File organisation/ directory structure

The type of file organisation is, amongst other aspects, defined by the conditions of a specific project which might determine a certain storage location or structure. In this regard, data protection or ethical issues might be relevant, e.g. copyright and licenses, type and number of different data used in a project, and expected volume of files.

Storage location

A suitable location for data storage has enough capacity, it is part of a backup management, enables all parties and all entitled users access and prohibits unlawful access.

Directory structure

Ideally, a directory structure is hierarchical and reflects content requirements.

The following categories can be considered:

• part projects, work packages or functional units within a project
• persons
• date or period (e.g. months, quarters)
• Contents respectively types of folders (e.g.
• research data (or file types),
• publications and reports,
• data analyses,
• conferences,
• literature,
• project cordination)
• File types (formats)

Examples for folders in a system (cf. Long 2009):

• „mailbox directory“: a folder for important e-mails that need to be saved. Your own e-mail box may soon be overloaded and too complex, and you might find it helpful to set up such a folder.

• „to clean directory“: a folder for the temporary storage of files that still need to be moved to their “proper” location.

• „hold then delete directory“: a folder for files that are no longer needed but should not be deleted because they might yet be needed one day...

• „posted directory“: a folder for files that have been sent to external parties. Versions of files are sometimes posted that are still being edited. It might be important for a later point in time to save exactly this particular version of a file.

Versioning

In order to track and distinguish different stages of file editing, versioning should be consequently defined. Versioning needs to be transparent to the file producer and all the users.

Versions can be identified via the file name, the file itself or a specific versioning document.

Example - Versioning in file or specific document:

Example - Versioning in file name

• Versioning by means of numbers (v1, v2, v3) and distinction according to minor and major modifications:
• example for major changes = samplefile _v1 -> ssamplefile_v2
• example for minor changes = samplefile_v1.1 -> samplefile_v1.2
• Confusing additional naming should be avoided and the file name should be kept consistent:
• instead of “Original1“, “edit“, “edit“, “final“,“final2“, rather “samplefile_v1“, “samplefile_v1.1“, “ samplefile_v1.2“, “samplefile_v2“, „samplefile_v2.1“

Please also note

• Each change should be documented.
• Outdated versions should be stored separately, otherwise rejected/deleted.
• The original copy should always be saved.
• An auto-backup software can replace storage or archiving of different versions.
• A control software can check versioning (e.g Subversion, TortoiseSVN).
• Or you can use software with an automated editing history.

File naming

File naming constitutes an important part of archiving.

Clear file naming

• are necessary to clearly distinguish files in a directory structure,
• avoid problems when several users are working on the same files,
• facilitate findability, particularly when files have accidentally been moved and
• serve transparency, since files are shown in structured order.

Regulations for the chosen method of naming files:

• The method should adhere to limitations of the chosen software (e.g. significance of capital and small letters, special characters or blank spaces).
• File names should be related to their content so that files can also be found regardless of storage location. This is particularly important in the case of accidental file relocation.
• File naming should be uniform and consistent.
• File names should not be too long.

Possible problems too long file names and paths:

• Blank spaces and special characters (& * % $ £ ] { ! @) – except for underscore and hyphen and umlaut should be avoided.
• Use clear names, no generic names (e.g. „dates“, “txt“), to avoid mistakes.
• Avoid long names and too many levels to keep below the maximum length allowed by Windows (255 characters).

Example for an error message that appears if the maximum length of characters Is exceeded.	Examples for errors that can appear if a file name contains too many umlaut characters.

Possible characters in file names:

• words relating to content or short titles (short, relevant and sufficient to allocate contents; ca. 25 characters) [2010-08-11_bioassay_toxicity_V1.sps]
• name or clear abbreviation of editor /creator/responsible person
• name of research team/department
• date of creation/ change/ publication (format year-month-day: YYYY-MM-DD or YYYY-MM or YYYY-YYYY or YYMMDD (2006-03-24_file description)
• state of editing (e.g. original, draft, corrected, excerpt, filtered)
• version number
• project number
• usage of appendages which mark the software used to create the file (xls/.xlsx for Excel; .txt for text files; for other see File formats (only available in german))

Naming research data and research materials

Below, important labels for respective files are highlighted in red.

