own policy and I'll give you the link to that later. The other point that's really important is that the role of the policy unit is to provide advice on the development of policy and procedure and to assist with the drafting and editing and formatting of documents. So nobody would in a university be making up a policy or drafting it on their own. They would have to do it in reference to the university's governance and policy operation. Wow, that worked. Just to reinforce the concept, there's a link here. This is the workflow for the University of Melbourne's policy process and you can see by this that it's unlikely that a policy will be drafted, reviewed, approved and implemented in under say a year. You might be lucky into it shorter but it does take time and it would be unrealistic to be too anxious about getting it finished. Current status of data management policies. Now what I'm going to do now is I'm going to look at the policy statement, its format and content. And I'm going to take you through this and try and explain how the different aspects work and what you have to consider when you're developing the policy. To do this, I used institutional websites and their web pages to demonstrate the aspects and approaches taken by the institutions to create policy statements and to provide the operational framework. Again, there's a link there to the ANS pages on data management. All of the, obviously all of the examples I use are available from the public web pages website. What's the procedure? Now we're going to keep using these terms so I thought perhaps I should go back and just make this clear again. Policy is the what, the procedures are the how. Operating practices can be formal or informal, specific to the department or applicable. So you know there's lots of levels of this. So in navigating that process at your institution you have to just do that bit of work to work out what fits where and whether an informal policy is better or more useful on the day than a formal policy. Whether it's an intro message. Now, all universities have policies on research and research practice. Typical ones are IP ethics and they're already existence and in place. This is the Research Data Management Policy for Wollongong University. As you can see it has the data proved, approved by Custodian and the Responsible Department of Research and Innovation. If you'll remember, I pointed out to you that the university works within the environment created by references and legislation. Here the External Australian Code for Responsible Conduct, December 2007. National Statements on Ethics, Policy, These are all have to be taken into account when the university creates an operational policy to be operated internally. If you go up you can see that the university has a code of practice for research policy, copyright, IP, records management, privacy. So you have the external and you have the internal. The thing that you have to consider when you are developing a policy that the approval of policy is quite sometimes perfect time is that you have to have the policy that you're drafting has to be checked against these to make sure that it doesn't contradict any of those things that it is in fact in line with the existing policies and legislation. In fact, I think we went back, if you looked at the way your university assesses or considers and looks at policy before its implementation phase, they will always go for legal review. The legal department always takes part in the development of new policies because they have to conform. Common policy topics, these are like the index to the policy. When you start your policy you've got to have a look at it. Now there are all, some policies will have 10 of these. This is quite short and concise. As you see by the dereference, it relates to research and development. It's got policy principles, application. I'm sure you've seen these things before. Definitions, roles and responsibilities and it goes on that way to just say who's doing what with what. And so common topics, if you're starting your policy you start with, well okay, policy principles, applications and you could think through those topics. What do I know about this? How do I think it should operate within my university? What would fit with the other structures and policies that are currently in place? Okay, current data management policies. When I went through all of the Australian institutions' websites and found the existing data management policies I read through them and analysed them and it becomes clear that what a data management policy that exists at the moment do is attempt to make clear who does what and to build on the requirements of Australian Code of Practice for research. Now, I'm sure everybody's familiar with this. The Code gives direction on all aspects of research including references to data and data management. And again, Anne has an awareness guide on this. Code of Practice has this thing where it says sufficient data and tool must be retained to justify the outcomes if they're challenged. Research should be challenged and this means that the data should be retained and accessible for five years. Well, I think if you've got a system that allows you to document and retain the data for five years and make it accessible for five years, you have the basis of an effective research data management process. You can see in this cartoon the implications. He can only probably reproduce his research outcomes if it was the right season and the right time. Code of Practice contends with the potential value of data and material for further research should be considered, particularly where the research would be difficult or impossible to repeat. Again, you know, if you look at that cartoon, Newton would have to be under the tree when the apples were ready to fall and the rest of the year we couldn't repeat the experiment. Anne's and the Code. Anne's has no formal responsibility for the Code that says, as previously described, the Code consigns with the aims of Anne's and promotes and is interested in increasing awareness of the Code. Institutional data management policies. Now, existing policies make about five consistent statements and that's the nub of the policy. They firstly acknowledge the requirements of the Code. They aim at good or best practice in data management. They assign the responsibility to the whole of the institution, rather than the researcher or the department or faculty, and they state that the data should be available for reuse if appropriate. Now, you can see that these all make sense in relation to managing the data and implementing it. They also, a couple of them, require the completion of a data management plan and are supported by data management guidelines. Now, I hope we'll be able to link into this site. This site says effective research data management, as most of the toolkits do, is about enhancing the reuse. It has a research data management plan, which is a tool, and that goes through provides the templates, index to different aspects of data management, and gives you links to policy. That's a very useful thing to have. The other aspect that they have to have is develop some policy a bit further, is this