For Researchers Dr Joanna Goodger Information Hertfordshire With Bill Worthington, Sara Hajnassiri, and Mohamed Hansraj
AN INTRODUCTION
Research Councils The majority of data is funded by the Government through Research Councils and Universities. There are seven Research Councils, each covering a broad disciplinary area. Research Councils UK (RCUK) is a strategic partnership between the Councils, enabling them to work together and enhance the overall performance and impact of UK research. Each council operates through grant programmes and some run their own research institutes and data centres. The requirements laid down by the research councils cover various aspects of the research life cycle.
Research Councils ESRC and BBSRC are leading development of Research Data Management Plans within their funding process as well as in their policies: All RCs fund at least one dedicated data centre Researchers are required to offer to deposit their data in a data centre, to deposit it locally and make it available to others. During funding, researchers need to include a data review and a data sharing plan. Funding can be sought for sharing resources, for buying existing data, and commissioning new data. Time limits are set for deposition and retention periods. Secondary users must acknowledge their data sources.
UH Policy In response to these government policy changes, UH has reviewed its data policy UPR12. The University of Hertfordshire s research data policy states; The University owns data generated by its research; it expects data to be managed and shared in a robust and professional manner; and it places the responsibility for proper research data management with the Principal Investigator. The University s data policy also has the following clauses from the main policy; 4.2 The University recognises the value of data as an institutional resource and considers that value to be increased through the widespread and appropriate use of data and by virtue of data quality. 4.3 The University considers the value of data to be diminished through misuse, misinterpretation, or unnecessary access restrictions.
What this means to Researchers The changes to RCUK and to UH data policies means that we are now required to think about preserving and reusing our data from the funding stages of the project. A data management plan is required as part of the funding application Once published, data should be anonymised and preserved in open access repositories, preferably RCUK data centres.
Activity Why you should but maybe can t share your data? Spend a couple of minutes thinking about the benefits of sharing your data, having other data available for you, and what you might not be able to share. Then we ll combine your answers and see if we can resolve the issues.
Benefits to Researchers More data is available for contrast and comparison More opportunities for collaboration as data is discovered Credit awarded on data as well as related publications Data preserved in managed archives Data is not lost when researchers leave research Student project data are managed assisting audit and reuse
PLANNING YOUR PROJECT
Research Data Plan The University of Hertfordshire and many funding bodies want assurance that you have systems and protocols in place to protect your data during the working, sharing, and preserving stages of your project. They require a Research Data Management Plan. Sometimes this is incorporated into the electronic grant application process e.g. Je-system. Others, like UH require you to complete a DMPonline Template. https://dmponline.dcc.ac.uk/
Research Data Plan To assist in preparation of basic Data Management Plans (DMP) at the funding application stage, the DCC presents DMPonline. These templates are designed to make researchers think about our projects and how to manage the data throughout the whole project life cycle.
Research Data Plan The DMP covers the whole project lifecycle. The plan is divided into 6 sections, which cover the lifecycle but are divided into aspects of the project, not chronological steps. Preservation and reuse Publishing Collection of new and existing data Sharing with collaborators Local storage and analysis
Activity What happens to your data during a project? The where, when, how, what, and why of your data. Spend a couple of minutes thinking about the choices you need to make about your data, throughout your project. Then we ll combine your ideas and compare them to the DMP.
Research Data Plan Introduction and context Data types, formats, standards, and capture methods Legal and ethical issues Short-term storage and data management Access, data sharing and re-use Deposit and long-term preservation
Research Data Plan Once you have received funding, you will be allocated an R1 number. At this point you should complete the UH Template, 29% of which is contextual information that is not required by you funder as it is part of a larger application export your UH DMP as pdf email this document to dmp@herts.ac.uk from your herts email account setting the subject as R1 number PI surname. Your DMP is stored in the Document Management System (DMS) in a secure area. This DMP can be updated at any point using the same subject. The DMS maintains previous editions automatically.
PROJECT LIFECYCLE
Introduction and context Basic Project Info. Thesis Title UH or Research Council? Duration Related Policies UH and STFC policies: open after publication as your research is public funded through UH & RCUK Roles and Responsibilities Yours and your supervisory team Additional help from IT staff, HelpDesk and other team members
Introduction and context UH Template Questions - Short description of the project s fundamental aims and purpose - Funding body requirements relating to the creation of data - Institutional or research group guidelines - Other policy-related dependencies - Date of creation - Aims and purpose of this plan - Target audience for this plan - Does this version supersede an earlier plan? - Contact details and expertise of nominated data managers /named individuals - Glossary of terms
Data types, formats, standards and capture methods What data will be created? Note the type and volume of data that will be created e.g. transcripts, measurements, imaging etc. Explain how you will capture the data. e.g. in a numbered, dated notebook. What formats do you propose to use and why? e.g. Microsoft Access, Excel or SPSS, as they re in widespread use. DICE DMP Breakdown
Activity What is data? What does data mean to you? Spend a couple of minutes thinking about what data you will be working with, throughout your project. Then we ll combine your ideas and compare them to the DMP.
