Research Data Management Notes

Research Data Management

Acknowledgement to the Curtin University Library guides Dr Ben Dwyer Senior Research Fellow Curtin Medical School Lecture adapted from Dr Brioni Moore

Learning Outcomes

• What is research data, and what forms does it exist in?
• Data integrity: What is it? How do we maintain it? How do we minimize risks?
• Understand the roles of data management plans.
• The role of publishing research.
• Understand why different types of research have varying retention requirements.

Research Data

• Any information that has been collected, observed, generated, or created to validate original research findings.
• Formats:
  • Digital
  • Non-digital

Examples of Research Data

• Documents (text, Word)
• Excel spreadsheets
• Laboratory notebooks, diaries, field notebooks
• Questionnaires, transcripts
• Audiotapes, videotapes
• Photographs, films
• Protein or genetic sequences
• Spectra
• Blood films
• Clinical records

Data Integrity

• Overall accuracy, completeness, and consistency of data.
• Also refers to the safety of data in regards to regulatory compliance.
• Integrity is maintained through the use of processes, rules, and standards implemented during the research design phase.
• If data integrity is secure, stored data will remain complete, accurate, and reliable.

Aspects of Data Integrity

• Physical integrity
  • Protection of data wholeness and accuracy as it is stored and received.
• Logical integrity
  • Keeps data unchanged as it is used in different ways in a database.
• Data integrity is NOT data security or data quality.

Data Integrity Risks

• Human error
• Transfer error
• Bugs and viruses
• Compromised hardware

Minimising Risks to Data Integrity

• Limiting access to data and changing permissions to restrict changes to information by unauthorised parties.
• Validating data to make sure it is correct at the time of gathering and use.
• Backing up data.
• Using logs to keep track of when data is added, modified, or deleted.
• Conducting regular audits.
• Using error detection software.

Data Management Plans

• A document outlining how you intend to handle your data.
• Key questions addressed in a data management plan:
  • What type of data will you collect?
  • How and where will you store the data?
  • Who will have access to it?
  • How will you address legislative requirements?
  • What agreements are in place regarding ownership?
  • How, where, and when will you share the data at the end of the project?
• A data management plan is required for:
  • HDR students
  • Obtaining ethics approval for human or animal ethics
  • To obtain access to the research drive

Curtin University Data Management Framework

• Pre Research
  • Risk Management Plan
  • Data Management Plan
  • Ethical Clearance
  • Training and Induction
• Research
  • Data Collection and Analysis
  • Metadata Generation
  • Storage & Access
• Post Research
  • Data Curation
  • Policy Compliance Monitoring
  • Risk Monitoring & Communication
  • Publish Data
  • Register Data (RDA)
  • Ongoing Curation
  • Usage Monitoring
• Responsibilities:
  • Primarily Institutional Responsibility
  • Primarily Researcher Responsibility
  • Joint Responsibility

Ownership

• Data ownership refers to intellectual property rights over the data created through research.
• May define roles around data management.
• A complex issue that usually involves the researcher, institution, funding agency, and any participating human subjects.
• At Curtin University:
  • Students retain ownership of their data.
  • Researchers need to make clear agreements with all collaborators as to the ownership of research data prior to the study starting.

File Management

• Having well-considered plans for the organisation of any data will improve its management, access, and use.
• Key methods to consider:
  • File naming
  • Folders and directory structures
  • File versioning
  • File formats

File Names

• Good file names:
  • Concise and meaningful
  • Descriptive
  • Date format
  • Full name
  • Sentence case
• Bad file names:
  • Not descriptive of topic or contents
  • Abbreviations
  • Bad length
  • Novelty characters
• Examples:
  • Good: Smith2011 Interview.doc, Chapter 1 Draft20042021.doc, Experiment 1.1v1180521.xml
  • Bad: final assignment.doc, Document22.docx, Experiment.xml, Lit review.txt

FAIR Principles

• The data you create as a researcher has incredible value.
• Findable
  • Properly describe what the data is; place in a searchable place; make it easy for computers and humans to locate it.
• Accessible
  • Use non-proprietary, standardised, and automated methods to supply data to those who need it.
• Interoperable
  • Store and provide data in widely used and accessible file formats.
• Reusable
  • Make clear how the data was collected.

Publication

• All research data has value beyond the original project.
• Publishing data improves the impact of your research and makes the data more findable, accessible, and reusable.
• Outcomes of publication:
  • Improve body of knowledge in the discipline area.
  • Reliability
  • Citations
  • Professional connections

Retention Requirements

• Every researcher must retain their research data for a period of time, with the duration of time varying based on the nature of the research.
• Retention periods, based on description:
  • “Major” research data: Retain permanently
  • “Minor” research data involving humans or animals and utilising high-risk materials: Minimum of 50 years
  • Data classed as minor but involving clinical trials: Minimum of 25 years
  • Data classed as minor but involving children: Minimum of 7 years
  • Short-term research undertaken by students for assessment purposes: 12 months

References

• Research Data Management. Curtin University Library LibGuides. [Available online: http://www.libguides.library.curtin.edu.au]
• Australian National Data Service. Research data management in practice. Available online: http://www.ands.org.au/guides/rdm-in-practice]