File name elements in general

• ID or study number: assignment to study
• abbreviation for type of data or material: interview, video, questionnaire etc.
• connective number of datatypes: 001 ff.
• serial number: different  files according to  type,e.g. a video consists of several files ): a, b, c. ff.
• version number if changes are made and documented, for instance by  anonymisimg 

[ID or study number]_[abbreviation for type of data or  material]_[connective number]_[ version]_[. Serial numner]

Qualitative interviews "int"

Possible elements of the name:
[ID or study number]_[abbreviation for  interview]_[connective nummber]_[version]

Example: file name for the first in a series of interviews

• RTF transcript: s01_int01_v01 (rtf-file, saved as s01_int_001_v01.rtf)
• MP3 audio file: s01_int01_v01 (mp3 file, ss01_int_001_v01.mp3)
• WAV audio file: s01_int01_v01 (wav-file, saved as s01_int_001_v01.wav)

Example: several audio files per interview  

[ID or study number]_[abbreviation for interview]_[connective number]_[Version]_[ identifier for series]

• audio file 1, interview first part: S01_int01_v01_a
• audio file 1, interview second Part : S01_int01_v01_b
• audio file 1, interview third part: S01_int01_v01_c

Focus groups "fg"

Example: sample labelling for the first transcript of a focus group

[ID or study number]_[abbr. For focus group]_[connective number]_[version]

s01_fg01_v001 (rtf-file, saved as s01_fg_001_v01.rtf)

Image and video files "pic", "vid"

Example: first image and video file

[ID or study number]_[abbr. for image or video]_[number]_[version]

• first image file from study s01: s01_pic001_v01
• first video file from study s01: s01_vid001_v01

Example: an image is connected to a speccific incident or event (e.g. picture of a focus group or a person)

[ID or study number]_[image or video]_[ connective number]_[version]_[identifier for series]

• Fifth picture out of a series of 10 images, belongs to interview no. 8, from study s01: s01_pic008_v01e
  In this case, ‘e’ represents the fifth letter in the alphabet for the fifth picture.

Example: the picture belongs to an Interview, and naming is in accordance with the naming of the respective interview with image file extension (.tiff)

• File name for the first conducted interview in study s01, linked transcript and image file:
• RTF transcript: S01_int001_v01.rtf
• audio file : S01_int001_v01.mp3
• image file: S01_int001_v01.tiff

Observation "obs", field notes "notes"

Example: notes taken during ainterpretation of a poem in a lesson

[ID or study number]_[type of lesson material]_[connective number]

• poem: s01_poem01
• notes /transcript of observations: s01_obs01
• field notes: s01_notes01

Quantitative data "ds", "syn", "op"

[ID or study number]_[identifier for dataset]_[version]
e.g. s5600_ds01_v01

[ID or study number]_[identifier for syntax]_[version]
e.g. s5600_syn01_v01

[ID or study number]_[identifier for output]_[version]
e.g. s5600_op01

Materials

[ID or study number]_[data type]_[version]

Examples:

• method report: s01_mb_v01
• questionnaire: s01_fb_v01
• code plan: s01_cp_v01

Further information and sources

Kaps, R. und Volker S. 2017.  Vertiefungsworkshop zum Forschungsdatenmanagement. Dateienorganisation und Projektablage   (only available in German, Workshop im Rahmen des Weiterbildungsprogramms der Leibniz Universität Hannover, 9. März 2017).

Long, J. S. (2009). The Workflow of Data Analysis Using Stata. Texas: Stata Press.

Research Data Management Training   by the University of Edinburgh - "Organising Data".

last update: 20.07.2018