Data types, formats, standards and capture methods Project Desc. and process Basic description for context Define data Software, Documents, Formats Existing and New Data Archival data, Catalogues, New obs. Proprietary? Project only for 6mth Metadata Headers: table column headings, image information, paper reference, software version.
Data types, formats, standards and capture methods UH Template Questions - Give a short overview description of the data being generated or reused in this research - How will you manage integration between the data being gathered during the project and pre-exiting data source? - How will you capture or create new data? - Which file formats will you use and why? - What directory and file naming convention will you use? - Are there any tools or software needed to create/process/visualise these data? - Are there appropriate computing hardware, facilities, and resources to manage, store, and analyse these data? - Are the datasets which you will be capturing/creating self-explanatory, or understandable in isolation?
Data types and formats Images / Photos Plots Code Tables Transcripts Audio-Visual
Data types and formats Images Raw, Processed, Plotted, Photos, Scans, CAD Formats Uses Considerations FITS, JPG, PNG, BMP, PS Reuse, paper, talk, poster, archive, web Use, size, longevity Tables Catalogues, Query results, Calculations, Measurements Text files, FITS, spread sheets Code input, spectra, plot, paper, CDS Use, metadata, accessibility Source code Models, simulations, scripts, inputs, outputs, instructions.c,.pl,.py,.idl, README, Make file, input, output Third party edit, run. paper, web User friendly; functions, size Interviews Audio, Video, Written Transcript.txt,.odt,.doc., mp3,.mp4,.avi Producing transcripts, further analysis Format, longevity, security, metadata
Data metadata How will the data be documented and described? What contextual details are needed? e.g. a description of the capture methods and data analysis. How will you capture this? e.g. in papers, in a database, in a readme text file, in file properties/headers. Which standards will you use and why? e.g. refer to data centre recommendations for metadata, controlled vocabularies, documentation. Are there any encoding guidelines you should follow? DICE DMP Breakdown
Data metadata UH Template Questions - What contextual details are needed to make the data you capture or collect meaningful? - How will you create or capture these metadata? - What form will the metadata take? - Why have you chosen particular standards and approaches for metadata and contextual documentation?
Legal and ethical issues How you will manage ethics and intellectual property? How will you safeguard the privacy of research participants? e.g. by negotiating informed consent. Will there be any restrictions and why? e.g. delays while you seek a patent, embargoes as right of first use. DICE DMP Breakdown
Legal and ethical issues Legal Issues Do you have copyright issues? Is there a patent pending on your work? Is the data personal? Ethics Who owns your data? UH? STFC? Third party company? How will the data be licenced? How will you deal with disputes?
Legal and ethical issues UH Template Questions - Are there ethical and privacy issues that may prohibit sharing some or all of the data? - How may they be resolved? - Is the data that you capture / create personal data in terms of the Data Protection Act (1998) or equivalent legislation if outside the UK? - What action have you taken to comply with your obligation under the Data Protection Act (1998) or equivalent legislation if outside the UK? - Will the data be covered by copyright or the Database Right? Give details. - Who owns the copyright and other intellectual property? - How will the database be licensed?
Legal and ethical issues How will you manage your data? How will you store and back-up the data? e.g. University storage with IT backup, mirror data on partner's server. DICE DMP Breakdown
Activity How is your data at risk? What precautions do you have in place to safeguard your data? Spend a couple of minutes thinking about how your data could be lost, damaged or stolen? Then we ll combine your ideas and compare them to the DMP.
Short-term storage and data management Storage Where will your data be stored? How will it be transmitted? Back-up Where will you back-up? Who will do it? And how often? Security Keeping sensitive data private Safe from loss or theft
Short-term storage and data management laptop Cluster External HD DVD, Tape UH PC local drive Networked drives U: and X: UH server 5G
Sharing c Access only to UH members Versioning OS independent Set file structure DMS Access only to UH members Undefined file structure No versioning Research Drives Send large files using the UH server Web based only Open and Free Zend.To Safeguarding data with Research Data Management
Short-term storage and data management Backing up should be an automatic part of your everyday research activities. In 2005, an electrical fault in the electronics and laser research building at the University of Southampton cost 50-100M including temporary building hire and transfer of work to Holland. Image if a fire or similar disaster happened at UH How much would it cost you Storing your data on the UH network means that it is stored at de Havilland and at college lane. Mountbatten Building, So ton Uni.
