Research Data Management

Having a well-considered plan for the structure and organisation of your research data will improve it's management, access, and re-use. The organisation of data is especially important in team projects where more than one person will be accessing and analysing the data.

The key methods and approaches to consider are:

1. File naming
2. Folders and directory structures
3. File versioning
4. File formats

Digital file names are important for identifying and finding a digital file. Researchers should develop and communicate a clear system of naming files so everyone involved in the research project understands and can appropriately apply the file naming rules to create and locate files.

The most important things to remember about file naming are:

Descriptiveness - Good naming conventions should provide useful cues to the content and status of a file, including its version. 

Consistent application - By selecting an appropriate naming convention for files as early as possible and follow it throughout the research, the benefit from file naming systems will be maximised. 

The following examples highlight basic principles of file naming.

Good file names:

20191004 Registry of participants - Survey.doc

ROBERTSON Thomas Logan - 2011 Interview.mp4

• File names are concise and meaningful
• Sentence case including a capital letter for names and proper nouns
• Full name with family name in UPPER CASE
• Date format: yyyymmdd. When referring to year only, always use four digits
• Terms are separated with a dash avoiding punctuation.

• Bad length. 25 to 50 characters is generally adequate
• Does not describe the subject or topic of the document’s contents
• Avoids abbreviations - temp could mean template, temporary etc. Use the full word instead
• Do not use the following characters within the document name: . , ; : = \ / * ? " < >

Like file naming, systems to organise folder and file directories require coherency and consistency.

Coherency - Anyone using the folders should be aware that there is a system and what it means.

Consistency - Anyone using the folders should be consistent in creating folder names in line with the system, but also in keeping the relevant files in the appropriate folders.

This will ensure it is easy to locate, organise, navigate and understand the context of all files and versions. 

Other concepts to consider include:

• File hierarchy refers to the number of levels or sub-folders in the directory. It is usually useful to have a maximum depth of 4 folders

• Folder direction determines how folders are nested and which way is most useful (e.g. Results/2012 or 2012/Results)

• Ambiguity or overlapping categories, especially at the top-level, can cause confusion

Over the duration of a research project, a dataset will undergo many changes. They may be as simple as adding more sets of findings, or as major as the addition of a new dimension or type of measurement. This is important for 2 reasons:

1. These changes are certain to change the conclusions derived from the dataset. In order to maintain integrity about where your conclusions came from, it's important to know which version of your data you're addressing.
2. The changes you make might ultimately not be useful. If this happens, you may need to go back to an earlier version of the dataset.

One of the tools used to address these needs is versioning. This refers to a system of keeping the old versions of a file and tracking the changes made in each version.

The most basic forms of versioning are manual systems. These usually contain two important elements:

1. The user adds a sequential number to the file name to indicate which version of the file it is
2. A change table in each document where versions, dates, authors and details of changes to the file are recorded

These are outlined in the ARDC Versioning guide and the UK Data Service Version Control and Authenticity page linked below.

While a manual system can work for many research projects, they can become difficult to use once your needs become complex or multiple people begin working on the same dataset, as explained in the video below. In these cases you should consider using version control software - Git is the best known and most widely used of this type of software.