Research Data Management - Research Guide

You may collect and retain data throughout your research project.

It is important that you design your data on the understanding/assumption that it will be shared:

• When creating data, describe it as you go - make sure you document sources, methods and reagents so that others will understand 
• Always document terminology and acronyms at the time of data collection, so that the meaning is clear

NB: In spreadsheets and tables:

• Do not merge cells
• Do not use cell colours, bolding or underlining to denote meaning - add a column instead
• Ensure you are consistent and have sensible layouts, labels and terminology
• Include only one piece of information per cell
• Consistently denote any blank cells "NULL" or "N/A" when intentionally left empty

Ideally, keep your data in a format this is:

• Machine-readable
• Nonproprietary
• Unencrypted
• Uncompressed
• In common usage by the research community and/or widely used within the discipline
• Interoperable among diverse platforms and applications
• Available royalty-free
• Fully and independently implementable by multiple sofware providers on multiple platforms without any intellectural property restrictions for necessary technology
• Developed and maintained by an open standards organisation with a well-defined, inclusive process for evolution of the standard

Organise files and any folders so that they make sense (and do not have a 'Miscellaneous' folder!) and employ a good File Naming Convention (FNC) that contains useful metadata to help you and others better identify, understand and navigate your work.

When naming files, try to include the following:

• Start with a date and in ISO 8601 format i.e. YYYYMMDD, so you can sort chronologically
• A short description that includes the project name and analysis undertaken
• The name or initials of the creator
• Version (v1, v2, v3 etc.) to keep track of incremental work
• Anything else that will be important in understanding the file

Some other recommendations:

• Ensure file names are unique and recognisable
• Incorporate descriptions to indicate why the version is different
• Use underscores instead of spaces
• Avoid special characters as these may be incompatible with certain Operating Systems
• Remember that some operating systems are case sensitive

This seems like a lot, but it can be achieved. An example of a good file name: 20230531_DOHProject_TissueData_RJ_v1.xlsx

If this becomes complex, use a readme.txt file to record the system.

You will need to ensure that your research data is stored securely for the life of the project and for any required retention period. Research funding bodies may mandate where research data must be stored including institutional, national or international repositories.

When deciding on data storage and access, always consider how sensitive the data is. Care should be taken when storing all sensitive or confidential data, especially if it relates to research with human subjects.

Store your data in more than one place and in more than one medium. Storing your data in multiple places and mediums ensures that data files can be restored, if they are corrupted, damaged or go missing.

Store the data in a format that is machine readable, avoiding proprietary formats if you can.

If your data can be shared, consider storing your data in a data repository or archive.