Short-term storage and data management rsync Updates the changes to files between two directories and servers /usr/bin/rsync -avu /data/jgoodger/ /local/data/ /usr/bin/rsync [options] [src] [dest] crontab Timed schedule to perform tasks your rsync for example SHELL=/bin/tcsh PATH=/sbin:/bin:/usr/sbin:/usr/bin MAILTO=a.user@herts.ac.uk 17 3 * * * /bin//usr/bin/rsync -avu /data/auser/ /local/data/
Short-term storage and data management Windows Backup and restore Mac Time machine Set an automated backup through control panel Back up your entire content to another disk or to the net.
Short-term storage and data management Most data needs some level of security: - Sensitive Personal Information - Proprietary data - New discoveries - Revolutionary code / software All of it needs to be accessible, but secure during storing, sharing, and publishing. If you lost it, who would be able to access your data?
Short-term storage and data management Keep your data secure in an encrypted folder Bitlocker is available on Windows 7 +, Truecrypt works on any operating system. Open Source Encryption that works with Windows, Mac and Linux Pack your files into an encrypted volume Send by email, shared drive, cloud storage, web space Password access Variable encryption algorithms available
Short-term storage and data management UH Template Questions - What is the ballpark size of the data being collected / created? - Where physically will you store the data during the project s lifetime? - What media will you use for primary storage during the project s lifetime? - What software will be used in storing and processing these data? - How will you backup the data during the project s lifetime? - How regularly will backups be made? - Who is responsible for backups? - How will you manage access restrictions and data security during the project s lifetime?
Data sharing and access What are the plans for data sharing and access? Who is expected to use the completed dataset(s) and for what purpose? How will the data be developed with future users in mind? e.g. choose appropriate formats. How will you make the data available? e.g. deposit in a data centre, forward copies on request, create website, publish a book. DICE DMP Breakdown
Activity What happens when you re finished? After you ve published, what happens to your data? Spend a couple of minutes thinking about where your data should be stored, who should have access, and what would happen if something happened to you? Then we ll combine your ideas and compare them to the DMP.
Data sharing and access Data sharing and Reuse Who else wants your data? Why might they not have it? Access to Data How and when will you release your data. Project timetable Timing Limits on pub dates? Special Journal or Conference Publication Embargo or Patent Pending?
Data sharing and access UH Template Questions - Which groups or organisations are likely to be interested in the data that you will create / capture? - How do you anticipate your new data being reused? - Are you under obligation or do you have plans to share all or part of the data that you create / capture? - If not, why will you not share you data? - If you can, how and when will you make the data available? - What is the process for gaining access to the data? Will access be chargeable? - Does the original data collector / creator / PI, retain the right to use the data before opening it up to wider use? Give details. - Are there any embargo periods for political / commercial / patent reasons? Give details. - How will you implement permission, restrictions, and embargoes?
Deposit and long-term preservation What is the strategy for long-term preservation and sustainability? What are the plans for sustainability? e.g. choose open standards, deposit in data centre. Which repository / data centre have you identified as a place to deposit data? Show you've consulted them. How will you prepare data for preservation and sharing? Show time and resource budgeted in. DICE DMP Breakdown
Deposit and long-term preservation Selection Which data will be kept / made public? Which tools are independently valuable? How will sensitive data by managed? Location and Schedule Where will your data be published? In a national or subject specific archive? At UH? How long should the data be kept? Metadata What metadata and documents will also be archived? How will this data be created?
Deposit and long-term preservation Working Data Publication Archiving Journal Currently, selection, methods, algorithms, results, plots, and conclusions are in papers, published in journals and open archived in the ArXiv. All Data Paper Supporting Data Archive (ArXiv) UHRA National Archive In the future you ll need to select supporting data, including material with independent scientific merit for publication online in open access archives; either subject specific or in the UHRA.
Deposit and long-term preservation What is kept depends on the decisions made by the government, RCUK and the journals. Probably supporting material and data that has scientific merit, but could be all of it. Be prepared! Keep clear and useful notes on your work; Annotate your code so others (including your future self) can make sense of it Keep a README of instructions for reduction, analysis or code procedures Clearly name the published results isolate them in or copy them to a directory Version control your codes, results, plots and drafts so you can compare at least Make a note of results/conclusions of dead ends
Deposit and long-term preservation UH Template Questions - What is the long-term strategy for maintaining, curating, and archiving these data? - On what basis will data be selected for the long-term preservation? - Will or should data be kept beyond the life of the project? - How long will or should these data be kept beyond the life of the project? - Which archive / repository / central database / data centre have you identified as a place to deposit data? - How will you manage sensitive data over the longer term? - What metadata / documentation will be submitted alongside the datasets or created on deposit / transformation in order to make the data re-useable? - How will this metadata / documentation be created, and by whom? - Will you include links to published materials and/or outcomes? Give details. - How will you address the issue of persistent citation?