Organization & description

Organizing your files and folders - electronic or paper - is a time-consuming activity but, in the end, it proves to be a considerable help: it makes it easier and faster to identify, locate, retrieve or even recover your data.

A good organization consists in particular of :

Structure your folders hierarchically (classification tree)
Use clear, consistent and most meaningful naming conventions and rules

Ideally, these rules should be defined as soon as possible in the life of the project, before creating too many files and folders. To avoid being overwhelmed, remember to sort, arrange or reorganize them regularly !

When you are conducting a research project as part of a team, it is essential that the whole group agree on the structure of folders and the name of the files to be adopted. This can be recorded and documented to allow everyone to access and find stored and shared data via the same collaborative workspace.

Folder and file tree structure

Organizing your files and folders is a task that should be thought out and carried out at the very beginning of the research project. It is important to choose a clear, consistent and common organization for the different data sets.

Precise naming rules are necessary to more easily and quickly locate and identify the files you are looking for. They allow their classification, avoid problems during file transfer or sharing, and facilitate their medium and long-term conservation.

This is why it is highly recommended to adopt a single logic and to choose unique and meaningful names. Ideally, one should not need to open a document to know what it is about !

Such rules are all the more important when conducting a research project within a team; they promote practical harmonization among all!

Literally "data on data", metadata is data used to define or describe another data regardless of its medium. Metadata can be used to identify, inform or reuse data.

Each set of data collected or created should be accompanied by comprehensive metadata, respecting the standards, rules and conventions of a discipline, and also machine readable.

While data custodians and, increasingly, researchers are aware that good metadata is essential for accessing and reusing research data, it is complex to determine precisely what metadata to capture. It should be kept in mind that only quality metadata supports the FAIR data principles.

Minimum set of metadata

An international consortium of libraries and services specializing in information sciences, DataCite has developed a metadata schema for publishing and quoting research data. In its Metadata Schema Documentation for the Publication and Citation of Research Data (Version 4.2., 2018), the minimum set of mandatory metadata to be used when describing (data sets) is as follows :

• Identify (from the dataset)
• Creator
• Title
• Publisher
• PublicationYear
• ResourceType

Some other metadata are optional :

• Subject
• Contributor
• Date
• RelatedIdentifier
• Description
• GeoLocation

An example of a data set description (xml format) is available : Metadata Example for a simple dataset.

Disciplinary metadata

Many disciplines have developed specific metadata standards. A complete list is available on the Digital Curation Center website and on the FAIRsharing.org website

Data Documentation Initiative (DDI)

This initiative is an international standard for describing data produced in the social, behavioural, economic and health sciences. DDI standards allow data to be documented, discovered and interoperable. The specifications and tools are available on the DDI website.

Metadata Standards Directory Working Group

The Metadata Standards Inventory Working Group is supported by individuals and organizations involved in the development, implementation and use of metadata for scientific data. The overarching objective is to develop an open and collaborative repository of metadata standards for scientific data to address data infrastructure and organization challenges.