one was a bit so, guidelines. So, if you're the researcher and you're attempting to do research and you want to know what do I have to do with it, the guidelines would lay it all out, like the storage, the format, and hopefully if it wouldn't make it easy for you, it would at least make it clear. Examples of policy. The first thing that any policy does is it states its purpose and aim. This example from the University of Wollongong states that this policy forms part of the institution's governance for research data management by recognising that data is a valuable product of research, and its aim is to establish a best-practice culture of data management within the institution. If we look at the actual reference, you come down here. The purpose of the policy is much longer. It talks about assisting with increasing, making existing available data available for purposes. So it had quite a number of things on it, as well as it's a very comprehensive statement of the purpose of the policy. Responsibility for research data. The data management policy makes the institution responsible. And again, here is an example from the Queensland University of Technology, which it says, and I quite like this myself, the director of e-research is responsible for overseeing the management of the research. The heads of school have to make sure there's guidelines and arrangements within the school to the department that meets with it and provides faculty for safe and secure storage. The director of the library and information services and learning group have to be responsible for the development of the maintenance of centrally supported repository, access to secure access and maintaining procedures. I'm going to read all of it, but this is a very clear statement who is responsible for what in relation to research data management. And your policies should always aim at trying to be as clear and unambiguous as possible, but of course it has to express the situation that the way your university wants to manage it. You may not want this detailed assignment of responsibilities. Access to research data. And of course you can't really have reusability unless the principal is contained in your policy that data should be made available for use by the researchers before the research and less precluded by conditions under which they are obtained. Now, all of the policies that are currently in place have this statement. You have to acknowledge that if the research is owned by someone else or if it has private and confidentiality, you just can't hand it out. But if it's the statements or are consistent in that, if it is possible and there are no barriers to sharing, access should be provided. Research data plans. Now, in a number of, a couple of institutions, they've decided that good practice in research management is the result of good planning at the onset of the research project. Now they continue through life life. And then they have made it compulsory to develop a research data management plan. Other institutions have taken the view that where they have assigned responsibility to the department and the researcher and have provided a toolkit, the sort of thing I showed you from the University of Western Australia, then they should allow the researcher to make the decision to fill in the research management plan and encourage it and provide training and support. So there's two approaches you can take and this is very much a decision for your university to decide whether or not you want to do that. Where to start? Well, firstly there has to be a commitment to research data management. There has to be a clear understanding of what is currently possible or available within the university institution to support research data management. Because you may have a metadata hub, you may have a library and who can support data management. You may have parts of it, but you may not have all of the things available that were shown in the DCC slide. So you may in fact have to think about what is it that we have that we can deploy. You will have to work, as I've made perfectly clear, you will have to work with the institution's governance unit to draft the policy. And I imagine that that process will be detailed and demanding and you'll have to be patient. The policy statement could be approved or drafted and approved after the support mechanisms such as data storage and tools are in place and have been tested. Or the policy could state the aim and then be amended as the systems mature. So there's a number of choices. It's not just quite run out and right, good it's today we'll write a policy. Examples of institutions that currently have policy procedures and tools. They're listed and there's a link. Now we have what we call a product catalog, in fact a list, it's a list of our projects and it lists projects, institutions, people and data management resources. Now if you have a look, we list the institution and we put a link to that part of the processes that they have in place. Now Griffith University has a code for responsible practice which is very much like research data policy. La Trobe University has a policy that doesn't have a data management tool or guidelines. Monash University of course has all three as does Queensland, Melbourne, Newcastle and Ongone. Each at a regular intervals I go through the websites looking for new things and in July Curtin and I think ECU have put up on their library site data management tools. As a policy person if I were asked, Francis, design a policy, a research data, draft this research data policy, I would, one of the ways I would start would be by finding examples of policy. I would have a look at the what the university said they were going to do and all of these are quite, they are subtly different, they're similarities but they are different and they reflect as it does the language and the process around it, the university's organisational culture and the way it likes to express its decision making. I would then look at the procedures and the guidelines and the tool kits and then I would think about how each part of that, the bot, the how and the support mechanisms work and are integrated and then I would start drafting, I would reflect on the situation within the current, the institute I was working in and then I would start to develop the policy. So I don't think you have to spend forever doing this but I have found it in the past to be a very useful practice. Now ANS resources, we have, I just showed you the product catalogue, we have the ANS web pages on data management which are comprehensive and we have another tool which I've mentioned which is the outline of a research data management policy. Now you have to use the university, I'm laboring this, I know you have to use your own template but you could do read through this and see the sorts of statements and the sorts of things that you might be able to apply when drafting your policy. Along with those the university ANS has me as a results and I'm always happy to talk to people, to exchange emails or support, I recognise that this is a pretty fast going over what is really quite a complex area and I'm more than happy to help and so you can contact me at any time.